Universidad Complutense de Madrid

UNIVERSIDAD COMPLUTENSE DE MADRID
FACULTAD DE VETERINARIA
DEPARTAMENTO DE NUTRICIÓN, BROMATOLOGÍA Y TECNOLOGÍA DE LOS
ALIMENTOS
TESIS DOCTORAL
Prevención de la sobremaduración y de la formación de aminas biógenas en

quesos mediante tratamientos de altas presiones
MEMORIA PARA OPTAR AL GRADO DE DOCTOR

PRESENTADA POR
Javier Calzada Gómez
Directores
Manuel Núñez Gutiérrez

Ana del Olmo Sánchez
Madrid, 2015
©Javier Calzada Gómez, 2015

Departamento de Nutrición, Bromatología y Tecnología de los Alimentos
PREVENCIÓN DE LA SOBREMADURACIÓN Y DE LA FORMACIÓN

DE AMINAS BIÓGENAS EN QUESOS MEDIANTE TRATAMIENTOS
DE ALTAS PRESIONES

Madrid, 2014
PREVENCIÓN DE LA SOBREMADURACIÓN Y DE LA FORMACIÓN

DE AMINAS BIÓGENAS EN QUESOS MEDIANTE TRATAMIENTOS
DE ALTAS PRESIONES
Memoria presentada por Javier Calzada Gómez para la obtención del grado
de Doctor por la Universidad Complutense de Madrid
Director: Prof. Dr. Manuel Núñez Gutiérrez

Codirector: Dra. Ana del Olmo Sánchez
Departamento de Tecnología de Alimentos
Instituto Nacional de Investigación y Tecnología Agraria y Alimentaria (INIA)
EL DOCTORANDO VºBº DEL DIRECTOR VºBº DEL CODIRECTOR

Manuel Núñez Gutiérrez, Profesor de Investigación del Departamento de
Tecnología de Alimentos del INIA, y la Dra. Ana del Olmo Sánchez
CERTIFICAN:
Que la Tesis doctoral titulada “Prevención de la sobremaduración y de

la formación de aminas biógenas en quesos mediante tratamientos de
altas presiones” de la que es autor Javier Calzada Gómez, ha sido
realizada bajo su dirección en el Departamento de Tecnología de
Alimentos del INIA, y cumple las condiciones exigidas para optar al
grado de Doctor por la Universidad Complutense de Madrid.
Madrid, 28 de Octubre de 2014
Fdo. Manuel Núñez Gutiérrez Fdo. Ana del Olmo Sánchez

Agradecimientos.
En primer lugar, me gustaría agradecer a mi director de Tesis, Manolo, el brindarme la
oportunidad de realizar este trabajo y por toda su confianza durante estos años. Por
supuesto a mi codirectora Anen, porque sin ella, sin su ayuda y su cariño, todo hubiese
sido más difícil y nunca hubiese conseguido llegar hasta aquí, y por todos los buenos
momentos que hemos pasado juntos superando los numerosos obstáculos y problemas
que nos hemos encontrado, juntos seguiremos superándolo todo.
A Pilar Gaya por su trabajo con el HPLC y sobre todo y más importante por su cariño
y alegría.
A Marta por todo lo que hemos pasado, en el departamento y fuera, y por su ayuda,
cariño y su amistad. A Sonia por su simpatía, su ayuda y su disposición a echar
siempre una mano. A Roci y Rakel por su amistad, que siempre me ha ayudado a tirar
hacia adelante con una sonrisa (Roci vuelve!!!). A Izaskun, Natalia y África por su
contagiosa alegría y porque sigáis siempre así. A Iria por su apoyo, su simpatía y esos
momentos de cigarritos y desahogos (mucho ánimo, que ya queda menos). Y a todos mis
compañeros del departamento por los buenos momentos y por hacer más llevadera esta
etapa: Juan, Marga, Máximo, Tomi, Joaquín, Pilar, Sagrario, Olga, Nerea, Chiqui,
Lucía, Eva, Chema, Ángela, Dani y Susana. Sin olvidarme de Ana GI por su ayuda
siempre con todos los papeleos y demás trámites. También agradecer su apoyo a las
“vecinas” de leguminosas: Isabel, Carmen C, Mercedes, Carmen B y especialmente a
Blanca y Merche por todo su cariño. Gracias a todos por vuestro apoyo.
Son demasiadas las personas a las que me gustaría agradecer y dedicar esta Tesis.
Desde que entré en el departamento de Tecnología de Alimentos del INIA, son muchas
las personas con las que he compartido momentos que nunca olvidaré y de las que he
aprendido tantas cosas en el trabajo y en la vida. A “Sita” por enseñarme esos primeros
pasos en micro, y por su amistad y todos los buenos momentos también quiero agradecer
a Pi, Victoria, Eugenia, Teresa, Nuri, Toñi, Ana Gómez… por todo lo compartido.
A mis amigos del INIA con los que tantos buenos momentos he compartido: Nines,
Isabelita, Mila, Isabel, Jesús, José Luis y Enrique, sin olvidarme de Evaristo que
debería de estar compartiendo este momento.
A Manuela, de Veterinaria, por su paciencia y su ayuda con todas las dudas y trámites
necesarios para llegar hasta aquí. A Isabel por su ayuda y su apoyo y a todos los que
me habéis dado ánimos para finalizar esta Tesis.
A Mary, John, Gabriel, Casi y Vlad, por todo vuestro apoyo y por que con esa pizca de
mala leche, siempre sacáis una sonrisa en los momentos duros.
A mis amigos, que no sois pocos y que os tengo abandonados, pero que sé que siempre
estáis para lo que sea. Por todo el apoyo y todos los buenos y malos momentos que
hemos pasado durante todos estos años en los que nos han pasado tantas cosas. A
Pedrito, Ramos, Manu, Jaime, Mati, Yeyo, Osky, Maca, Borjita, Juncal, Tranchete,
Eric, Pablo, Alvarito, Sandrita, Luigy, Laura, Carmela y a Julián, el culpable de que
fuese a parar en el INIA. Sé que faltáis mucha gente con la que he compartido grandes
momentos y aquí os dejo este espacio …………………….. ……………, para que
lo rellenéis y tengáis vuestro hueco.
A mis padres, por su apoyo y su cariño, por siempre estar dispuestos a echar una mano,
por comprender que les haya tenido un poco abandonados y por todo lo que me han
dado. A mi abuela para que siga adelante con mucho ánimo y dando guerra.
A todos, gracias, se os quiere.
“Solo hay dos cosas infinitas, el universo y la estupidez

humana, y de la primera no estoy seguro”
A. Einstein.
ÍNDICE GENERAL.
Capítulo 1. Introducción. 1
1.1. El queso. 3
1.1.1. Definición, historia y producción. 3
1.1.1.1. Definiciones de queso. 3
1.1.1.2. Antecedentes históricos. 4
1.1.1.3. Producción y consumo. 6
1.1.2. Elaboración del queso. 8
1.1.2.1. La leche. 8
1.1.2.2. La coagulación de la leche. 12
1.1.2.3. La maduración del queso. 14
1.1.2.3.1. Glicolisis de la lactosa. Catabolismo de lactato y citrato. 14
1.1.2.3.2. Lipolisis y catabolismo de los ácidos grasos libres. 17
1.1.2.3.3. Proteolisis y catabolismo de los aminoácidos. 21
1.1.2.3.4. Sobremaduración. 26
1.2. Tipos de quesos. 27
1.2.1. Queso Azul. 28
1.2.2. Torta del Casar. 32
1.2.3. Queso Brie. 34
1.2.4. Queso Arzúa-Ulloa. 36
1.3. Aminas biógenas. 37
1.3.1. Aminas biógenas en alimentos. 39
1.3.2. Formación de aminas biógenas. 43
1.3.3. Factores que influyen en la formación de aminas biógenas. 46
1.3.4. Papel fisiológico y toxicología de las aminas biógenas. 50
1.3.5. Control de la formación de aminas biógenas. 53
1.4. Altas presiones hidrostáticas. 55
1.4.1. Modo de acción sobre microorganismos y enzimas. 59
1.4.2. Efecto de las altas presiones en queso. 61
1.4.2.1. Efecto sobre los microorganismos. 62
1.4.2.2. Efecto sobre las enzimas. 64
1.4.2.3. Efecto sobre la textura y la microestructura. 66
1.5. Bibliografía. 68
Objetivos de la tesis. 85
Capítulo 2. Proteolysis and biogenic amine buildup in high-pressure treated ovine 87

milk blue-veined cheese.
Journal of Dairy Science. (2013) 96: 4816-4829.
i
Capítulo 3. High-pressure processing decelerates lipolysis and formation of volatile 103
compounds in ovine milk blue-veined cheese.
Journal of Dairy Science . (2013) 96: 7500-7510.
Capítulo 4. Using high-pressure processing for reduction of proteolysis and 117

prevention of over-ripening of raw milk cheese.
Food and Bioprocess Technology . (2014) 7: 1404-1413.
Capítulo 5. Reducing biogenic-amine-producing bacteria, decarboxylase activity, 129

and biogenic amines in raw milk cheese by high-pressure treatments.
Applied and Environmental Microbiology . (2013) 79: 1277-1283.
Capítulo 6. High-pressure processing for the control of lipolysis, volatile 139

compounds and off-odours in raw milk cheese.
Food and Bioprocess Technology . (2014) 7: 2207-2217.
Capítulo 7. Effect of high-pressure-processing on the microbiology, 153

proteolysis, texture and flavour of Brie cheese during ripening and refrigerated
storage.
International Dairy Journal. (2014) 37: 64-73.
Capítulo 8. Effect of high-pressure-processing on the lipolysis and 165
volatile compounds of Brie cheese during ripening and refrigerated storage.
International Dairy Journal . (2014) 39: 232-239.
Capítulo 9. Effect of high-pressure processing on the microbiology, 185

proteolysis, biogenic amines and flavour of cheese made from unpasteurized milk.
Food & Bioprocess Technology . (2014) doi: 10.1007/s11947-014-1406-7.
Capítulo 10. Effect of high-pressure-processing on the lipolysis, volatile 201

compounds, odour and colour of cheese made from unpasteurized milk.
Food & Bioprocess Technology . (2014). Enviado.
Capítulo 11. Discusión general. 243
Capítulo 12. Conclusiones. 285
Capítulo 13. Resumen ampliado. 289
Capítulo 14. Extended abstract. 309
ii
ÍNDICE DE TABLAS.
Tabla 1 Composición de la leche de vaca, oveja y cabra. 9
Tabla 2 Etapas en la elaboración del queso. 12
Tabla 3 Características de las principales aminas biógenas. 38
Tabla 4 Concentración (mg/kg) de aminas biógenas en diferentes 42

alimentos.
Tabla 5 Tendencias de las características microbiológicas y químicas del 250
queso azul tratado por alta presión con respecto al queso control.
Tabla 6 Tendencias de las características sensoriales del queso azul 251
tratado por alta presión con respecto al queso control.
Tabla 7 Valores de las características microbiológicas, químicas y 253
sensoriales en queso azul control de 180 días, quesos tratados
por alta presión de 180-360 días y queso control de 360 días.
Tabla 8 Tendencias de las características microbiológicas y químicas de la 258
Torta del Casar tratada por alta presión con respecto al queso
control.
Tabla 9 Tendencias de las características de textura y sensoriales de la 259
Torta del Casar tratada por alta presión con respecto al queso
control.
Tabla 10 Valores de las características microbiológicas y químicas en Torta 261
del Casar control de 60 días, quesos tratados por alta presión de
60-240 días y queso control de 120-240 días (sobremadurado).
Tabla 11 Valores de las características de textura y sensoriales en Torta del 262
Casar control de 60 días, quesos tratados por alta presión de 60-
240 días y queso control de 120-240 días (sobremadurado).
queso Brie tratado por alta presión con respecto al queso control.
Tabla 13 Tendencias de las características de textura, color y sensoriales del 269
queso Brie tratado por alta presión con respecto al queso control.
Tabla 14 Valores de las características microbiológicas y químicas en queso 270
Brie control de 30 días, quesos tratados por alta presión de 30-
120 días y queso control de 60-120 días (sobremadurado).
Tabla 15 Valores de las características de textura, color y sensoriales en 271
queso Brie control de 30 días, quesos tratados por alta presión de
30-120 días y queso control de 60-120 días (sobremadurado).
queso Arzúa-Ulloa tratado por alta presión con respecto al queso
control.
iii

Tabla 17 Tendencias de las características de textura, color y sensoriales del 277
queso Arzúa-Ulloa tratado por alta presión con respecto al queso
control.
Tabla 18 Valores de las características microbiológicas y químicas en queso 278
Arzúa-Ulloa control de 21 y 60 días, quesos tratados por alta
presión de 60-240 días y queso control de 120-240 días.
Tabla 19 Valores de las características de textura, color y sensoriales en 279
queso Arzúa-Ulloa control de 21 y 60 días, quesos tratados por
alta presión de 60-240 días y queso control de 120-240 días.
iv

ÍNDICE DE FIGURAS.
Figura 1 Fragmentos de vasijas para el desuerado (b, d) y esquema de la 5

reconstrucción de los fragmentos de la vasija (a, c). (Salque et al.,
2013).
Figura 2 Friso de la lechería. (Friso sumerio, 3000 a.C.). 6
Figura 3 Producción (en toneladas) de queso durante el año 2012. 7

Figura 4 Consumo per cápita (kg/persona y año) de queso en el mundo 8
durante el año 2011.
Figura 5 Micela de caseína obtenida por microscopía electrónica de 10
barrido. (Dalgleish et al., 2004).
Figura 6 Esquema de la glicolisis y del catabolismo de lactato y citrato. 15
(Sarantinopoulos et al., 2001, Marilley & Casey, 2004).
Figura 7 Esquema de la lipolisis y del catabolismo de ácidos grasos libres. 18
(Collins et al., 2003).
Figura 8 Esquema de la proteolisis y del catabolismo de aminoácidos. 22
(Yvon & Rijnen, 2001, McSweeney, 2004).
Figura 9 Corte de queso azul (Roncari-blue). 28
Figura 10 Cuña de Torta del Casar. 32
Figura 11 Esquema de migración y gradientes en queso Brie. 34
Figura 12 Cuña de queso Arzúa-Ulloa. 36
Figura 13 Reacciones de formación de las aminas biógenas. 44
Figura 14 Ruta y reacciones de formación de diaminas y poliaminas. 45
Figura 15 Esquema de sistemas de presurización directa (A) e indirecta (B). 56

Figura 16 Ejemplo de algunos productos comerciales tratados por altas 57
presiones.
Figura 17 Aumento del número de equipos industriales de altas presiones 58
distribuidos en el mundo.
v

 Abreviaturas
a* tendencia al rojo
ADC arginina descarboxilasa
AGL ácidos grasos libres
aw actividad de agua
αS1-CN alfa-s1-caseína
αS2-CN alfa-s2-caseína
α-LA alfa-lactoalbúmina
b* tendencia al amarillo
BAL-NI Bacterias ácido lácticas no iniciadoras
β-CN beta-caseína
β-LG beta-lactoglobulina
C10:0 ácido cáprico
C12:0 ácido laúrico
C14:0 ácido mirístico
C16:0 ácido palmítico
C18:0 ácido esteárico
C18:1 ácido oleico
C18:2 ácido linoleico
C18:3 ácido linolénico
C4:0 ácido butírico
C6:0 ácido caproico
C8:0 ácido caprílico
CADC cetoácido descarboxilasa
CADH cetoácido deshidrogenasa
DAO diamino oxidasa
DAT diamino N-acetiltranferasa
DM dry matter
DOP denominación de origen protegida
ES extracto seco
FAO Food and Agriculture Organization of the United Nations
FDA Food and Drug Administration
FDC fenilalanina descarboxilasa
GC gas chromatography
GC-MS gas chromatography-mass spectrography
GDL glucono-δ-lactona
GRAS Generally Recognized as Safe
γ-CN gamma-caseína
vii

HADH hidroxiácido deshidrogenasa
HDC histidina descarboxilasa
HPLC high pressure liquid chromatography
HPP high pressure processing
HTST High Temperature Short Time
κ-CN kappa-caseína
L* luminosidad
LDC lisina descarboxilasa
LDH L-lactato deshidrogenasa
LPL lipoproteína lipasa
λ-CN lambda-caseína
MAO monoamino oxidasa
MAT metionina adenosiltransferasa
nd no detectado
NS no significativo
ODC ornitina descarboxilasa
OPA orto-ftalaldehído (técnica espectrofotométrica)
PAO poliamino oxidasa
PCA principal component analysis
PCP pirrolidona carboxilil aminopeptidasa
Pep C aminopeptidasa C
Pep D dipeptidasa D
Pep DA dipeptidasa DA
Pep G aminopeptidasa G
Pep N aminopeptidasa N
Pep T tripeptidasa T
Pep V dipeptidasa V
PepA glutamil aminopeptidasa
PepF oligoendopeptidasas intracelulares F
PepI prolin aminopeptidasa
PepL leucil aminopeptidasa
PepO oligoendopeptidasas intracelulares O
PepP aminopeptidasa P
PepQ prolidasa Q
PepR prolinasa R
PepX prolildipeptidil aminopeptidasa
PrtP proteinasa asociada a la pared celular
p-κ-CN para-kappa-caseína
SAMDC S-adenosil-L-metionina descarboxilasa
SPME solid phase micro extraction
TDC tirosina descarboxilasa
viii

TrDC triptófano descarboxilasa
UA unidades de área
ufc unidades formadoras de colonias
UHT Ultra High Temperature
ix

PREVENCIÓN DE LA SOBREMADURACIÓN Y DE LA FORMACIÓN DE

AMINAS BIÓGENAS EN QUESOS MEDIANTE TRATAMIENTOS DE
ALTAS PRESIONES
RESUMEN de TESIS
Departamento de Tecnología de Alimentos
Instituto Nacional de Investigación y Tecnología Agraria y Alimentaria, Madrid
Director: Prof. Dr. Manuel Nuñez Gutiérrez, Departamento de Tecnología de Alimentos,

INIA, Madrid.
Codirector: Dra. Ana del Olmo Sánchez, Departamento de Tecnología de Alimentos, INIA,
Madrid.
xi
Resumen
El queso desarrolla sus características organolépticas durante la maduración, debido

a la acción de diferentes microorganismos y enzimas sobre los componentes de la
cuajada. Los microorganismos y enzimas continúan ejerciendo su efecto durante el
almacenamiento en refrigeración dando lugar a un desequilibrio de los compuestos
responsables del aroma y sabor, lo que ocasiona una pérdida de calidad que limita la
vida útil y provoca el rechazo por parte del consumidor. Un grupo de enzimas
procedentes de microorganismos que pueden estar presentes en el queso son las
descarboxilasas, que en presencia de aminoácidos libres son responsables de la
formación de aminas biógenas. La acumulación de estos compuestos en el queso puede
provocar efectos adversos para la salud del consumidor.
Las altas presiones hidrostáticas se emplean como método de pasteurización no

térmica en diferentes alimentos para eliminar los microorganismos patógenos y
alterantes, manteniendo en mayor medida las propiedades nutricionales y organolépticas
de los alimentos que los tratamientos térmicos. En queso se ha estudiado además el
efecto de las altas presiones con el fin de acelerar la maduración. Únicamente en queso
fresco se han estudiado los tratamientos de altas presiones con el fin de alargar la vida
útil.
En el presente trabajo de investigación se ha evaluado el efecto de la aplicación de

altas presiones hidrostáticas sobre la proteolisis, lipolisis y catabolismo de aminoácidos y
ácidos grasos libres en cuatro variedades de queso (queso azul, Torta del Casar, Brie y
Arzúa-Ulloa) a lo largo de un prolongado periodo de maduración y almacenamiento en
refrigeración, con el fin de alargar la vida útil y evitar la acumulación de aminas biógenas.
En queso azul de leche pasteurizada de oveja, madurado con Penicillium roqueforti

en su interior, se aplicaron 400 y 600 MPa a las 3 (400-3S y 600-3S), 6 (400-6S y 600-6S) y
9 semanas (400-9S y 600-9S) de maduración. Los análisis se realizaron inmediatamente
después de los tratamientos y a los 3, 6, 9 y 12 meses. Los tratamientos de altas
presiones redujeron los niveles de microorganismos, llegándose a niveles no detectables
de P. roqueforti en los quesos tratados con 600 MPa. La fuerte proteolisis primaria
característica de esta variedad de queso no pudo evitarse con los tratamientos de altas
presiones y únicamente los tratamientos de 600 MPa aplicados a las 3 y 6 semanas
consiguieron reducir los niveles de péptidos hidrófilos. Por otro lado, los tratamientos de
xii
Resumen
600 MPa y el de 400 MPa aplicado a las 3 semanas consiguieron reducir la proteolisis
secundaria, con niveles de aminoácidos y proteolisis global similares a los del queso
control de 180 días. Las altas presiones redujeron la formación de tiramina, pero no de
putrescina, triptamina y β-feniletilamina. Los tratamientos de altas presiones

disminuyeron la formación de todos los grupos de ácidos grasos libres, excepto en los
quesos 400-6S y 400-9S. En cuanto a los compuestos volátiles, las altas presiones
consiguieron reducir los niveles de ésteres, alcoholes, hidrocarburos, compuestos
bencénicos y nitrogenados, especialmente en los quesos tratados con 600 MPa, en los
cuales además se redujo la formación de cetonas. En los quesos tratados a las 3 semanas
disminuyeron los niveles de ácidos volátiles, compuestos azufrados y terpenos.
Únicamente el tratamiento de 600 MPa aplicado a las 3 semanas redujo el ligero
aumento de intensidad de sabor que se dio en el queso control. Todos los quesos
tratados mantuvieron la misma calidad de sabor que el queso control, excepto el queso
tratado con 600 MPa a las 3 semanas.
En Torta del Casar elaborada con leche cruda de oveja y cuajo vegetal de cardo
(Cynara cardunculus), se aplicaron 400 y 600 MPa a las 3 (400-3S y 600-3S) y 5 semanas
(400-5S y 600-5S) de maduración. Los análisis se realizaron inmediatamente después de
los tratamientos y a los 2, 4, 6 y 8 meses. Las altas presiones redujeron los niveles de
microorganismos, en especial los tratamientos de 600 MPa, y evitaron el aumento de pH
que se dio en el queso control. Los tratamientos de 600 MPa consiguieron frenar la
proteolisis primaria, reduciendo la hidrólisis de caseínas y la formación de péptidos
hidrófilos. La proteolisis secundaria únicamente se frenó con los tratamientos de 600
MPa, reduciendo el tratamiento de 600 MPa aplicado a las 3 semanas la formación de
aminoácidos libres y los tratamientos de 600 MPa la proteolisis global. Las altas
presiones redujeron la actividad tirosina descarboxilasa, causando los tratamientos de
600 MPa un descenso de los niveles de tiramina e histamina, mientras que todos los
tratamientos consiguieron reducir los niveles de cadaverina y, excepto el tratamiento de
400 MPa aplicado a las 3 semanas, los de putrescina, triptamina y β-feniletilamina. Todos
los tratamientos provocaron una reducción de la actividad esterasa al final del

almacenamiento en refrigeración, así como de los niveles de ácidos grasos libres de
cadena corta, media y larga, de los ácidos de cadena ramificada y de ácido propanoico.
xiii
Resumen
Por lo que respecta a los compuestos volátiles, las altas presiones consiguieron reducir
los niveles de ésteres, aldehídos y especialmente de compuestos azufrados, que
alcanzaron niveles muy elevados en el queso control. Los tratamientos de 600 MPa
redujeron los niveles de alcoholes y los tratamientos aplicados a las 3 semanas los de
ácidos volátiles. Todos los tratamientos evitaron la fuerte pérdida de calidad de sabor y
olor que se dio en el queso control, reduciéndose también el aumento de intensidad de
olor, olor pútrido y olor rancio. Los tratamientos de 600 MPa frenaron el incremento de
intensidad de sabor que se dio en el queso control.
En queso Brie de leche de vaca pasteurizada, con Penicillium camemberti en su

superficie, se aplicaron 400 y 600 MPa a las 2 (400-2S y 600-2S) y 3 semanas (400-3S y
600-3S) de maduración. Los análisis se realizaron inmediatamente después de los
tratamientos y a los 1, 2, 3 y 4 meses. Los tratamientos de altas presiones redujeron los
niveles de microorganismos, llegándose a niveles no detectables de P. camemberti en los
quesos tratados con 600 MPa, y evitaron el fuerte aumento de pH que se dio en el queso
control. La proteolisis primaria se vio reducida por los tratamientos de 600 MPa,
disminuyendo la hidrólisis de αS- y β-caseína, evitando todos los tratamientos el fuerte
aumento de péptidos hidrófobos que se dio en el queso control. Sin embargo, las altas
presiones no consiguieron frenar la proteolisis secundaria, estimada por los niveles de
aminoácidos libres y la proteolisis global. Las altas presiones redujeron los niveles de
ácidos grasos libres de cadena corta y ramificada. Además, los tratamientos aplicados a
las 2 semanas redujeron los niveles de ácidos grasos de cadena media y larga y los
tratamientos de 600 MPa los niveles de ácido etanoico. En cuanto a los compuestos
volátiles, las altas presiones consiguieron limitar el incremento de cetonas y compuestos
bencénicos, nitrogenados y azufrados. Todos los tratamientos evitaron el descenso de
elasticidad y firmeza, mientras que sólo los tratamientos aplicados a las 2 semanas
limitaron el descenso de fracturabilidad. Las altas presiones evitaron la pérdida de
luminosidad y el aumento de la tendencia al rojo y al amarillo en el interior del queso.
Las altas presiones provocaron la pérdida de la capa superficial de moho haciendo que
los parámetros de color de la superficie fuesen muy diferentes a los del queso control.
Todos los tratamientos evitaron el aumento de la intensidad de sabor y olor y de sabor
amargo, así como la pérdida de calidad de sabor y olor que se dio en el queso control.
xiv
Resumen
En el queso Arzúa-Ulloa elaborado con leche cruda de vaca, se aplicaron 400 y 600
MPa a las 2 (400-2S y 600-2S) y 3 semanas (400-3S y 600-3S) de maduración. Los análisis
se realizaron inmediatamente después de los tratamientos y a los 2, 4, 6 y 8 meses. Los
tratamientos de altas presiones redujeron los niveles de microorganismos, especialmente
en los quesos tratados con 600 MPa, y evitaron el incremento de pH que se dio en el
queso control. Las altas presiones no consiguieron frenar la proteolisis primaria pero sí la
secundaria, reduciendo los niveles de aminoácidos libres y la proteolisis global. Las altas
presiones disminuyeron la actividad tirosina descarboxilasa y los niveles de tiramina.
Además, los tratamientos de 600 MPa redujeron los niveles de putrescina e histamina.
Las altas presiones disminuyeron los niveles de ácido etanoico y de los ácidos grasos de
cadena corta. Por lo que respecta a los compuestos volátiles, las altas presiones evitaron
parcialmente el descenso de ácidos volátiles, éteres y compuestos azufrados. En los
quesos tratados con 600 MPa se evitó el descenso de cetonas e hidrocarburos. Los
tratamientos de 600 MPa frenaron la pérdida de luminosidad del interior del queso,
aunque redujeron la tendencia al rojo y no afectaron a la tendencia al amarillo. Los
tratamientos de altas presiones no evitaron el incremento de la intensidad del sabor ni
de sabor umami que se dio en el queso control, pero mantuvieron la calidad del sabor.
Financiación para el desarrollo de la presente Tesis Doctoral:
La presente Tesis Doctoral se realizó en el marco del proyecto AGL 2009-07801,

“Prevención de la sobremaduración y de la formación de aminas biógenas en quesos
mediante tratamientos por altas presiones” del Ministerio de Ciencia e Innovación.
Durante la realización de la presente Tesis, su autor fue beneficiario de una beca F.P.I.
(Formación de Personal Investigador) del Ministerio de Ciencia e Innovación, con
referencia BES-2010-030444 asociada al proyecto AGL 2009-07801.
xv
ESTRUCTURA DE LA TESIS
La presente tesis doctoral consta de los siguientes apartados:
 Una Introducción (Capítulo 1) redactada en español, que incluye una

revisión bibliográfica sobre el queso, las aminas biógenas y las altas
presiones hidrostáticas, y los Objetivos principales del trabajo de
investigación.
 Nueve capítulos temáticos (Capítulos 2 a 10) redactados en inglés y

presentados en formato de publicaciones en revistas científicas, que
constituyen el cuerpo de la tesis.
 Una Discusión general (Capítulo 11) y una sección de Conclusiones

(Capítulo 12), redactadas en español.
 Dos resúmenes ampliados, uno redactado en español (Capítulo 13) y el

segundo redactado en inglés (Capítulo 14).
xvii

Capítulo 1. Introducción.

1

Fotografía: equipo de altas presiones modelo “Hiperbaric 135” empleado para la
presurización de los quesos, de Hiperbaric (Burgos).
Introducción
1.1. El queso
El queso es un sistema complejo en el que se suceden un gran número de

reacciones bioquímicas que hacen que en un momento dado se alcance el equilibrio
correcto de los compuestos que dan al producto final sus características organolépticas
típicas. Los agentes responsables de estos cambios son los microorganismos y enzimas
presentes en la leche, los microorganismos y enzimas añadidos (en particular los
cultivos o fermentos lácticos y el cuajo u otras enzimas coagulantes) y los
microorganismos contaminantes que acceden a la leche o al queso durante la
maduración. Factores intrínsecos como el pH y la actividad de agua, y externos como la
temperatura y la humedad relativa modulan la actividad de los agentes implicados en la
maduración del queso.
1.1.1. Definición, historia y producción
1.1.1.1. Definiciones de queso
Queso es el nombre genérico que se da a un grupo de productos alimenticios

basados en la coagulación de la leche y separación del suero. Según el Diccionario de
la Lengua de la Real Academia Española el queso se define como “el producto obtenido
por maduración de la cuajada de la leche con características propias para cada uno de
los tipos según su origen o método de fabricación”.
Una definición más completa sería la dada por la FAO en el Codex Alimentarius
para Leche y Productos Lácteos, en el apartado 2 de la Norma General para el Queso
(CODEX STAN 283-1978), donde se define de la siguiente manera:
“2.1 Se entiende por queso el producto blando, semiduro, duro y extra duro,
madurado o no madurado, y que puede estar recubierto, en el que la
proporción entre las proteínas de suero y la caseína no sea superior a la de la
leche, obtenido mediante:
a. Coagulación total o parcial de la proteína de la leche, leche desnatada, leche

parcialmente desnatada, nata, nata de suero, o suero de mantequilla, o de
cualquier combinación de estos materiales, por acción del cuajo u otros
coagulantes idóneos, y por escurrimiento parcial del suero que se desprende
3

Introducción
como consecuencia de dicha coagulación, respetando el principio de que la

elaboración del queso da lugar a una concentración de la proteína láctea
(especialmente la porción de caseína) y que por consiguiente, el contenido de
proteína del queso deberá ser evidentemente más alto que el de la mezcla de
los materiales lácteos ya mencionados en base a la cual se elaboró el queso; y/o
b. Técnicas de elaboración que comportan la coagulación de la proteína de la

leche y/o de productos obtenidos de la leche que dan un producto final que
posee las mismas características físicas, químicas y organolépticas que el
producto definido en el apartado a.
2.1.1 Se entiende por queso sometido a maduración el queso que no está listo
para el consumo poco después de la fabricación, sino que debe mantenerse
durante cierto tiempo a una temperatura y en unas condiciones tales que se
produzcan los cambios bioquímicos y físicos necesarios y característicos del
queso en cuestión.
2.1.2 Se entiende por queso madurado por mohos un queso curado en el que la
maduración se ha producido principalmente como consecuencia del desarrollo
característico de mohos por todo el interior y/o sobre la superficie del queso.
2.1.3 Se entiende por queso sin madurar el queso que está listo para el consumo
poco después de su fabricación.”
1.1.1.2. Antecedentes históricos
No se sabe con exactitud cuándo o dónde comenzó la elaboración de los primeros

quesos. Se supone que esta práctica está relacionada con la historia de la
domesticación de animales productores de leche hace 8.000 – 10.000 años, en el
Neolítico. Según la leyenda, un mercader árabe que realizaba un largo viaje por el
desierto conservó leche en un recipiente fabricado a partir del estómago de un cordero
y cuando fue a consumirla vio que estaba coagulada y que se había separado el suero.
Los primeros productos lácteos fermentados probablemente se produjeron por una
combinación fortuita de sucesos, como la aparición de un grupo de bacterias con la
capacidad de crecer en la leche y producir suficiente ácido para bajar el pH hasta el
4

Introducción
punto isoeléctrico de las caseínas junto con unas condiciones y temperatura apropiadas
para que estas crecieran y dieran lugar al producto fermentado. Originariamente el
queso debió elaborarse como método para conservar la leche mediante coagulación
ácida espontánea. Cuando el gel de leche coagulada por acidificación se rompía, este se
separaba en suero y cuajada, siendo el suero ácido una bebida refrescante, mientras
que la cuajada podía tomarse fresca o bien almacenarse para su posterior consumo.
Posteriormente se observó que la leche coagulaba en contacto con determinados
productos de origen animal o vegetal. Las
enzimas proteolíticas están ampliamente
distribuidas, pero una primera fuente pudo
ser el estómago de los rumiantes. Al
encontrar restos de cuajada en el
estómago de mamíferos lactantes, después
de ser sacrificados, se pudieron empezar a
utilizar los estómagos como fuente de
estas enzimas, que siguen empleándose en
la actualidad. Las ventajas de convertir la
leche en queso, tales como su estabilidad
durante el almacenamiento y la facilidad
de transporte, hicieron que los procesos
Figura 1. Fragmentos de vasijas para el
de elaboración de queso se fueran desuerado (b, d) y esquema de la
estableciendo en las antiguas civilizaciones reconstrucción de los fragmentos de la vasija
(a, c). (Salque et al., 2013).
(Fox & McSweeney, 2004). Existen
evidencias arqueológicas (Figura 1) sobre la elaboración de queso que datan del año
5500 a.C. en el norte de Europa (Salque et al., 2013). Los primeros testimonios gráficos
son el Friso de la Lechería (Figura 2), friso sumerio datado en el tercer milenio a.C., y
diversos murales de tumbas egipcias de alrededor del año 2000 a.C.
En la antigua Grecia el queso era un alimento habitual entre campesinos y pastores.

En el siglo VIII a.C. Homero describe en la Odisea a Polifemo elaborando quesos de
leche de oveja y en la Ilíada hace referencia a quesos elaborados con savia de higuera
como coagulante. Es de la palabra griega formos (cesta donde se depositaba el queso y
5

Introducción
Figura 2. Friso de la lechería. (Friso sumerio, 3000 a.C.).
se separaba el suero) de donde viene el nombre que se da al queso en francés, italiano

y catalán (fromage, formaggio y formatge). El consumo del queso en Grecia se hacía
condimentándolo con miel, hierbas aromáticas y aceite.
En Roma se siguió perfeccionando la forma de hacer queso, como expuso

Columella en el siglo I d.C. en el libro VII del tratado “De re rustica”. Más tarde Plinio el
Viejo en su obra “Naturalis historia”, describe la diversidad de quesos de la época,
donde describe el queso Lunense, de 100 libras de peso. Aquí el queso era consumido
por todas las clases sociales y los soldados romanos lo recibían como parte de su
ración. De la palabra latina caseus (carente de suero) deriva el nombre de queso en
español, queixo en gallego o queijo en portugués.
Los avances en la elaboración del queso en Europa progresaron lentamente en los

siglos posteriores a la caída del Imperio Romano. Durante la Edad Media, la mejora y
conservación de las prácticas queseras tuvo lugar en los monasterios, de donde
proceden muchos de los quesos que conocemos en la actualidad.
La llegada de la Época Renacentista implicó el auge del comercio y el aumento de

la población urbana, convirtiendo el queso en producto importante para la economía.
En esta época el queso comenzó a comercializarse fuera de las zonas de producción,
llegando hasta el Nuevo Mundo con las colonizaciones.
Durante el siglo XIX el queso se convirtió en un producto gastronómico símbolo de

exquisitez y distinción. Con la llegada del siglo XX y los avances tecnológicos y
bacteriológicos, el sector se industrializó, haciendo posible el aumento de la producción
y del comercio y consumo a nivel mundial.
1.1.1.3. Producción y consumo
En la actualidad el queso es uno de los principales productos agroalimentarios, con

una producción mundial de aproximadamente 20,6 x 106 toneladas en el año 2012 y un
6

Introducción
aumento medio anual de alrededor del 1,8 % en los últimos 10 años. Como se recoge
en la Figura 3, el continente con mayor producción es Europa, con aproximadamente
10,4 x 106 toneladas (50,5 % de la producción mundial) pero el país con mayor
producción es Estados Unidos con 5,3 x 106 toneladas (25,5 % de la producción
mundial).
Figura 3. Producción (en toneladas) de queso durante el año 2012.
América del
Norte
5.721.097
Asia
1.484.715
América del Sur
1.085.972
Europa
10.450.508 África
977.886
Oceanía
América central
615.340
y Caribe
266.197
Países productores.
6.000.000
5.000.000
4.000.000
3.000.000
2.000.000
1.000.000
EE.UU. Alemania Francia Italia Países Bajos

Polonia Egipto Rusia Argentina Canadá
Fuente: FAOSTAT.
En cuanto al consumo per cápita (Figura 4) en el año 2011, la media mundial se

estima en 2,9 kg/persona y año, situándose Grecia a la cabeza de consumo con 26,7
kg/persona y año.
7

Introducción
En España la producción de queso durante el año 2012 fue de 213.732 toneladas,

situándose como decimonoveno productor a nivel mundial (FAOSTAT, 2014). Por otro
lado el consumo per cápita en el año 2013 fue de 8,2 kg (MAGRAMA, 2014).
Figura 4. Consumo per cápita (kg/persona y año) de queso en el mundo durante el año 2011.
Fuente: FAOSTAT.
1.1.2. Elaboración del queso
La transformación de la leche en queso comprende una serie de etapas:

tratamiento o preparación de la leche, acidificación, coagulación, corte de la cuajada y
desuerado, moldeo, prensado, salado y maduración. Dichas etapas pueden diferir según
la variedad. A continuación se describen las principales características de la leche y las
etapas de la transformación de la leche en queso.
1.1.2.1. La leche
La leche es una dispersión coloidal formada por agua, grasa, proteínas, azúcares,
vitaminas y minerales. Su componente mayoritario es el agua. La concentración del
resto de los componentes depende de factores como la especie (Tabla 1) y, dentro de
una misma especie, de la raza, la alimentación, la estación del año, la fase de lactancia e
incluso el número diario de ordeños.
Las proteínas de la leche se clasifican en dos grupos principales, las caseínas y las
proteínas del suero. Las caseínas precipitan a 20 ºC cuando se alcanza un pH de 4,6 (su
8

Introducción
punto isoeléctrico), mientras que las proteínas del suero permanecen solubles en esas
condiciones. Las proteínas del suero representan entre el 14 y el 26 % de la proteína
total de la leche y se dividen en dos fracciones, lactoalbúminas y lactoglobulinas. La
fracción de las lactoalbúminas está formada principalmente por α-lactoalbúmina (α-LA)
y β-lactoglobulina (β-LG), además de proteínas menores. La β-LG representa
aproximadamente el 50 % de las proteínas séricas en la leche de vaca y su punto
isoeléctrico se alcanza a un pH de aproximadamente 5,2. Su estructura es globular y es
muy resistente a la proteolisis. La α-LA representa aproximadamente el 20 % de las
proteínas séricas de la leche de vaca, su estructura también es globular y su punto
isoeléctrico se alcanza a un pH de aproximadamente 4,8 (Fox, 2009). La fracción de las
lactoglobulinas está compuesta principalmente por inmunoglobulinas.
Tabla 1. Composición de la leche de vaca, oveja y cabra.
Vaca Oveja Cabra
Proteína % 2,35-3,80 4,75-7,20 2,61-4,09
Grasa % 3,50-5,02 3,60-9,97 3,48-5,63
Lactosa % 4,40-4,90 4,11-5,51 4,10-4,76
Extracto seco % 11,44-13,72 14,40-20,70 11,60-14,96
Fuente: Malacarne et al., 2002, Park et al., 2007, Raynal-Ljutovac et al., 2008, Yuksel et al., 2012.
La grasa de la leche se encuentra en forma de glóbulos cuyo diámetro puede ir

desde 0,1 a 20 µm. Alrededor del 99 % de la grasa de la leche se encuentra
compartimentada en estos glóbulos. La membrana que rodea a los glóbulos está
formada por fosfolípidos, colesterol, lipoproteínas, glicoproteínas y enzimas. El interior
de los glóbulos de grasa está constituido en su mayoría (98 %) por triglicéridos, aunque
también contienen diglicéridos, monoglicéridos, ácidos grasos libres, lípidos polares y
esteroles, así como trazas de vitaminas liposolubles y β-caroteno. El peso molecular de
los triglicéridos puede ir desde 470 a 890 g/mol y la colocación de los ácidos grasos en
las posiciones del glicerol no es aleatoria (Gresti et al., 1993, Blasi et al., 2008, Lopez,
2011). Los ácidos grasos de cadena inferior a 10 carbonos suelen localizarse
9

Introducción
mayoritariamente en la posición sn-3 mientras que el ácido mirístico tiende

preferentemente a colocarse en la posición sn-2. El ácido palmítico suele encontrase
distribuido por igual entre las posiciones sn-1 y sn-2 mientras que el oleico, aunque se
distribuye entre las tres posiciones, suele encontrarse mayoritariamente en la posición
sn-1.
Las caseínas son las principales

proteínas de la leche y se encuentran
formando agregados moleculares llamados
micelas, cuyo diámetro puede ir desde 50 a
500 nm (Figura 5). Las micelas están
formadas aproximadamente por un 92 %
200 nm
de caseínas y un 8 % de compuestos
inorgánicos como fosfatos, calcio y otras Figura 5. Micela de caseína obtenida por
microscopía electrónica de barrido. (Dalgleish et
sales y cationes. Las micelas están al., 2004).
formadas por cuatro caseínas principales:
la alfa S1 (αS1-CN), alfa S2 (αS2-CN), beta (β-CN) y kappa (κ-CN). La mayoría de los
modelos estructurales indican que la κ-CN está situada principalmente en la superficie,
formando una capa protectora que proporciona estabilidad a la micela. La cantidad de
κ-CN es inversamente proporcional al tamaño de la micela. La β-CN se encuentra
distribuida principalmente en el interior, mientras que las αS-CNs se encuentran
distribuidas por toda la micela (Dalgleish et al., 2004, Dalgleish & Corredig, 2012, Bijl et
al., 2014). La estabilidad térmica de las caseínas y por tanto la de la leche, así como el
color blanco y la coagulación por efecto del cuajo, se deben a la estructura y
propiedades de la micela de caseína (Fox & Brodkorb, 2008). Desde el punto de vista de
la estabilidad de la micela, la κ-CN es la caseína más importante ya que es la única
soluble en presencia de calcio a la concentración que este se encuentra en la leche y es
capaz de estabilizar diez veces su peso de caseínas sensibles al calcio (Farrell et al.,
2004, Fox & Brodkorb, 2008). Las principales proteínas de la leche (αS1-CN, αS2-CN, β-
CN, κ-CN, α-LA y β-LG) presentan microheterogeneidades que pueden deberse a cinco
factores: variabilidad en el grado de fosforilación, puentes disulfuro, grado de
glicosilación, polimorfismo genético e hidrólisis producida por la plasmina. La plasmina
10

Introducción
es una enzima capaz de hidrolizar la β-CN dando péptidos C-terminales que son las
llamadas gamma caseínas (γ-CN) y péptidos N-terminales que son las proteosa-
peptonas. Los fragmentos N-terminales de la hidrólisis de la αS1-CN por la acción de la
plasmina son las lambda caseínas (λ-CN). Las variantes debidas al polimorfismo
genético ocurren durante la síntesis de estas proteínas por sustitución o deleción de
aminoácidos. Existen variantes genéticas en leche de vaca, oveja, cabra y otras las
especies. En leche de vaca se han identificado 8 variantes de αS1-CN, 4 de αS2-CN, 12 de
β-CN, 11 de κ-CN, 3 de α-LA y 11 de β-LG (Caroli et al., 2009, Fox, 2009). Las cuatro
caseínas principales se diferencian entre ellas en el número de residuos aminoacídicos
que las forman (199 la αS1-CN, 207 la αS2-CN, 209 la β-CN y 169 la κ-CN), peso molecular
(23.612 la αS1-CN, 25.228 la αS2-CN, 23.980 la β-CN y 19.005 la κ-CN), la cantidad de
grupos fosfato que llevan unidos, el grado de glicosilación y el punto isoeléctrico (Fox,
2009). Su proporción varía según el tipo de leche, siendo el porcentaje respecto a la
caseína total de 37 y 33 % de αS1-CN, 7 y 14 % de αS2-CN, 42 y 30 % de β-CN y 9 y 14 %
de κ-CN en leche de vaca y oveja respectivamente (Bramanti et al., 2003). La cantidad
de proteína de la leche, así como la proporción de caseínas y la estructura de las
micelas de caseína tienen una gran influencia en las características tecnológicas de la
leche.
La lactosa, el principal carbohidrato de la leche, es un disacárido formado por una

molécula de α- o β-glucosa y otra de β-galactosa. Durante la elaboración del queso la
mayor parte de la lactosa se pierde en el suero, mientras que la parte que queda
retenida en la cuajada es empleada por las bacterias lácticas para producir ácido láctico.
En la leche también se pueden encontrar, aunque en muy baja concentración, glucosa,
galactosa y oligosacáridos.
La leche contiene minerales esenciales (potasio, sodio, calcio, magnesio, cloro y

ésteres de fosfato), de los cuales sodio, potasio y cloro pueden aparecer como iones
libres. Las concentraciones de calcio, magnesio, fosfato y citrato dependen del
contenido de caseínas de la leche. La leche también contiene diversas vitaminas
liposolubles como la A, D y E, y vitaminas hidrosolubles como la B1, B2, B3, B5, B6, B12 y C.
Se han descrito más de 70 enzimas endógenas en la leche, algunas de las cuales

tienen características tecnológicas reconocidas, mientras que de otras se desconoce
11

Introducción
cuál es su papel. Estas enzimas pueden encontrarse libres, asociadas a los glóbulos de
grasa o a las micelas de caseína. La leche contiene, entre otras, proteinasas como la
plasmina, que es una proteinasa alcalina, o la catepsina D, que es una proteinasa ácida
cuya concentración está relacionada con el número de células somáticas (Fox & Kelly,
2006). Otras enzimas con características tecnológicas conocidas son las lipasas y
esterasas, entre las que destaca la lipoproteína lipasa. También se encuentran lisozima,
xantina oxidorreductasa, catalasa, lactoperoxidasa, amilasa y aldolasa.
1.1.2.2. La coagulación de la leche
Como ya se ha indicado, la elaboración de quesos conlleva una serie de etapas o

pasos que dependen del tipo de queso a elaborar (Tabla 2). La coagulación de la leche
es la única etapa común a todas las variedades de queso y da como resultado un gel de
proteína en el que quedan atrapados los glóbulos de grasa (Fox & McSweeney, 2004).
Tabla 2. Etapas en la elaboración del queso.
1 Preparación de la leche (pasteurización, homogeneización, microfiltración...). Opcional.

2 Adición de cultivo iniciador y/o cultivos adjuntos. Opcional.
3 Coagulación (enzimática, ácida o mixta).
4 Corte de la cuajada.
5 Tratamiento de la cuajada (calentamiento, lavado, salado...). Opcional.
6 Desuerado.
7 Moldeado.
8 Prensado. Opcional
9 Maduración. Opcional
La coagulación de la leche puede ser ácida y/o enzimática. La coagulación ácida se

puede realizar por adición de bacterias que fermentan la lactosa a ácido láctico, por
adición de ácidos o por adición de glucono-δ-lactona (GDL) que por hidrólisis da ácido
glucónico, provocando la acidificación de la leche. La acidificación de la leche provoca
cambios en las propiedades fisicoquímicas de las micelas de caseína. A medida que el
pH desciende, el fosfato cálcico coloidal de las micelas se va disolviendo provocando la
desestabilización de estas y la liberación de las caseínas al medio. Cuando el pH se va
aproximando al punto isoeléctrico de las caseínas se van formando agregados hasta
12

Introducción
que se obtiene el gel láctico desmineralizado (Lucey & Singh, 1997). El gel formado por
acidificación es un gel firme, friable y poco contráctil que retiene mucha agua.
La coagulación enzimática se realiza mediante la adición de proteasas a la leche.

Los cuajos tradicionalmente empleados en la elaboración de queso son los de origen
animal, que contienen quimosina (EC 3.4.23.4) como principal enzima (80-90 %) y (10-
20 %) pepsina (EC 3.4.23.1). Este cuajo se obtiene del estómago de mamíferos lactantes.
La quimosina es una proteasa aspártica y su actividad proteolítica comparada con la de
otras proteinasas es muy baja, pero es muy activa en la hidrólisis del enlace Phe105-
Met106 de la κ-caseina (Horne & Banks, 2004). La pepsina es una enzima muy parecida a
la quimosina aunque menos específica. Es una proteasa muy ácida, con un pH óptimo
de actividad en torno a 2 y que a valores de pH superiores a 6,6 se inhibe. Existen
también cuajos microbianos obtenidos de hongos o bacterias (EC 3.4.23.9) como:
Rhizomucor miehei, Aspergillus oryzae o Irpex lactis cuya especificidad es comparable a
la de la quimosina, pero que tienen mayor estabilidad térmica (Kumar et al., 2010). La
quimosina recombinante es una preparación enzimática obtenida de microorganismos
modificados genéticamente como Aspergillus niger, Kluyveromyces lactis y Escherichia
coli, a los que se les ha introducido el gen de la quimosina. Debido a la limitación en la
producción de cuajo natural, al aumento de la producción de quesos y a la aprobación
de la quimosina recombinante como GRAS (Generally Recognized as Safe) por la FDA
(Food and Drug Administration) de Estados Unidos en los años 90, en la actualidad más
del 50 % de los quesos fabricados en el mundo son elaborados con este tipo de
quimosina (Kumar et al., 2010). También pueden emplearse proteasas de origen vegetal,
de diversas fuentes como hojas o ramas de higuera, frutas (kiwi o melón), raíces
(jengibre) y flores, como las cardosinas o cinarasas extraídas del cardo (Cynara
cardunculus) (Shah et al., 2014). Las cardosinas son proteinasas aspárticas que, al igual
que la quimosina, hidrolizan el enlace Phe105-Met106 de la κ-caseina (Roseiro et al.,
2003). La hidrólisis de la κ-caseína provoca la ruptura de la capa protectora de la micela
de caseína en macropéptido y para-κ-caseína. En presencia de suficiente calcio estas
micelas modificadas comienzan a unirse unas con otras en largas cadenas ramificadas,
que acaban formando una estructura tridimensional que da lugar a la cuajada (Horne &
Banks, 2004, Osintsev & Qvist, 2004). La mayor parte del cuajo empleado en la
13

Introducción
elaboración del queso se pierde con el suero, mientras que la parte de cuajo que queda
retenido en la cuajada continúa con su actividad pudiendo hidrolizar la β-caseína y la
αS1-caseína. La αS2-caseína parece ser resistente a la hidrólisis por la quimosina y la para-
κ-caseína, a pesar de tener varios enlaces adecuados para su actividad, tampoco parece
que se hidrolice con esta enzima.
1.1.2.3. La maduración del queso
Una de las etapas más importantes en el proceso de elaboración de la mayoría de

los quesos es la maduración, proceso por el cual el queso alcanza las características de
sabor, aroma y textura típicas de cada variedad. Cada tipo de queso tiene un tiempo y
unas condiciones específicas de maduración (temperatura y humedad). El tiempo de
maduración puede ir desde las 1-2 semanas para el queso Mozzarella hasta 2 años para
Parmigiano-Reggiano. La maduración es un proceso complejo que normalmente
implica cambios en la microbiota, incluyendo la muerte y lisis del cultivo iniciador, el
desarrollo de microbiota endógena de la leche y en algunos casos el crecimiento de
microbiota secundaria, añadida o contaminante accidental. La actividad metabólica de
la microbiota secundaria influye en el desarrollo del sabor y en algunos casos, como en
los quesos con mohos en su superficie, en la textura. La maduración implica a su vez
una serie de cambios bioquímicos que pueden deberse a las enzimas del cuajo, las
endógenas de la leche, las de los cultivos iniciadores, las de la microbiota secundaria
añadida, las de la microbiota de la leche y las enzimas exógenas añadidas para acelerar
la maduración (McSweeney, 2004). Las reacciones bioquímicas que se dan durante la
maduración se pueden clasificar en tres apartados principales:
1.1.2.3.1. Glicolisis de la lactosa. Catabolismo de lactato y citrato
Durante la formación de la cuajada, la lactosa es fermentada por el cultivo iniciador

dando ácido láctico (principalmente el isómero L), lo que provoca una disminución del
pH. La acidificación tiene una gran influencia en la textura debido a la desmineralización
de la micela de caseína. También tiene influencia en la proteolisis debido a que la
micela de caseína desmineralizada es más susceptible a la proteolisis y, además, a pH
bajo la cuajada puede retener más quimosina.
14

Introducción
La lactosa que queda retenida en la cuajada es rápidamente metabolizada a L-

lactato por el cultivo iniciador. Al cabo de unos días apenas se detecta lactosa. El lactato
formado es a su vez un importante sustrato para una serie de reacciones (Figura 6) que
ocurren durante la maduración del queso (McSweeney & Fox, 2004) y que se describen
brevemente a continuación:
OH OH
C O OH H2C H2C
HO O H O OH
CH2 H
H
HO C C O OH
OH H O OH H
H H H
CH2 H OH H OH
C O OH Lactosa
Citrato
OH OH
H2C H2C
HO O OH H O OH
H H
OH H OH H
H H HO H
H OH H OH
Galactosa Glucosa
Tagatosa-6-P Glucosa-6-P
Dihidroxiacetona-P Gliceraldehído-3-P
CO2
Oxalacetato Piruvato
Acetato CO2
-Acetolactato CO2
Acetil-CoA
Propionato, H3C CH CO OH CO2
Acetato, Propionibacterium
OH Diacetilo
LD
CO2 y H2O Lactato H Acetoína
Acetato
ty ium
BA
um
Penicillium
DL-lactato
bu rid
L-
ric
Acetaldehído
ro st
O2
ty Clo
Acetato 2,3-butanodiol
Butirato y CO2
y H2
CO2 y H2O Etanol
Figura 6. Esquema de la glicolisis y del catabolismo de lactato y citrato (Sarantinopoulos et

al., 2001, Marilley & Casey, 2004). (LDH: Lactato deshidrogenasa. BAL-NI: Bacterias lácticas no
iniciadoras).
· En la mayoría de los quesos se produce la racemización del L-lactato a D-lactato

por la acción de bacterias lácticas de la microbiota secundaria. Esta racemización suele
producirse por la acción de la enzima L-lactato deshidrogenasa (LDH), que oxida el L-
15

Introducción
lactato formando piruvato, el cual es reducido a D-lactato por efecto de la D-LDH

obteniéndose la mezcla racémica de DL-lactato. La racemización probablemente no
afecta al sabor final del queso, pero esta mezcla es menos soluble que el L-lactato puro
lo que provoca que se formen cristales de lactato cálcico que pueden provocar rechazo
por el consumidor.
· En presencia de oxígeno algunas bacterias lácticas pueden oxidar el lactato dando

acetato, etanol, formato y/o CO2 aunque esta ruta no es muy común debido a la
limitación de oxígeno que existe en el queso. Las bacterias del género Pediococcus en
presencia de una elevada concentración de oxígeno pueden formar 1 mol de acetato y
1 mol de CO2 a partir de 1 mol de lactato y 1 mol de O2. El pH óptimo para que se dé
esta oxidación es entre 5 y 6. Esta reacción depende por tanto de las cepas presentes,
así como de la concentración de lactato y de la disponibilidad de oxígeno. El acetato
suele encontrarse en la mayoría de los quesos y contribuye al sabor.
· En los quesos madurados con mohos en la superficie (como Camembert o Brie) el

catabolismo oxidativo es muy fuerte. Los mohos superficiales (Penicillium camemberti)
catabolizan rápidamente el lactato a H2O y CO2 provocando un aumento del pH en la
superficie, lo que causa en el queso un gradiente de pH y la difusión del lactato hacia la
superficie.
· El lactato también puede ser catabolizado anaeróbicamente por Clostridium
tyrobutyricum produciéndose butirato e hidrógeno, lo que provoca el defecto conocido

como "hinchazón tardía" que se da en algunos quesos de pasta dura.
· En los quesos tipo Suizo el lactato es metabolizado por Propionibacterium
freudenreichii subsp. shermanii con formación de propionato, acetato, CO2 y H2O. El

CO2 migra a través de la cuajada hasta los puntos más débiles en los que se acumula,
formándose los característicos ojos de este tipo de quesos.
Los niveles de citrato en la leche son bajos y además la mayor parte se pierde, al
igual que la lactosa, en el suero. Las pequeñas cantidades de citrato residual retenido en
la cuajada pueden ser metabolizadas por los cultivos iniciadores mesófilos, formándose
diacetilo, acetato, acetoína, 2,3-butanodiol y CO2. El CO2 producido durante el
metabolismo del citrato es el responsable de los característicos ojos de los quesos de
16

Introducción
tipo holandés (Edam, Gouda) y de defectos en queso Cheddar y Cottage. Debido a la

formación de diacetilo, acetato, acetoína y 2,3-butanodiol el metabolismo del citrato es
muy importante en el aroma y sabor de quesos de tipo holandés, Cheddar y Cottage
(McSweeney & Fox, 2004).
1.1.2.3.2. Lipolisis y catabolismo de los ácidos grasos libres
La fracción lipídica es esencial para la correcta formación del sabor durante la

maduración del queso. La mayor parte de la grasa de la leche son triglicéridos
(aproximadamente el 98 %), en los que los ácidos grasos están colocados en posiciones
no aleatorias del glicerol. Durante la maduración esta grasa puede ser oxidada o
hidrolizada. Los ácidos poliinsaturados son especialmente sensibles a la oxidación,
dando aldehídos insaturados que provocan defectos del sabor como la rancidez. La
oxidación lipídica no ocurre de manera significativa en el queso, debido probablemente
a su bajo potencial redox y a la presencia de antioxidantes naturales. La hidrólisis
enzimática de los triglicéridos (lipolisis) ocurre en los quesos durante la maduración
dando lugar a ácidos grasos libres, diglicéridos, monoglicéridos y, en última instancia,
glicerol (Figura 7). La mayoría de los quesos tienen un grado de lipolisis bajo o
moderado y un exceso de ácidos grasos libres provoca defectos de sabor. Únicamente
los quesos madurados por mohos o los de pasta dura italianos se caracterizan por un
elevado grado de lipolisis.
La lipolisis en el queso se debe a la presencia de enzimas lipolíticas (hidrolasas),

que rompen el enlace éster que existe entre el ácido graso y el glicerol, obteniéndose el
ácido graso libre. Estas enzimas lipolíticas pueden clasificarse como lipasas y esterasas,
que se diferencian por tres características principales: longitud de la cadena del aciléster
que hidrolizan, naturaleza físico-química del sustrato y la cinética enzimática.
Las esterasas hidrolizan preferentemente acilésteres de 2 a 8 átomos de carbono,

mientras que las lipasas hidrolizan preferentemente los de 10 o más átomos de
carbono. Las esterasas hidrolizan sustratos solubles en soluciones acuosas mientras que
las lipasas hidrolizan sustratos emulsionados. Las esterasas tienen una cinética de tipo
Michaelis-Menten clásica, mientras que las lipasas únicamente se activan cuando se
encuentran en una interfase hidrófoba/hidrófila y siguen una cinética de Michaelis-
Menten interfacial.
17

Introducción
En general las enzimas lipolíticas son específicas de las posiciones exteriores (sn-1 y
sn-3) del triglicérido. Inicialmente los triglicéridos son hidrolizados a 1,2- y 2,3-
diglicéridos y después a 2-monoglicéridos. Las lipasas del queso pueden proceder de la
leche, de algunos preparados de cuajo, del cultivo iniciador, de cultivos adjuntos, de la
microbiota secundaria o bien pueden ser lipasas exógenas añadidas.
Lipasas
CH3
O
C OH O
O CH3
O
H3C C OH
C
O CH3  ó hidroxiácidos
O
C
O
H2O
Triglicéridos
O
Diglicéridos
Monoglicéridos H3C C
Glicerol O
 ó lactonas
O
H3C C
Ácidos grasos OH
R SH -oxidación
parcial
Tioles CH3CH2OH
Etanol co2
O
O O H3C C CH3
H3C C Metilcetonas
H3C C
SR O CH2 CH3
Tioésteres Etilésteres
OH
H3C CH CH3
2-Alcoholes
Figura 7. Esquema de la lipolisis y del catabolismo de ácidos grasos libres. (Collins et al., 2003).
La leche contiene una potente lipasa endógena, la lipoproteína lipasa (LPL) que
normalmente nunca alcanza su máxima actividad en la leche. La LPL en la leche está
asociada con las micelas de caseína y la membrana lipoproteica de los glóbulos de
18

Introducción
grasa. Si esta membrana se rompe por agitación, homogeneización o un mal manejo de

la leche, puede tener lugar una excesiva lipolisis que da lugar a la aparición de sabores
defectuosos. Esta enzima no tiene una gran especificidad en cuanto al tipo de ácido
graso, pero tiene mayor tendencia por la hidrólisis de los ácidos grasos situados en las
posiciones sn-1 y sn-3 de los mono-, di- y triglicéridos donde suelen encontrarse ácidos
grasos de cadena corta o media (Collins et al., 2003). La LPL es más activa en quesos de
leche cruda que en quesos de leche pasteurizada. Un tratamiento de 72 ºC durante 15
segundos provoca una considerable inactivación de la LPL, pero puede seguir
contribuyendo a la lipolisis en algunos quesos de leche pasteurizada ya que se requiere
un tratamiento de 78 ºC durante 10 segundos para su completa inactivación (Deeth,
2006).
Normalmente los cuajos comerciales no tienen actividad lipolítica, pero algunas

pastas de cuajo empleadas en la elaboración de algunas variedades de quesos, como
los de pasta dura italianos, contienen esterasa pregástrica. Esta esterasa es altamente
específica para la esterificación de ácidos grasos de cadena corta situados en la
posición sn-3.
Otra fuente de enzimas lipolíticas son los microorganismos, que pueden proceder
de cultivos iniciadores, cultivos adjuntos o ser microorganismos contaminantes. Las
lipasas y esterasas de las bacterias lácticas son los principales agentes lipolíticos de gran
variedad de quesos elaborados con leche pasteurizada. Las bacterias lácticas poseen
enzimas capaces de hidrolizar un amplio rango de ésteres de ácidos grasos de los
mono-, di- y triglicéridos. Las bacterias lácticas son muy poco lipolíticas en comparación
con otros microorganismos, aunque su presencia en gran número en el queso durante
un largo periodo de maduración puede hacer que se liberen niveles significativos de
ácidos grasos. Las lipasas y esterasas de las bacterias lácticas son en general
intracelulares, por lo que deben ser liberadas mediante la lisis de las células para ejercer
su actividad, aunque en algún caso se han encontrado esterasas asociadas a la
superficie celular (Gobbetti et al., 1997). Las bacterias propiónicas tienen una actividad
lipolítica de 10 a 100 veces mayor que las bacterias lácticas. En los quesos con mohos,
se ha visto que los del género Penicillium tienen una gran actividad lipolítica. Las
19

Introducción
lipasas y esterasas procedentes de microorganismos son muy diferentes y dependen de

la especie e incluso de la cepa.
Las lipasas exógenas pueden ser de origen animal o microbiano. Estas enzimas se
emplean de forma comercial en algunos quesos de pasta dura italianos y de forma
experimental en otras variedades, para conseguir alcanzar sabores característicos y
acelerar la maduración (Hernández et al., 2009).
Los principales ácidos grasos libres procedentes de la lipolisis se pueden clasificar

según el tamaño de la cadena como de cadena corta: butírico (C4:0), caproico (C6:0) y
caprílico (C8:0); de cadena media: cáprico (C10:0), laúrico (C12:0) y mirístico (C14:0); y de
cadena larga: palmítico (C16:0), esteárico (C18:0), oleico (C18:1), linoleico (C18:2) y linolénico
(C18:3). También se encuentran ácidos carboxílicos que no provienen de la lipolisis como
el acético y el propiónico y ácidos ramificados como el isobutírico y el isovalérico. Los
ácidos grasos libres pueden contribuir directamente al sabor del queso, especialmente
los de cadena corta y cadena media. Así el C4:0 aporta sabor rancio y típico de queso, el
C6:0 aporta sabor picante y típico de queso azul y el C8:0 da sabor a cera, jabón, cabra y
moho, pero fundamentalmente son importantes sustratos de reacciones que dan lugar
a compuestos responsables del aroma y sabor del queso (Curioni & Bosset, 2002).
Los ácidos grasos libres formados durante la maduración del queso pueden dar
lugar a cetonas, ésteres, lactonas, aldehídos y alcoholes (Figura 7). La formación de
metilcetonas se produce mediante la β-oxidación del ácido graso libre y posterior
descarboxilación del correspondiente cetoácido. Por sucesivas reacciones de β-
oxidación se pueden obtener metilcetonas de cadena corta a partir de ácidos grasos
libres de cadena larga. Estas metilcetonas procedentes del catabolismo de ácidos
grasos libres son muy típicas de quesos madurados con mohos, especialmente la 2-
heptanona y la 2-nonanona que contribuyen al aroma característico de queso azul y
son las principales metilcetonas del queso Camembert. Las metilcetonas también son
importantes en variedades de queso no maduradas con mohos. Las metilcetonas
pueden a su vez ser reducidas, dando los correspondientes alcoholes secundarios que
aportan al queso aromas a fresco, hierba y afrutado (Kinsella & Hwang, 1976, Qian et
al., 2002). Los ésteres son compuestos que forman parte del aroma del queso y se
forman por reacción de un ácido y un alcohol, por esterificación química o enzimática.
20

Introducción
Esta última puede ser llevada a cabo por la enzima carboxilesterasa, que está presente
en la mayoría de los microorganismos implicados en la elaboración de queso. Los
ésteres se encuentran en gran variedad de quesos y contribuyen aportando aromas
afrutados, florales y dulces (Curioni & Bosset, 2002). Las lactonas aportan al queso
aromas de melocotón, coco y albaricoque y se forman a partir de los hidroxiácidos de la
grasa. Para las δ-lactonas y las γ-lactonas de cadena larga se ha propuesto un
mecanismo de formación directa en un solo paso del hidroxiácido que se encuentra
esterificado en el triglicérido, mecanismo que estaría únicamente influido por la
temperatura (Alewijn et al., 2007). Otro mecanismo para la formación de lactonas podría
ser la hidrólisis del hidroxiácido y posterior esterificación cíclica, dependiente de la
presencia de microorganismos y de la temperatura (Rehman et al., 2000, Curioni &
Bosset, 2002). Los aldehídos lineales pueden formarse por β-oxidación de ácidos grasos
libres insaturados y se caracterizan por aportar al queso aromas frescos y herbales, pero
cuando su concentración supera cierto límite pueden dar lugar a aromas desagradables
(Curioni & Bosset, 2002, Collins et al., 2003).
1.1.2.3.3. Proteolisis y catabolismo de los aminoácidos
La proteolisis es el fenómeno bioquímico más complejo y en la mayoría de

variedades el más importante que tiene lugar durante la maduración del queso,
influyendo tanto en el sabor como en la textura. Los péptidos de pequeño tamaño y los
aminoácidos liberados pueden contribuir directamente al sabor. Además, estos últimos
sirven como sustrato de reacciones en las que se forman nuevos compuestos
responsables del aroma y sabor. La proteolisis contribuye a la textura mediante la
hidrólisis de la matriz de proteínas del queso, disminuyendo la actividad de agua e
indirectamente aumentando el pH (Upadhyay et al., 2004). Durante la proteolisis se
produce la hidrólisis de las caseínas y de los péptidos de tamaño elevado, dando lugar a
péptidos de tamaño intermedio (proteolisis primaria), que a su vez son hidrolizados
para dar péptidos pequeños y en última instancia aminoácidos libres (proteolisis
secundaria). Este proceso se debe a la acción de enzimas (proteasas y peptidasas) que
pueden proceder de la leche, del cuajo residual y de microorganismos (bacterias lácticas
del cultivo iniciador, cultivos adjuntos y otras bacterias presentes en la leche), además
de enzimas exógenas que pueden ser añadidas para acelerar la maduración.
21

Introducción
Caseínas
Proteasas
Péptidos de alto peso

molecular
Proteasas y
peptidasas
Péptidos de bajo peso

molecular
Peptidasas
NH3 CO2
Desaminación Descarboxilasas
Cetoácidos Aminoácidos Aminas
oxidativa
as
as
Cetoácido 1
in
m Aminoácido
as
sa
as
an
Li
Tr
Cetoácido 2
2-
-CADH
Tirosina Triptófano Metionina H

AD
DC
H
CA

Acetil-CoA Hidroxiácido
Fenol Indol Metanotiol Aldehído O
xi
da
d.
ci
Re
ón
Alcohol Ácido
carboxílico
Éster
Tioéster
Figura 8. Esquema de la proteolisis y del catabolismo de aminoácidos (Yvon & Rijnen, 2001,
McSweeney, 2004). (α-CADC: α-cetoácido descarboxilasa. α–CADH: α-cetoácido deshidrogenasa.
2-HADH: 2-hidroxiácido deshidrogenasa).
De las numerosas enzimas de la leche, la plasmina es la más importante desde el

punto de vista de la proteolisis. Aunque tratamientos térmicos intensos pueden
provocar la inactivación de la plasmina, con tratamientos HTST o UHT la plasmina
mantiene entre un 20 y un 40 % de su actividad (Ismail & Nielsen, 2010). La plasmina es
una proteinasa sérica alcalina asociada a la micela de caseína, con un pH y temperatura
22

Introducción
óptimos de 7,5 y 37 ºC. La plasmina actúa preferentemente sobre la β-CN, la cual posee
de 15 a 17 enlaces (dependiendo de la variante) susceptibles de ser hidrolizados, pero
únicamente 3 enlaces son hidrolizados de manera significativa: Lys28-Lys29, Lys105-His106
y Lys107-Glu108. La αS2-CN es también un buen sustrato para la plasmina. En tampón la
plasmina puede hidrolizar la αS2-CN hasta en 14 enlaces distintos, 12 de los cuales son
enlaces Lys-X y 2 son Arg-X. Sin embargo, no está bien establecido qué péptidos
resultan de la hidrólisis de la αS2-CN por plasmina en leche o queso debido a la gran
variabilidad en el grado de fosforilación (Larsson et al., 2006). La αS1-CN es menos
susceptible de ser hidrolizada por la plasmina que la α S2- y la β-CN y se han identificado
12 enlaces Lys-X y 5 Arg-X susceptibles de ser hidrolizados por la plasmina, dando
como resultado las λ-CN. La κ-CN y las proteínas séricas son muy resistentes a la
hidrólisis por plasmina.
Tras el desuerado parte del cuajo queda retenido en la cuajada jugando un

importante papel durante la proteolisis inicial de las caseínas. La β-CN en tampón es
hidrolizada en 7 enlaces distintos por la quimosina, mientras que en queso esta
hidrólisis es muy limitada y mayoritariamente afecta al enlace Leu192-Tyr193. De todos los
enlaces hidrolizados de la αS1-CN por la quimosina, en queso sólo tres sufren esta
hidrólisis: el Phe23-Phe24 (el más susceptible de ser atacado), Leu98-Leu99 y Leu101-Lys102.
La αS2-CN parece ser bastante resistente a la quimosina, mientras que la para-κ-CN a
pesar de tener varios enlaces susceptibles de ser hidrolizados se ha visto que es muy
resistente (Exterkate et al., 1997, Upadhyay et al., 2004).
Las cardosinas se caracterizan por ser más proteolíticas que la quimosina. Estas
enzimas vegetales tienen una actividad más inespecífica, hidrolizando más enlaces
peptídicos de las caseínas y dando lugar a mayor cantidad de péptidos hidrófobos de
sabor amargo (Roseiro et al., 2003).
Otra fuente de enzimas proteolíticas son los microorganismos presentes en el

queso. Sus enzimas pueden ser intracelulares y extracelulares. Las extracelulares están
asociadas a la pared celular y pueden considerarse como enzimas inmovilizadas (Visser,
1993), mientras que las intracelulares necesitan ser liberadas a la matriz del queso por
medio de la lisis celular. Las bacterias lácticas del cultivo iniciador poseen una
proteinasa asociada a la pared celular (PrtP), oligoendopeptidasas intracelulares (PepO y
23

Introducción
PepF), al menos tres aminopeptidasas generales (PepN, PepC y PepG), glutamil

aminopeptidasa (PepA), pirrolidona carboxilil aminopeptidasa (PCP), leucil
aminopeptidasa (PepL), prolildipeptidil aminopeptidasa (PepX), prolin aminopeptidasa
(PepI), aminopeptidasa P (PepP), prolinasa (PepR), prolidasa (PepQ), dipeptidasas
generales (PepV, PepD, PepDA) y tripeptidasas generales (PepT), además de un sistema
de transporte de péptidos y aminoácidos. Todo este sistema proteolítico es necesario
para que las bacterias lácticas alcancen poblaciones elevadas en la leche. En quesos de
leche cruda se encuentran además bacterias lácticas no pertenecientes al cultivo
iniciador. Estas bacterias lácticas aportan al queso su actividad proteolítica
complementando la de las bacterias lácticas del cultivo iniciador (Sousa et al., 2001,
Savijoki et al., 2006). En algunas variedades de queso se añaden también cultivos
adjuntos que aportan a los quesos determinadas características. Los microorganismos
adjuntos principales son P. roqueforti en quesos azules (Roquefort, Gorgonzola), P.
camemberti en quesos madurados con mohos en su superficie (Camembert, Brie),
Brevibacterium linens y otros microorganismos corineformes y distintas especies de
levaduras en quesos de corteza lavada (Munster, Limburger) y Propionibacterium
freudenreichii subsp. shermanii en los quesos de tipo Suizo. Cada microorganismo
aporta al queso sus enzimas, favoreciendo la proteolisis. De B. linens se ha aislado una
proteinasa extracelular y una aminopeptidasa intracelular. En los quesos de tipo Suizo,
las cepas de Propionibacterium spp. son muy poco proteolíticas aunque tienen una
gran actividad peptidolítica (Sousa et al., 2001). Los mohos del género Penicillium se
caracterizan por tener dos proteasas extracelulares, de las cuales una es una
metaloproteasa y la otra es una proteasa aspártica. Estos mohos también presentan
exopeptidasas extracelulares así como metaloaminopeptidasas extracelulares y
peptidasas intracelulares. Son considerados como muy proteolíticos debido
principalmente a la acción de la proteasa aspártica y de la metaloproteasa (Cantor et al.,
2004, Spinnler & Gripon, 2004).
Los aminoácidos liberados durante la proteolisis pasan a ser sustratos de

reacciones de formación de compuestos responsables del sabor y aroma del queso
(Figura 8). De forma general se considera que estas reacciones pueden darse
principalmente a través de dos rutas diferentes. Una de las rutas se inicia mediante una
24

Introducción
eliminación catalizada por liasas que cortan la cadena lateral de los aminoácidos. Esta
ruta se ha observado en los aminoácidos aromáticos (tirosina y triptófano) y en la
metionina, formándose en un solo paso fenol, indol y metanotiol respectivamente. La
otra ruta empieza principalmente con una transaminación catalizada por transaminasas
y se ha observado en aminoácidos aromáticos, de cadena ramificada y en metionina.
Las transaminasas están ampliamente distribuidas entre los microorganismos y

emplean el piridoxal-5’-fosfato como cofactor. Estas enzimas transfieren el grupo amino
de un aminoácido a un α-cetoácido (normalmente ácido α-cetoglutárico) formándose el
α-cetoácido correspondiente al aminoácido y un nuevo aminoácido correspondiente al
α-cetoácido original. Estos α-cetoácidos son productos intermedios de la ruta de la
aminotransferasa y pueden dar lugar a hidroxiácidos, aldehídos, alcoholes, ácidos
carboxílicos y metanotiol (Yvon & Rijnen, 2001, Smit et al., 2005). La reducción de los α-
cetoácidos a hidroxiácidos se ha observado en muchas bacterias lácticas. Se han
identificado varias 2-hidroxiácido deshidrogenasas, siendo la más conocida la lactato
deshidrogenasa cuya especificidad está principalmente restringida al piruvato. Otro
grupo de enzimas son las llamadas hidroxiisocaproato-deshidrogenasas, debido a que
tienen preferencia por el α-cetoisocaproato aunque catalizan otros sustratos empleando
NADH como donante de hidrógeno. La descarboxilación de los α-cetoácidos da lugar a
aldehídos que aportan a los quesos aromas a malta y chocolate. Esta transformación se
produce principalmente por la acción de levaduras, aunque se ha visto también en
algunas bacterias. La mayoría de las cepas de lactococos producen pequeñas
cantidades de aldehídos, lo que sugiere que tienen poca actividad o que estos
aldehídos son transformados rápidamente por reducción a alcoholes o por oxidación a
ácidos carboxílicos (Yvon & Rijnen, 2001, Smit et al., 2005). La reacción entre ácido y
alcohol da como resultado la formación de ésteres, que aportan al queso notas dulces,
afrutadas y florales. La descarboxilación oxidativa de los α-cetoácidos da como
resultado la formación de ácidos carboxílicos. Esta reacción está catalizada por α-
cetoácido deshidrogenasas y genera acetil-CoA, que posteriormente es oxidado,
liberándose los ácidos carboxílicos. La formación de ácidos carboxílicos a partir de
aminoácidos no parece ser muy común, aunque algunas levaduras, bacterias lácticas y
propionibacterias los producen bajo ciertas condiciones (Yvon & Rijnen, 2001). Los α-
25

Introducción
cetoácidos de la fenilalanina y de la metionina pueden ser transformados químicamente

en benzaldehído y metiltioacetaldehído, respectivamente. El 2-metilpropanal puede
formarse mediante una reacción química a partir del α-cetoácido de la leucina. Otra
importante reacción química que sufren los aminoácidos es la degradación de Strecker,
que da como producto 3-metilbutanal cuando el sustrato es la leucina. Otras rutas del
catabolismo de aminoácidos son la desaminación oxidativa del aminoácido,
formándose amoniaco y un α-cetoácido, y la descarboxilación, que implica la conversión
de un aminoácido en su correspondiente amina biógena liberándose CO2. Las aminas
biógenas generalmente tienen aromas fuertes y desagradables, y son además
compuestos con actividad fisiológica que pueden tener efectos tóxicos (Silla Santos,
1996, Curtin & McSweeney, 2004) como posteriormente se describe en la sección 1.3.
1.1.2.3.4. Sobremaduración
La calidad de los quesos viene determinada por su sabor, aroma, textura y

apariencia visual. El aroma y el sabor característico de cada variedad de queso se debe
al correcto equilibrio y concentración de compuestos responsables del aroma y sabor
(Mulder, 1952). Estos compuestos pueden ser volátiles o no volátiles. Los volátiles
contribuyen principalmente al aroma, pero también influyen en el sabor y los no
volátiles contribuyen de forma significativa al sabor. Durante la maduración se forman
estos compuestos debido a la glicolisis, lipolisis, proteolisis y catabolismo del lactato,
citrato, aminoácidos y ácidos grasos.
La sobremaduración es un fenómeno producido por un exceso de cualquiera de los

procesos que tienen lugar durante la maduración, especialmente la proteolisis,
provocando un desequilibrio en la composición de los compuestos responsables del
aroma y sabor. Este desequilibrio de compuestos provoca que las cualidades del queso
no sean las adecuadas, causando una menor aceptación por parte del consumidor y las
consecuentes pérdidas económicas del productor (Wick et al., 2004). Algunas
variedades de queso experimentan una intensa proteolisis debido al empleo de enzimas
coagulantes de elevada actividad proteolítica, tales como las proteasas extraídas del
cardo (Cynara cardunculus) empleadas en la elaboración de algunas variedades de
queso como la Torta del Casar o el queso de la Serena. En otros quesos, como el azul o
el Brie, los responsables de la elevada proteolisis son los mohos pertenecientes al
26

Introducción
género Penicillium que poseen proteasas y peptidasas de elevada actividad. Una

elevada proteolisis también puede provocar defectos de textura debido a que la
hidrólisis de las caseínas durante el almacenamiento disminuye las interacciones
proteína-proteína provocando una disminución de la firmeza. En la práctica, el tiempo
de maduración provoca un aumento de la firmeza debido fundamentalmente a la
pérdida de humedad, por lo que el reblandecimiento del queso debido a la
sobremaduración sólo tendría lugar en quesos envasados de forma que conserven la
humedad (Lucey et al., 2003). Una excesiva lipolisis también puede provocar
sobremaduración debido a la liberación de altos niveles de ácidos grasos. Apenas se
han realizado trabajos de investigación para controlar este problema más allá del
empleo de temperaturas de refrigeración para reducir la actividad enzimática. Durante
la maduración en la industria existe un control efectivo de la temperatura, ya que un
aumento de temperatura durante esta fase provocaría una maduración acelerada que
reduciría la vida útil del producto. La sobremaduración se puede dar durante la
conservación del producto en establecimientos comerciales y hogares, donde el control
de la temperatura no siempre es el adecuado. El fenómeno de la sobremaduración
limita la vida útil del queso y en algunos casos hace que dentro de la vida útil
establecida por el fabricante, el queso llegue al consumidor con peor calidad que la
deseada por el fabricante (Wick et al., 2004).
1.2. Tipos de quesos
En la actualidad existe una gran diversidad de quesos, habiéndose sugerido la

existencia de más de 1000 variedades. En la Universidad de Wisconsin se ha realizado
una lista con 1400 variedades de queso (www.cdr.wisc.edu). A nivel de divulgación, en la
página web www.mundoquesos.com se encuentran descritas 2420 variedades de
quesos. Muchas de estas variedades son muy similares, por lo que podrían considerarse
variantes en vez de variedades. La producción de queso tiene una larga historia y
tradición, lo que se refleja en la gran variedad de quesos existente.
Para proteger y preservar esta diversidad tradicional, en 1883 se empezó a plantear

el concepto de denominación de origen protegida (DOP). Esta denominación se da a los
quesos que son elaborados con una materia prima determinada, en una zona
27

Introducción
geográfica delimitada y mediante el empleo de unas técnicas dadas (McSweeney et al.,

2004). Se han realizado diferentes intentos de clasificación de quesos, basándose en la
textura (extraduro, duro, semiduro, semiblando y blando), método de coagulación
(coagulación enzimática, ácida o mixta), especie de procedencia de la leche (vaca, oveja,
cabra…), tratamiento aplicado a la leche (cruda, pasteurizada, termizada o microfiltrada),
tiempo de maduración (tierno, semicurado, curado, viejo y añejo), contenido graso
(extragraso, graso, semigraso y desnatado) y en los cultivos iniciadores y secundarios
empleados (madurados por mohos en el interior, por mohos en superficie, de corteza
lavada…).
A continuación se describen las características principales de los quesos objeto del

estudio realizado en la presente Tesis. La elección de estos quesos se ha realizado en
base al alto grado de proteolisis y al riesgo de formación de aminas biógenas. Entre los
quesos elegidos se encuentra uno de leche de oveja pasteurizada con mohos en su
interior (queso azul, Roncari-blue), uno de leche cruda de oveja coagulado con cuajo
vegetal (Torta del Casar), uno de leche de vaca pasteurizada con mohos en su superficie
(queso Brie) y uno de leche cruda de vaca coagulado con cuajo animal (queso Arzúa-
Ulloa).
1.2.1. Queso azul
El queso azul se caracteriza por el

crecimiento del moho Penicillium
roqueforti en su interior, que le confiere
su apariencia, aroma y sabor típicos.
Muchos países fabrican sus propios tipos
de queso azul, cada uno con sus
diferentes características. Pueden
elaborarse con distintos tipos de leche de
vaca, oveja, cabra o mezcla, cruda o
pasteurizada. Los quesos azules son en
Figura 9. Corte de queso azul (Roncari-blue).
general muy heterogéneos, con
pronunciados gradientes de pH, sal y actividad de agua (aw). Presentan un alto grado de
lipolisis y de proteolisis en comparación con otros quesos. La maduración tiene lugar en
28

Introducción
su mayor parte a temperaturas entre 6 y 12 ºC y puede durar de 3 a 6 meses. El pH de

los quesos azules puede abarcar desde 4,6 hasta 5,3, dependiendo de la variedad. La
conversión de la lactosa en ácido láctico se produce por la acción de las bacterias
lácticas del cultivo iniciador. Durante la maduración el pH puede aumentar hasta 6,5 en
el centro y hasta 5,9 en la corteza. El pH del interior aumenta más rápidamente que en
la corteza debido a que el nivel de NaCl es menor, lo cual favorece el crecimiento de P.
roqueforti. Este aumento de pH se debe al metabolismo del ácido láctico a CO2 por
mohos y levaduras y a la elevada proteolisis seguida de la producción de amoniaco por
desaminación de los aminoácidos. El salado se puede hacer por inmersión en salmuera
o por aplicación de sal seca sobre la superficie. En ambos casos se produce un
gradiente de NaCl de la superficie al interior, que se va equilibrando lentamente
durante la maduración. El crecimiento de P. roqueforti está muy influenciado por la
composición gaseosa del queso, ya que la concentración de O2 disminuye rápidamente
generando un ambiente anaerobio en casi todo el queso excepto en las fisuras y
canales de penetración. La especie P. roqueforti está particularmente bien adaptada
para crecer en este ambiente de baja concentración de O2 combinado con una alta
concentración de CO2.
La lipolisis en los quesos azules es muy intensa. En otras variedades una lipolisis
excesiva puede causar la aparición de rancidez, pero en los quesos azules los ácidos
grasos libres son neutralizados cuando aumenta el pH. Por lo general, el nivel de ácidos
grasos libres totales aumenta con la maduración, especialmente después de la
esporulación de los mohos, aunque se ha observado en ocasiones un descenso al final
de la maduración (Prieto et al., 2000). Este descenso puede deberse a la conversión de
los ácidos grasos en metilcetonas. Normalmente los niveles de ácidos grasos de
tamaño entre C12:0 – C18:3 en los quesos azules son más elevados que los de tamaño C4:0
– C10:0 (Larsen & Jensen, 1999). La degradación de lípidos en los quesos azules es
debida principalmente a las enzimas de P. roqueforti que produce dos lipasas
extracelulares, una ácida y otra alcalina. El pH óptimo de la lipasa ácida es 6 pero
mantiene una estabilidad máxima a pH comprendido entre 3,7 y 6,0. Su temperatura
óptima es de 35-40 ºC pero mantiene un 37 % de su actividad máxima a 5 ºC. El pH
óptimo para la lipasa alcalina es de 8,8-9,0 a 30 ºC, y de 9,0-10,0 a 20 ºC, pero sigue
29

Introducción
teniendo una considerable actividad a niveles de pH entre 4,5 y 11,0. Aunque P.

roqueforti domina por lo general en la degradación de los lípidos, existen otros agentes
lipolíticos. La actividad lipolítica de las bacterias lácticas es generalmente baja y no tiene
mucha influencia en los quesos azules pero también puede influir en la lipolisis la
presencia de levaduras, ya que las levaduras que aparecen en quesos azules tienen al
menos actividad esterasa y pueden hidrolizar los ácidos grasos de cadena corta de los
triglicéridos.
La proteolisis en los quesos azules también es muy intensa en comparación con

otros tipos de quesos. Las caseínas son hidrolizadas en más enlaces y en mayor
proporción, por lo que apenas quedan caseínas intactas al final de la maduración. Se
forma mayor número de péptidos diferentes que en los quesos de pasta semidura y se
libera una mayor cantidad de aminoácidos como resultado de las peptidasas de P.
roqueforti y de las levaduras y en menor grado de las bacterias lácticas. Inicialmente la
proteolisis es debida, al igual que en otros quesos, al cuajo residual, plasmina y enzimas
del cultivo iniciador. Después de un tiempo la proteolisis es dominada por P. roqueforti,
que produce dos proteasas extracelulares: una metaloproteasa y una proteasa aspártica.
La actividad de estas dos enzimas es máxima en el momento en que el moho empieza a
esporular. La metaloproteasa es activa a pH entre 4,5 y 8,5 y su pH óptimo para la
hidrólisis de caseínas es de 5,5. La metaloproteasa tiene una especificidad muy amplia,
pudiendo hidrolizar la αS1-CN y la β-CN en tampón dando lugar a 8 y 9 péptidos
diferentes, respectivamente. Además la metaloproteasa se caracteriza por hidrolizar la
β-CN en el enlace Pro90-Glu91, lo cual no es muy común. La proteasa aspártica es
estable a pH comprendido entre 3,5 y 6,0, y presenta dos valores de pH óptimos para la
hidrólisis de caseínas que son 3,5 y 5,5. La proteasa aspártica, al igual que la quimosina,
tiene preferencia por el enlace Phe23-Phe24 de la αS1-CN, y posteriormente el péptido C-
terminal formado es hidrolizado para dar 4 ó 5 nuevos péptidos. La β-CN puede ser
hidrolizada por esta enzima para dar 5 péptidos diferentes. Se ha establecido una
relación entre el desarrollo de P. roqueforti, la actividad de la proteasa aspártica y la
mayor liberación de péptidos amargos. Penicillium roqueforti produce una
metaloaminopeptidasa alcalina extracelular con un pH óptimo de 8,0 que es específica
de los aminoácidos hidrófobos y por tanto su actividad reduce el amargor. Además este
30

Introducción
moho posee varias exopeptidasas, una carboxipeptidasa ácida extracelular con amplia
especificidad que libera aminoácidos básicos, ácidos e hidrófobos y que también
contribuye a la reducción del amargor. Penicillium roqueforti tiene también gran
número de peptidasas intracelulares que pueden ser alcalinas, carboxi- y
aminopeptidasas. La actividad proteolítica así como el nivel de proteasas y peptidasas
varía según la cepa de P. roqueforti empleada. En los quesos azules el catabolismo de
los aminoácidos es llevado a cabo por las bacterias lácticas y los mohos. Los
aminoácidos más comunes en este tipo de quesos son el ácido glutamico, leucina,
valina y lisina. La desaminación oxidativa de los aminoácidos es debida a P. roqueforti
en el interior del queso y a otros microorganismos en la superficie. Esta actividad
produce amoniaco, que contribuye al sabor del queso azul. La concentración de aminas
varía ampliamente, apareciendo por lo general la tiramina en mayor concentración que
la triptamina y la histamina (Cantor et al., 2004).
En los quesos azules se producen, principalmente por P. roqueforti, un gran

número de compuestos volátiles y no volátiles que influyen en el sabor y aroma. El
sabor del queso azul deriva principalmente de la degradación de los lípidos. Los ácidos
grasos libres dan sabor y aroma al queso y son a su vez transformados a otros
compuestos. Los ácidos (C4:0-C12:0) tienen notas de sabor rancio y picante y suelen
encontrarse en los quesos azules en altas concentraciones. El alto pH neutraliza estos
ácidos haciendo que estos aporten su sabor pero sin que se dé el defecto de rancidez.
Las metilcetonas son los compuestos mayoritarios en los quesos azules y su
concentración en el queso está relacionada con la intensidad de sabor a “queso azul”.
Las metilcetonas mayoritarias en estos quesos son la 2-heptanona y la 2-nonanona,
aunque también son importantes la 2-pentanona y la 2-undecanona. En general las
metilcetonas aportan al queso aromas florales, afrutados y a moho. Estos compuestos
pueden ser reducidos a alcoholes secundarios en condiciones anaerobias. Los
principales alcoholes secundarios son 2-heptanol, 2-nonanol y 2-pentanol. Sus sabores
son similares a los aportados por las correspondientes metilcetonas, pero en alta
concentración pueden dar un exceso de sabor a moho. Otros compuestos que aportan
aromas afrutados y florales son los ésteres y las lactonas. Los aldehídos suelen aparecer
en baja concentración y no se conoce bien su impacto en el sabor de estos quesos.
31

Introducción
1.2.2. Torta del Casar
La Torta del Casar es un queso de

pasta blanda, amparado bajo la
denominación de origen protegida (DOP)
de la Torta del Casar aprobada por el
reglamento de la Comisión Europea (EC
1491/2003). Este queso se caracteriza por
el empleo en su elaboración de cuajo
vegetal natural, obtenido de las flores
desecadas del cardo (Cynara cardunculus),
Figura 10. Cuña de Torta del Casar.
sin adición de ningún cultivo iniciador. Se
elabora en 36 términos municipales de la provincia de Cáceres, a partir de leche cruda
de ovejas de raza merina o entrefina. El periodo mínimo de maduración es de 60 días y
debe realizarse a temperaturas comprendidas entre 4 y 12 ºC, y entre 75 y 90 % de
humedad relativa. El empleo de cuajo vegetal hace que este queso tenga un sabor
ligeramente amargo y una textura entre blanda y untuosa. El pH final de estos quesos
debe estar comprendido entre 5,2 y 5,9. El salado se puede realizar mediante inmersión
en salmuera o por aplicación de sal seca sobre la superficie (BOE, 2002). El empleo de
leche cruda sin adición de un cultivo iniciador hace que los microorganismos presentes
en la leche jueguen un importante papel en el proceso de maduración de este queso.
Existe una gran variabilidad en el producto final, provocado por las diferencias de
microorganismos presentes en la leche cruda. Además de bacterias lácticas, se
encuentran en estos quesos enterococos, coliformes y micrococos así como levaduras y
mohos (Ordiales et al., 2013a). Debido al empleo de leche cruda también pueden
encontrarse microorganismos patógenos como Listeria, Salmonella o Staphylococcus
aureus, que normalmente desaparecen al cabo de los 60 días de maduración.
Los agentes responsables de la lipolisis en estos quesos son la LPL endógena de la

leche y las lipasas y esterasas de los microorganismos presentes en la leche. La LPL no
tiene una gran especificidad en cuanto al tipo de ácido graso, pero tiene mayor
tendencia por la hidrólisis de los ácidos grasos situados en las posiciones sn-1 y sn-3 de
los mono-, di- y triglicéridos donde suelen encontrarse ácidos grasos de cadena corta o
32

Introducción
media (Collins et al., 2003). La diversidad de microorganismos hace que el grado de

lipolisis en estos quesos sea muy variado. Por lo general las bacterias lácticas tienen
poca actividad lipolítica, pero al encontrarse en gran número pueden tener un efecto
considerable. Por otro lado, la aportación de otros microorganismos que se encuentren
en menor número puede ser importante debido a una mayor capacidad lipolítica. La
liberación de ácidos grasos contribuye al sabor y aroma del queso, por su presencia y
por ser sustratos de posteriores reacciones.
La proteolisis de esta variedad de queso viene dada por la acción de la plasmina, de

las enzimas de los microorganismos presentes y, principalmente, de las cardosinas del
cuajo retenido. El cuajo vegetal empleado en la elaboración de estos quesos se
caracteriza por tener menor actividad coagulante y mayor actividad proteolítica. Esta
actividad proteolítica es menos específica que la de la quimosina, pudiendo hidrolizar
αS1-CN, β-CN y al menos una de las γ-CN, dando mayor número de péptidos diferentes.
La actividad proteolítica del extracto de las flores del cardo depende de la variedad, fase
de maduración de la flor, parte de la flor, tiempo de secado y contenido final de
humedad. Este extracto contiene dos enzimas, cardosina A y cardosina B. La cardosina A
se asemeja en su actividad a la quimosina, mientras que la cardosina B se asemeja a la
pepsina (Verissimo et al., 1995). Estas proteinasas tienen mayor afinidad por las zonas
hidrófobas de las caseínas, provocando una mayor liberación de péptidos hidrófobos
que dan al queso su característico sabor amargo. La αS1-CN puede ser atacada en 9
enlaces distintos, siendo el enlace Phe23-Val24 el más susceptible. La αS2-CN se hidroliza
principalmente en los enlaces Phe88-Tyr89 y Ser9-Ser10, mientras que la β-CN lo hace en
los enlaces Leu127-Thr128 y Leu190-Tyr191.
Además de los compuestos volátiles más comunes en los quesos como los
alcoholes, cetonas, aldehídos y esteres, en la Torta del Casar cabe destacar la presencia
de los ácidos 2-metilpropanoico y 3-metilbutanoico, procedentes de la desaminación
de la valina y de la leucina respectivamente, y también la presencia de los ácidos acético
y propiónico procedentes de la actividad microbiana, siendo estos de mayor
importancia en el aroma de este tipo de quesos que los ácidos procedentes de la
lipolisis (Delgado et al., 2010, Ordiales et al., 2013b).
33

Introducción
1.2.3. Queso Brie
El queso Brie pertenece, al igual

que el Camembert, al grupo de
quesos madurados por mohos en
superficie. Se caracteriza por estar
recubierto de una capa blanca,
formada por el micelio del moho
Penicillium camemberti. La presencia
Figura 11. Esquema de migración y gradientes en
de este moho confiere al queso una
queso Brie.
apariencia visual y un aroma y sabor
característicos. Este queso se elabora con leche de vaca, que puede ser cruda o
pasteurizada. La maduración se realiza a temperaturas comprendidas entre 11 y 13 ºC y
en torno al 90 % de humedad relativa. Se añade a la leche un cultivo iniciador de
bacterias lácticas y P. camemberti y opcionalmente se pueden añadir como cultivos
adjuntos Geotrichum candidum y Kluyveromyces lactis (Spinnler & Gripon, 2004). Las
bacterias lácticas fermentan la lactosa, formándose ácido láctico y disminuyendo el pH.
El descenso del pH junto a la presencia de oxígeno permite al moho P. camemberti
colonizar la superficie degradando el lactato y produciendo CO2, con la consiguiente
elevación del pH en la superficie. Este aumento del pH superficial hace que se produzca
un gradiente de pH y una migración del lactato hacia la corteza. El pH de la superficie
puede aumentar hasta 7,0 al final de la maduración, mientras que en el interior el
aumento es menor. El elevado pH de la superficie permite que los microorganismos
sensibles al pH ácido se puedan establecer en la corteza y favorece la actividad de las
enzimas. Un efecto del gradiente de pH es la migración de calcio y fosfatos hacia la
superficie del queso, donde por efecto del elevado pH se forma fosfato cálcico
insoluble. Estos gradientes y la migración de minerales provocan cambios reológicos
que dan a este queso su textura característica.
La lipolisis en quesos Camembert y Brie es consecuencia principalmente de las

enzimas de P. camemberti, acompañado de las lipasas y esterasas de las bacterias
lácticas del cultivo iniciador. Penicillium camemberti produce grandes cantidades de
una lipasa extracelular alcalina, cuyo pH óptimo está en torno a 9,0 aunque mantiene un
34

Introducción
50 % de su actividad a pH 6,0. Esta enzima es el principal agente lipolítico en queso Brie

y es más activa en triglicéridos con ácidos de bajo peso molecular. Geotrichum
candidum sintetiza dos lipasas, una de las cuales libera preferentemente ácido oleico, y
la otra ácidos insaturados de 18 carbonos situados en la posición sn-2 del triglicérido.
Aunque estos quesos sufren una intensa proteolisis, esta no llega a ser tan intensa
como en los quesos azules. La proteolisis es producida por tres agentes, el cuajo, la
plasmina y las proteinasas de los microorganismos, siendo las de P. camemberti las más
activas. En estos quesos se ha observado una fuerte degradación de la αS1-CN en todo
el queso, mientras que la β-CN sólo se degrada fuertemente en la parte externa.
Penicillium camemberti tiene una gran actividad proteolítica debido a la producción de
endo- y exopeptidadasas extracelulares y sintetiza también grandes cantidades de una
metaloproteinasa y de una proteinasa aspártica. Durante la maduración esta actividad
es muy baja en el centro del queso, mientras que en la parte externa se produce un
gran aumento después de 6-7 días de maduración, cuando P. camemberti empieza a
crecer. La metaloproteinasa y la aspartil proteinasa alcanzan su concentración máxima
en el queso a los 15 días y después empiezan a descender. Debido a que el moho sólo
se encuentra en la superficie, el grado de proteolisis es mayor en esta zona que en el
interior. El moho produce grandes cantidades de aminoácidos debido a la síntesis de
exopeptidasas extracelulares. Por otro lado G. candidum sintetiza proteinasas intra- y
extracelulares, cuya producción depende mucho de la cepa. La actividad enzimática de
G. candidum y de las bacterias lácticas es mucho menor que la de P. camemberti,
siendo este último el principal responsable de la proteolisis.
Al igual que en queso azul, las metilcetonas juegan un papel importante en el

aroma de los quesos madurados por mohos en superficie. El principal agente en la
formación de las metilcetonas es un sistema enzimático de P. camemberti, que
mediante la β-oxidación de los ácidos grasos libres da lugar a la formación este tipo de
compuestos. El 1-octen-3-ol es un compuesto característico de este tipo de quesos, que
aporta aroma a champiñón crudo y puede llegar a representar entre el 5-10 % de los
compuestos volátiles en queso Camembert. Los ácidos grasos tienen también un
importante papel en estos quesos, sobre todo los de cadena media y corta. Al igual que
en el queso azul también se pueden encontrar lactonas, que aportan al queso notas
35

Introducción
afrutadas. También se pueden encontrar, aunque en menor concentración, otros

compuestos responsables del aroma tales como aldehídos, compuestos azufrados,
aminas, ésteres y terpenos.
1.2.4. Queso Arzúa-Ulloa
El queso Arzúa-Ulloa es un queso

gallego, elaborado con leche de vaca, que se
encuentra amparado bajo la DOP de Arzúa-
Ulloa aprobada por el reglamento de la
Comisión Europea (UE 20/2010). La zona de
producción comprende 32 municipios de las
provincias de A Coruña, Lugo y Pontevedra.
La leche empleada en la elaboración de este
queso debe ser de las razas frisona, pardo
alpina, rubia gallega y/o sus cruces. Puede Figura 12. Cuña de queso Arzúa-Ulloa.
emplearse cruda o pasteurizada. Para su coagulación se emplea extracto de cuajo

animal y fermentos lácticos. La cuajada obtenida debe ser lavada con agua, con el fin de
reducir la acidez. El salado de estos quesos puede realizarse sobre la cuajada
directamente en la cuba o introduciendo los quesos en una solución de salmuera. Los
quesos Arzúa-Ulloa se caracterizan por ser quesos de pasta semiblanda, de sabor suave
con aroma lácteo que recuerda a la mantequilla. Dentro de la DOP de Arzúa-Ulloa se
distinguen tres tipos de queso: Arzúa-Ulloa, Arzúa-Ulloa de granja y Arzúa-Ulloa
curado. El de granja se caracteriza por que la leche debe proceder de vacas que se
encuentran en la propia explotación en la que se elabora el queso. El periodo de
maduración debe ser como mínimo de 6 días, excepto en el queso curado que debe
ser como mínimo de 6 meses. La maduración se realiza a temperaturas inferiores a 15
ºC y entre un 75 y 90 % de humedad relativa. El pH de estos quesos debe de estar
comprendido entre 5,0 y 5,5.
En general se considera que los quesos de esta variedad elaborados con leche
cruda suelen tener un sabor más amargo y una mayor intensidad que los de leche
pasteurizada (Menéndez et al., 2000). El queso Arzúa-Ulloa de leche cruda tiene gran
variabilidad de sabor debido a las diferencias en la microbiota de la leche. La proteolisis
36

Introducción
y lipolisis en estos quesos depende del tipo de leche empleada. En los quesos de leche
pasteurizada son las enzimas del cultivo iniciador junto con las enzimas resistentes a la
pasteurización de la leche las responsables de la maduración. En los quesos de leche
cruda además de estas enzimas se encuentran las enzimas de los microorganismos
inicialmente presentes en la leche, como cepas de Enterococcus y Micrococcus que
presentan actividad proteolítica y lipolítica (Centeno et al., 1995).
Los quesos Arzúa-Ulloa de leche cruda, además de sufrir una mayor proteolisis y
lipolisis, tienen mayor número y concentración de compuestos volátiles relacionados
con el aroma y sabor que los elaborados con leche pasteurizada. Los compuestos más
abundantes suelen ser los alcoholes, seguidos de cetonas y ésteres en los quesos de
leche cruda y las cetonas y aldehídos en los de leche pasteurizada. El hecho de que los
ácidos grasos y los ésteres sólo se encuentren en los quesos de leche cruda y que la
abundancia de cetonas y alcoholes secundarios sea mayor, está en consonancia con la
lipolisis más intensa en estos quesos. Además, la mayor abundancia de aldehídos de
cadena ramificada y de alcoholes en los quesos de leche cruda indica que el
catabolismo de aminoácidos de cadena ramificada es mayor en estos quesos, como
consecuencia de la actividad de los microorganismos procedentes de la leche
(Rodríguez-Alonso et al., 2009).
1.3. Aminas biógenas
Las aminas biógenas son compuestos básicos nitrogenados de bajo peso

molecular, con actividad biológica. Su síntesis y degradación forma parte del
metabolismo normal de animales, plantas y microorganismos (ten Brink et al., 1990). Las
aminas (Tabla 3) pueden ser alifáticas (cadaverina, putrescina, espermina y espermidina),
aromáticas (tiramina y β-feniletilamina) o heterocíclicas (histamina y triptamina).
Se pueden clasificar según el número de grupos amino que contienen como

mono-, di- o poliaminas (Silla Santos, 1996). La putrescina, espermidina y espermina se
pueden considerar como un grupo diferenciado, debido a las funciones que
desempeñan en el organismo y a la participación de la putrescina en la síntesis de la
espermidina y espermina (Kalač, 2009).
37

Introducción
A pesar de formar parte del metabolismo normal y de regular diversas funciones

indispensables en el organismo, la ingesta de niveles elevados de aminas biógenas
puede tener efectos tóxicos, tales como aumento de la presión sanguínea, dolores de
cabeza, calambres abdominales y urticaria.
Tabla 3. Características de las principales aminas biógenas.

Aminoácido Amina
Fórmula Estructura química Categoría
precursor biógena
NH2
N Monoamina
Histidina Histamina C5H9N3
N heterocíclica
H
NH2 Monoamina
Triptófano Triptamina C10H12N2
heterocíclica
N
H
NH2
Monoamina
Tirosina Tiramina C8H11NO
aromática
HO
NH2
Monoamina
Fenilalanina β-feniletilamina C8H11N
aromática
Diamina
NH2
Ornitina Putrescina C4H12N2 H 2N
alifática
Diamina
Lisina Cadaverina C5H14N2 H2N NH2
alifática
Ornitina y NH NH2 Poliamina

H2N NH
Espermina C10H26N4
Arginina alifática
Ornitina y NH Poliamina
Espermidina C7H19N3 NH2
H2N
Arginina alifática
Las aminas biógenas pueden estar presentes como componentes endógenos de

alimentos frescos, como frutas y verduras. También pueden aparecer y acumularse
como resultado de la actividad enzimática microbiana, que en algunos casos se asocia a
técnicas de manipulación poco higiénicas (ten Brink et al., 1990). En los alimentos las
38

Introducción
aminas biógenas se forman principalmente por descarboxilación de aminoácidos

(Loizzo et al., 2013). Entre los alimentos con mayores niveles de aminas biógenas se
encuentran el pescado y productos derivados del pescado, productos lácteos, carne y
productos cárnicos, vegetales fermentados y bebidas alcohólicas como el vino y la
cerveza (Ladero et al., 2010).
1.3.1. Aminas biógenas en alimentos
La mayor incidencia de casos de intoxicación por aminas biógenas está asociada al

consumo de pescado y productos derivados (Silla Santos, 1996). En el pescado la amina
más abundante es la histamina, aunque también se han encontrado putrescina,
cadaverina, tiramina, espermina y espermidina, además de β-feniletilamina y triptamina
en bajas concentraciones. La familia de los Scombridae, familia a la que pertenecen el
atún y el bonito, es la más habitualmente implicada en casos de intoxicación por
histamina (escombroidosis). Estos pescados se caracterizan por tener una musculatura
rica en histidina libre, entre 1 g/kg en el arenque y 15 g/kg en el atún, que puede ser
convertida en histamina (Shalaby, 1996). Parte de la microbiota habitual del pescado
tiene la capacidad de producir histamina a bajas temperaturas. El contenido de aminas
biógenas en pescado se ha propuesto como índice de frescura, contaminación o
deterioro del producto (Visciano et al., 2012).
El queso es, tras el pescado y sus derivados, el alimento más frecuentemente

asociado a intoxicaciones por aminas biógenas y por lo tanto también uno de los más
estudiados (Stratton et al., 1991). Como ya se ha indicado, el resultado final de la
proteolisis es la formación de aminoácidos libres, que junto a la presencia de
microorganismos con actividad descarboxilasa puede llevar a la acumulación de
grandes cantidades de aminas biógenas (Silla Santos, 1996). Debido a la gran variedad
de quesos que existe, los niveles de aminas biógenas son muy variables. Un motivo de
la gran variabilidad de aminas biógenas en los quesos es el empleo de leche cruda o
pasteurizada, ya que la leche cruda puede contener microorganismos, tanto gram
positivos como gram negativos, con actividad descarboxilasa que son eliminados
mediante los tratamientos de pasteurización. Algunos cultivos iniciadores pueden
contener cepas de bacterias lácticas con actividad descarboxilasa (Burdychova &
Komprda, 2007). Otro factor que provoca variabilidad en los niveles de aminas en el
39

Introducción
queso es la diferencia en el tiempo de maduración cada variedad de queso: a mayor

tiempo de maduración mayor concentración de aminas biógenas se puede alcanzar
(Komprda et al., 2008a). Como se puede observar en la Tabla 4, la concentración de
aminas biógenas en queso puede abarcar desde cantidades traza hasta niveles de 2101
mg/kg de cadaverina, 1585 mg/kg de tiramina y 129 mg/kg de triptamina en queso
azul de leche pasteurizada (Novella-Rodríguez et al., 2003), 1042 mg/kg de histamina y
876 mg/kg de putrescina en queso azul de leche cruda (Fernández et al., 2007a), 202
mg/kg de espermidina y 40 mg/kg de espermina en queso Cheddar (Bardócz et al.,
1993) y 54,3 mg/kg de β-feniletilamina en queso de tipo holandés (Komprda et al.,
2007).
La carne y los productos cárnicos son un sustrato rico en aminoácidos libres. Las
aminas biógenas están presentes en la carne fresca de forma natural, pero su presencia
en altas concentraciones es un índice de incorrecta manipulación (Ruiz-Capillas &
Jiménez-Colmenero, 2004a). El procesado de la carne, cortado, loncheado o picado,
puede favorecer la contaminación por microorganismos con actividad descarboxilasa, lo
que junto con el almacenamiento a temperaturas poco adecuadas (> 5 ºC) o durante
largos periodos puede permitir la formación de aminas biógenas. El contenido en
aminas biógenas de la carne se ha propuesto como índice de calidad y frescura del
producto (Ruiz-Capillas & Jiménez-Colmenero, 2004a). Los productos cárnicos con
mayor concentración de aminas biógenas suelen ser los productos fermentados. La
variabilidad observada en este tipo de productos depende de la adición o no de
cultivos iniciadores de bacterias lácticas, del empleo de distintos ingredientes como sal
y aditivos, y de los tiempos y condiciones de fermentación (Shalaby, 1996). En los
productos cárnicos fermentados la amina biógena más abundante suele ser la tiramina,
pudiendo llegar a concentraciones de más de 600 mg/kg en chorizo (Hernández-Jover
et al., 1997). Además de tiramina se puede encontrar histamina, putrescina, cadaverina,
triptamina, β-feniletilamina, espermidina y espermina. La putrescina y la cadaverina
llegan a alcanzar niveles de hasta 415 mg/kg y 658 mg/kg respectivamente en chorizo,
debido probablemente a la contaminación con microorganismos gram negativos
(Hernández-Jover et al., 1997, Latorre-Moratalla et al., 2012).
40

Introducción
En frutas y verduras frescas, al igual que en otros alimentos frescos, las aminas
biógenas se suelen encontrar en concentraciones bajas. La presencia de altas
concentraciones está asociada a contaminación microbiana, largos periodos de
almacenamiento o mala manipulación (Moret et al., 2005). Los productos vegetales
fermentados tienen una gran variabilidad de concentraciones de aminas biógenas,
según el microorganismo implicado en la fermentación y el tiempo y condiciones de
fermentación. Se han observado niveles de hasta 951 y 529 mg/kg de tiramina y
putrescina respectivamente en chucrut (Kalač et al., 1999) y de 592, 550 y 486 mg/kg de
histamina, cadaverina y espermidina respectivamente en salsa de soja (Lu et al., 2009).
Las bebidas fermentadas como el vino y la cerveza pueden contener aminas

biógenas. Las aminas biógenas más abundantes en el vino son la histamina, tiramina y
putrescina. Su formación y acumulación depende de las condiciones de elaboración,
tiempo de maceración con la piel, duración y condiciones de la fermentación, y
crecimiento de bacterias y levaduras endógenas en el mosto (Gardini et al., 2005,
Beneduce et al., 2010). Las aminas se pueden formar durante la fermentación alcohólica
por acción de las levaduras, aunque sus niveles suelen ser bajos al finalizar esta fase. Sin
embargo, se puede producir un aumento considerable durante la fermentación
maloláctica por parte de las bacterias lácticas y durante el almacenamiento y añejado
(Lonvaud-Funel, 2001, Ancin-Azpilicueta et al., 2008). El rango de concentraciones de
aminas biógenas en vino puede abarcar desde niveles no detectables hasta valores de
47,3 mg/L de putrescina (Landete et al., 2005).
41

42
Tabla 4. Concentración (mg/kg) de aminas biógenas en diferentes alimentos.
Alimento Tiramina Histamina Triptamina Putrescina Cadaverina Feniletilamina Espermina Espermidina Referencia
Productos lácteos
Queso azul (LP) 0 - 1585 0 - 377 0 - 129 3 - 257 0 - 2101 0 - 40 0 - 19 0 - 72 (Novella-Rodríguez et
Introducción
Queso curado (LC) 0 - 609 0 - 391 0 - 34 0 - 670 1 - 369 0 - 30 0 - 22 0 - 40 al., 2003)

(Fernández et al.,
Queso azul (LC) 0 - 1052 0 - 1042 nd 0 - 876 0 – 757 0 - 27 - nd
2007a)
Cheddar (LP) nd nd nd 654,9 nd nd 40,0 202,2 (Bardócz et al., 1993)
Queso tipo holandés
299,8 1,8 - 60,8 2,0 54,3 0,2 0,3 (Komprda et al., 2007)
(LP)
Emmental (LP) 19,3 19,6 nd 16,8 12,6 nd nd nd (Mayer et al., 2010)
Gouda (LP) 27,7 26,1 104,4 nd nd 7,0 16,3 28,8 (Mayer et al., 2010)
Pescado
Atún fresco 0 – 10,6 0 – 9,5 0 – 5,8 0 – 4,8 0 – 9,5 0 – 1,7 7,3 – 37,0 1,2 – 11,7 (Veciana Nogues et al.,
Atún enlatado 0 – 40,5 0 – 40,5 0 – 12,9 0 – 2,2 0 – 12,0 0 – 7,3 2,2 -35,2 1,5 – 9,9 1997)
Atún nd 25,6 nd 1,7 - nd nd nd (Du et al., 2002)
Rihaakuru 0 - 50,8 0 - 5478,1 < 5,00 0 - 290,1 0 - 291,1 0 - 23,9 0 - 379,6 0 - 79,9 (Naila et al., 2011)
Pulpo 0 – 14,5 1,3 – 9,1 0 – 3,4 2,8 – 94,1 0,1 – 164 nd 0 – 4,2 0 – 1,6 (Hu et al., 2012)
Calamar 0 – 24,3 0 – 7,4 0 – 4,0 2,7 – 22,1 1,9 – 17,3 nd 1,3 – 6,4 0 – 1,8 (Hu et al., 2012)
Productos cárnicos
Salchichón 53 - 513 0 - 151 0 - 65 5 - 400 0 - 342 0 - 35 7 - 42 1 – 14 (Hernández-Jover et al.,
Fuet 32 – 743 0 - 358 0 - 88 2 - 222 5 - 51 0 - 34 9 - 30 1 - 11 1997)
Jamón curado 4,87 1,73 nd 0,92 5,52 nd 44,28 5,44 (Ruiz-Capillas &
Jiménez-Colmenero,
Chorizo 214,45 15,53 nd 185,14 229,33 nd 39,94 8,15
2004b)
nd: no determinado; ‐: no detectado; LP: leche pasteurizada; LC: leche cruda.

Introducción
En la cerveza la putrescina, espermidina, espermina y agmatina, suelen considerarse

como constituyentes naturales, ya que estas aminas biógenas se encuentran en la
malta. Histamina, tiramina y cadaverina se consideran procedentes de la actividad de
bacterias lácticas contaminantes durante la elaboración de la cerveza. La mayor
formación de aminas biógenas en la cerveza se produce durante la fermentación,
aunque también pueden producirse durante otras etapas de la elaboración (Bokulich &
Bamforth, 2013) y durante el almacenamiento en botella, debido a una mala
pasteurización (Kalač & Křı́žek, 2003). En algunos tipos de cerveza especiales la
fermentación secundaria en botella puede dar lugar a altas concentraciones de aminas
biógenas. La mayor variabilidad en el contenido en aminas biógenas se ha observado
en cervezas elaboradas por fermentación espontánea, abarcando niveles de tiramina
que van desde menos de 1 mg/L hasta 68 mg/L (Izquierdo-Pulido et al., 1996b, Romero
et al., 2003, Loret et al., 2005).
1.3.2. Formación de aminas biógenas
Las monoaminas (histamina, tiramina, triptamina y β-feniletilamina) y las diaminas

(cadaverina y putrescina) se forman principalmente como resultado de la
descarboxilación directa de aminoácidos. Las reacciones de descarboxilación están
mediadas por enzimas descarboxilasa, e implican la eliminación del grupo carboxilo del
aminoácido precursor, dando así lugar a la amina correspondiente. (Figura 13 y Figura
14). Las enzimas que median esta reacción son la histidina descarboxilasa (HDC; EC
4.1.1.22), tirosina descarboxilasa (TDC; EC 4.1.1.25), triptófano descarboxilasa (TrDC; EC
4.1.1.28), fenilalanina descarboxilasa (FDC; EC 4.1.1.53), lisina descarboxilasa (LDC; EC
4.1.1.18) y ornitina descarboxilasa (ODC; EC 4.1.1.17), que dan lugar a histamina,
tiramina, triptamina, β-feniletilamina, cadaverina y putrescina, respectivamente. Existe
además otra ruta alternativa para la formación de putrescina, llamada “deureación”
(Komprda et al., 2008b), en la que la arginina es descarboxilada por la enzima arginina
descarboxilasa (ADC; EC 4.1.1.19) dando lugar a la agmatina (Figura 14), la cual por
acción de la agmatina ureohidrolasa (agmatinasa; EC 3.5.3.11) pierde urea dando lugar a
la putrescina (Hillary & Pegg, 2003). La arginina puede convertirse en ornitina por
acción de la arginasa (EC 3.5.3.1). Las enzimas descarboxilasa pertenecen al grupo de
enzimas piridoxal-fosfato-dependientes, pero algunas de ellas pueden emplear como
43

Introducción
cofactor piruvato en vez de piridoxal-5’-fosfato (Coton et al., 1998; Recsei & Snell, 1985;
Kamath et al., 1991). En presencia de pequeñas cantidades de ornitina, la ornitina
descarboxilasa puede formar cadaverina mediante la descarboxilación de la lisina
(Karovičová & Kohajdová, 2005).
O
NH2
OH TDC
NH2
HO
HO CO2
Tirosina Tiramina
O
HDC NH2
N OH N
NH2 N
N
H CO2 H
Histidina Histamina
O
TrDC
NH2
OH
NH2
N CO2 N
H H
Triptófano Triptamina
O
OH NH2
FDC
NH2
CO2
Fenilalanina Feniletilamina
Figura 13. Reacciones de formación de las aminas biógenas. (TDC: tirosina descarboxilasa.
HDC: histidina descarboxilasa. TrDC: triptófano descarboxilasa. FDC: fenilalanina
descarboxilasa).
44

Introducción
La síntesis de poliaminas (espermidina y espermina) es un proceso más complejo

en el cual intervienen también reacciones de descarboxilación (Bardócz, 1995). Dicha
síntesis se produce a partir de metionina y putrescina. La metionina proporciona los
grupos aminopropilo necesarios para convertir la putrescina en espermidina, y esta a su
vez en espermina.
O
H2N LDC H2N NH2
OH
NH2
Lisina CO2 Cadaverina
O
S
H3C OH
NH2
Metionina
NH O
O Arginasa
MAT H2N NH OH
H2N OH
NH2
+ N NH2 Arginina
- NH3 NH2
O CH3 Ornitina
O
+ N
C S H2N ADC CO2
O N C O
N
HO OH CO2 H2N NH
ODC NH2
S-adenosilmetionina H2N NH
asa Agmatina
SAMDC atin
CO2 Agm
NH2
N H2N PAO
NH2
CH3 Putrescina
O
+ + N
H3N S
N
N Espermidina sintasa N'-Acetilespermidina
HO OH
S-adenosilmetionina DAT
descarboxilada
NH2
H2N NH
PAO
Espermidina
N
NH2 Espermina sintasa N'-Acetilespermina
CH3
O
N
S
N DAT
N NH NH2
HO OH H2N NH
Espermina
5 '-metiltioadenosina
Figura 14. Ruta y reacciones de formación de diaminas y poliaminas. (LDC: lisina descarboxilasa.
MAT: metionina adenosiltransferasa. SAMDC: S-adenosil-L-metionina descarboxilasa. ADC:
arginina descarboxilasa. ODC: ornitina descarboxilasa. PAO: poliamino oxidasa. DAT: diamino N-
acetiltransferasa).
Mediante la acción de la enzima metionina adenosiltransferasa y posterior

descarboxilación la metionina se transforma en S-adenosilmetionina descarboxilada, la
cual en presencia de putrescina y de espermidina sintasa da lugar a espermidina, y en
presencia de espermidina y de espermina sintasa da lugar a la espermina. Esta reacción
puede sufrir un proceso de retroconversión (Figura 14) mediante la enzima diamino N-
acetiltranferasa (DAT; EC 2.3.1.57) dando N’-acetilespermina o N’-acetilespermidina que
45

Introducción
mediante la acción de una poliamino oxidasa (PAO) dan lugar a espermidina y

putrescina, respectivamente (Hillary & Pegg, 2003, Wallace et al., 2003).
1.3.3. Factores que influyen en la formación de aminas biógenas
La presencia de microorganismos descarboxilasa positivos es determinante para la

formación de aminas biógenas en alimentos. No obstante, otros muchos factores como
la disponibilidad de aminoácidos libres, pH, NaCl, temperatura y presencia del cofactor
piridoxal-5’-fosfato, pueden también influir sobre la síntesis de aminas biógenas
(Giuffrida et al., 2006). La disponibilidad de aminoácidos es un factor limitante para la
síntesis de aminas biógenas, ya que como se ha descrito anteriormente las aminas
biógenas se forman por descarboxilación de aminoácidos (Tabla 3, Figuras 13 y 14). El
pH del medio puede afectar tanto al crecimiento microbiano como a la actividad
enzimática de las descarboxilasas (cuyo pH óptimo es de 5,0), influyendo además en la
regulación de la transcripción de los genes de la descarboxilasa (tdcA) y del
transportador (tyrP) que se expresan únicamente a valores de pH ácidos (Linares et al.,
2009, Linares et al., 2011). La formación de aminas biógenas por los microorganismos
podría ser un mecanismo para contrarrestar el pH del medio cuando se alcanzan
condiciones ácidas (Koessler et al., 1928, Bover Cid et al., 2008, Loizzo et al., 2013). Así la
acidificación del medio favorece la actividad descarboxilasa provocando un aumento de
la concentración de aminas (Chander et al., 1988, Chander et al., 1989, Fernández et al.,
2007b), aunque otros estudios indican una reducción de estas debido a un menor
crecimiento microbiano (Maijala, 1994, Gardini et al., 2001, Tabanelli et al., 2012). La
presencia de altas concentraciones de NaCl en el medio causa una disminución del
crecimiento microbiano y de la formación de aminas biógenas (Chander et al., 1988,
Chander et al., 1989, Sumner et al., 1990, Gardini et al., 2001, Tabanelli et al., 2012). La
temperatura tiene un marcado efecto sobre la formación de aminas biógenas, ya que
influye en el crecimiento y multiplicación microbianos y en la actividad enzimática de las
descarboxilasas (Chander et al., 1988, Chander et al., 1989, Gardini et al., 2001, Tabanelli
et al., 2012). Así por ejemplo en un estudio realizado con queso madurado durante 56
días a 16 ºC se obtuvo una concentración total de aminas de 800 mg/kg, mientras que
después 112 días a 10 ºC la concentración de aminas fue de 350 mg/kg (Pachlová et al.,
2012). En otro estudio realizado con extracto de histidina descarboxilasa libre de células,
46

Introducción
se obtuvo la máxima actividad descarboxilasa a temperaturas próximas a 50 ºC

(Tabanelli et al., 2012). Las descarboxilasas son enzimas piridoxal-fosfato-dependientes,
que requieren de este compuesto como cofactor para poder ejercer su actividad (Curtin
& McSweeney, 2004). Sin embargo, hay algunos estudios en los que este cofactor tuvo
menor influencia sobre la formación de aminas que el resto de factores estudiados
(Moreno-Arribas & Lonvaud-Funel, 1999, Gardini et al., 2005, Marcobal et al., 2006).
Un factor indispensable para la formación y acumulación de aminas biógenas en

alimentos es la presencia de microorganismos con actividad descarboxilasa. Son
muchos los microorganismos capaces de producir aminas en alimentos, incluyendo
bacterias gram positivas, gram negativas, levaduras y mohos. En queso las aminas
biógenas, especialmente tiramina e histamina, son formadas principalmente por las
bacterias lácticas (Linares et al., 2011), que pueden formar parte del cultivo iniciador o
pertenecer a la microbiota endógena de la leche (Burdychova & Komprda, 2007,
Komprda et al., 2008b). En un caso de intoxicación por histamina debido al consumo de
queso Suizo se identificó una cepa de Lactobacillus buchneri como la principal
responsable (Sumner et al., 1990). En un estudio de quesos de tipo holandés se aislaron
17 cepas de Lactobacillus y 19 de Enterococcus, y se vio que 14 de las cepas de
Enterococcus eran tirosina descarboxilasa positivas, pertenecientes a las especies
Enterococcus durans, faecalis, faecium y casseliflavus. De las cepas de Lactobacillus
aisladas 3 fueron histidina descarboxilasa positivas, entre las cuales se identificó una
cepa de Lactobacillus helveticus que pertenecía al cultivo iniciador (Burdychova &
Komprda, 2007). En queso Montasio se aislaron 1237 bacterias lácticas y 200
enterobacterias, de las cuales 182 bacterias lácticas y 181 enterobacterias mostraron
capacidad de descarboxilar al menos uno de los aminoácidos ensayados (Marino et al.,
2008). En queso azul (Marino et al., 2000) se aislaron 104 cepas de Enterobacteriaceae y
todas ellas mostraron capacidad para descarboxilar al menos dos aminoácidos. Se
observó además que la mayoría de estos aislados pertenecían al género Enterobacter
(especies gergoviae, cloacae y aerogenes), seguido de Serratia liquefaciens y Escherichia
coli. Las cepas de Enterobacter aerogenes fueron las que produjeron mayor cantidad de
cadaverina, y las de Enterobacter cloacae de putrescina. Histamina y tiramina se
produjeron en menor cantidad y únicamente un aislado de Arizona spp. produjo una
47

Introducción
concentración de tiramina superior a 100 ppm. En un estudio realizado con diferentes

variedades de quesos españoles se obtuvieron 694 aislados, de los cuales 125 se
clasificaron como bacterias gram positivas con actividad descarboxilasa y 44 como gram
negativas con actividad descarboxilasa. De las 125 bacterias gram positivas con
actividad descarboxilasa, se vio que 117 eran formadoras de tiramina y únicamente 10
de histamina, identificándose 63 como Enterococcus faecalis. De los 44 aislados de
bacterias gram negativas con actividad descarboxilasa, 43 eran formadoras de
histamina y únicamente 2 de tiramina, identificándose 20 de estos aislados como Hafnia
alvei. Además se aislaron 2 mohos formadores de histamina, uno de los cuales fue
identificado como Geotrichum candidum (Roig-Sagués et al., 2002). En leche y quesos
de distintas regiones de Francia se aislaron 173 bacterias gram negativas, de las cuales
el 64 % presentó capacidad de producir al menos una amina biógena, principalmente
putrescina o cadaverina. Los principales microorganismos productores de histamina
fueron Morganella morganii, Hafnia alvei, Klebsiella spp. y Enterobacter spp., mientras
que los principales productores de tiramina fueron fundamentalmente especies de los
géneros de Enterobacter, Proteus y Pseudomonas (Coton et al., 2012). En general se
considera que la mayoría de las bacterias gram negativas descritas como contaminantes
de la leche, como Escherichia coli, Hafnia alvei, Morganella morganii o Pseudomonas,
pueden tener la capacidad de producir histamina, putrescina o cadaverina, mientras que
los principales microorganismos formadores de aminas biógenas en queso son las
bacterias gram positivas, siendo las bacterias lácticas las principales formadoras de
histamina y tiramina (Linares et al., 2012). Algunas levaduras han sido descritas como
productoras de aminas biógenas. Así, el empleo de las levaduras Pichia jadinii, Yarrowia
lipolytica o Debaryomyces hansenii en la elaboración de queso Raclette provocó un
aumento en la concentración de aminas biógenas (Wyder et al., 1999). En queso
Pecorino Crotonese se aislaron cepas de Debaryomyces hansenii capaces de
descarboxilar ornitina e histidina, así como cepas de Yersinia lipolytica capaces de
descarboxilar ornitina, tirosina, lisina y fenilalanina (Gardini et al., 2006)
En pescado los principales microorganismos productores de histamina son

Morganella morganii, Klebsiella pneumoniae y Hafnia alvei (López-Sabater et al., 1994b,
Özoğul, 2004). Se han aislado además cepas capaces de producir aminas biógenas
48

Introducción
pertenecientes a los géneros Enterobacter, Staphylococcus, Proteus, Acinetobacter,

Rahnella, Plesiomonas, Pseudomonas, Serratia, Shewanella y Aeromonas (López-Sabater
et al., 1994a, Tsai et al., 2004, Allen et al., 2005, Dalgaard et al., 2006). Algunos de estos
estudios se han realizado sobre muestras implicadas en episodios de intoxicación, como
por ejemplo atún ahumado en frío en Dinamarca en 2004 (Emborg & Dalgaard, 2006) o
iwashi maruboshi (sardina secada) en Japón en 2002 (Kanki et al., 2004).
En carne y productos cárnicos se debe distinguir entre productos frescos y

fermentados. En carne fresca, la cadaverina se ha asociado principalmente a la presencia
de Enterobacteriaceae, mientras que la putrescina se ha asociado con altos recuentos
de microorganismos aerobios totales (Ruiz-Capillas & Jiménez-Colmenero, 2004a). En
productos cárnicos frescos se han realizado distintos estudios en función del tipo de
envasado durante distintos tiempos de conservación, con el fin de identificar los
microorganismos productores de aminas biógenas. En estos estudios se han aislado e
identificado diferentes cepas de Pseudomonas, Serratia, Staphylococcus, Enterobacter,
Yersinia, Escherichia coli, Morganella, Proteus, Klebsiella, Citrobacter y Hafnia (Durlu-
Özkaya et al., 2001, Balamatsia et al., 2006, Curiel et al., 2011). Por otro lado, en
productos cárnicos fermentados los principales microorganismos productores de
aminas biógenas suelen ser las bacterias lácticas, siendo la tiramina la amina más
abundante (Ruiz-Capillas & Jiménez-Colmenero, 2004a, Latorre-Moratalla et al., 2012).
Se han aislado microorganismos productores de aminas biógenas procedentes del
cultivo iniciador o de la microbiota endógena de la carne, que fueron identificados
como Lactobacillus, Lactococcus, Leuconostoc, Weissella, Enterococcus, Staphylococcus,
Carnobacterium, Citrobacter, Oenococcus, Enterobacter, Klebsiella, Proteus, Serratia y
Morganella, e incluso Pseudomonas (Bover-Cid et al., 2001, Suzzi & Gardini, 2003, de las
Rivas et al., 2008, Latorre-Moratalla et al., 2010).
En los productos vegetales también debemos distinguir entre producto fresco y

fermentado. En vegetales frescos las aminas biógenas se pueden formar principalmente
por la presencia de distintas cepas de Enterobacteriaceae y Pseudomonas. Se han
aislado e identificado distintas cepas de Klebsiella, Pantoea, Hafnia, Enterobacter,
Serratia, Morganella y Aeromonas con capacidad de producir histamina, putrescina y/o
cadaverina durante el almacenamiento en refrigeración de vegetales (Halász et al., 1994,
49

Introducción
Lavizzari et al., 2010). En productos fermentados los principales productores de aminas

biógenas suelen ser las bacterias lácticas, como Lactobacillus y Leuconostoc, que
pueden pertenecer al cultivo iniciador o formar parte de la microbiota endógena del
producto fresco (Shalaby, 1996, Penas et al., 2010).
El vino puede contener aminas biógenas que, al igual que en otros productos
fermentados, suelen formarse por la actividad enzimática de las bacterias lácticas
responsables de la fermentación maloláctica. Uno de los principales microorganismos
formadores de aminas en vino es Oenococcus oeni, junto con cepas de Lactobacillus,
Leuconostoc y Pediococcus (Gardini et al., 2005, Landete et al., 2005, Beneduce et al.,
2010, Spano et al., 2010). En la cerveza la formación de aminas biógenas sucede
principalmente por la actividad de las bacterias lácticas durante la fermentación, pero
también pueden estar implicadas algunas cepas de Enterobacteriaceae y de
Saccharomyces (Bokulich & Bamforth, 2013). Se han identificado cepas de Lactobacillus
y Pediococcus con actividad descarboxilasa en diversos tipos de cerveza (Izquierdo-
Pulido et al., 1996a, Kalač & Křı́žek, 2003).
1.3.4. Papel fisiológico y toxicología de las aminas biógenas
Las aminas biógenas juegan importantes papeles en la fisiología de animales,

plantas y microorganismos. Las aminas biológicamente más activas son la histamina y la
tiramina, y en menor grado las poliaminas (espermidina y espermina).
La histamina está normalmente presente en gran variedad de tejidos en mamíferos

y organismos invertebrados. En humanos se puede encontrar en distintas
concentraciones en cerebro, pulmones, estómago, intestino delgado y grueso, útero y
uréteres (Ladero et al., 2010). La histamina es principalmente sintetizada por mastocitos,
basófilos, plaquetas, neuronas histaminérgicas y células enterocromafines,
almacenándose intracelularmente en vesículas y pudiendo secretarse al medio (Maintz
& Novak, 2007). La histamina modula distintas funciones, interactuando con receptores
específicos (H1, H2, H3 y H4) en células diana. Los receptores H1 se encuentran
principalmente en el sistema nervioso central, células del musculo liso y células
endoteliales. Están implicados en la vasodilatación, broncoconstricción, permeabilidad
vascular, procesos cognitivos y de atención, y pueden inducir la liberación de adrenalina
y noradrenalina, siendo además responsables de la percepción del dolor y el picor en
50

Introducción
picaduras de insectos (Hill et al., 1997, Coruzzi et al., 2001, Bongers et al., 2011). Los
receptores H2 están ampliamente distribuidos en diferentes tejidos, entre los que se
encuentran el cerebro, mucosa gástrica y tejido cardiaco. Su función principal es la
secreción de ácido gástrico, así como la relajación del músculo liso uterino, vascular y
de las vías aéreas. También pueden inhibir algunas funciones del sistema inmune (Hill et
al., 1997, Bongers et al., 2011). Los receptores H3 se encuentran principalmente en el
sistema nervioso central y en menor medida en el periférico. En el sistema nervioso
central provocan la inhibición de la liberación y formación de histamina, mientras que
en el periférico inhiben la liberación de diversos neurotransmisores como serotonina,
noradrenalina, acetilcolina y dopamina, e inhiben además la secreción de ácido gástrico
en el estómago (Coruzzi et al., 2001, Bongers et al., 2011). Los receptores H4, los más
recientemente identificados, se han encontrado en mastocitos y leucocitos, y están
implicados en funciones relacionadas con procesos inflamatorios y alérgicos (Huang &
Thurmond, 2008, Bongers et al., 2011).
La tiramina y la β-feniletilamina se encuentran en el organismo en cantidades traza

y están implicadas en el mantenimiento de la actividad neuronal (Silla Santos, 1996). La
tiramina y la β-feniletilamina provocan aumento de la presión sanguínea y aumento de
la frecuencia cardiaca de forma indirecta mediante la liberación de noradrenalina de las
vesículas en las que se encuentra almacenada (Knoll et al., 1996, Khwanchuea et al.,
2008). La tiramina puede aumentar el nivel de glucosa en sangre (Ladero et al., 2010).
La putrescina, espermidina y espermina son factores esenciales para el crecimiento

y funcionamiento normal celular, regulando la expresión génica mediante la
modificación de la estructura del ADN y la modulación de las rutas de las señales de
transducción (Kalač, 2009, Ladero et al., 2010). Estas aminas intervienen en el desarrollo,
maduración y mantenimiento del tracto intestinal (Kalač, 2009), participando además en
la protección del epitelio gástrico inhibiendo la secreción de ácido (Medina et al., 2003).
Las aminas biógenas están implicadas en diversas funciones fisiológicas en

humanos y animales, pero su ingesta en niveles elevados puede tener efectos tóxicos.
La aparición de los síntomas suele ser rápida, entre 5 minutos y 2 horas tras la ingesta, y
los efectos no suelen durar más de 12 horas.
51

Introducción
En mamíferos existe un eficiente sistema de detoxificación en el tracto intestinal y

en condiciones normales las aminas exógenas son rápidamente detoxificadas por
acción de las amino oxidasas. No obstante, en individuos con un sistema de
detoxificación deficiente, bajo tratamiento con fármacos inhibidores de las amino
oxidasas o debido a un consumo excesivo de aminas biógenas, pueden darse episodios
de intoxicación (Silla Santos, 1996). La principal vía de detoxificación es la oxidación,
aunque también puede producirse por metilación o acetilación (Ladero et al., 2010). Las
enzimas responsables de la oxidación son la monoamino oxidasa (MAO) y la diamino
oxidasa (DAO). Existen dos isoformas primarias de la MAO conocidas como MAO-A y
MAO-B. La MAO-A desamina la serotonina en el sistema nervioso central y las
monoaminas ingeridas en el tracto intestinal, mientras que la MAO-B se encuentra
principalmente en hígado y músculo, y desamina la dopamina y la β-feniletilamina. La
histamina y las diaminas son desaminadas por la DAO en el tracto intestinal,
proporcionando protección frente a pequeñas cantidades (McCabe-Sellers et al., 2006).
El efecto tóxico de las aminas se manifiesta con síntomas generales como dolor de
cabeza, migraña, hipertensión o hipotensión, taquicardia, urticaria, enrojecimiento y
calambres abdominales. La histamina provoca la dilatación de los vasos sanguíneos
periféricos provocando hipotensión, enrojecimiento y dolores de cabeza, y también
provoca la contracción del músculo liso intestinal dando lugar a calambres
abdominales, diarrea y vómitos (Silla Santos, 1996). Los síntomas de la intoxicación por
histamina pueden confundirse con reacciones alérgicas ya que induce aumento de
secreciones en las vías aéreas, enrojecimiento facial, urticaria, dificultad respiratoria,
taquicardia y espasmos bronquiales (Ladero et al., 2010). La tiramina y la β-
feniletilamina pueden provocar migrañas y en ocasiones naúseas, vómitos y problemas
respiratorios. En casos extremos pueden desembocar en una crisis hipertensiva,
hemorragia cerebral y fallo cardiaco (Shalaby, 1996, Ladero et al., 2010). Además de
estos efectos tóxicos hay estudios que indican que podrían estar implicadas en otros
procesos patológicos. Así, la tiramina puede favorecer la adherencia de Escherichia coli
O157:H7 al epitelio intestinal (Lyte, 2004). Las diaminas putrescina y cadaverina podrían
estar implicadas en procesos neoplásicos y cancerígenos a nivel colo-rectal. En cuanto a
las poliaminas espermidina y espermina, son potenciadores del efecto tóxico de las
52

Introducción
otras aminas biógenas, aumentando la permeabilidad de la pared intestinal y

favoreciendo una mayor absorción de las aminas. Otro problema que plantean las
poliaminas y diaminas es su capacidad para formar N-nitrosaminas, agentes
mutagénicos, carcinogénicos y teratogénicos (Shalaby, 1996, Ladero et al., 2010), en
presencia de nitritos y bajo las condiciones ácidas del estómago. La putrescina y la
cadaverina pueden transformarse por efecto de la temperatura durante el cocinado en
pirrolidina y piperidina respectivamente, que pueden reaccionar con nitritos dando N-
nitropirrolidina y N-nitropiperidina. La formación de nitrosaminas se ve por tanto
favorecida por la abundancia de aminas y nitratos o nitritos, pH ácido, temperaturas
altas y ciertos tratamientos o procesado del alimento como asado, fritura o ahumado.
Existen diversos estudios sobre la toxicidad de las aminas biógenas (Naranjo, 1966, Til
et al., 1997), pero debido a la gran cantidad de factores que pueden influir sobre la
toxicidad y efecto de estos compuestos, como la presencia de otras aminas, la presencia
de inhibidores o potenciadores y la eficiencia del mecanismo de detoxificación
individual, resulta muy difícil establecer un límite máximo de seguridad para el consumo
de aminas en alimentos. No existe una legislación que regule el límite máximo de
aminas biógenas en alimentos, excepto para los niveles de histamina en pescado. La
legislación europea (EC 2073/2005) establece un límite máximo de histamina de 200
mg/kg en pescado fresco y de 400 mg/kg en productos curados derivados del pescado,
mientras que la FDA de Estados Unidos establece 50 ppm, como nivel orientativo de
pescado en mal estado (FDA, 2011).
1.3.5. Control de la formación de aminas biógenas
La acumulación de aminas biógenas en alimentos puede dar lugar a alteraciones

(olores y sabores desagradables) y a reacciones tóxicas y adversas. Se han desarrollado
distintos métodos con el fin de prevenir la formación de aminas biógenas en alimentos.
El más empleado, no sólo para evitar la formación de aminas sino también para evitar
otros riesgos microbiológicos, es el control o mantenimiento de temperaturas de
conservación entre 0 y 5 ºC, lo que reduce la tasa de crecimiento de los
microorganismos productores de aminas biógenas y disminuye la actividad
descarboxilasa. En algunos casos, como en atún de aleta amarilla, a estas temperaturas
sigue existiendo actividad descarboxilasa y se puede producir un ligero aumento de
53

Introducción
aminas biógenas (Du et al., 2002, Naila et al., 2010). En productos fermentados la
temperatura necesaria para su elaboración puede favorecer la formación de aminas
biógenas. El empleo de técnicas correctas de manipulación y de medidas higiénicas
durante el procesado del producto, contribuyen a prevenir la formación de aminas
biógenas al reducir la contaminación del producto por microorganismos descarboxilasa
positivos. El empleo de cultivos iniciadores que carezcan de actividad descarboxilasa
también es un procedimiento útil. En el caso del queso, la pasteurización de la leche
elimina los microorganismos descarboxilasa positivos, pero en algunas variedades de
queso es necesario el empleo de leche cruda para mantener sus cualidades
características.
Determinados conservantes y aditivos (como el sorbato potásico, ácido cítrico,

ácido succínico, D-sorbitol y ácido málico) son capaces de inhibir el crecimiento de
microorganismos descarboxilasa positivos, controlando así la formación de aminas
biógenas en el alimento (Naila et al., 2010). Otros aditivos que resultan también eficaces
son el NaCl, que limita el crecimiento microbiano y la formación de aminas biógenas
(Gardini et al., 2001, Tabanelli et al., 2012), y la glucono-δ-lactona, que disminuye el pH
y limita el crecimiento microbiano (Maijala et al., 1993). El nitrito sódico es capaz de
limitar el crecimiento microbiano y la formación de aminas (Kurt & Zorba, 2009), pero
puede aumentar el riesgo de formación de nitrosaminas en el alimento. Se ha
propuesto también el empleo de ingredientes naturales con actividad antimicrobiana
como el ajo y el jengibre para el control de aminas biógenas en alimentos (Naila et al.,
2010).
Mediante el envasado a vacío, en atmósfera modificada o en envases activos, se

puede también controlar el crecimiento microbiano y la formación de aminas biógenas
(Balamatsia et al., 2006). Otras técnicas más complejas incluyen la irradiación y la
aplicación de altas presiones hidrostáticas. La irradiación permite alargar la vida útil de
los alimentos al reducir la carga microbiana, limitando así también la formación de
aminas. Dosis de 20 KGrays sobre disoluciones acuosas de aminas inducen su
descomposición hasta un 95 %, pero se requieren más estudios sobre los compuestos
de degradación resultantes y los posibles efectos sobre los alimentos (Kim et al., 2004).
Las altas presiones hidrostáticas, como se describirá más adelante en detalle,
54

Introducción
constituyen un método de pasteurización no térmica de los alimentos, que logra reducir

la carga microbiana y alargar la vida útil del producto en función de la intensidad de
presurización (Huang et al., 2014), y que limita también la formación de aminas
biógenas.
Además de las técnicas anteriormente descritas, basadas principalmente en el

control de los microorganismos productores de aminas, existen también métodos
basados en la degradación de las aminas biógenas presentes en el alimento. Se ha
propuesto el empleo de cepas de bacterias lácticas, Brevibacterium, Geotrichum o
Micrococcus, capaces de oxidar aminas a aldehídos, dando lugar a peróxido de
hidrógeno y otras aminas o amoniaco. Estos microorganismos pueden incorporarse al
alimento durante su procesado o bien emplearse como parte del cultivo iniciador en
productos fermentados (Leuschner et al., 1998, Herrero-Fresno et al., 2012).
1.4. Altas presiones hidrostáticas
La aplicación de altas presiones hidrostáticas como procedimiento para conservar

la leche y reducir el número de bacterias presentes en la misma se describió ya a finales
del siglo XIX (Hite, 1899). Sin embargo, debido a su alto coste y dificultades técnicas su
empleo en alimentos no adquirió importancia hasta la década de los 80. El procesado
de alimentos con altas presiones (HPP, High Pressure Processing) implica someter al
producto una vez sumergido en un líquido transmisor, generalmente agua debido a su
baja compresibilidad, a presiones de entre 100 y 600 MPa (aunque pueden alcanzarse
niveles de hasta 1400 MPa), con o sin aplicación de calor (Bermúdez-Aguirre &
Barbosa-Cánovas, 2011, Mújica-Paz et al., 2011).
Esta tecnología se fundamenta en dos principios:
· La ley de Pascal o principio isostático, por la que una presión externa aplicada
sobre un fluido incompresible confinado se transmite de forma instantánea y uniforme
en todas las direcciones. Según esto, la presión se transmite de forma instantánea y
uniforme a todas las zonas del producto, independientemente de su tamaño y forma,
haciendo que todo él reciba el mismo tratamiento (Martínez-Rodríguez et al., 2012,
Huang et al., 2014).
55

Introducción
· El principio de Le Chatelier, que indica que un sistema en equilibrio al sufrir un

cambio (de concentración, presión, volumen o temperatura) se ajusta para contrarrestar
dicho cambio. Así, al aumentar la presión se verán favorecidas las reacciones que
impliquen una reducción de volumen (Eisenmenger & Reyes-De-Corcuera, 2009).
Según su funcionamiento, los equipos de altas presiones pueden ser:
· Equipos de tipo discontinuo. Estos equipos se emplean para presurizar productos

líquidos o sólidos y para ello el producto se introduce en la cámara de presurización
envasado en un material flexible. La forma de generar la presión en la cámara puede ser
directa (Figura 15 A), mediante la compresión del fluido por un pistón, o indirecta
(Figura 15 B), mediante una bomba de alta presión o un intensificador que introduce
agua en la cámara provocando el aumento de la presión (Yaldagard et al., 2008). Una
vez terminado el tratamiento, la cámara se despresuriza y se retira el producto. Este tipo
de equipos tienen la ventaja de que el producto se trata envasado, evitando así su
contaminación posterior.
Figura 15. Esquema de sistemas de presurización directa (A) e indirecta (B).
· Equipos de tipo semicontinuo. Estos equipos se emplean para el tratamiento de

productos líquidos, que pueden ser bombeados a la cámara de presurización. En este
tipo de equipos la presión se genera de forma directa y una vez finalizado el
tratamiento el producto es bombeado a un tanque estéril para su posterior envasado
(Chawla et al., 2011).
56

Introducción
Existe además un sistema, no empleado en alimentos, en el que mediante el

calentamiento del fluido de transmisión en la cámara se consigue aumentar la presión
(San Martin et al., 2002).
Figura 16. Ejemplo de algunos productos comerciales tratados por altas presiones.
Las altas presiones afectan a enlaces no covalentes, como fuerzas de Van der
Waals, interacciones electrostáticas y puentes de hidrógeno (Stewart et al., 2006, Chawla
et al., 2011, Mújica-Paz et al., 2011). Por ello, la estructura de moléculas de alto peso
molecular, como proteínas o carbohidratos, puede ser alterada por las altas presiones,
mientras que moléculas más pequeñas, como compuestos volátiles, pigmentos y
vitaminas, no resultan afectadas (Huang et al., 2014). Los tratamientos por altas
presiones logran la pasteurización no térmica del producto, así como la inactivación de
enzimas, obteniéndose un producto microbiológicamente seguro que mantiene sus
57

Introducción
características nutricionales y organolépticas mejor que los sometidos a tratamientos

térmicos (Martínez-Rodríguez et al., 2012).
El empleo de las altas presiones como método no térmico de pasteurización en

alimentos ha sido aprobado por la FDA y el Departamento de Agricultura de Estados
Unidos (Huang et al., 2014). Además, el empleo de tratamientos combinados de altas
presiones y temperatura ha sido aprobado por la FDA como método para la
esterilización comercial de alimentos de baja acidez (Bermúdez-Aguirre & Barbosa-
Cánovas, 2011, Mújica-Paz et al., 2011). Durante la presurización, debido al trabajo de
compresión contra las fuerzas intermoleculares, se produce un calentamiento
adiabático que aumenta la temperatura del producto y del fluido transmisor (así por
ejemplo, el agua sufre un aumento de 3 ºC por cada 100 MPa). Se produce también un
desplazamiento del pH, generalmente hacia valores ácidos, que sin embargo una vez
retirada la presión vuelve a su valor inicial (Bermúdez-Aguirre & Barbosa-Cánovas, 2011,
Mújica-Paz et al., 2011). Los tratamientos por altas presiones presentan también la
ventaja de reducir costes y tiempo de procesado frente a los tratamientos térmicos
convencionales, además de la mejora en la calidad del producto (Martínez-Rodríguez et
al., 2012).
180
160
140
Número de equipos
120
100
80
60
40
20
0
Año
Figura 17. Aumento del número de equipos industriales de altas presiones distribuidos en el mundo.
Existe una amplia gama de productos tratados por altas presiones que se
comercializan en todo el mundo (Figura 16). En los últimos años esta tecnología ha
58

Introducción
adquirido gran importancia, como se puede ver por el número de equipos distribuidos
a nivel mundial (Figura 17).
1.4.1. Modo de acción sobre microorganismos y enzimas
El aumento de la vida útil de los alimentos tratados por altas presiones se debe
principalmente a la inactivación de los microorganismos y las enzimas presentes en los
mismos. El efecto de las altas presiones sobre los microorganismos depende del nivel
de presión alcanzado, del tiempo de aplicación, de las características de los
microorganismos, de la temperatura del proceso y de la composición del medio
(Stewart et al., 2006). La inactivación de los microorganismos se produce por la
acumulación de daños en distintas partes de la célula que cuando exceden la capacidad
de reparación, acaba dando lugar a la lisis microbiana. En algunos casos el daño celular
puede ser reparado si las condiciones del medio son favorables (Rendueles et al., 2011).
Uno de los primeros sitios afectados por las altas presiones es la membrana celular. La
presurización induce una disminución de fluidez de los lípidos y fosfolípidos de
membrana, llegando incluso a su solidificación, provocando una disminución de la
permeabilidad y la pérdida de funcionalidad e integridad de la membrana. Esto mismo
sucede en el interior de la célula, en la membrana de algunos orgánulos (Rendueles et
al., 2011, Huang et al., 2014). Las altas presiones provocan la disociación de los
ribosomas, causando una reducción o inhibición de la síntesis de proteínas y enzimas
(Abe, 2007). Las altas presiones pueden causar además daños estructurales y
morfológicos, como la separación entre la membrana y la pared celular y la compresión
de las vacuolas. Las altas presiones también afectan a la funcionalidad del material
genético, inhibiendo los mecanismos de replicación y transcripción del ADN (Huang et
al., 2014). Otro efecto importante inducido por las altas presiones es la
desnaturalización de proteínas, incluyendo proteínas de membrana y enzimas. La
desnaturalización de las proteínas de membrana puede limitar el flujo de protones,
impidiendo la regulación del pH intracelular. La desnaturalización de las proteínas por
las altas presiones se debe a la modificación de su estructura y su efecto depende del
nivel estructural de la molécula. La estructura primaria viene determinada por la
secuencia de aminoácidos unidos mediante enlaces covalentes, que no se ven afectados
por las altas presiones. La estructura secundaria viene determinada por la formación de
59

Introducción
puentes de hidrógeno, que son poco susceptibles a las presiones pero pueden verse
afectados a presiones superiores a 300 MPa, siendo el efecto irreversible por encima de
700 MPa. Se ha visto además que la β-lámina es más barorresistente que la α-hélice. La
estructura terciaria está determinada por interacciones hidrofóbicas, puentes disulfuro y
fuerzas de Van der Waals, que son enlaces muy susceptibles a la presión, de tal modo
que presiones superiores a 200 MPa provocan ya cambios significativos. La estructura
cuaternaria aparece en proteínas oligoméricas y se debe a la presencia de distintas
interacciones, como puentes de hidrógeno, puentes disulfuro e interacciones
electrostáticas. Los puentes de hidrógeno presentan una alta resistencia a la presión,
pero no así los puentes disulfuro y las interacciones electrostáticas, por lo que la
estructura cuaternaria resulta muy susceptible a la presurización (Rendueles et al., 2011,
Huang et al., 2014). El efecto de las altas presiones sobre las enzimas es altamente
dependiente del tipo de enzima y de las condiciones del medio (San Martin et al., 2002).
Las enzimas pueden activarse o inactivarse, dependiendo de la intensidad de la presión
aplicada y del tipo de enzima (Yaldagard et al., 2008, Huang et al., 2014). Como ya se ha
descrito, los puentes de hidrógeno pueden llegar a verse afectados por las altas
presiones, lo que puede resultar en cambios de la estabilidad y actividad de la enzimas.
De este modo, algunas enzimas pueden ver incrementada su actividad por la
modificación de la velocidad del paso limitante o la modulación de la especificidad y
así, por ejemplo, la tripsina sometida a un tratamiento de 300 MPa durante 300 minutos
ve incrementada su actividad en un 36% (Eisenmenger & Reyes-De-Corcuera, 2009, Kim
et al., 2013). Sin embargo, en general se considera que presiones superiores a 400 MPa
pueden inducir cambios de la estructura tridimensional, alterando el centro activo y
provocando la inactivación de la enzima. Se ha descrito que algunas enzimas, como la
termolisina, se inactivan hasta un 50 % con presiones de 100 MPa (Kim et al., 2013),
mientras que otras, como la polifenoloxidasa de ciruela, resisten hasta 900 MPa
(Weemaes et al., 1998).
La resistencia de los microorganismos a las altas presiones es muy variable y

depende principalmente del tipo de microorganismo, la temperatura y la composición
de la matriz en la que se encuentra (Stewart et al., 2006). Valores bajos de actividad de
agua ejercen un efecto baroprotector en los microorganismos (Morales et al., 2006).
60

Introducción
Además a temperaturas de 4 ºC algunos microorganismos son menos resistentes a la

presión que a 20 ºC (San Martin et al., 2002). Las altas presiones normalmente tienen
mayor efecto en microorganismos con mayor grado de organización y complejidad
estructural. Los microorganismos procariotas son generalmente más resistentes que los
eucariotas. Los protozoos y los parásitos son especialmente sensibles a la presión, y
presiones relativamente bajas logran su inactivación (Rivalain et al., 2010). Los mohos y
levaduras tienen una resistencia intermedia, siendo los micelios de los mohos muy
sensibles, mientras que las esporas son más resistentes (Martínez-Rodríguez et al.,
2014). Los virus poseen un amplio rango de resistencia a las altas presiones debido su
gran diversidad estructural, siendo los poliovirus los más barorresistentes, resistiendo
presiones de hasta 600 MPa (Grove et al., 2008). La resistencia de las formas vegetativas
de las bacterias depende de su fisiología, morfología y fase en la que se encuentran,
siendo en general las gram positivas más resistentes que las gram negativas, los cocos
más resistentes que los bacilos, y las bacterias en fase estacionaria más resistentes que
las bacterias en fase exponencial (Chawla et al., 2011, Demazeau & Rivalain, 2011b). Se
ha visto además que los microorganismos que poseen mayor proporción de ácidos
grasos insaturados en la membrana celular son más resistentes a la presión (Rendueles
et al., 2011). Por otro lado, las esporas bacterianas son extremadamente resistentes a las
altas presiones, por lo que se han estudiado distintas estrategias para su destrucción,
como la eliminación directa mediante la combinación de altas presiones con otros
parámetros (como altas temperaturas) o la inducción de la germinación de las esporas
seguida de la inactivación de las formas vegetativas (Nguyen Thi Minh et al., 2010,
Demazeau & Rivalain, 2011a, Reineke et al., 2011).
1.4.2. Efecto de las altas presiones en queso
El empleo de altas presiones en la elaboración del queso se enfocó inicialmente al

tratamiento de la leche para eliminar los microorganismos presentes en esta.
Posteriormente, las altas presiones empezaron a aplicarse directamente sobre la
cuajada o el queso, con el fin inactivar microorganismos no deseados (patógenos o
alterantes), prolongar el tiempo de conservación del queso y/o modificar el proceso de
maduración. Dependiendo de la variedad de queso y de la intensidad del tratamiento
de presurización, la maduración puede ser acelerada mediante la lisis de los
61

Introducción
microorganismos y liberación al medio de sus enzimas o frenada mediante la lisis de los

microorganismos y la inactivación de las enzimas (Martínez-Rodríguez et al., 2012),
alargando la vida útil como se ha visto en queso fresco (Evert-Arriagada et al., 2012,
Evert-Arriagada et al., 2014).
1.4.2.1. Efecto sobre los microorganismos
La capacidad de las altas presiones para inactivar microorganismos depende del

tipo de microorganismo y del medio en el que se encuentre. Debido a las diferencias
físico-químicas que se dan entre las distintas variedades de queso y a los cambios que
experimenta cada variedad durante el proceso de maduración, el efecto de la
presurización sobre los microorganismos es muy diverso. La capacidad de las altas
presiones para reducir la presencia de microorganismos contaminantes del queso ha
sido demostrada por diversos autores (Rendueles et al., 2011, Martínez-Rodríguez et al.,
2012, Mota et al., 2013). En pasta de queso Cheddar (100 g de queso y 42 g de solución
salina), tratamientos de 400 MPa a una temperatura de 10 ºC lograron reducciones de
los niveles de P. roqueforti de 2 ud. log, mientras que aplicados a 20 ºC o 30 ºC
alcanzaron reducciones de 6 ud. log (O'Reilly et al., 2000). En este mismo estudio, se
observaron reducciones de S. aureus y E. coli desde niveles de 107 ufc/g hasta niveles
no detectables mediante tratamientos de 400 MPa a 30 ºC o de 600 MPa a 20 ºC. En
quesos de leche cruda (Arqués et al., 2006), se han conseguido reducciones de bacterias
coliformes de hasta 5,5 ud. log mediante tratamientos de 400 MPa aplicados 2 días
después de la fabricación, así como reducciones de 4,0 ud. log de bacterias gram
negativas, de 1,8 ud. log de Micrococcaceae y de 1,5 ud. log de Staphylococcus
coagulasa positivos. La presurización a 300 o 400 MPa aplicada 50 días después de la
fabricación indujo reducciones de 2,3 o 6,4 ud. log de bacterias gram negativas,
respectivamente; además, el tratamiento de 400 MPa redujo por debajo del límite de
detección los niveles de bacterias coliformes. En otro estudio, la aplicación de 600 MPa
al día siguiente de la fabricación consiguió reducir en 4,8 ud. log los niveles de bacterias
psicrótrofas, en 2,6 ud. log los de Enterobacteriaceae y Listeria spp. y en 2,4 ud. log los
de Micrococcaceae (Delgado et al., 2012). Por otro lado, tratamientos de 200 o 300 MPa
aplicados a queso de 1 día lograron reducciones de Enterobacteriaceae de entre 1 y 2
ud. log, mientras que 400 o 500 MPa redujeron los niveles entre 3 y 4 ud. log;
62

Introducción
tratamientos superiores a 300 MPa aplicados a queso de 15 días consiguieron niveles

no detectables de Enterobacteriaceae (Juan et al., 2007b).
La presurización no sólo afecta a los microorganismos patógenos y alterantes del

queso sino también a los cultivos iniciadores y cultivos secundarios, pudiendo
emplearse para controlar la maduración. Se ha observado que tratamientos de 405 MPa
durante 3 minutos aplicados a quesos Cheddar elaborados con distintas
concentraciones de sal, dieron como resultado reducciones de entre 1 y 4 ud. log de las
bacterias lácticas del cultivo iniciador (Ozturk et al., 2013). Tratamientos de 400 y 600
MPa, en queso azul de 42 días, consiguieron reducir los niveles de P. roqueforti,
levaduras, Enterococcus, Lactobacillus y bacterias lácticas secundarias, alcanzando
reducciones de hasta 3 ud. log para P. roqueforti y bacterias lácticas secundarias con el
tratamiento de 600 MPa (Voigt et al., 2010). En queso Cheddar de 1 mes (Wick et al.,
2004), tratamientos superiores a 300 MPa consiguieron reducir los niveles de
Lactococcus lactis del cultivo iniciador y tratamientos superiores a 200 MPa redujeron
los niveles de Lactobacillus. En el caso de los quesos presurizados a 500 y 800 MPa, los
niveles de Lactobacillus se redujeron por debajo del límite de detección. Durante el
almacenamiento posterior de los quesos, los niveles de Lactococcus lactis aumentaron
únicamente en los quesos presurizados a 400 MPa, mientras que los niveles de
Lactobacillus aumentaron en los quesos presurizados a 200 y 300 MPa. Sin embargo, en
queso Cheddar de 4 meses fueron necesarios tratamientos superiores a 500 MPa para
conseguir reducciones significativas de los niveles de Lactococcus lactis. En quesos de
leche de oveja, la presurización a 200 o 300 MPa aplicada a queso de 1 día indujo
reducciones en los niveles de Lactococcus de tan sólo 1 ud. log, mientras que
tratamientos con 400 y 500 MPa lograron reducciones de 2 y 5 ud. log respectivamente,
para Lactococcus, y de 2 ud. log para Lactobacillus. Estos mismos tratamientos de 400 y
500 MPa aplicados a queso de 15 días, redujeron los niveles de Lactobacillus en 2 y 5
ud. log respectivamente, y en 2 ud. log los niveles de Lactococcus (Juan et al., 2007b).
Tratamientos con 300 MPa en quesos de leche de oveja, aplicados a queso de 1 o 15
días, indujeron reducciones en los niveles de Lactobacillus de 3 y 1 ud. log
respectivamente (Juan et al., 2007a). Tratamientos de 400 y 600 MPa aplicados a
quesos de 1 o 30 días, redujeron significativamente los niveles de bacterias lácticas en
63

Introducción
quesos de leche cruda de cabra (Delgado et al., 2012). En quesos de leche cruda de
oveja tratados con altas presiones a los 2 días de la fabricación, se obtuvieron
reducciones de bacterias lácticas, lactobacilos y enterococos de 1,4, 0,6 y 2,1 ud. log,
respectivamente, para el tratamiento de 300 MPa, y de 2,0, 1,6 y 2,7 ud. log,
respectivamente, para 400 MPa (Arqués et al., 2006). En quesos en salmuera la
aplicación de 200 MPa a queso de 15 días no produjo reducciones significativas,
mientras que con 500 MPa se consiguió reducir en 8, 2 y 4 ud. log los niveles de
Lactococcus, Lactobacillus y bacterias lácticas secundarias, respectivamente
(Moschopoulou et al., 2010). Las altas presiones también se han empleado con el
objetivo de atenuar los cultivos iniciadores empleados en la elaboración de queso. De
esta forma, mediante el empleo de 200 MPa se consiguió que cultivos de Lactococcus
lactis, empleados como adjuntos, no acidificasen durante la elaboración de queso
Cheddar y sí aumentasen la proteolisis secundaria durante la maduración (Upadhyay et
al., 2007). Estos mismos resultados se obtuvieron también en queso Feta al tratar un
cultivo indicador formado por diferentes cepas de bacterias lácticas (Maniou et al.,
2013).
1.4.2.2. Efecto sobre las enzimas
El control de la maduración mediante las altas presiones no depende únicamente

de la presencia de microorganismos viables, sino también de la actividad de las enzimas
presentes, que pueden proceder de la leche o de microorganismos lisados. A una
temperatura de 8 ºC, tratamientos inferiores a 800 MPa y 60 minutos, no produjeron
ningún efecto sobre la plasmina en queso Cheddar, así como tampoco tratamientos
inferiores a 400 MPa aplicados a 20 ºC. Tratamientos de 600 y 800 MPa a 20 ºC,
aplicados entre 15 y 60 minutos consiguieron una leve inactivación (≤ 15 %) de la
plasmina, mientras que con estos mismos tratamientos aplicados a 30 ºC se consiguió
una inactivación de entre el 31 y 48 % (Huppertz et al., 2004, Rynne et al., 2008). En
quesos de leche de oveja, la actividad de la plasmina no se vio afectada por
tratamientos de hasta 500 MPa (Juan et al., 2007b) y al ser tratada en tampón la
plasmina no se vio afectada por presiones de hasta 800 MPa (Malone et al., 2003). La
quimosina en queso Cheddar no se vio afectada por presiones inferiores a 400 MPa,
independientemente de la temperatura a la que se aplicase el tratamiento, mientras que
64

Introducción
tratamientos de 600 y 800 MPa provocaron una inactivación de entre el 87 y el 96 %

(Huppertz et al., 2004, Rynne et al., 2008). Tratamientos superiores a 400 MPa aplicados
a queso de 1 día de leche de oveja provocaron una importante reducción de la
actividad de la quimosina (Juan et al., 2007b), mientras que en tampón esta misma
enzima no se vio afectada por presiones de hasta 400 MPa, manteniendo cerca de un
50 % de su actividad a presiones de 800 MPa (Malone et al., 2003). En un estudio con
diferentes tipos de quesos (Gouda, Camembert y Kurpiowski), diferentes presiones (200
a 800 MPa) y diferentes tiempos de maduración, se observó que la actividad proteolítica
disminuyó con la aplicación de tratamientos superiores a 200 MPa y que la actividad
amino- y endopeptidásica se redujeron con tratamientos de 600 MPa. La enzima lisina-
aminopeptidasa mostró mayor resistencia a las altas presiones en queso Camembert. En
los quesos Camembert y Gouda se encontró actividad caseinolítica residual después de
ser tratados con 800 MPa a las 2 semanas de maduración (Reps et al., 2003). En queso
Edam se observó un ligero aumento de la actividad proteolítica en los quesos
presurizados a 200 MPa, mientras que los tratados a 400 MPa mostraron un ligero
descenso de esta actividad (Iwańczak & Wiśniewska, 2005). La actividad lactato
deshidrogenasa (LDH) en quesos de leche de oveja aumentó con la aplicación de 300 y
400 MPa un día después de la fabricación, mientras que la aplicación de 500 MPa
provocó un descenso de la actividad debido a la inactivación parcial de esta enzima
(Juan et al., 2007b, 2008). Igualmente, en quesos en salmuera, la actividad LDH aumentó
con tratamiento a 200 MPa y se redujo con tratamiento a 500 MPa (Moschopoulou et
al., 2010). En quesos de leche cruda de oveja, la actividad aminopeptidásica se redujo
significativamente con tratamientos de 300 y 400 MPa aplicados 2 días después de la
fabricación, mientras que la aplicación de 300 MPa a los 50 días provocó un aumento
significativo respecto al control (Garde et al., 2007). La aplicación de tratamientos de
entre 200 y 500 MPa un día después de la fabricación en quesos de leche de oveja
redujo la actividad aminopeptidásica, mientras que estos mismos tratamientos
aplicados a los 15 días provocaron un aumento de actividad, excepto en los quesos
tratados con 500 MPa (Juan et al., 2007b). La aplicación de tratamientos de 200 y 500
MPa a quesos en salmuera 15 días después de su fabricación no provocó ningún efecto
significativo sobre la actividad aminopeptidásica (Moschopoulou et al., 2010). En queso
Hispánico de 15 días de maduración tratado con 400 MPa disminuyó la actividad
65

Introducción
aminopeptidasa (Ávila et al., 2006), aunque no se vio ninguna diferencia significativa de

actividad esterasa respecto al control (Ávila et al., 2007). Tratamientos de 600 MPa a
temperatura ambiente consiguieron reducir la actividad de la lipasa en tampón, pero
fueron necesarios 700 MPa a 45 ºC para conseguir su total inactivación (Seyderhelm et
al., 1996). Mediante estudios en tampón, se ha visto que tratamientos a partir de 400
MPa provocan la inactivación de la proteasa asociada a la envoltura celular, mientras
que tratamientos de 100 MPa aumentan su actividad. Las aminopeptidasas se ven
afectadas de distinta manera por las altas presiones. Así, se ha visto que en tampón
tratamientos de 400 MPa son capaces de inactivar la aminopeptidasa N, mientras que la
aminopeptidasa A no se ve afectada por tratamientos de hasta 800 MPa y la
aminopeptidasa C aumenta su actividad con tratamientos de hasta 700 MPa (Malone et
al., 2003).
1.4.2.3. Efecto sobre la textura y la microestructura
Las altas presiones pueden provocar cambios en la estructura de la red de caseínas

del queso, modificando su textura. La aplicación de tratamientos de 405 MPa a quesos
Cheddar elaborados con distintas concentraciones de sal, hizo que los valores de
dureza y fracturabilidad de los quesos tratados fuesen menores que los de sus
respectivos controles no tratados (Ozturk et al., 2013). En quesos de leche de oveja, la
aplicación de tratamientos de 300 MPa a diferentes tiempos de maduración provocó
que los quesos tratados el día 1 después de la fabricación presentaran valores más altos
de deformabilidad que los quesos no tratados y que los presurizados el día 15 (Juan et
al., 2008). En quesos de leche cruda de cabra, tratados con 400 y 600 MPa a diferentes
tiempos de maduración, se observó que únicamente los tratados a día 1 sufrieron
reducciones significativas de dureza, gomosidad y masticabilidad respecto al control, no
apreciándose diferencias entre la aplicación de 400 y 600 MPa (Delgado et al., 2012). En
quesos de leche cruda de oveja, tratados con 300 y 400 MPa 2 días después de la
fabricación, los valores de fracturabilidad y elasticidad se vieron significativamente
incrementados, mientras que la dureza únicamente aumentó en los quesos tratados con
400 MPa (Garde et al., 2007). En quesos Cheddar tratados con 345 y 483 MPa durante
tiempos de 3 y 7 minutos, los valores de firmeza fueron significativamente menores que
los del queso control; los quesos tratados durante 3 minutos con 345 MPa y los
66

Introducción
tratados 7 minutos con 483 MPa presentaron valores de elasticidad inferiores al control
y no se vio afectada la cohesividad (Serrano et al., 2005). En estos quesos se analizó
además su microestructura mediante microscopía electrónica de barrido. En el queso
control se observó una red porosa de proteínas, con aspecto de esponja, en la que se
englobaban glóbulos grasos de diferentes formas y tamaños, distribuidos sin
uniformidad. La matriz de proteína de los quesos tratados fue sin embargo más
compacta y continua, más semejante a la de los quesos curados. Estos cambios de
estructura se relacionaron con la intensidad y el tiempo de los tratamientos,
observándose zonas compactas y zonas esponjosas en los quesos tratados con 345
MPa y obteniéndose una estructura en general más compacta en el queso tratado con
483 MPa durante 7 minutos. Cuando se aplicaron tratamientos de entre 50 y 400 MPa
durante 5 o 15 minutos a quesos en salmuera, se observó que los tratamientos de 50 y
100 MPa no afectaron a la textura ni a la microestructura de los quesos, mientras que
los quesos presurizados a 200 y 400 MPa presentaron valores de firmeza, elasticidad y
gomosidad inferiores a los del queso control (Koca et al., 2011). Mediante el análisis de
la microestructura por microscopía confocal de barrido por láser, se pudo apreciar una
red porosa de proteínas, con glóbulos de grasa de diferentes tamaños y formas tanto
en el queso control como en los tratados con 50 y 100 MPa. Los quesos tratados con
200 y 400 MPa mostraron una red de proteínas más compacta, con glóbulos de grasa
de menor tamaño y distribuidos de forma más uniforme. Por otro lado, la aplicación de
las altas presiones durante 15 minutos dio lugar a una estructura más densa y compacta
que su aplicación durante 5 minutos.
67

Introducción
1.5. Bibliografía
Abe, F. (2007). Exploration of the effects of high hydrostatic pressure on microbial
growth, physiology and survival: Perspectives from piezophysiology. Bioscience
Biotechnology and Biochemistry 71, 2347-2357.
Alewijn, M., Smit, B. A., Sliwinski, E. L. & Wouters, J. T. M. (2007). The formation
mechanism of lactones in Gouda cheese. International Dairy Journal 17, 59-66.
Allen, D. G., Green, D. P., Bolton, G. E., Jaykus, L. A. & Cope, W. G. (2005). Detection
and identification of histamine-producing bacteria associated with harvesting and
processing mahimahi and yellowfin tuna. Journal of Food Protection 68, 1676-1682.
Ancin-Azpilicueta, C., Gonzalez-Marco, A. & Jimenez-Moreno, N. (2008). Current
knowledge about the presence of amines in wine. Critical Reviews in Food Science
and Nutrition 48, 257-275.
Arqués, J. L., Garde, S., Gaya, P., Medina, M. & Nuñez, M. (2006). Inactivation of
microbial contaminants in raw milk La Serena cheese by high-pressure treatments.
Journal of Dairy Science 89, 888-891.
Ávila, M., Garde, S., Gaya, P., Medina, M. & Nuñez, M. (2006). Effect of high-pressure
treatment and a bacteriocin-producing lactic culture on the proteolysis, texture,
and taste of Hispánico cheese. Journal of Dairy science 89, 2882-2893.
Ávila, M., Calzada, J., Garde, S. & Nuñez, M. (2007). Effect of a bacteriocin-producing
Lactococcus lactis strain and high-pressure treatment on the esterase activity and
free fatty acids in Hispánico cheese. International Dairy Journal 17, 1415-1423.
Balamatsia, C. C., Paleologos, E. K., Kontominas, M. G. & Savvaidis, I. N. (2006).
Correlation between microbial flora, sensory changes and biogenic amines
formation in fresh chicken meat stored aerobically or under modified atmosphere
packaging at 4 degrees C: possible role of biogenic amines as spoilage indicators.
Antonie Van Leeuwenhoek International Journal of General and Molecular Microbiology
89, 9-17.
Bardócz, S., Grant, G., Brown, D. S., Ralph, A. & Pusztai, A. (1993). Polyamines in
food - implications for growth and health. Journal of Nutritional Biochemistry 4, 66-
71.
Bardócz, S. (1995). Polyamines in food and their consequences for food quality and
human health. Trends in Food Science & Technology 6, 341-346.
Beneduce, L., Romano, A., Capozzi, V., Lucas, P., Barnavon, L., Bach, B., Vuchot, P.,
Grieco, F. & Spano, G. (2010). Biogenic amine in wines. Annals of Microbiology 60,
573-578.
Bermúdez-Aguirre, D. & Barbosa-Cánovas, G. V. (2011). An update on high
hydrostatic pressure, from the laboratory to industrial applications. Food
Engineering Reviews 3, 44-61.
68

Introducción

Bijl, E., de Vries, R., van Valenberg, H., Huppertz, T. & Van Hooijdonk, T. (2014).
Factors influencing casein micelle size in milk of individual cows: Genetic variants
and glycosylation of kappa-casein. International Dairy Journal 34, 135-141.
Blasi, F., Montesano, D., De Angelis, M., Maurizi, A., Ventura, F., Cossignani, L.,
Simonetti, M. S. & Damiani, P. (2008). Results of stereospecific analysis of
triacylglycerol fraction from donkey, cow, ewe, goat and buffalo milk. Journal of
Food Composition and Analysis 21, 1-7.
BOE. (2002). Orden APA/1144/2002. Boletín Oficial del Estado del 23 de Mayo de 2002.
123, 18578-18586.
Bokulich, N. A. & Bamforth, C. W. (2013). The microbiology of malting and brewing.
Microbiology and Molecular Biology Reviews 77, 157-172.
Bongers, G., de Esch, I. & Leurs, R. (2011). Molecular pharmacology of the four
histamine receptors. Advances in experimental medicine and biology 709, 11-19.
Bover Cid, S., Miguélez-Arrizado, M. J., Becker, B., Holzapfel, W. H. & Vidal-Carou,
M. C. (2008). Amino acid decarboxylation by Lactobacillus curvatus CTC273 affected
by the pH and glucose availability. Food Microbiology 25, 269-277.
Bover-Cid, S., Hugas, M., Izquierdo-Pulido, M. & Vidal-Carou, M. C. (2001). Amino
acid-decarboxylase activity of bacteria isolated from fermented pork sausages.
International Journal of Food Microbiology 66, 185-189.
Bramanti, E., Sortino, C., Onor, M., Beni, F. & Raspi, G. (2003). Separation and
determination of denatured alpha(s1)-, alpha(s2)-, beta- and kappa-caseins by
hydrophobic interaction chromatography in cows', ewes' and goats' milk, milk
mixtures and cheeses. Journal of Chromatography A 994, 59-74.
Burdychova, R. & Komprda, T. (2007). Biogenic amine-forming microbial communities
in cheese. Fems Microbiology Letters 276, 149-155.
Cantor, M. D., van den Tempel, T., Hansen, T. K. & Ardö, Y. (2004). Blue cheese. In
Cheese: Chemistry, Physics and Microbiology, pp. 175-198: Academic Press.
Caroli, A. M., Chessa, S. & Erhardt, G. J. (2009). Invited review: Milk protein
polymorphisms in cattle: Effect on animal breeding and human nutrition. Journal of
Dairy Science 92, 5335-5352.
Centeno, J. A., Cepeda, A. & Rodríguez-Otero, J. L. (1995). Identification and
preliminary characterization of strains of enterococci and micrococci isolated from
Arzúa raw cows'-milk cheese. Nahrung-Food 39, 55-62.
Chander, H., Batish, V. K., Babu, S. & Bhatia, K. L. (1988). Studies on optimal
conditions for amine production by E. coli. Milchwissenschaft-Milk Science
International 43, 90-91.
Chander, H., Batish, V. K., Babu, S. & Singh, R. S. (1989). Factors affecting amine
production by a selected strain of Lactobacillus bulgaricus. Journal of Food Science 54,
940-942.
Chawla, R., Patil, G. R. & Singh, A. K. (2011). High hydrostatic pressure technology in
dairy processing: a review. Journal of Food Science and Technology-Mysore 48, 260-268.
69

Introducción
CODEX STAN 238-1978. Norma general del codex para el queso. CODEX
ALIMENTARIUS.
Collins, Y. F., McSweeney, P. L. H. & Wilkinson, M. G. (2003). Lipolysis and free fatty
acid catabolism in cheese: a review of current knowledge. International Dairy Journal
13, 841-866.
Coruzzi, G., Morini, G., Adami, M. & Grandi, D. (2001). Role of histamine H-3
receptors in the regulation of gastric functions. Journal of Physiology and
Pharmacology 52, 539-553.
Coton, E., Rollan, G. C. & Lonvaud-Funel, A. (1998). Histidine carboxylase of
Leuconostoc oenos 9204: Purification, kinetic properties, cloning and nucleotide
sequence of the hdc gene. Journal of Applied Microbiology 84, 143-151.
Coton, M., Delbes-Paus, C., Irlinger, F., Desmasures, N., Le Fleche, A., Stahl, V.,
Montel, M. C. & Coton, E. (2012). Diversity and assessment of potential risk factors
of Gram-negative isolates associated with French cheeses. Food Microbiology 29, 88-
98.
Curiel, J. A., Ruiz-Capillas, C., de las Rivas, B., Carrascosa, A. V., Jiménez-
Colmenero, F. & Muñoz, R. (2011). Production of biogenic amines by lactic acid
bacteria and enterobacteria isolated from fresh pork sausages packaged in different
atmospheres and kept under refrigeration. Meat Science 88, 368-373.
Curioni, P. M. G. & Bosset, J. O. (2002). Key odorants in various cheese types as
determined by gas chromatography-olfactometry. International Dairy Journal 12,
959-984.
Curtin, Á. C. & McSweeney, P. L. H. (2004). Catabolism of amino acids in cheese
during ripening. In Cheese: Chemistry, Physics and Microbiology, pp. 435-454:
Academic Press.
Dalgaard, P., Madsen, H. L., Samieian, N. & Emborg, J. (2006). Biogenic amine
formation and microbial spoilage in chilled garfish (Belone belone belone) - effect
of modified atmosphere packaging and previous frozen storage. Journal of Applied
Microbiology 101, 80-95.
Dalgleish, D. G., Spagnuolo, P. A. & Goff, H. D. (2004). A possible structure of the
casein micelle based on high-resolution field-emission scanning electron
microscopy. International Dairy Journal 14, 1025-1031.
Dalgleish, D. G. & Corredig, M. (2012). The structure of the casein micelle of milk and
its changes during processing. Annual Review of Food Science and Technology, Vol 3 3,
449-467.
de las Rivas, B., Ruiz-Capillas, C., Carrascosa, A. V., Curiel, J. A., Jímenez-
Colmenero, F. & Muñoz, R. (2008). Biogenic amine production by Gram-positive
bacteria isolated from Spanish dry-cured "chorizo" sausage treated with high
pressure and kept in chilled storage. Meat Science 80, 272-277.
Deeth, H. C. (2006). Lipoprotein lipase and lipolysis in milk. International Dairy Journal
16, 555-562.
70

Introducción

Delgado, F. J., González-Crespo, J., Cava, R., García-Parra, J. & Ramírez, R. (2010).
Characterisation by SPME-GC-MS of the volatile profile of a Spanish soft cheese
P.D.O. Torta del Casar during ripening. Food Chemistry 118, 182-189.
Delgado, F. J., González-Crespo, J., Cava, R. & Ramírez, R. (2012). Changes in
microbiology, proteolysis, texture and sensory characteristics of raw goat milk
cheeses treated by high-pressure at different stages of maturation. LWT-Food
Science and Technology 48, 268-275.
Demazeau, G. & Rivalain, N. (2011a). High hydrostatic pressure and biology: a brief
history. Applied Microbiology and Biotechnology 89, 1305-1314.
Demazeau, G. & Rivalain, N. (2011b). The development of high hydrostatic pressure
processes as an alternative to other pathogen reduction methods. Journal of Applied
Du, W. X., Lin, C. M., Phu, A. T., Cornell, J. A., Marshall, M. R. & Wei, C. I. (2002).
Development of biogenic amines in yellowfin tuna (Thunnus albacares): Effect of
storage and correlation with decarboxylase-positive bacterial flora. Journal of Food
Science 67, 292-301.
Durlu-Özkaya, F., Ayhan, K. & Vural, N. (2001). Biogenic amines produced by
Enterobacteriaceae isolated from meat products. Meat Science 58, 163-166.
Eisenmenger, M. J. & Reyes-De-Corcuera, J. I. (2009). High pressure enhancement of
enzymes: A review. Enzyme and Microbial Technology 45, 331-347.
EC 1491/2003. Reglamento (CE) Nº 1491/2003 de la comisión de 25 de agosto de 2003
por el que se completa el Reglamento (CE) nº 2400/96 (Ficodindia dell'Etna, Monte
Etna, Colline di Romagna, Pretuziano delle Colline Teramane, Torta del Casar,
Manzana de Girona or Poma de Girona). Diario Oficial de la Unión Europea. 26-8-
2003. L214/6 - L214/7.
EC 2073/2005. Reglamento (CE) nº 2073/2005 de la comisión de 15 de noviembre de
2005 relativo a los criterios microbiológicos aplicables a los productos alimenticios.
Diario Oficial de la Unión Europea. 22-12-2005. L338/1 – L338/26.
Emborg, J. & Dalgaard, P. (2006). Formation of histamine and biogenic amines in cold-
smoked tuna: An investigation of psychrotolerant bacteria from samples
implicated in cases of histamine fish poisoning. Journal of Food Protection 69, 897-
906.
Evert-Arriagada, K., Hernández-Herrero, M. M., Juan, B., Guamis, B. & Trujillo, A. J.
(2012). Effect of high pressure on fresh cheese shelf-life. Journal of Food Engineering
110, 248-253.
Evert-Arriagada, K., Hernández-Herrero, M. M., Guamis, B. & Trujillo, A. J. (2014).
Commercial application of high-pressure processing for increasing starter-free
fresh cheese shelf-life. LWT-Food Science and Technology 55, 498-505.
Exterkate, F. A., Lagerwerf, F. M., Haverkamp, J. & van Schalkwijk, S. (1997). The
selectivity of chymosin action on alpha(s1)- and beta-caseins in solution is
modulated in cheese. International Dairy Journal 7, 47-54.
FAOSTAT (2014). http://faostat3.fao.org/faostat-gateway/go/to/home/E.
71

Introducción
Farrell, H. M., Jimenez-Flores, R., Bleck, G. T. & other authors (2004). Nomenclature
of the proteins of cows' milk - Sixth revision. Journal of Dairy Science 87, 1641-1674.
FDA. (2011). Scombrotoxin (histamine) formation. In Fish and Fishery Products
Hazards and Controls Guidance. U. S. Department of Health and Human Services. pp.
113-152.
Fernández, M., Linares, D. M., del Río, B., Ladero, V. & Alvarez, M. A. (2007a). HPLC
quantification of biogenic amines in cheeses: correlation with PCR-detection of
tyramine-producing microorganisms. Journal of Dairy Research 74, 276-282.
Fernández, M., Linares, D. M., Rodríguez, A. & Alvarez, M. A. (2007b). Factors
affecting tyramine production in Enterococcus durans IPLA 655. Applied Microbiology
and Biotechnology 73, 1400-1406.
Fox, P. F. & McSweeney, P. L. H. (2004). Cheese: An Overview. In Cheese: Chemistry,
Physics and Microbiology, pp. 1-18: Academic Press.
Fox, P. F. & Kelly, A. L. (2006). Indigenous enzymes in milk: Overview and historical
aspects - Part 1. International Dairy Journal 16, 500-516.
Fox, P. F. & Brodkorb, A. (2008). The casein micelle: Historical aspects, current
concepts and significance. International Dairy Journal 18, 677-684.
Fox, P. F. (2009). Milk: An overview. In Milk proteins: From expression to food, pp. 1-
54. San Diego: Academic Press.
Garde, S., Arqués, J. L., Gaya, P., Medina, M. & Nuñez, M. (2007). Effect of high-
pressure treatments on proteolysis and texture of ewes' raw milk La Serena cheese.
International Dairy Journal 17, 1424-1433.
Gardini, F., Martuscelli, M., Caruso, M. C., Galgano, F., Crudele, M. A., Favati, F.,
Guerzoni, M. E. & Suzzi, G. (2001). Effects of pH, temperature and NaCl
concentration on the growth kinetics, proteolytic activity and biogenic amine
production of Enterococcus faecalis. International Journal of Food Microbiology 64, 105-
117.
Gardini, F., Zaccarelli, A., Belletti, N., Faustini, F., Cavazza, A., Martuscelli, M.,
Mastrocola, D. & Suzzi, G. (2005). Factors influencing biogenic amine production
by a strain of Oenococcus oeni in a model system. Food Control 16, 609-616.
Gardini, F., Tofalo, R., Belletti, N., Iucci, L., Suzzi, G., Torriani, S., Guerzoni, M. E. &
Lanciotti, R. (2006). Characterization of yeasts involved in the ripening of Pecorino
Crotonese cheese. Food Microbiology 23, 641-648.
Giuffrida, D., Ziino, M., Verzera, A., Condurso, C. & Romeo, V. (2006). Biogenic
amines in a typical "pasta filata" Italian cheese. Acta Alimentaria 35, 435-443.
Gobbetti, M., Smacchi, E. & Corsetti, A. (1997). Purification and characterization of a
cell surface-associated esterase from Lactobacillus fermentum DT41. International
Dairy Journal 7, 13-21.
Gresti, J., Bugaut, M., Maniongui, C. & Bezard, J. (1993). Composition of molecular
species of triacylglycerols in bovine milk fat. Journal of Dairy Science 76, 1850-1869.
72

Introducción

Grove, S. F., Forsyth, S., Wan, J., Coventry, J., Cole, M., Stewart, C. M., Lewis, T.,
Ross, T. & Lee, A. (2008). Inactivation of hepatitis A virus, poliovirus and a
norovirus surrogate by high pressure processing. Innovative Food Science &
Emerging Technologies 9, 206-210.
Halász, A., Baráth, A., Simon-Sarkadi, L. & Holzapfel, W. (1994). Biogenic amines and
their production by microorganisms in food. Trends in Food Science & Technology 5,
42-49.
Hernández, I., Barrón, L. J. R., Virto, M., Pérez-Elortondo, F. J., Flanagan, C., Rozas,
U., Nájera, A. I., Albisu, M., Vicente M. S. & de Renobales, M. (2009). Lipolysis,
proteolysis and sensory properties of ewe's raw milk cheese (Idiazabal) made with
lipase addition. Food Chemistry 116, 158-166.
Hernández-Jover, T., Izquierdo-Pulido, M., Veciana-Nogués, M. T., Mariné-Font, A.
& Vidal-Carou, M. C. (1997). Biogenic amine and polyamine contents in meat and
meat products. Journal of Agricultural and Food Chemistry 45, 2098-2102.
Herrero-Fresno, A., Martínez, N., Sánchez-Llana, E., Díaz, M., Fernández, M., Cruz
Martin, M., Ladero, V. & Alvarez, M. A. (2012). Lactobacillus casei strains isolated
from cheese reduce biogenic amine accumulation in an experimental model.
Hill, S. J., Ganellin, C. R., Timmerman, H., Schwartz, J. C., Shankley, N. P., Young, J.
M., Schunack, W., Levi, R. & Haas, H. L. (1997). International union of
pharmacology .13. Classification of histamine receptors. Pharmacological Reviews 49,
253-278.
Hillary, R. A. & Pegg, A. E. (2003). Decarboxylases involved in polyamine biosynthesis
and their inactivation by nitric oxide. Biochimica Et Biophysica Acta-Proteins and
Proteomics 1647, 161-166.
Hite, B. H. (1899). The effect of pressure in the preservation of milk. Bulletin West
Virginia University Agricultural Experiment Station 58, 15 - 35.
Horne, D. S. & Banks, J. M. (2004). Rennet-induced Coagulation of Milk. In Cheese:
Chemistry, Physics and Microbiology, pp. 47-70: Academic Press.
Hu, Y., Huang, Z. Y., Li, J. & Yang, H. (2012). Concentrations of biogenic amines in fish,
squid and octopus and their changes during storage. Food Chemistry 135, 2604-2611.
Huang, H. W., Lung, H. M., Yang, B. B. & Wang, C. Y. (2014). Responses of
microorganisms to high hydrostatic pressure processing. Food Control 40, 250-259.
Huang, J. F. & Thurmond, R. L. (2008). The new biology of histamine receptors. Current
Allergy and Asthma Reports 8, 21-27.
Huppertz, T., Fox, P. F. & Kelly, A. L. (2004). Susceptibility of plasmin and chymosin in
Cheddar cheese to inactivation by high pressure. Journal of Dairy Research 71, 496-
499.
Ismail, B. & Nielsen, S. S. (2010). Invited review: Plasmin protease in milk: Current
knowledge and relevance to dairy industry. Journal of Dairy Science 93, 4999-5009.
Iwańczak, M. & Wiśniewska, K. (2005). Effect of high pressures on the process of
Edam cheese proteolysis. High Pressure Research 25, 43-50.
73

Introducción
Izquierdo-Pulido, M., Font-Fábregas, J., Carceller-Rosa, J. M., Mariné-Font, A. &

Vidal-Carou, C. (1996a). Biogenic amine changes related to lactic acid bacteria
during brewing. Journal of Food Protection 59, 175-180.
Izquierdo-Pulido, M., Hernández-Jover, T., Mariné-Font, A. & Vidal-Carou, M. C.
(1996b). Biogenic amines in European beers. Journal of Agricultural and Food
Chemistry 44, 3159-3163.
Juan, B., Barron, L. J. R., Ferragut, V., Guamis, B. & Trujillo, A. J. (2007a). Changes in
the volatile composition of a semihard ewe milk cheese induced by high-pressure
treatment of 300 MPa. Journal of Agricultural and Food Chemistry 55, 747-754.
Juan, B., Ferragut, V., Buffa, M., Guamis, B. & Trujillo, A. J. (2007b). Effects of high
pressure on proteolytic enzymes in cheese: Relationship with the proteolysis of
ewe milk cheese. Journal of Dairy Science 90, 2113-2125.
Juan, B., Ferragut, V., Guamis, B. & Trujillo, A. J. (2008). The effect of high-pressure
treatment at 300 MPa on ripening of ewes' milk cheese. International Dairy Journal
18, 129-138.
Kalač, P., Špička, J., Křı́žek, M., Steidlová, S. & Pelikánová, T. (1999). Concentrations
of seven biogenic amines in sauerkraut. Food Chemistry 67, 275-280.
Kalač, P. & Křı́žek, M. (2003). A review of biogenic amines and polyamines in beer.
Journal of the Institute of Brewing 109, 123-128.
Kalač, P. (2009). Recent advances in the research on biological roles of dietary
polyamines in man. Journal of Applied Biomedicine 7, 65-74.
Kamath, A. V., Vaaler, G. L. & Snell, E. E. (1991). Pyridoxal phosphate-dependent
histidine descarboxylases - cloning, sequencing, and expression of genes from
Klebsiella planticola and Enterobacter aerogenes and properties of the overexpressed
enzymes. Journal of Biological Chemistry 266, 9432-9437.
Kanki, M., Yoda, T., Ishibashi, M. & Tsukamoto, T. (2004). Photobacterium
phosphoreum caused a histamine fish poisoning incident. International Journal of
Food Microbiology 92, 79-87.
Karovičová, J. & Kohajdová, Z. (2005). Biogenic amines in food. Chemical Papers 59, 70-
79.
Khwanchuea, R., Mulvany, M. J. & Jansakul, C. (2008). Cardiovascular effects of
tyramine: Adrenergic and cholinergic interactions. European Journal of Pharmacology
579, 308-317.
Kim, J. H., Ahn, H. J., Jo, C., Park, H. J., Chung, Y. J. & Byun, M. W. (2004). Radiolysis
of biogenic amines in model system by gamma irradiation. Food Control 15, 405-408.
Kim, N., Maeng, J. S. & Kim, C. T. (2013). Effects of medium high pressure treatments
on protease activity. Food Science and Biotechnology 22, 289-294.
Kinsella, J. E. & Hwang, D. (1976). Biosynthesis of flavors by Penicillium roqueforti.
Biotechnology and Bioengineering 18, 927-938.
74

Introducción

Knoll, J., Miklya, N., Knoll, B., Markó, R. & Rácz, D. (1996). Phenylethylamine and
tyramine are mixed-acting sympathomimetic amines in the brain. Life Sciences 58,
2101-2114.
Koca, N., Balasubramaniam, V. M. & Harper, W. J. (2011). High-pressure effects on the
microstructure, texture, and color of white-brined cheese. Journal of Food Science 76,
E399-E404.
Koessler, K. K., Hanke, M. T. & Sheppard, M. (1928). Production of histamine,
tyramine, bronchospastic and arteriospastic substances in blood broth by pure
cultures of microorganisms. Journal of Infectious Diseases 43, 363-377.
Komprda, T., Smělá, D., Novická, K., Kalhotka, L., Šustová, K. & Pechová, P. (2007).
Content and distribution of blogenic amines in Dutch-type hard cheese. Food
Chemistry 102, 129-137.
Komprda, T., Burdychová, R., Dohnal, V., Cwiková, O. & Sládková, P. (2008a). Some
factors influencing biogenic amines and polyamines content in Dutch-type semi-
hard cheese. European Food Research and Technology 227, 29-36.
Komprda, T., Burdychová, R., Dohnal, V., Cwiková, O., Sládková, P. & Dvořáčková,
H. (2008b). Tyramine production in Dutch-type semi-hard cheese from two
different producers. Food Microbiology 25, 219-227.
Kumar, A., Grover, S., Sharma, J. & Batish, V. K. (2010). Chymosin and other milk
coagulants: sources and biotechnological interventions. Critical Reviews in
Biotechnology 30, 243-258.
Kurt, S. & Zorba, O. (2009). The effects of ripening period, nitrite level and heat
treatment on biogenic amine formation of "sucuk" - A Turkish dry fermented
sausage. Meat Science 82, 179-184.
Ladero, V., Calles-Enríquez, M., Fernández, M. & Alvarez, M. A. (2010). Toxicological
effects of dietary biogenic amines. Current Nutrition & Food Science 6, 145-156.
Landete, J. M., Ferrer, S., Polo, L. & Pardo, I. (2005). Biogenic amines in wines from
three Spanish regions. Journal of Agricultural and Food Chemistry 53, 1119-1124.
Larsen, M. D. & Jensen, K. (1999). The effects of environmental conditions on the
lipolytic activity of strains of Penicillium roqueforti. International Journal of Food
Larsson, M., Zakora, M., Dejmek, P. & Ardö, Y. (2006). Primary proteolysis studied in
a cast cheese made from microfiltered milk. International Dairy Journal 16, 623-632.
Latorre-Moratalla, M. L., Bover-Cid, S., Talon, R., Aymerich, T., Garriga, M., Zanardi,
E., Ianeri, A., Fraqueza, M. J., Elias, M., Drosinos, E. H., Lauková, A. & Vidal-
Carou, M. C. (2010). Distribution of aminogenic activity among potential
autochthonous starter cultures for dry fermented sausages. Journal of Food
Protection 73, 524-528.
Latorre-Moratalla, M. L., Bover-Cid, S., Veciana-Nogués, M. T. & Vidal-Carou, M. C.
(2012). Control of biogenic amines in fermented sausages: role of starter cultures.
Frontiers in Microbiology 3, 169, pp. 1-9.
75

Introducción
Lavizzari, T., Breccia, M., Bover-Cid, S., Vidal-Carou, M. C. & Veciana-Nogues, M. T.

(2010). Histamine, cadaverine, and putrescine produced in vitro by
Enterobacteriaceae and Pseudomonadaceae isolated from spinach. Journal of Food
Leuschner, R. G., Heidel, M. & Hammes, W. P. (1998). Histamine and tyramine
degradation by food fermenting microorganisms. International Journal of Food
Linares, D. M., Fernández, M., Martín, M. C. & Álvarez, M. A. (2009). Tyramine
biosynthesis in Enterococcus durans is transcriptionally regulated by the
extracellular pH and tyrosine concentration. Microbial Biotechnology 2, 625-633.
Linares, D. M., Martín, M. C., Ladero, V., Álvarez, M. A. & Fernández, M. (2011).
Biogenic amines in dairy products. Critical Reviews in Food Science and Nutrition 51,
691-703.
Linares, D. M., Del Río, B., Ladero, V., Martínez, N., Fernández, M., Martín, M. C. &
Álvarez, M. A. (2012). Factors influencing biogenic amines accumulation in dairy
products. Frontiers in Microbiology 3, 180, pp. 1-10.
Loizzo, M. R., Menichini, F., Picci, N., Puoci, F., Spizzirri, U. G. & Restuccia, D.
(2013). Technological aspects and analytical determination of biogenic amines in
cheese. Trends in Food Science & Technology 30, 38-55.
Lonvaud-Funel, A. (2001). Biogenic amines in wines: role of lactic acid bacteria. FEMS
Microbiology Letters 199, 9-13.
Lopez, C. (2011). Milk fat globules enveloped by their biological membrane: Unique
colloidal assemblies with a specific composition and structure. Current Opinion in
Colloid & Interface Science 16, 391-404.
López-Sabater, E. I., Rodriguez-Jerez, J. J., Hernández-Herrero, M. & Mora-Ventura,
M. T. (1994a). Evaluation of histidine decarboxilase activity of bacteria isolated
from sardine (Sardina pilchardus) by an enzymic method. Letters in Applied
López-Sabater, E. I., Rodríguez-Jerez, J. J., Roig-Sagués, A. X. & Mora-Ventura, M. A.
T. (1994b). Bacteriological quality of tuna fish (Thunnus thynnus) destined for
canning: Effect of tuna handling on presence of histidine decarboxilase bacteria
and histamine level. Journal of Food Protection 57, 318-323.
Loret, S., Deloyer, P. & Dandrifosse, G. (2005). Levels of biogenic amines as a measure
of the quality of the beer fermentation process: Data from Belgian samples. Food
Chemistry 89, 519-525.
Lu, Y. M., Chen, X. H., Jiang, M., Lv, X., Rahman, N., Dong, M. S. & Gujun, Y. (2009).
Biogenic amines in Chinese soy sauce. Food Control 20, 593-597.
Lucey, J. A. & Singh, H. (1997). Formation and physical properties of acid milk gels: a
review. Food Research International 30, 529-542.
Lucey, J. A., Johnson, M. E. & Horne, D. S. (2003). Invited review: Perspectives on the
basis of the rheology and texture properties of cheese. Journal of Dairy Science 86,
2725-2743.
76

Introducción

Lyte, M. (2004). The Biogenic amine tyramine modulates the adherence of Escherichia
coli O157:H7 to intestinal mucosa. Journal of Food Protection 67, 878-883.
MAGRAMA (2014). http://www.magrama.gob.es/
Maijala, R. (1994). Histamine and tyramine production by a lactobacillus strain
subjected to external pH decrease. Journal of Food Protection 57, 259-262.
Maijala, R. L., Eerola, S. H., Aho, M. A. & Hirn, J. A. (1993). The efffect of GDL-
induced pH decrease on the formation of biogenic amines in meat. Journal of Food
Maintz, L. & Novak, N. (2007). Histamine and histamine intolerance. American Journal
of Clinical Nutrition 85, 1185-1196.
Malacarne, M., Martuzzi, F., Summer, A. & Mariani, P. (2002). Protein and fat
composition of mare's milk: some nutritional remarks with reference to human and
cow's milk. International Dairy Journal 12, 869-877.
Malone, A. S., Wick, C., Shellhammer, T. H. & Courtney, P. D. (2003). High pressure
effects on proteolytic and glycolytic enzymes involved in cheese manufacturing.
Maniou, D., Tsala, A., Moschopoulou, E., Giannoglou, M., Taoukis, P. & Moatsou, G.
(2013). Effect of high-pressure-treated starter on ripening of Feta cheese. Dairy
Science & Technology 93, 11-20.
Marcobal, A., Martín-Álvarez, P. J., Moreno-Arribas, M. V. & Muñoz, R. (2006). A
multifactorial design for studying factors influencing growth and tyramine
production of the lactic acid bacteria Lactobacillus brevis CECT 4669 and
Enterococcus faecium BIFI-58. Research in Microbiology 157, 417-424.
Marilley, L. & Casey, M. G. (2004). Flavours of cheese products: metabolic pathways,
analytical tools and identification of producing strains. International Journal of Food
Marino, M., Maifreni, M., Moret, S. & Rondinini, G. (2000). The capacity of
Enterobacteriaceae species to produce biogenic amines in cheese. Letters in Applied
Marino, M., Maifreni, M., Bartolomeoli, I. & Rondinini, G. (2008). Evaluation of
amino acid-decarboxylative microbiota throughout the ripening of an Italian PDO
cheese produced using different manufacturing practices. Journal of Applied
Martínez-Rodríguez, Y., Acosta-Muñiz, C., Olivas, G. I., Guerrero-Beltrán, J.,
Rodrigo-Aliaga, D. & Sepulveda, D. R. (2012). High hydrostatic pressure
processing of cheese. Comprehensive Reviews in Food Science and Food Safety 11, 399-
416.
Martínez-Rodríguez, Y., Acosta-Muñiz, C., Olivas, G. I., Guerrero-Beltrán, J.,
Rodrigo-Aliaga, D., Mujica-Paz, H., Welti-Chanes, J. & Sepulveda, D. R. (2014).
Effect of high hydrostatic pressure on mycelial development, spore viability and
enzyme activity of Penicillium Roqueforti. International Journal of Food Microbiology
168, 42-46.
77

Introducción
Mayer, H. K., Fiechter, G. & Fischer, E. (2010). A new ultra-pressure liquid

chromatography method for the determination of biogenic amines in cheese.
Journal of Chromatography A 1217, 3251-3257.
McCabe-Sellers, B. J., Staggs, C. G. & Bogle, M. L. (2006). Tyramine in foods and
monoamine oxidase inhibitor drugs: A crossroad where medicine, nutrition,
pharmacy, and food industry converge. Journal of Food Composition and Analysis 19,
S58-S65.
McSweeney, P. L. H. (2004). Biochemistry of cheese ripening: Introduction and
overview. In Cheese: Chemistry, Physics and Microbiology, pp. 347-360: Academic
Press.
McSweeney, P. L. H. & Fox, P. F. (2004). Metabolism of residual lactose and of lactate
and citrate. In Cheese: Chemistry, Physics and Microbiology, pp. 361-371:
Academic Press.
McSweeney, P. L. H., Ottogalli, G. & Fox, P. F. (2004). Diversity of cheese varieties: An
overview. In Cheese: Chemistry, Physics and Microbiology, pp. 1-23: Academic
Press.
Medina, M. A., Urdiales, J. L., Rodríguez-Caso, C., Ramírez, F. J. & Sánchez-Jiménez,
F. (2003). Biogenic amines and polyamines: Similar biochemistry for different
physiological missions and biomedical applications. Critical Reviews in Biochemistry
and Molecular Biology 38, 23-59.
Menéndez, S., Centeno, J. A., Godı́nez, R. & Rodrı́guez-Otero, J. L. (2000). Effects of
Lactobacillus strains on the ripening and organoleptic characteristics of Arzúa-Ulloa
cheese. International Journal of Food Microbiology 59, 37-46.
Morales, P., Calzada, J., Rodríguez, B., de Paz, M., Gaya, P. & Nuñez, M. (2006). Effect
of cheese water activity and carbohydrate content on the barotolerance of Listeria
monocytogenes Scott A. Journal of Food Protection 69, 1328-1333.
Moreno-Arribas, V. & Lonvaud-Funel, A. (1999). Tyrosine decarboxylase activity of
Lactobacillus brevis IOEB 9809 isolated from wine and L. brevis ATCC 367. FEMS
Microbiology Letters 180, 55-60.
Moret, S., Smela, D., Populin, T. & Conte, L. S. (2005). A survey on free biogenic
amine content of fresh and preserved vegetables. Food Chemistry 89, 355-361.
Moschopoulou, E., Anisa, T., Katsaros, G., Taoukis, P. & Moatsou, G. (2010).
Application of high-pressure treatment on ovine brined cheese: Effect on
composition and microflora throughout ripening. Innovative Food Science &
Emerging Technologies 11, 543-550.
Mota, M. J., Lopes, R. P., Delgadillo, I. & Saraiva, J. A. (2013). Microorganisms under
high pressure - Adaptation, growth and biotechnological potential. Biotechnology
Advances 31, 1426-1434.
Mulder, H. (1952). Taste and flavour forming substances in cheese. Netherlands Milk and
Dairy Journal 6, 157 - 167.
78

Introducción

Mújica-Paz, H., Valdez-Fragoso, A., Samson, C. T., Welti-Chanes, J. & Torres, J. A.
(2011). High-pressure processing technologies for the pasteurization and
sterilization of foods. Food and Bioprocess Technology 4, 969-985.
Naila, A., Flint, S., Fletcher, G., Bremer, P. & Meerdink, G. (2010). Control of biogenic
amines in food-existing and emerging approaches. Journal of Food Science 75, R139-
R150.
Naila, A., Flint, S., Fletcher, G. C., Bremer, P. J. & Meerdink, G. (2011). Biogenic
amines and potential histamine - Forming bacteria in Rihaakuru (a cooked fish
paste). Food Chemistry 128, 479-484.
Naranjo, P. (1966). Toxicity of histamine: lethal doses. In Histamine and Anti-
Histaminics, pp. 179-201. Edited by M. Rocha e Silva: Springer Berlin Heidelberg.
Nguyen Thi Minh, H., Dantigny, P., Perrier-Cornet, J.-M. & Gervais, P. (2010).
Germination and inactivation of Bacillus subtilis spores induced by moderate
hydrostatic pressure. Biotechnology and bioengineering 107, 876-883.
Novella-Rodríguez, S., Veciana-Nogués, M. T., Izquierdo-Pulido, M. & Vidal-Carou,
M. C. (2003). Distribution of biogenic amines and polyamines in cheese. Journal of
Food Science 68, 750-755.
O'Reilly, C. E., O'Connor, P. M., Kelly, A. L., Beresford, T. P. & Murphy, P. M. (2000).
Use of hydrostatic pressure for inactivation of microbial contaminants in cheese.
Applied and Environmental Microbiology 66, 4890-4896.
Ordiales, E., Benito, M. J., Martín, A., Casquete, R., Serradilla, M. J. & Córdoba, M. D.
(2013a). Bacterial communities of the traditional raw ewe's milk cheese "Torta del
Casar" made without the addition of a starter. Food Control 33, 448-454.
Ordiales, E., Martín, A., Benito, M. J., Hernández, A., Ruiz-Moyano, S. & Córdoba,
M. D. (2013b). Role of the microbial population on the flavor of the soft-bodied
cheese Torta del Casar. Journal of Dairy Science 96, 5477-5486.
Osintsev, A. M. & Qvist, K. B. (2004). Study of the mechanism of the proteolytic stage
of enzymatic coagulation of milk casein. Colloid Journal 66, 192-196.
Özoğul, F. (2004). Production of biogenic amines by Morganella morganii, Klebsiella
pneumoniae and Hafnia alvei using a rapid HPLC method. European Food Research
and Technology 219, 465-469.
Ozturk, M., Govindasamy-Lucey, S., Jaeggi, J. J., Johnson, M. E. & Lucey, J. A. (2013).
The influence of high hydrostatic pressure on regular, reduced, low and no salt
added Cheddar cheese. International Dairy Journal 33, 175-183.
Pachlová, V., Buňka, F., Flasarová, R., Válková, P. & Buňková, L. (2012). The effect of
elevated temperature on ripening of Dutch type cheese. Food Chemistry 132, 1846-
1854.
Park, Y. W., Juarez, M., Ramos, M. & Haenlein, G. F. W. (2007). Physico-chemical
characteristics of goat and sheep milk. Small Ruminant Research 68, 88-113.
79

Introducción
Penas, E., Frias, J., Sidro, B. & Vidal-Valverde, C. (2010). Impact of fermentation
conditions and refrigerated storage on microbial quality and biogenic amine
content of sauerkraut. Food Chemistry 123, 143-150.
Prieto, B., Franco, I., Fresno, J. M., Bernardo, A. & Carballo, J. (2000). Picon Bejes-
Tresviso blue cheese: an overall biochemical survey throughout the ripening
process. International Dairy Journal 10, 159-167.
Qian, M., Nelson, C. & Bloomer, S. (2002). Evaluation of fat-derived aroma
compounds in Blue cheese by dynamic headspace GC/olfactometry-MS. Journal of
the American Oil Chemists Society 79, 663-667.
Raynal-Ljutovac, K., Lagriffoul, G., Paccard, P., Guillet, I. & Chilliard, Y. (2008).
Composition of goat and sheep milk products: An update. Small Ruminant Research
79, 57-72.
Recsei, P. A. & Snell, E. E. (1985). Pyruvoyl-dependent histidine decarboxylases -
mechanism of cleavage of the proenzyme from Lactobacillus buchneri. Journal of
Biological Chemistry 260, 2804-2806.
Rehman, S. U., Banks, J. M., Brechany, E. Y., Muir, D. D., McSweeney, P. L. H. & Fox,
P. F. (2000). Influence of ripening temperature on the volatiles profile and flavour
of Cheddar cheese made from raw or pasteurised milk. International Dairy Journal
10, 55-65.
Reineke, K., Mathys, A. & Knorr, D. (2011). The impact of high pressure and
temperature on bacterial spores: Inactivation mechanisms of bacillus subtilis above
500 MPa. Journal of Food Science 76, M189-M197.
Rendueles, E., Omer, M. K., Alvseike, O., Alonso-Calleja, C., Capita, R. & Prieto, M.
(2011). Microbiological food safety assessment of high hydrostatic pressure
processing: A review. LWT-Food Science and Technology 44, 1251-1260.
Reps, A., Kolakowski, P., Wisniewska, K. & Krasowska, M. (2003). The effect of high
pressures on proteolytic enzymes in ripened cheeses. Milchwissenschaft-Milk Science
Rivalain, N., Roquain, J. & Demazeau, G. (2010). Development of high hydrostatic
pressure in biosciences: Pressure effect on biological structures and potential
applications in Biotechnologies. Biotechnology Advances 28, 659-672.
Rodríguez-Alonso, P., Centeno, J. A. & Garabal, J. I. (2009). Comparison of the volatile
profiles of Arzúa-Ulloa and Tetilla cheeses manufactured from raw and
pasteurized milk. LWT - Food Science and Technology 42, 1722-1728.
Roig-Sagués, A. X., Molina, A. P. & Hernández-Herrero, M. M. (2002). Histamine and
tyramine-forming microorganisms in Spanish traditional cheeses. European Food
Research and Technology 215, 96-100.
Romero, R., Bagur, M. G., Sánchez-Viñas, M. & Gázquez, D. (2003). The influence of
the brewing process on the formation of biogenic amines in beers. Analytical and
Bioanalytical Chemistry 376, 162-167.
80

Introducción

Roseiro, L. B., Barbosa, M., Ames, J. M. & Wilbey, R. A. (2003). Cheesemaking with
vegetable coagulants - the use of Cynara L. for the production of ovine milk
cheeses. International Journal of Dairy Technology 56, 76-85.
Ruiz-Capillas, C. & Jiménez-Colmenero, F. (2004a). Biogenic amines in meat and meat
products. Critical Reviews in Food Science and Nutrition 44, 489-499.
Ruiz-Capillas, C. & Jiménez-Colmenero, F. (2004b). Biogenic amine content in Spanish
retail market meat products treated with protective atmosphere and high pressure.
European Food Research and Technology 218, 237-241.
Rynne, N. M., Beresford, T. P., Guinee, T. P., Sheehan, E., Delahunty, C. M. & Kelly,
A. L. (2008). Effect of high-pressure treatment of 1 day-old full-fat Cheddar cheese
on subsequent quality and ripening. Innovative Food Science & Emerging Technologies
9, 429-440.
Salque, M., Bogucki, P. I., Pyzel, J., Sobkowiak-Tabaka, I., Grygiel, R., Szmyt, M. &
Evershed, R. P. (2013). Earliest evidence for cheese making in the sixth millennium
BC in northern Europe. Nature 493, 522-525.
San Martin, M. F., Barbosa-Canovas, G. V. & Swanson, B. G. (2002). Food processing
by high hydrostatic pressure. Critical Reviews in Food Science and Nutrition 42, 627-
645.
Sarantinopoulos, P., Kalantzopoulos, G. & Tsakalidou, E. (2001). Citrate metabolism
by Enterococcus faecalis FAIR-E 229. Applied and Environmental Microbiology 67, 5482-
5487.
Savijoki, K., Ingmer, H. & Varmanen, P. (2006). Proteolytic systems of lactic acid
bacteria. Applied Microbiology and Biotechnology 71, 394-406.
Serrano, J., Velazquez, G., Lopetcharat, K., Ramirez, J. A. & Torres, J. A. (2005).
Moderately high hydrostatic pressure processing to reduce production costs of
shredded cheese: Microstructure, texture, and sensory properties of shredded
milled curd cheddar. Journal of Food Science 70, S286-S293.
Seyderhelm, I., Boguslawski, S., Michaelis, G. & Knorr, D. (1996). Pressure induced
inactivation of selected food enzymes. Journal of Food Science 61, 308-310.
Shah, M. A., Mir, S. A. & Paray, M. A. (2014). Plant proteases as milk-clotting enzymes
in cheesemaking: a review. Dairy Science & Technology 94, 5-16.
Shalaby, A. R. (1996). Significance of biogenic amines to food safety and human health.
Food Research International 29, 675-690.
Silla Santos, M. H. (1996). Biogenic amines: their importance in foods. International
Journal of Food Microbiology 29, 213-231.
Smit, G., Smit, B. A. & Engels, W. J. M. (2005). Flavour formation by lactic acid
bacteria and biochemical flavour profiling of cheese products. Fems Microbiology
Reviews 29, 591-610.
Sousa, M. J., Ardo, Y. & McSweeney, P. L. H. (2001). Advances in the study of
proteolysis during cheese ripening. International Dairy Journal 11, 327-345.
81

Introducción
Spano, G., Russo, P., Lonvaud-Funel, A., Lucas, P., Alexandre, H., Grandvalet, C.,
Coton, E., Coton, M., Barnavon, L., Bach, B., Rattray, F., Bunte, A., Magni, C.,
Ladero, V., Alvarez, M., Fernández, M., Lopez, P., de Palencia, P. F., Corbi, A.,
Trip, H. & Lolkema, J.S. (2010). Biogenic amines in fermented foods. European
Journal of Clinical Nutrition 64, S95-S100.
Spinnler, H. E. & Gripon, J. C. (2004). Surface mould-ripened cheeses. In Cheese:
Stewart, D. I., Kelly, A. L., Gulinee, T. P. & Beresford, T. P. (2006). High pressure
processing: review of application to cheese manufacture and ripening. Australian
Journal of Dairy Technology 61, 170-178.
Stratton, J. E., Hutkins, R. W. & Taylor, S. L. (1991). Biogenic amines in cheese and
other fermented foods: A review. Journal of Food Protection 54, 460-470.
Sumner, S. S., Roche, F. & Taylor, S. L. (1990). Factors controlling histamine
production in swiss cheese inoculated with lactobacillus buchneri. Journal of Dairy
Science 73, 3050-3058.
Suzzi, G. & Gardini, F. (2003). Biogenic amines in dry fermented sausages: a review.
Tabanelli, G., Torriani, S., Rossi, F., Rizzotti, L. & Gardini, F. (2012). Effect of
chemico-physical parameters on the histidine decarboxylase (HdcA) enzymatic
activity in Streptococcus thermophilus PRI60. Journal of Food Science 77, M231-M237.
ten Brink, B., Damink, C., Joosten, H. & JHJ., H. i. t. V. (1990). Occurrence and
formation of biologically-active amines in foods. International Journal of Food
Til, H. P., Falke, H. E., Prinsen, M. K. & Willems, M. I. (1997). Acute and subacute
toxicity of tyramine, spermidine, spermine, putrescine and cadaverine in rats. Food
and Chemical Toxicology 35, 337-348.
Tsai, Y. H., Kung, H. D., Lee, T. M., Lin, G. T. & Hwang, D. F. (2004). Histamine-
related hygienic qualities and bacteria found in popular commercial scombroid fish
fillets in Taiwan. Journal of Food Protection 67, 407-412.
UE 20/2010. Reglamento (UE) Nº 20/2010 de la comisión de 12 de enero de 2010 por el
que se inscribe una denominación en el Registro de Denominaciones de Origen
Protegidas y de Indicaciones Geográficas Protegidas [Arzúa-Ulloa (DOP)]. Diario
Oficial de la Unión Europea. 13-1-2010. L8/1 – L8/2.
Upadhyay, V. K., McSweeney, P. L. H., Magboul, A. A. A. & Fox, P. F. (2004).
Proteolysis in cheese during ripening. In Cheese: Chemistry, Physics and
Microbiology, pp. 391-VIII: Academic Press.
Upadhyay, V. K., Huppertz, T., Kelly, A. L. & McSweeney, P. L. H. (2007). Use of high
pressure treatment to attenuate starter bacteria for use as adjuncts for Cheddar
cheese manufacture. Innovative Food Science & Emerging Technologies 8, 485-492.
Veciana Nogues, M. T., Marine Font, A. & Vidal Carou, M. C. (1997). Biogenic amines
in fresh and canned tuna. Effects of canning on biogenic amine contents. Journal of
Agricultural and Food Chemistry 45, 4324-4328.
82

Introducción

Verissimo, P., Esteves, C., Faro, C. & Pires, E. (1995). The vegetable rennet of Cynara
cardunculus L. contains 2-proteinases with chymosin and pepsin-like specificities.
Biotechnology Letters 17, 621-626.
Visciano, P., Schirone, M., Tofalo, R. & Suzzi, G. (2012). Biogenic amines in raw and
processed seafood. Frontiers in Microbiology 3, 188, pp. 1-10.
Visser, S. (1993). Proteolytic enzymes and their relation to cheese ripening and flavor:
an overview. Journal of Dairy Science 76, 329-350.
Voigt, D. D., Chevalier, F., Qian, M. C. & Kelly, A. L. (2010). Effect of high-pressure
treatment on microbiology, proteolysis, lipolysis and levels of flavour compounds
in mature blue-veined cheese. Innovative Food Science & Emerging Technologies 11,
68-77.
Wallace, H. M., Fraser, A. V. & Hughes, A. (2003). A perspective of polyamine
metabolism. Biochemical Journal 376, 1-14.
Weemaes, C., Ludikhuyze, L., Van den Broeck, I. & Hendrickx, M. (1998). High
pressure inactivation of polyphenoloxidases. Journal of Food Science 63, 873-877.
Wick, C., Nienabert, U., Anggraeni, O., Shellhammer, T. H. & Courtney, P. D. (2004).
Texture, proteolysis and viable lactic acid bacteria in commercial cheddar cheeses
treated with high pressure. Journal of Dairy Research 71, 107-115.
Wyder, M. T., Bachmann, H. P. & Puhan, Z. (1999). Role of selected yeasts in cheese
ripening: An evaluation in foil wrapped Raclette cheese. Food Science and
Technology-Lebensmittel-Wissenschaft & Technologie 32, 333-343.
Yaldagard, M., Mortazavi, S. A. & Tabatabaie, F. (2008). The principles of ultra high
pressure technology and its application in food processing/preservation: A review
of microbiological and quality aspects. African Journal of Biotechnology 7, 2739-2767.
Yuksel, Z., Avci, E., Uymaz, B. & Erdem, Y. K. (2012). General composition and protein
surface hydrophobicity of goat, sheep and cow milk in the region of Mount Ida.
Small Ruminant Research 106, 137-144.
Yvon, M. & Rijnen, L. (2001). Cheese flavour formation by amino acid catabolism.
83

Objetivos
 Objetivos de la Tesis
El presente trabajo se realizó en el Departamento de Tecnología de Alimentos del

Instituto Nacional de Investigación y Tecnología Agraria y Alimentaria (INIA), integrado
en el proyecto AGL 2009-07801 “Prevención de la sobremaduración y de la formación
de aminas biógenas en quesos mediante tratamientos por altas presiones” del Plan
Nacional de I+D+I. Durante su realización, su autor fue beneficiario de una beca de
Formación de Personal Investigador del Ministerio de Ciencia e Innovación, con
referencia BES-2010-030444.
Los fundamentos de la investigación se basan en la existencia de quesos con una

intensa proteolisis, debida al empleo de coagulantes que contienen enzimas de elevada
actividad proteolítica o a la inoculación de mohos del género Penicillium productores
de proteinasas extracelulares de elevada actividad proteolítica. Ello da lugar durante la
conservación en refrigeración a un exceso de maduración o sobremaduración,
responsable de que el producto llegue al consumidor con menor calidad de la deseada
por el fabricante, lo que disminuye la aceptabilidad dentro de su vida comercial.
Además, en algunos quesos, en particular si se elaboran con leche cruda, se encuentran
microorganismos con actividad descarboxilasa capaces de generar niveles elevados de
aminas biógenas, con efectos perjudiciales sobre la salud del consumidor.
Para controlar ambos efectos indeseables, se planteó el presente trabajo de

investigación con dos objetivos principales: prevenir la sobremaduración de los quesos
alargando su vida útil y evitar la acumulación de aminas biógenas en los quesos. El
instrumento empleado para ello fue la aplicación de tratamientos de altas presiones
hidrostáticas (400 y 600 MPa) a cuatro variedades de quesos elaborados en España, en
diferentes momentos de su periodo de maduración.
Para poder valorar la consecución de estos objetivos, se monitorizó el efecto de los

diferentes tratamientos de altas presiones, a lo largo de un periodo prolongado de
conservación, sobre la proteolisis, la lipolisis, la formación de compuestos volátiles, la
textura, el color, las características organolépticas y la formación de aminas biógenas en
queso azul, Torta del Casar, queso Brie y queso Arzúa-Ulloa.
85

Capítulo 2.
Proteolysis and biogenic amine buildup in high-pressure
treated ovine milk blue-veined cheese.

87

Fotografía: corte de queso azul (Roncari-blue).

J. Dairy Sci. 96:4816–4829
http://dx.doi.org/10.3168/jds.2012-6409
© American Dairy Science Association®, 2013.
Proteolysis and biogenic amine buildup in high-pressure

treated ovine milk blue-veined cheese
J. Calzada, A. Del Olmo, A. Picon, P. Gaya, and M. Nuñez1
Departamento de Tecnología de Alimentos, Instituto Nacional de Investigación y Tecnología Agraria y Alimentaria, 28040 Madrid, Spain
ABSTRACT MPa) at an early ripening stage (3 wk), which affected

biochemical changes and sensory characteristics.
Penicillium roqueforti plays an important role in the Key words: blue cheese, high pressure, biogenic
ripening of blue-veined cheeses, mostly due to lactic amine, proteolysis
acid consumption and to its extracellular enzymes. The
strong activity of P. roqueforti proteinases may bring
about cheese over-ripening. Also, free amino acids at INTRODUCTION
high concentrations serve as substrates for biogenic
amine formation. Both facts result in shorter product Indigenous milk enzymes, coagulant enzymes, lactic
shelf-life. To prevent over-ripening and buildup of bio- acid bacteria (LAB) from starter cultures, and other
genic amines, blue-veined cheeses made from pasteur- added or contaminating microorganisms and their en-
ized ovine milk were high-pressure treated at 400 or zymes degrade milk proteins to peptides and free AA
600 MPa after 3, 6, or 9 wk of ripening. Primary and during cheese manufacture and ripening (Visser, 1993).
secondary proteolysis, biogenic amines, and sensory Peptides contribute to cheese flavor, in most cases
characteristics of pressurized and control cheeses were positively, although high levels of some hydrophobic
monitored for a 90-d ripening period, followed by a 270- peptides have been associated with cheese bitterness
d refrigerated storage period. On d 90, treatments at (Gomez et al., 1997). Free AA are directly involved in
400 MPa had lowered counts of lactic acid bacteria and cheese taste and serve also as precursors of compounds
P. roqueforti by less than 2 log units, whereas treatments responsible for flavor and aroma (Yvon and Rijnen,
at 600 MPa had reduced lactic acid bacteria counts by 2001). In blue-veined cheeses, Penicillium roqueforti is
more than 4 log units and P. roqueforti counts by more a major proteolytic agent, indirectly because of lactic
than 6 log units. No residual α-casein (CN) or κ-CN acid consumption, which brings about a rise in cheese
were detected in control cheese on d 90. Concentrations pH favorable for many chemical reactions, and di-
of β-CN, para-κ-CN, and γ-CN were generally higher in rectly due to its highly active extracellular proteinases
600 MPa cheeses than in the rest. From d 90 onwards, (Gripon et al., 1977).
hydrophilic peptides were at similar levels in pressur- As ripening continues during storage, distribution,
ized and control cheeses, but hydrophobic peptides and retail, the cheese purchased by the consumer may
and the hydrophobic-to-hydrophilic peptide ratio were be of a stronger or different flavor than the manufac-
at higher levels in pressurized cheeses than in control turer intended (Wick et al., 2004). Cheeses suffering
cheese. Aminopeptidase activity, overall proteolysis, extensive proteolysis are particularly prone to undesir-
and free amino acid contents were generally higher in able effects such as over-ripening associated to off-flavor
control cheese than in pressurized cheeses, particularly development, which bring about shorter product shelf-
if treated at 600 MPa. Tyramine concentration was life. In addition, the free AA generated at high levels
lower in pressurized cheeses, but tryptamine, phenyle- in strongly proteolyzed cheeses may serve as substrates
thylamine, and putrescine contents were higher in some for bacterial decarboxylases, resulting in the accumula-
of the pressurized cheeses than in control cheese. Dif- tion of biogenic amines (BA), a group of compounds of
ferences in sensory characteristics between pressurized notable public health significance (Silla Santos, 1996).
and control cheeses were generally negligible, with the Tyramine is a potent vasoconstrictor with an effect
only exception of treatment at high pressure level (600 on healthy individuals usually limited to headache or
migraine, whereas histamine may cause urticaria, hy-
potension, headache, flushing, and abdominal cramps
(Taylor, 1986; Til et al., 1997). In spite of the wide
consensus on their toxicity, acceptable levels of BA
Received November 23, 2012.
Accepted April 10, 2013. in foods have not been established in most countries,
1
Corresponding author: nunez@inia.es partly because of the variable sensitivity of individuals
4816
PROTEOLYSIS AND BIOGENIC AMINES IN BLUE CHEESE 4817
to these compounds, which may be affected by factors Roquefort, has not been studied. The objective of the
such as alcohol consumption. Concentrations of 100 to present work was to investigate the effect of HP treat-
800 mg of tyramine per kilogram and 30 mg of phenyl- ments at 400 or 600 MPa, applied after 3, 6 or 9 wk
ethylamine per kilogram have been suggested as toxic of ripening, on the primary and secondary proteolysis,
(ten Brink et al., 1990), but only an upper limit of 100 the formation of BA, and the sensory characteristics of
mg of histamine per kilogram has been set for fish and ovine milk blue-veined cheese.
fish products.
Monoamines tyramine, phenylethylamine, histamine,
MATERIALS AND METHODS
and tryptamine are respectively formed through the
decarboxylation of tyrosine, phenylalanine, histidine, Cheese-Making and HP Treatments
and tryptophan, whereas diamines cadaverine and
putrescine derive from lysine and ornithine or argi- Pasteurized (73°C for 16 s) ovine milk was used for
nine via agmatine (Silla Santos, 1996). Enterococci are the manufacture of blue-veined cheese in duplicate trials
generally considered the main tyramine formers and carried out on consecutive days. Lactic cultures (Flora
heterofermentative lactobacilli the main producers of Danica, 120 g; Chr. Hansen, Tres Cantos, Spain), P.
histamine, but other cheese-borne bacteria may be roqueforti (Choozit PA, 1 dose; Danisco, Sassenage,
also involved in BA formation in cheese (Edwards and France), and CaCl2 (120 g, in aqueous solution) were
Sandine, 1981; Joosten and Northolt, 1987; Pircher et added to milk. Milk (1,200 L) was coagulated at 32°C
al., 2007). for 35 min with animal rennet (Naturen Premium, 200
Even though milk hygienization procedures, such mL; Chr. Hansen). Curd was cut into 1.5-cm cubes and
as bactofugation and pasteurization, lower the levels held at 32°C for 20 min. Whey was drained out and
of decarboxylase-positive bacteria, postpasteuriza- the nonpressed curd was distributed into cylindrical
tion contamination of milk and curd by adventitious molds. Cheeses, 18 cm in diameter and 10 cm high,
bacteria-harboring AA decarboxylases, and their sub- were turned over 5 times while held at 21°C for 24 h,
sequent growth and metabolic activity during cheese followed by another 48 h at 15°C. Afterward, they were
ripening, usually results in BA buildup. High-pressure salted by rubbing dry salt onto all the surfaces and
(HP) treatments efficiently inactivate microbial con- pierced from the 2 flat surfaces. Ripening took place at
taminants in milk and cheese (O’Reilly et al., 2000; 10°C and 93% relative humidity until d 30 and at 5°C
Arqués et al., 2006). They offer the advantage of ap- from d 30 to d 90. Refrigerated storage at 3°C lasted
plication after cheese manufacture, when the contami- until d 360.
nation of the cheese interior is no longer possible, with Cheeses from each trial were pressurized for 5 min
the only exception being blue-veined cheeses, which are at 400 or 600 MPa after ripening for 3, 6 or 9 wk and
pierced some days after manufacture. The effect of HP coded as 400W3, 600W3, 400W6, 600W6, 400W9, and
treatments on the characteristics of cheeses made from 600W9, respectively. Before HP treatments, cheeses
bovine milk (O’Reilly et al., 2003; Evert-Arriagada et were vacuum-packaged in CN300 bags (Cryovac Grace
al., 2012), ovine milk (Arqués et al., 2007; Juan et al., S.A., Barcelona, Spain). A 120-L capacity isostatic
2007), and caprine milk (Saldo et al., 2002; Delgado press (NC Hyperbaric, Burgos, Spain) was used for HP
et al., 2012) has been investigated. Pressurization may treatments. Come-up times to reach 400 and 600 MPa
also be useful to impede BA formation during cheese were 1.82 and 2.65 min and depressurization times were
ripening. To our knowledge, HP treatments with this 6 and 8 s, respectively. Temperature of the water used
objective have been assayed only on a pasteurized as transmitting fluid remained under 13°C during the
goat milk cheese, but BA contents of control cheese process. After HP treatments cheeses were unpackaged,
in the experiment were so low that it was not possible ripening and storage proceeded under the same condi-
to establish differences with respect to the pressurized tions as for control cheese. A total of 44 cheeses per
cheese (Novella-Rodríguez et al., 2002). trial (1 per treatment and sampling date) were used for
The effect of HP treatments on the chemical charac- analytical determinations.
teristics of blue-veined cheeses has been studied only
on an Irish cow milk blue cheese, which was pressurized Microbiological Analyses
on d 42 of ripening and stored afterward at 4°C for
28 d (Voigt et al., 2010). However, no information is Representative cheese samples (10 g) were homog-
available on the effect of HP treatments on BA for- enized with 90 mL of a sterile 2% (wt/vol) sodium
mation in blue-veined cheeses. Moreover, the effect of citrate solution at 45°C in a Colworth Stomacher 400
pressurization on the characteristics of ovine milk blue- (A. J. Seward Ltd., London, UK). Decimal dilutions
veined cheeses, which include reputed varieties such as of samples were prepared in sterile 0.1% peptone
Journal of Dairy Science Vol. 96 No. 8, 2013
4818 CALZADA ET AL.
solution. Total viable counts, LAB, enterococci, and ternal standard area and expressed as milligrams of
lactobacilli were determined in duplicate on plates protein per grams of cheese DM.
of PCA (Biolife, Milano, Italy) incubated for 48 h at Hydrophilic and hydrophobic peptides in the water-
30°C, M17 agar (acidified at pH 5.7 with acetic acid; soluble fraction of cheese were determined on dupli-
Biolife) incubated for 48 h at 30°C, KF Streptococcus cate samples by reverse-phase HPLC according to a
agar (Oxoid, Basingstoke, UK) incubated for 48 h at previously described method (Lau et al., 1991), using
37°C, and Rogosa agar (acidified at pH 5.4 with acetic a Beckman System Gold chromatograph (Beckman
acid; Biolife) incubated anaerobically for 48 h at 37°C, Instruments España S.A.) equipped with a diode array
respectively. Micrococcaceae were determined in dupli- detector module 168 with detection wavelength at 280
cate on plates of mannitol salt agar (Oxoid) incubated nm. Peaks with retention times from 5.5 to 14.6 min
for 72 h at 30°C, coagulase-positive staphylococci were were considered to correspond to hydrophilic peptides
determined on Baird-Parker agar (Oxoid) with RPF and those with retention times from 14.6 to 20.5 min to
Supplement II (Biolife) incubated at 37°C for 48 h, hydrophobic peptides. Results were expressed in arbi-
gram-negative bacteria were determined on McConkey trary units (AU), calculated as units of chromatogram
agar (Biolife) incubated for 24 h at 30°C, and coliforms area per milligram of cheese DM.
were determined on VRBA agar (Oxoid) incubated for Free AA and BA were simultaneously extracted from
24 h at 37°C. Molds and yeasts were determined in duplicate samples according to a previously described
duplicate on plates of Chloramphenicol Glucose Agar method (Krause et al., 1995). Analysis of individual
(Scharlab, Barcelona, Spain) incubated for 72 h at free AA was carried out by reverse-phase HPLC after
25°C. Microbial counts were expressed in log cfu per derivatization with Waters AccQ Fluor Reagent, using
gram of cheese. a Waters AccQ Tag (Waters, Milford, MA) column.
Quantitative analysis of individual BA after derivatiza-
Physicochemical and Enzymatic Determinations tion with dabsyl chloride was carried out by reverse-
phase HPLC using a System Gold HPLC apparatus
Proteins were analyzed by capillary gel electrophore- (Beckman Coulter, Palo Alto, CA) equipped with a
sis according to a previously described method (Garde Nova-pack C18 column (Waters). A standard mixture
et al., 2002), with some modifications, on an automated of BA (Sigma-Aldrich) was used for their identification
P/ACE MDQ capillary electrophoresis apparatus con- and quantification.
trolled by a 32 Karat Software (Beckman Instruments Aminopeptidase activity released into the cheese was
España S.A., Madrid, Spain). Briefly, 5 g of cheese were measured on duplicate samples of an extract obtained
mixed with 25 mL of 2% sodium citrate solution and by homogenizing 10 g of cheese with 20 mL of 100 mM
homogenized for 1 min in an Ultra-Turrax T-10 blender sodium phosphate buffer, pH 7, at 20°C for 2 min in
(IKA, Staufen, Germany) at high speed on ice. Twenty an Ultra-Turrax T-10 blender, followed by centrifug-
microliters of cheese homogenate was mixed with 170 ing (10,000 × g, 20 min, 4°C) and filtering through
μL of 100 mM Tris-HCl buffer (pH 9.0) containing 1% Whatman No. 2 paper (GE Healthcare UK Ltd., Buck-
SDS, 10 μL of 2-mercaptoethanol, and 4 μL of a 10 inghamshire, UK). Lysine p-nitroanilide (Lys-p-NA)
kDa internal standard (Beckman Instruments España and leucine p-nitroanilide (Leu-p-NA) were used as
S.A.), and heated at 95°C for 10 min before injection at substrates. One activity unit corresponds to the activ-
5 kV. Electrophoretic separation was performed at 15 ity of enzyme (s) producing 1 nmol of p-nitroaniline per
kV for 30 min after a 4-min ramp in a bare-fused silica minute per gram of cheese.
capillary column (Beckman Instruments España S.A.) Overall proteolysis determined by the o-phthaldial-
of 50 μm internal diameter and 30 cm total length, dehyde test was analyzed in duplicate, as previously
in SDS-buffer gel (Beckman Instruments España S.A.). described (Church et al., 1983), and expressed as the
To calculate the molecular weight of peaks monitored absorbance at 340 nm. Cheese DM was determined in
at 214 nm, the coefficient of relative time mobility to triplicate after cheese grinding with sand by drying to
the internal standard was compared with those of a a constant weight in an oven at 102°C. Cheese pH was
mixture of 10, 20, 35, 50, 100, 150, and 225 kDa protein measured in triplicate directly by means of a Crison
standards (SDS-molecular weight protein size standard; penetration electrode (model 52–3,2; Crison Instru-
Beckman Instruments España S.A.). Commercial stan- ments S.A., Barcelona, Spain) coupled to a Crison
dards (Sigma-Aldrich, Alcobendas, Spain) of bovine GPL 22 pH-meter. A cheese sag index was defined and
α-CN, β-CN, κ-CN, α-LA, β-LG, serum albumin, and calculated as the percentage of sagging in the central
lactoferrin were used for the identification of proteins. point of the flat surface with respect to the height at
Protein peaks were quantified with respect to the in- the edge of the cheese.

Sensory Evaluation and remained fairly constant in the rest of the cheeses
(Table 1), in which the mortality caused by HP treat-
Fifteen trained panelists evaluated the cheeses on ments had been more marked.
d 90, 180, 240, and 360 for flavor quality (overall ac- Penicillium roqueforti counts reached 7.73, 7.92, and
ceptance), flavor intensity (overall intensity), and the 7.74 log cfu/g on d 21, 42, and 63, respectively. Its
flavor attributes acid, bitter, salty, sweet, and umami population declined by 5.33, 0.74, and 0.71 log units
on a 0- to 10-point scale, using a horizontal line an- following HP treatment at 400 MPa after 3, 6, and 9 wk
chored in the middle and at both ends, as previously of ripening, respectively, and fell below detection level
described (Garde et al., 2006). Cheeses were cut in immediately after HP treatment in all 600 MPa cheeses,
representative triangular slices (15–20 g), which were independently of the day of pressurization (data not
held for 2 h at 20 to 22°C before sensory evaluation. shown). Decreases reported for P. roqueforti in Irish
Four cheeses per session (1 control and 3 experimental blue-veined cheese pressurized on d 42 at 400 and 600
cheeses, manufactured on the same day), coded with MPa were 2.16 and 2.68 log units, respectively (Voigt
random 3-digit numbers, were randomly presented to et al., 2010). The high P. roqueforti mortality observed
panelists. Bread and water were used as rinsing agents in the present work for HP treatment on d 21 may be
between cheeses. ascribed to the physiological status of the strain at that
time, with profuse mycelium growth but still with a low
Statistical Treatment level of spores, presumably more baroresistant forms.
Underestimation of survivors, in particular of suble-
An ANOVA with HP treatment (6 treatments and thally injured cells when plated on a selective agar,
control) and ripening time as main effects was performed can be also responsible for the low P. roqueforti count
on the analytical variables by means of SPSS Win 14.0 in 400W3 cheese immediately after treatment on d 21.
program (SPSS Inc., Chicago, IL). Calculation of corre- The differences in P. roqueforti mortality caused by HP
lations and comparison of means by Tukey’s test, with treatments on d 42 in both works can be associated to
the significance assigned at P < 0.05, were carried out the use of different strains in cheese manufacture. From
using the same program. d 90 onwards, P. roqueforti declined gradually in con-
trol and 400 MPa cheeses and was not detected in any
RESULTS AND DISCUSSION of the 600 MPa cheeses (Table 1). The population of
non-Penicillium fungi, mostly Geotrichum and yeasts,
Microbial Groups reached a maximum of 7.07 log cfu/g on d 42, and was
lowered by 2.91, 1.73, and 0.32 log units in 400W3,
Total viable counts and LAB were significantly (P 400W6, and 400W9 cheeses, respectively, immediately
< 0.05) affected by HP treatments. Both the pres- after pressurization with respect to control cheese (data
sure level and, at a lesser degree, the ripening time at not shown). Similarly to P. roqueforti, these microor-
pressurization influenced mortality rates, which were ganisms were not detected on d 90 in 600 MPa cheeses.
almost coincident for total viable counts and LAB, an From d 90 onwards, non-Penicillium fungi declined
expected result given that LAB were the predominant gradually in control cheese and were occasionally de-
microbial group. Counts of LAB, which reached a maxi- tected in pressurized cheeses (Table 1). Yeast counts
mum of 9.54 log cfu/g on d 1, decreased by 1.06, 4.69, were not lowered by pressurization in Irish blue-veined
0.43, 3.98, 0.12, and 3.99 log units in 400W3, 600W3, cheese treated at 400 MPa, and declined by 2.10 log
400W6, 600W6, 400W9, and 600W9 cheeses, respec- units in cheese treated at 600 MPa (Voigt et al., 2010).
tively, immediately after pressurization with respect to Lactobacilli and enterococci counts were below 5 log
control cheese (data not shown). Counts of LAB recov- cfu/g in all cheeses from d 1 to 90. They remained below
ered by approximately 0.5 log units after treatment at 5 log cfu/g in control cheese and in cheeses treated at
400 MPa, but no recovery was observed in 600 MPa 400 MPa from d 90 to 360, and were only occasionally
cheeses during the rest of the ripening period. In pres- detected in 600 MPa cheeses, at levels below 4 log cfu/g
surized Irish blue-veined cheese, counts of lactococci (data not shown). In Irish blue-veined cheese, counts of
declined by 1.6 log units at 400 MPa and recovered nonstarter LAB and enterococci were below 6 log cfu/g
by up to 1.9 log units during storage, whereas at 600 in control and 400 MPa cheeses, and below 4 log cfu/g
MPa they declined by 1.9 log units and did not recover in 600 MPa cheeses (Voigt et al., 2010). In the present
during storage (Voigt et al., 2010). In the present work, study, Micrococcaceae, staphylococci, gram-negative
LAB counts declined during refrigerated storage, from bacteria, and coliforms were detected occasionally, at
d 90 to 360, in control, 400W6, and 400W9 cheeses, counts below 3 log cfu/g (data not shown).

Table 1. Counts1 of the main microbial groups in control ovine milk blue cheese and pressurized cheeses 4820
Microbial
group Days Control cheese 400W32 600W32 400W62 600W62 400W92 600W92
Lactic acid bacteria 90 8.24 ± 0.02c 6.34 ± 0.59b 3.58 ± 0.39a 7.98 ± 0.23c 4.07 ± 0.14a 7.74 ± 0.03c 4.08 ± 0.27a
180 8.02 ± 0.03d 6.70 ± 0.14b 3.75 ± 0.04a 7.46 ± 0.10c 3.49 ± 0.04a 7.07 ± 0.05bc 3.31 ± 0.31a
270 7.64 ± 0.06d 6.71 ± 0.22c 4.04 ± 0.12b 6.31 ± 0.30c 3.15 ± 0.13a 6.60 ± 0.04c 3.86 ± 0.16ab
360 6.57 ± 0.13c 6.41 ± 0.32c 3.21 ± 0.70a 5.63 ± 0.02b 3.33 ± 0.32a 5.45 ± 0.17b 3.67 ± 0.06a
Penicillium roqueforti 90 7.53 ± 0.10c 5.96 ± 0.06b ND3a 5.80 ± 0.67b NDa 5.93 ± 0.04b NDa
180 7.10 ± 0.04d 4.36 ± 0.78b NDa 5.94 ± 0.33c NDa 5.77 ± 0.07c NDa
270 6.56 ± 0.23d 3.24 ± 0.72b NDa 5.61 ± 0.22c NDa 5.07 ± 0.47c NDa
360 5.66 ± 0.08c 2.79 ± 0.46b NDa 3.38 ± 0.31b NDa 2.61 ± 0.36b NDa
Non-Penicillium fungi 90 5.19 ± 0.76c 4.11 ± 0.47c NDa 4.71 ± 0.22c NDa 2.73 ± 0.42b NDa
180 4.02 ± 0.45b 3.65 ± 0.38b 3.14 ± 0.36b NDa NDa NDa NDa

270 3.49 ± 0.60bc 4.05 ± 1.19c 2.48 ± 0.28b NDa NDa 3.13 ± 0.66bc NDa
360 NDa 3.64 ± 0.46b NDa NDa NDa NDa 2.73 ± 0.65b
a–d
Means in the same row followed by different superscripts differ (P < 0.05).
1
Mean ± SE (n = 4) of duplicate determinations in 2 cheese-making experiments. Microbial counts are expressed as log colony-forming units per gram of cheese.
2
Treatments were pressurized at 400 MPa after 3 wk of ripening (400W3), 600 MPa after 3 wk (600W3), 400 MPa after 6 wk (400W6), 600 MPa after 6 wk (600W6), 400 MPa
after 9 wk (400W9), and 600 MPa after 9 wk (600W9).
3
ND = not detected.
CALZADA ET AL.
Table 2. Values of pH, DM, and sag index in control ovine milk blue cheese and pressurized cheeses
Item Days Control cheese 400W31 600W31 400W61 600W61 400W91 600W91
Cheese pH2 90 5.86 ± 0.06b 5.29 ± 0.04a 5.24 ± 0.06a 5.62 ± 0.05b 5.80 ± 0.06b 5.67 ± 0.04b 5.70 ± 0.04b
180 5.47 ± 0.02c 5.09 ± 0.02a 5.01 ± 0.01a 5.38 ± 0.02bc 5.43 ± 0.02bc 5.35 ± 0.02b 5.35 ± 0.03b
270 5.26 ± 0.01cd 5.11 ± 0.04b 4.94 ± 0.01a 5.23 ± 0.02c 5.19 ± 0.03bc 5.34 ± 0.03d 5.27 ± 0.02cd
360 5.34 ± 0.01c 5.15 ± 0.04b 5.06 ± 0.02a 5.36 ± 0.02c 5.38 ± 0.02c 5.34 ± 0.01c 5.40 ± 0.01c
DM (%)2 90 53.95 ± 0.54a 53.27 ± 0.75a 53.86 ± 0.50a 52.06 ± 0.74a 54.46 ± 0.42a 53.67 ± 0.31a 52.75 ± 1.13a
180 54.43 ± 0.36ab 52.95 ± 0.47a 52.98 ± 0.30a 53.39 ± 1.03ab 54.61 ± 0.70ab 54.85 ± 0.78ab 55.05 ± 0.43b
270 53.16 ± 0.69a 54.40 ± 0.61a 55.08 ± 0.19a 54.10 ± 0.76a 53.59 ± 0.84a 55.23 ± 0.44a 53.99 ± 0.80a
360 53.35 ± 0.27ab 53.60 ± 0.79ab 55.16 ± 0.57b 54.49 ± 1.16ab 52.96 ± 0.28a 54.54 ± 0.73ab 54.46 ± 0.46ab
Sag index2,3 90 15.04 ± 1.08a 36.24 ± 1.43c 34.21 ± 0.35c 27.16 ± 1.63b 25.49 ± 0.74b 30.25 ± 1.99bc 29.22 ± 0.47bc
180 13.51 ± 0.49a 32.15 ± 0.52d 29.45 ± 1.27cd 26.10 ± 0.78bc 21.62 ± 1.72b 23.92 ± 1.16bc 30.09 ± 0.66cd
270 14.84 ± 0.80a 31.94 ± 0.73c 28.30 ± 1.07bc 23.28 ± 0.78b 22.92 ± 1.49b 28.03 ± 1.35bc 27.96 ± 1.07bc
360 15.14 ± 0.56a 31.78 ± 1.75d 30.49 ± 1.54cd 24.06 ± 1.46b 21.68 ± 2.94b 24.61 ± 1.06bc 23.23 ± 1.55b
a–d
1
2
Mean ± SE (n = 6) of triplicate determinations in 2 cheese-making experiments.
3
Cheese sag index was calculated as 100 − (100 height at the central point/height at the edge).
Cheese pH, DM, and Sag Index only whey protein detected on d 90 in control cheese
was β-LG, at a level 99% lower than on d 1 (Table 3).
Control cheese pH, which had declined to 4.76 on d 1, Considerably higher levels of residual α-CN and β-CN
fell further to 4.63 by d 7, and rose afterward gradually are found in non-Penicillium ripened cheeses (Gomez et
to values of 5.65 on d 21, 6.10 on d 42, and 6.24 on d al., 1999). In addition to the rennet and starter LAB
63 (data not shown), most likely because of lactic acid proteinases acting in most cheese varieties, P. roque-
consumption by molds and yeasts. This pH pattern is forti proteinases (Modler et al., 1974; Trieu-Cuot et
similar to that recorded for Stilton cheese (Madkor et al., 1982) were most likely responsible for the intense
al., 1987). In pressurized cheeses, pH values immediately proteolysis observed during ripening of blue-veined
after HP treatments were similar to those recorded for cheese in the present work. During the refrigerated
the respective control cheeses, with differences always storage period proteolysis still progressed. Thus, β-CN
under 0.2 pH units. However, the pH evolved differently and para-κ-CN declined from d 90 to 360 in control
during ripening of 400W3 and 600W3 cheeses than in cheese by 96 and 80%, respectively, and γ-CN, the
the rest, with significantly (P < 0.05) lower values on degradation products resulting from primary proteoly-
d 90 in cheeses pressurized after 3 wk of ripening, most sis, by 81% (Table 4). From d 90 onwards, the highest
probably because of impaired lactic acid consumption levels of β-CN, para-κ-CN, and γ-CN were generally
by damaged P. roqueforti cells (Table 2). Dry matter found in the 600W3 cheese. Protein concentrations in
of control cheese, which was 46.63% on d 1, increased all cheeses during refrigerated storage were at such low
to 50.73 and 52.97% on d 21 and 63 (data not shown), levels, with the exception of γ-CN, that a significant
respectively, and to 53.95% on d 90 (Table 2). During contribution of residual proteins to the formation of
refrigerated storage, DM remained fairly constant, with peptides and free AA at this stage is to be precluded.
differences between cheeses under 3% at all sampling A less pronounced proteolysis, as shown by the lower
times. pH 4.6- and TCA-soluble N contents, was recorded for
The sag index, measuring the subsiding at the cen- Irish blue-veined cheese pressurized at 600 MPa than
tral point of the flat surface with respect to height at for control cheese or for 400 MPa cheese (Voigt et al.,
the edge, was significantly (P < 0.05) higher in all the 2010). The lower proteolysis in cheeses treated at 600
pressurized cheeses, independently of the pressure level MPa can be ascribed to a lower activity of chymosin
or the ripening time at pressurization, than in control and, probably, of P. roqueforti proteinases. Chymosin
cheese (Table 2). The sag index of control cheese did is partially inactivated by pressurization at 400 MPa
not vary during refrigerated storage, whereas gradual or higher pressures, but plasmin, a more baroresistant
shape recovery took place in all the pressurized cheeses enzyme, maintains full activity after pressurization at
from d 90 to 360, with increases in the sag index rang- 500 or 600 MPa (Malone et al., 2003; Juan et al., 2007).
ing from 3.1 to 6.0%. To our knowledge, no information is available on the
barotolerance of P. roqueforti proteinases.
Proteolysis Hydrophilic peptides increased gradually during the
ripening of control cheese, with levels attaining 6.79,
Changes in the concentration of proteins from milk 24.51, 25.78, and 27.10 AU respectively on d 1, 21,
to 1-d-old cheese can be explained by losses during 42, and 63, whereas the levels of hydrophobic peptides
whey drainage and by enzymatic hydrolysis. Losses in reached a maximum of 7.48 AU on d 21 (data not
whey were responsible for the decline observed in the shown) and declined thereafter to 5.66 AU on d 90. Both
contents of the water-soluble proteins α-LA, β-LG, and hydrophilic peptides and hydrophobic peptides suffered
serum albumin from milk to 1-d-old cheese (Table 3). minor variations from d 90 to 360 (Table 5). In the case
In the absence of proteolysis, contents of α-CN, β-CN, of hydrophilic peptides, the differences between control
and κ-CN should have increased from milk to 1-d-old and pressurized cheeses at the same sampling time
cheese, due to the retention and concentration of CN were generally not significant. However, hydrophobic
micelles in the curd. However, because of the activ- peptides were generally found at higher levels in pres-
ity of rennet and starter proteinases the contents of surized cheeses than in the respective control cheese.
α-CN, β-CN, and κ-CN referred to DM declined by Consequently, the hydrophobic peptides-to-hydrophilic
65, 11, and 71% (Table 3), respectively, from milk peptides ratio, which has been reported to correlate
to 1-d-old cheese. On d 90, α-CN and κ-CN were no well with cheese bitterness (Gomez et al., 1997), was
longer detectable in control cheese, whereas β-CN was lower in control cheese than in most of the pressurized
99% lower and para-κ-CN 79% lower than on d 1. A cheeses (Table 5). The highest levels of hydrophobic
more rapid hydrolysis of α-CN than of β-CN was also peptides and of the peptide ratio from d 90 to 360
recorded for Stilton cheese (Madkor et al., 1987). The were generally found for the 600W3 cheese. Contrary to
4822
Table 3. Main proteins1 in milk and control ovine milk blue cheese during ripening
Serum
Days α-CN β-CN κ-CN para-κ-CN γ-CN α-LA β-LG albumin
Milk 59.52 ± 2.95 205.47 ± 13.21 15.23 ± 1.34 ND2 7.97 ± 1.26 2.19 ± 0.10 49.15 ± 2.65 5.54 ± 0.67
1 20.76 ± 3.69c 183.67 ± 19.42c 4.36 ± 0.36c 28.35 ± 2.34c 7.95 ± 1.02a 0.87 ± 0.08b 7.37 ± 0.79b 0.80 ± 0.15c
21 1.49 ± 0.48b 25.05 ± 2.44b 0.95 ± 0.25b 15.25 ± 1.46b 44.87 ± 2.80c NDa 0.21 ± 0.03a 0.13 ± 0.08b
42 1.12 ± 0.21b 13.38 ± 1.72ab 0.43 ± 0.13b 9.64 ± 1.14ab 32.05 ± 2.40bc NDa 0.15 ± 0.06a 0.05 ± 0.03b
63 0.54 ± 0.19b 6.20 ± 1.38a 0.02 ± 0.01b 8.22 ± 0.96ab 30.05 ± 3.06b NDa 0.18 ± 0.07a NDa
90 NDa 1.22 ± 0.35a NDa 5.94 ± 0.82a 19.60 ± 4.21ab NDa 0.07 ± 0.03a NDa

a–c
Means for cheese samples on the same column followed by different superscripts differ (P < 0.05).
1
Mean ± SE (n = 4) of duplicate determinations in 2 cheese-making experiments. Proteins are expressed in milligrams per gram of milk or cheese DM.
2
ND = not detected.
CALZADA ET AL.
Table 4. Main caseins1 in control ovine milk blue cheese and pressurized cheeses
Casein Days Control cheese 400W32 600W32 400W62 600W62 400W92 600W92
β-CN 90 1.22 ± 0.35a 3.58 ± 0.84b 3.22 ± 0.76ab 0.63 ± 0.24a 2.40 ± 0.12ab 0.59 ± 0.21a 0.71 ± 0.14a
180 0.80 ± 0.22ab 1.20 ± 0.20ab 1.98 ± 0.30b 0.62 ± 0.17ab 0.66 ± 0.19ab 0.33 ± 0.09a 0.39 ± 0.17a
270 0.07 ± 0.05a 0.30 ± 0.11ab 0.85 ± 0.26b 0.16 ± 0.02a 0.35 ± 0.09ab 0.09 ± 0.04a 0.15 ± 0.05a
360 0.05 ± 0.03a 0.13 ± 0.06ab 0.17 ± 0.04ab 0.16 ± 0.07ab 0.32 ± 0.09b 0.07 ± 0.03ab 0.09 ± 0.02a
para-κ-CN 90 5.94 ± 0.82a 5.71 ± 1.12a 6.41 ± 0.77a 3.15 ± 0.62a 6.00 ± 0.68a 3.68 ± 0.83a 4.84 ± 0.30a
180 3.50 ± 0.95ab 2.94 ± 0.31ab 4.54 ± 0.65b 2.56 ± 0.29ab 3.97 ± 0.47ab 1.65 ± 0.49a 2.60 ± 0.40ab
270 1.28 ± 0.22ab 1.43 ± 0.18abc 2.95 ± 0.62c 0.60 ± 0.18a 2.29 ± 0.29bc 0.91 ± 0.28ab 1.71 ± 0.61abc
360 1.16 ± 0.24ab 1.31 ± 0.29ab 2.01 ± 0.35b 0.75 ± 0.16a 1.21 ± 0.32ab 0.60 ± 0.11a 1.25 ± 0.08ab
γ-CN 90 19.60 ± 3.21ab 14.41 ± 2.81ab 17.24 ± 2.46ab 9.99 ± 1.93a 22.08 ± 2.09b 11.90 ± 2.18a 12.31 ± 0.37ab
180 12.87 ± 2.92ab 8.95 ± 0.53ab 14.17 ± 1.97b 10.10 ± 1.49ab 13.14 ± 0.38ab 5.92 ± 1.26a 8.37 ± 1.13ab
270 4.66 ± 0.97ab 5.50 ± 0.86ab 10.42 ± 1.97c 2.82 ± 0.67a 7.54 ± 0.75bc 3.82 ± 0.52ab 4.80 ± 1.23ab
360 3.77 ± 0.48ab 3.85 ± 1.09ab 6.03 ± 0.98b 2.56 ± 0.29a 4.80 ± 0.70ab 2.79 ± 0.25a 5.18 ± 0.19ab
a–c
1
Mean ± SE (n = 4) of duplicate determinations in 2 cheese-making experiments. Caseins are expressed in milligrams per gram of cheese DM. α-Casein and κ-CN were not detected
on d 90 and afterward.
2
our results, no significant differences in the late-eluting wk 6 or 9 showed the highest overall proteolysis values,
region, corresponding to hydrophobic peptides, were and 600W3 cheese the lowest value.
found between control and pressurized Irish blue-veined Total free AA concentration increased in control
cheeses (Voigt et al., 2010). The shorter cheese ripen- cheese from 1.49 mg/g of DM on d 1 to 61.20 mg/g of
ing and storage periods in their work may explain the DM on d 90, a considerably higher concentration than
different finding. those reported for Gorgonzola cheese and other strongly
Aminopeptidase activity in control cheese reached proteolyzed cheeses of approximately the same ripening
maximum levels of 24.83 nmol p-nitroaniline per min- time (Gobbetti et al., 1997; Garde et al., 2002, 2007),
ute per gram on d 42 (with Leu-p-NA as substrate) and but lower than the concentration of free AA in Stilton
24.58 nmol p-nitroaniline per minute per gram on d cheese (Madkor et al., 1987). In the present work, free
63 (with Lys-p-NA as substrate; data not shown), and AA concentrations in pressurized cheeses immediately
then declined during the rest of the ripening period after treatment were close to those of the respective
to 19.61 and 19.43 nmol p-nitroaniline per minute per control cheese (data not shown). From d 90 to 360, total
gram, respectively, on d 90 (Table 6). Aminopeptidase free AA increased by 91% in control cheese and by 59
activity, which originates mostly from starter LAB but to 116% in pressurized cheeses. Similar to the pattern
also from other added or contaminating microorgan- observed for overall proteolysis, the concentration of to-
isms, was not determined in Irish blue-veined cheese tal free AA during refrigerated storage tended to reach
(Voigt et al., 2010). In Gorgonzola cheese, not only its maximum levels in control cheese and in 400W6 and
aminopeptidase activity but also carboxypeptidase and 400W9 cheeses, probably favored by conformational
iminopeptidase activities, attributed to P. roqueforti by changes in the CN of these cheeses that facilitate the
the authors, increased until d 86, the last sampling date access of enzymes to their substrates (O’Reilly et al.,
(Gobbetti et al., 1997). High-pressure treatments show 2003), whereas the lowest values were those of 600W3
differences in the inactivation of bacterial peptidases, cheese (Table 7). Early pressurization (after 3 wk of
depending not only on process parameters, but also on ripening) at a high pressure level (600 MPa) was the
bacterial species and assay substrates, and the effect most effective treatment in retarding secondary prote-
may vary for different peptidolytic enzymes within a olysis, even though the lowest aminopeptidase activity
bacterial strain (Malone et al., 2003). In the present values were not those of 600W3 cheese.
work, pressurization lowered aminopeptidase activity Aminopeptidase activities on Leu-p-NA and Lys-p-
by 20 to 30% at 400 MPa and by 40 to 50% at 600 NA as substrates correlated well with each other (r2
MPa immediately after treatment (data not shown), values = 0.922 and 0.982 on d 90 and 360, respectively),
in agreement with previous results for a nonmold- and this was also true for the correlation of overall pro-
ripened ovine milk cheese (Garde et al., 2007). On d teolysis with total free AA (r2 values = 0.782 and 0.942
90, aminopeptidase activity levels were significantly (P on d 90 and 360). However, a significant correlation
< 0.05) lower in all the pressurized cheeses than in con- of aminopeptidase activities with overall proteolysis or
trol cheese. Aminopeptidase activity declined further free AA concentration in cheeses was not observed (r2
during refrigerated storage, to levels under 20% of the values = <0.10 and <0.40 on d 90 and 360). It must
maximum values, and the differences between cheeses be taken into account that aminopeptidases were at
persisted until d 360 (Table 6). levels far from their maximum values during refriger-
Overall proteolysis, as determined by the o-phthal- ated storage, and also that aminopeptidases present in
dialdehyde method, increased in control cheese from a intact microbial cells not lysed by treatment at 400
value of 0.15 on d 1 to 0.89 on d 21, 2.93 on d 42, 5.45 MPa, which most likely are not recovered in the cheese
on d 63 (data not shown), and 6.89 on d 90 (Table extracts used for aminopeptidase assays, contribute to
7). Proteolysis values during ripening were consider- cheese secondary proteolysis too. The low correlation
ably higher than those reported for cheese varieties not of aminopeptidase activity to overall proteolysis and
mold-ripened (Garde et al., 2002, 2007). Proteolysis total free AA can be also attributed to the fact that
values in pressurized cheeses immediately after treat- carboxypeptidase and iminopeptidase activities coming
ment were similar to those of the respective control from P. roqueforti, which enhance overall proteolysis
cheese (data not shown). However, significant (P < and participate in free AA formation, were not evalu-
0.05) differences in the overall proteolysis values of ated when using Leu-p-NA and Lys-p-NA as substrates.
cheeses were recorded on d 90, the lowest value being
that of 600W3 cheese (Table 7). During refrigerated BA
storage, overall proteolysis increased by 51% in control
cheese, and by 37 to 68% in pressurized cheeses. On d Histamine, cadaverine, and spermine were not
360, control cheese and cheeses treated at 400 MPa on detected at any of the sampling times in any of the
4824
Table 5. Hydrophilic peptides, hydrophobic peptides, and the hydrophobic-to-hydrophilic peptide ratio in control ovine milk blue cheese and pressurized cheeses
Peptide Days Control cheese 400W31 600W31 400W61 600W61 400W91 600W91
Hydrophilic peptides2 90 28.16 ± 0.82a 36.64 ± 0.41d 27.42 ± 0.63a 32.58 ± 0.96bc 28.91 ± 1.22ab 30.70 ± 0.51abc 33.00 ± 0.08cd
180 28.08 ± 1.49a 31.76 ± 0.28a 29.40 ± 1.55a 26.68 ± 1.14a 29.20 ± 0.14a 27.93 ± 0.19a 27.13 ± 1.66a
270 29.39 ± 0.55ab 27.80 ± 0.35ab 25.86 ± 1.23a 30.19 ± 1.11b 27.90 ± 0.23ab 30.72 ± 0.86b 31.12 ± 1.05b
360 29.81 ± 0.87ab 30.10 ± 2.73ab 26.86 ± 0.11a 33.11 ± 0.53bc 27.55 ± 0.74a 36.05 ± 0.49c 31.93 ± 0.12ab
Hydrophobic peptides2 90 5.66 ± 0.14a 9.40 ± 0.36d 8.85 ± 0.36cd 7.16 ± 0.20b 7.72 ± 0.68bc 6.73 ± 0.06ab 6.70 ± 0.13ab
180 4.80 ± 0.19a 7.61 ± 0.30c 9.08 ± 0.55d 6.11 ± 0.13ab 7.12 ± 0.30bc 5.88 ± 0.40ab 5.24 ± 0.33a
270 4.44 ± 0.32a 4.14 ± 0.12a 6.53 ± 0.26bc 4.78 ± 0.96ab 5.35 ± 0.49ab 6.82 ± 0.35bc 7.33 ± 0.22c
360 4.54 ± 0.18a 6.54 ± 0.01b 8.33 ± 0.32d 6.81 ± 0.49bc 6.61 ± 0.16b 6.93 ± 0.07bc 7.85 ± 0.11cd
Ratio2 90 0.20 ± 0.01a 0.26 ± 0.01bc 0.32 ± 0.01d 0.22 ± 0.01ab 0.27 ± 0.01c 0.22 ± 0.01ab 0.20 ± 0.01a
180 0.17 ± 0.00a 0.24 ± 0.01b 0.31 ± 0.03c 0.23 ± 0.01ab 0.24 ± 0.01b 0.21 ± 0.02ab 0.19 ± 0.01ab

270 0.15 ± 0.01a 0.15 ± 0.00a 0.25 ± 0.00b 0.16 ± 0.03a 0.19 ± 0.02ab 0.22 ± 0.01b 0.24 ± 0.02b
360 0.15 ± 0.01a 0.22 ± 0.02c 0.31 ± 0.01d 0.21 ± 0.01bc 0.24 ± 0.01c 0.19 ± 0.00ab 0.25 ± 0.00c
a–d
1
2
Mean ± SE (n = 4) of duplicate determinations in 2 cheese-making experiments. Peptides, determined at 280 nm, are expressed in arbitrary units (AU), calculated as units of
chromatogram area per milligram of cheese DM.
CALZADA ET AL.
Table 6. Aminopeptidase activity on Leu-p-NA and Lys-p-NA in control ovine milk blue cheese and pressurized cheeses
Aminopeptidase activity Days Control cheese 400W31 600W31 400W61 600W61 400W91 600W91
Activity on Leu-p-NA2 90 19.61 ± 0.99d 12.34 ± 0.31c 11.34 ± 0.27bc 7.58 ± 0.12a 11.55 ± 0.70bc 11.39 ± 0.72bc 8.19 ± 0.32ab
180 6.66 ± 0.48d 4.38 ± 0.21bc 4.77 ± 0.52c 3.25 ± 0.52ab 2.63 ± 0.31a 2.86 ± 0.07ab 2.65 ± 0.05a
270 5.30 ± 0.83c 2.31 ± 0.07b 2.34 ± 0.15b 1.44 ± 0.07a 1.42 ± 0.01a 2.13 ± 0.10b 1.21 ± 0.01a
360 3.26 ± 0.27c 1.11 ± 0.18a 1.39 ± 0.05ab 1.85 ± 0.08b 0.92 ± 0.06a 1.77 ± 0.04b 0.89 ± 0.12a
Activity on Lys-p-NA2 90 19.43 ± 0.48e 10.92 ± 0.61cd 9.48 ± 0.34abc 8.16 ± 0.27ab 12.87 ± 0.59d 10.34 ± 0.99bcd 7.23 ± 0.09a
180 6.26 ± 0.42b 3.65 ± 0.08a 3.24 ± 0.35a 3.32 ± 0.36a 2.38 ± 0.23a 2.94 ± 0.16a 2.22 ± 0.08a
270 4.33 ± 0.67b 1.46 ± 0.02a 1.06 ± 0.03a 1.30 ± 0.10a 0.97 ± 0.02a 1.70 ± 0.13a 1.23 ± 0.07a
360 3.21 ± 0.27c 1.00 ± 0.19ab 1.02 ± 0.03ab 1.71 ± 0.13b 0.91 ± 0.09ab 1.66 ± 0.05ab 0.75 ± 0.06a
a–e
1
2
Mean ± SE (n = 4) of duplicate determinations in 2 cheese-making experiments. Activity is expressed in nanomole p-nitroaniline per minute per gram.
Table 7. Overall proteolysis and free AA in control ovine milk blue cheese and pressurized cheeses
Item Days Control cheese 400W31 600W31 400W61 600W61 400W91 600W91
Overall proteolysis2 90 6.89 ± 0.17bc 5.25 ± 0.62ab 4.03 ± 0.37a 6.83 ± 0.04bc 5.23 ± 0.05ab 7.06 ± 0.08c 6.55 ± 0.83bc
180 8.28 ± 0.54d 6.51 ± 0.04b 4.49 ± 0.12a 8.62 ± 0.87d 7.32 ± 0.09c 8.85 ± 0.40d 7.00 ± 0.32bc
270 8.65 ± 0.37c 6.88 ± 0.24b 4.81 ± 0.45a 8.83 ± 0.25c 7.85 ± 0.21bc 9.01 ± 0.12c 8.52 ± 0.37c
360 10.42 ± 0.29c 8.10 ± 0.14b 5.53 ± 0.42a 11.38 ± 0.51c 8.78 ± 0.16b 10.40 ± 0.17c 8.96 ± 0.63b
Total free AA2 90 61.20 ± 3.67c 42.09 ± 2.89b 25.21 ± 1.31a 64.68 ± 6.64c 59.64 ± 2.51c 63.04 ± 3.70c 59.66 ± 2.91c
180 88.61 ± 2.10c 60.10 ± 3.28b 36.46 ± 0.72a 90.96 ± 1.08cd 80.98 ± 0.55c 92.91 ± 1.58d 78.97 ± 5.74c
270 101.13 ± 3.89cd 73.55 ± 5.61b 38.77 ± 2.62a 108.28 ± 0.24d 84.73 ± 1.35b 101.86 ± 3.83cd 88.27 ± 10.03bc
360 116.76 ± 1.42d 81.45 ± 3.46b 54.33 ± 5.70a 116.68 ± 0.77d 95.80 ± 1.42c 123.48 ± 1.08d 94.81 ± 1.15c
a–d
1
2
Mean ± SE (n = 4) of duplicate determinations in 2 cheese-making experiments. Overall proteolysis estimated by the o-phthaldialdehyde method is expressed as the absorbance
at 340 nm. Total free AA are expressed in milligrams per gram cheese DM.
Table 8. Main biogenic amines in control ovine milk blue cheese and pressurized cheeses
Biogenic amine Days Control cheese 400W31 600W31 400W61 600W61 400W91 600W91
Tyramine2 180 7.15 ± 2.66ab 7.80 ± 1.09b 6.57 ± 1.70ab 4.91 ± 0.66a 5.78 ± 1.36a 6.24 ± 0.99ab 4.62 ± 1.21a
270 26.70 ± 2.92a 25.78 ± 6.30a 24.78 ± 5.85a 18.63 ± 2.83a 24.46 ± 3.84a 23.15 ± 2.45a 23.84 ± 3.37a
360 52.20 ± 9.56c 32.01 ± 2.55b 33.43 ± 2.43b 20.90 ± 1.68a 32.39 ± 1.92b 30.62 ± 5.86b 27.22 ± 1.79ab
Tryptamine2 180 61.88 ± 7.30ab 67.07 ± 9.71b 46.33 ± 5.06a 54.29 ± 6.01ab 42.85 ± 5.63a 69.40 ± 6.72b 57.91 ± 8.02ab
PROTEOLYSIS AND BIOGENIC AMINES IN BLUE CHEESE
270 70.59 ± 5.17ab 86.28 ± 2.85b 65.49 ± 12.44a 70.89 ± 3.72ab 73.59 ± 2.61ab 81.57 ± 3.68ab 74.76 ± 5.10ab
360 71.11 ± 3.82a 102.90 ± 7.91bc 76.77 ± 4.93a 74.69 ± 4.32a 104.40 ± 3.10bc 93.95 ± 2.61ab 120.09 ± 3.94c
Phenylethylamine2 180 13.25 ± 0.94c 8.13 ± 2.53ab 11.77 ± 3.38bc 5.07 ± 0.80a 8.18 ± 0.43ab 4.11 ± 0.56a 7.21 ± 2.07a
270 57.04 ± 6.51a 64.49 ± 2.55ab 61.05 ± 5.55ab 66.31 ± 1.60ab 54.50 ± 4.15a 80.68 ± 9.81b 75.31 ± 9.22b
360 61.44 ± 2.79a 74.22 ± 5.47b 77.71 ± 9.95bc 57.49 ± 4.49a 63.11 ± 5.58ab 70.70 ± 6.19ab 92.05 ± 5.31c
Putrescine2 180 17.28 ± 1.33bc 14.51 ± 1.68a 15.90 ± 2.01ab 12.67 ± 0.87a 19.52 ± 3.15c 13.03 ± 1.40a 18.70 ± 2.46bc
270 33.46 ± 2.09a 33.96 ± 2.85a 36.27 ± 2.58a 33.22 ± 1.30a 46.79 ± 3.96b 38.34 ± 1.45a 42.79 ± 2.81ab
360 32.97 ± 2.05ab 34.48 ± 3.11ab 39.50 ± 3.55bc 30.38 ± 0.54a 44.05 ± 2.81cd 31.00 ± 2.81a 49.70 ± 2.79d
a–d
1
2
Mean ± SE (n = 4) of duplicate determinations in 2 cheese-making experiments. Biogenic amines are expressed in milligrams per kilogram of cheese DM.

4825
4826 CALZADA ET AL.
cheeses. Spermidine was found on d 270 and 360 at low limit the formation of the respective biogenic amines,
concentrations, ranging from 5.06 to 14.03 mg/kg of similarly to tyrosine and histidine.
DM, in control cheese and in some of the HP-treated Biogenic amines formation through free AA decar-
cheeses (data not shown). Spermine and spermidine boxylation may constitute an alternative energy source
may be formed by starter LAB. These polyamines for cheese microbiota in the absence of fermentable
were detected at low levels, 0.20 mg/L of spermine and carbohydrates (Fernandez-García et al., 2000). Con-
0.75 mg/L of spermidine, in sterilized nonfat milk, and versely, some strains of Lactobacillus, Pediococcus, and
increased during the fermentation of milk with added Micrococcus are capable of degrading BA, such as tyra-
rennet by Lactococcus lactis to levels as high as 1.7 and mine and histamine, by means of monoamine oxidases,
10 mg/L, respectively (Santos et al., 2003). According preferably under aerobic conditions (Leuschner et al.,
to those authors, spermine reached a maximum after 12 1998). Accumulation of BA is thus the result of BA
h at 20°C and had practically disappeared 12 h later. formation and degradation by cheese microbiota.
Concentrations of spermidine and spermine in the pres- Concentrations of BA reported for blue-veined cheeses
ent study are in agreement with those found for pres- show a considerable variability, and do not seem to be
surized and control goat milk cheeses, which contained related to the use of raw or pasteurized milk. Roquefort
higher concentrations of spermidine, ranging from 14.7 cheese contained, on average, 19, 40, 65, and 158 mg/
to 26.4 mg/kg of DM, than of spermine, which ranged kg of histamine, tyramine, putrescine, and cadaverine,
from 0.9 to 3.5 mg/kg of DM (Novella-Rodríguez et al., respectively (De Vuyst et al., 1976), and Stilton cheese
2002). contained 39, 121, 11, and 126 mg/kg of histamine, ty-
Tyramine, which was not detected on d 90 in any ramine, phenyletylamine, and putrescine, respectively
of the cheeses, attained significantly (P < 0.05) lower (Baker et al., 1987). Roquefort cheese has been always
levels on d 360 in all the pressurized cheeses than in made from raw milk, and Stilton cheese was made
control cheese (Table 8). The low tyramine concentra- from unpasteurized milk at the time of the cited work
tions found in the present work and the absence of (Gkatzionis et al., 2009). High concentrations of BA,
histamine can be related to the low counts of entero- 490 mg/kg of histamine and 625 mg/kg of tyramine,
cocci and lactobacilli, potential tyrosine and histidine were found for Danish blue cheese (Ingles et al., 1985)
decarboxylase-positive bacteria, but also to the low and for Spanish blue cheese made from raw milk, which
cheese-ripening and storage temperatures used, 5°C contained 1,042 mg/kg of histamine and 1,052 mg/kg
from d 30 to 90, and 3°C from d 90 onwards, which of tyramine (Fernández et al., 2007). For Egyptian raw
impair BA formation (Stratton et al., 1991). Tyrosine milk blue cheese, up to 36, 2,220, 27, 12, and 16 mg/kg
and histidine concentrations in cheeses on d 360, which of DM of histamine, tyramine, putrescine, and cadaver-
ranged from 4.11 to 9.29 and from 3.80 to 9.86 mg/g ine, respectively, were found (Rabie et al., 2011). Those
of DM (data not shown), respectively, did not appear authors achieved significant decreases in histamine and
as limiting factors for tyramine and histamine forma- tyramine contents by cheese irradiation at 4 kGy, and
tion. Tryptamine, which was below detection level in in putrescine and cadaverine contents by irradiation at
all cheeses on d 90, reached higher concentrations in 6 kGy.
400W3, 600W6, and 600W9 cheeses by d 360 than
in control cheese. Phenylethylamine was found in all Sensory Characteristics
cheeses on d 90 at low concentrations, ranging from
4.96 to 13.28 mg/kg of DM (data not shown), and at- Flavor intensity scores of pressurized cheeses did not
tained higher levels in 400W3, 600W3, and 600W9 by d differ from those of control cheese, with the only excep-
360 cheeses than in control cheese (Table 8). Putrescine tion of 600W3 cheese scores, which were significantly
was also detected in all cheeses at low concentrations (P < 0.05) lower than those of control cheese (Table 9).
on d 90, ranging from 5.38 to 8.65 mg/kg of DM (data The scores obtained for flavor quality were also gener-
not shown), and reached higher concentrations in ally lower for 600W3 cheeses than for the rest (Table 9).
600W6 and 600W9 cheeses by d 360 than in control The low flavor scores of 600W3 cheese can be associated
cheese (Table 8). During the fermentation of sterilized to its higher levels of hydrophobic peptides and peptide
nonfat milk with added rennet by L. lactis, up to 0.82 ratio, and its lower levels of overall proteolysis and total
mg/L of tyramine and 0.15 mg/L of putrescine were free AA. In fact, significant (P < 0.05) r2 values of
formed, but phenylethylamine and tryptamine were not 0.758 and 0.808 were obtained for the regressions of
detected (Santos et al., 2003). In the present work, the flavor intensity scores on free AA contents in all cheeses
concentrations of phenylalanine, tryptophan, and argi- on d 270 and 360, respectively, and r2 values of 0.830
nine in control and pressurized cheeses did not seem to and 0.934 for the regressions of flavor quality scores on

free AA contents on d 270 and 360. Cheese treatment
at a high pressure level (600 MPa) at an early ripening
0.34ab
0.31ab
0.37ab
0.31ab
0.28ab
0.30b
0.29c
0.26c
600W91
stage (3 wk) negatively affected biochemical changes

and sensory characteristics. The flavor attributes acid,
±
±
±
±
±
±
±
±
6.88
6.80
7.45
7.13
6.86
7.45
7.09
6.57
bitter, salty, sweet, and umami attained similar scores
in control and pressurized cheeses and, with the excep-
tion of umami scores, which increased significantly (P
< 0.05) with time, did not vary during the refrigerated
0.49ab
0.44ab
b
0.27b
0.25b
0.38b
0.35b
0.28c
0.31
400W91
storage period (data not shown).

No sensory evaluation was carried out on Irish blue-
±
±
±
±
±
±
±
±
7.17
7.39
7.36
7.39
6.79
6.63
7.16
7.09
veined cheese (Voigt et al., 2010). In the case of Gor-

gonzola blue cheese, treatments at 600 and 700 MPa did
not reduce the acceptance by untrained panelists, and
no changes in the bitter, salty, and piquant attributes
ab
0.29ab
0.38ab
0.34ab
0.33bc
0.29b
0.27b
0.43b
0.26
600W61
were observed (Carminati et al., 2004). Conversely, a

negative effect of irradiation on the odor and taste of
±
±
±
±
±
±
±
±
6.86
7.41
7.11
7.31
6.74
6.51
6.55
6.63
Egyptian raw milk blue cheese was recorded when this

procedure was assayed to reduce the formation of BA
(Rabie et al., 2011).
0.21ab
0.26bc
b
0.28b
0.34b
0.29b
0.32b
0.21c
0.24
400W61
Mean ± SE of determinations in 2 cheese-making experiments by 15 trained panelists on a 0- to 10-point scale.
CONCLUSIONS
±
±
±
±
±
±
±
±
7.25
7.02
7.18
7.90
7.07
7.28
7.06
7.30
Pressurization of blue-veined cheese lowered micro-

bial counts, in particular treatments at 600 MPa, which
reduced LAB counts by more than 4 log units and P.
roqueforti counts by more than 6 log units. The levels
a
0.34a
0.36a
0.28a
0.35a
0.42a
0.46a
0.46a
0.29
600W31
of residual caseins were generally higher in 600 MPa

Table 9. Sensory characteristics of control ovine milk blue cheese and pressurized cheeses
±
±
±
±
±
±
±
±
cheeses than in the rest of the cheeses. Hydrophilic pep-

6.48
6.41
6.27
6.50
6.11
5.33
5.18
5.53
tides reached similar levels in pressurized and control

cheeses from d 90 onwards, but the level of hydrophobic
peptides tended to be higher in pressurized cheeses.
ab
0.40ab
0.40ab
Aminopeptidase activity, overall proteolysis, and free

0.23b
0.32b
0.40b
0.35b
0.36a
0.26
400W31
AA contents were generally higher in control cheese

±
±
±
±
±
±
±
±
than in pressurized cheeses. Tyramine concentration was

7.01
7.28
7.09
7.09
6.39
6.45
6.19
6.50
lower in pressurized cheeses than in control cheese, but

higher tryptamine, phenylethylamine, and putrescine
contents were found in some of the pressurized cheeses,
ab
0.19bc
0.22bc
in particular in cheese pressurized at 600 MPa after 9

0.17b
0.18b
0.19b
0.17b
0.21b
0.16
Control
cheese
wk of ripening, than in control cheese. Differences in

±
±
±
±
±
±
±
±
sensory characteristics between pressurized and control

6.77
7.30
7.23
7.55
7.12
7.19
6.49
7.03
cheeses were generally negligible, with the exception

of treatment at a high pressure level (600 MPa) at an
early ripening stage (3 wk), which negatively affected
Days
90
180
270
360
90
180
270
360
biochemical changes and sensory characteristics.
ACKNOWLEDGMENTS
This work was supported by project AGL2009-07801

Sensory characteristic
from the Ministry of Science and Innovation (MICINN;

2
Flavor intensity
Madrid, Spain). J. Calzada was the recipient of a MI-

Flavor quality2
CINN fellowship. The authors are grateful for the valu-

able help of V. Carbonero (Lácteas Toledo, Guadamur,
Spain) with cheese supply and of F. Purroy (NC Hy-
perbaric, Burgos, Spain) with high pressure treatments.
a–c

4828 CALZADA ET AL.
REFERENCES Ingles, D. L., J. F. Black, D. Gallimore, R. Tindale, and K. J. Shaw.

1985. Estimation of biogenic-amines in foods. J. Sci. Food Agric.
Arqués, J. L., S. Garde, E. Fernández-García, P. Gaya, and M. Nuñez. 36:402–406.
2007. Volatile compounds, odor, and aroma of La Serena cheese Joosten, H. M. L. J., and M. D. Northolt. 1987. Conditions allow-
high-pressure treated at two different stages of ripening. J. Dairy ing the formation of biogenic-amines in cheese. 2. Decarboxyl-
Sci. 90:3627–3639. ase properties of some nonstarter bacteria. Neth. Milk Dairy J.
Arqués, J. L., S. Garde, P. Gaya, M. Medina, and M. Nuñez. 2006. In- 41:259–280.
activation of microbial contaminants in raw milk La Serena cheese Juan, B., V. Ferragut, M. Buffa, B. Guamis, and A. J. Trujillo. 2007.
by high pressure treatments. J. Dairy Sci. 89:888–891. Effects of high pressure on proteolytic enzymes in cheese: Rela-
Baker, G. B., J. T. F. Wong, R. T. Coutts, and F. M. Pasutto. 1987. tionship with the proteolysis of ewe milk cheese. J. Dairy Sci.
Simultaneous extraction and quantitation of several bioactive 90:2113–2125.
amines in cheese and chocolate. J. Chromatogr. 392:317–331. Krause, I., A. Bockhardt, H. Neckermann, T. Henle, and H. Kloster-
Carminati, D., M. Gatti, B. Bonvini, E. Neviani, and G. Mucchetti. meyer. 1995. Simultaneous determination of amino acids and bio-
2004. High-pressure processing of Gorgonzola cheese: Influence on genic amines by reversed-phase high-performance liquid chroma-
Listeria monocytogenes inactivation and on sensory characteristics. tography of the dabsyl derivatives. J. Chromatogr. A 715:67–79.
J. Food Prot. 67:1671–1675. Lau, K. Y., D. M. Barbano, and R. R. Rasmussen. 1991. Influence of
Church, F. C., H. E. Swaisgood, D. H. Porter, and G. L. Catignani. pasteurization of milk on protein breakdown in Cheddar cheese
1983. Spectrophotometric assay using o-phtaldialdehyde for de- during aging. J. Dairy Sci. 74:727–740.
termination of proteolysis in milk and isolated milk proteins. J. Leuschner, R. G., M. Heidel, and W. P. Hammes. 1998. Histamine and
Dairy Sci. 66:1219–1227. tyramine degradation by food fermenting microorganisms. Int. J.
De Vuyst, A., W. Vervack, and M. Foulon. 1976. Détection d’amines Food Microbiol. 39:1–10.
non volatiles dans quelques fromages. Lait 56:414–422. Madkor, S., P. F. Fox, S. I. Shalabi, and N. H. Metwalli. 1987. Stud-
Delgado, F. J., J. Gonzalez-Crespo, R. Cava, and R. Ramirez. 2012. ies on the ripening of Stilton cheese: Proteolysis. Food Chem.
Changes in microbiology, proteolysis, texture and sensory charac- 25:13–29.
teristics of raw goat milk cheeses treated by high-pressure at dif- Malone, A. S., C. Wick, T. H. Shellhammer, and P. D. Courtney.
ferent stages of maturation. LWT-Food Sci. Technol. 48:268–275. 2003. High pressure effects on proteolytic and glycolytic enzymes
Edwards, S. T., and W. E. Sandine. 1981. Public health significance of involved in cheese manufacturing. J. Dairy Sci. 86:1139–1146.
amines in cheese. J. Dairy Sci. 64:2431–2438. Modler, H. W., J. R. Brunner, and C. M. Stine. 1974. Extracellular
Evert-Arriagada, K., M. M. Hernández-Herrero, B. Juan, B. Guamis, protease of Penicillium roqueforti. I. Production and characteris-
and A. J. Trujillo. 2012. Effect of high-pressure on fresh cheese tics of crude enzyme preparation. J. Dairy Sci. 57:523–527.
shelf-life. J. Food Eng. 110:248–253. Novella-Rodríguez, S., M. T. Veciana-Nogués, J. Saldo, and M. C.
Fernández, M., D. M. Linares, B. del Río, V. Ladero, and M. A. Al- Vidal-Carou. 2002. Effects of high hydrostatic pressure treatments
varez. 2007. HPLC quantification of biogenic amines in cheeses: on biogenic amine contents in goat cheeses during ripening. J.
Correlation with PCR-detection of tyramine-producing microor- Agric. Food Chem. 50:7288–7292.
ganisms. J. Dairy Res. 74:276–282. O’Reilly, C. E., A. L. Kelly, J. C. Oliveira, P. M. Murphy, M. A.
Fernandez-García, E., J. Tomillo, and M. Nuñez. 2000. Formation of E. Auty, and T. P. Beresford. 2003. Effect of varying high-pres-
biogenic amines in raw milk Hispánico cheese manufactured with sure treatment conditions on acceleration of ripening of Cheddar
proteinases and different levels of starter culture. J. Food Prot. cheese. Innov. Food Sci. Emerg. Technol. 4:277–284.
63:1551–1555. O’Reilly, C. E., P. M. O’Connor, A. L. Kelly, T. P. Beresford, and
Garde, S., J. L. Arqués, P. Gaya, M. Medina, and M. Nuñez. 2007. P. M. Murphy. 2000. Use of hydrostatic pressure for inactivation
Effect of high pressure treatments on the proteolysis and texture of microbial contaminants in cheese. Appl. Environ. Microbiol.
of ewes’ raw milk La Serena cheese. Int. Dairy J. 17:1424–1433. 66:4890–4896.
Garde, S., M. Ávila, P. Gaya, M. Medina, and M. Nuñez. 2006. Prote- Pircher, A., F. Bauer, and P. Paulsen. 2007. Formation of cadaver-
olysis of Hispánico cheese manufactured using lacticin 481-produc- ine, histamine, putrescine and tyramine by bacteria isolated from
ing Lactococcus lactis ssp. lactis INIA 639. J. Dairy Sci. 89:840– meat, fermented sausages and cheeses. Eur. Food Res. Technol.
849. 226:225–231.
Garde, S., J. Tomillo, P. Gaya, M. Medina, and M. Nuñez. 2002. Pro- Rabie, M. A., H. I. Siliha, S. M. El-Saidy, A. A. El-Badawy, and F. X.
teolysis in Hispánico cheese manufactured using a mesophilic start- Malcata. 2011. Effect of γ-irradiation upon biogenic amine forma-
er, a thermophilic starter and bacteriocin-producing Lactococcus tion in blue cheese during storage. Int. Dairy J. 21:373–376.
lactis ssp. lactis INIA 415 adjunct culture. J. Agric. Food Chem. Saldo, J., P. L. H. McSweeney, E. Sendra, A. L. Kelly, and B. Guamis.
50:3479–3485. 2002. Proteolysis in caprine milk cheese treated by high pressure
Gkatzionis, K., R. S. T. Linforth, and C. E. R. Dodd. 2009. Vola- to accelerate cheese ripening. Int. Dairy J. 12:35–44.
tile profile of Stilton cheeses: Differences between zones within a Santos, W. C., M. R. Souza, M. M. O. P. Cerqueira, and M. B. A.
cheese and dairies. Food Chem. 113:506–512. Gloria. 2003. Bioactive amines formation in milk by Lactococcus
Gobbetti, M., R. Burzigotti, E. Smacchi, A. Corsetti, and M. De An- in the presence or not of rennet and NaCl at 20 and 32 °C. Food
gelis. 1997. Microbiology and biochemistry of Gorgonzola cheese Chem. 90:219–230.
during ripening. Int. Dairy J. 7:519–529. Silla Santos, M. H. 1996. Biogenic amines: Their importance in foods.
Gomez, M. J., S. Garde, P. Gaya, M. Nuñez, and M. Medina. 1997. Int. J. Food Microbiol. 29:213–231.
Relationship between level of hydrophobic peptides and bitterness Stratton, J. E., R. W. Hutkins, and S. L. Taylor. 1991. Biogenic amines
in cheese made from pasteurized and raw milk. J. Dairy Res. in cheese and other fermented foods. J. Food Prot. 54:460–470.
64:289–297. Taylor, S. L. 1986. Histamine food poisoning: Toxicology and clinical
Gomez, M. J., E. Rodríguez, P. Gaya, M. Nuñez, and M. Medina. aspects. Crit. Rev. Toxicol. 17:91–128.
1999. Characteristics of Manchego cheese manufactured from raw ten Brink, B., C. Damink, H. M. L. J. Joosten, and J. H. Huis in ’t
and pasteurized ovine milk and with defined-strain and or com- Veld. 1990. Occurrence and formation of biologically-active amines
mercial mixed-strain starter cultures. J. Dairy Sci. 82:2300–2307. in foods. Int. J. Food Microbiol. 11:73–84.
Gripon, J.-C., M. J. Desmazeaud, D. Le Bars, and J. L. Bergère. Til, H. P., H. E. Falke, M. K. Prinsen, and M. I. Willems. 1997. Acute
1977. Role of proteolytic enzymes of Streptococcus lactis, Penicil- and subacute toxicity of tyramine, spermidine, spermine, putres-
lium roqueforti, and Penicillium caseicolum during cheese ripening. cine and cadaverine in rats. Food Chem. Toxicol. 35:337–348.
J. Dairy Sci. 60:1532–1538.

Trieu-Cuot, P., M.-J. Archieri-Haze, and J.-C. Gripon. 1982. Effect Wick, C., U. Nienaber, O. Anggraeni, T. H. Shellhammer, and P. D.
of aspartyl proteinases of Penicillium caseicolum and Penicillium Courtney. 2004. Texture, proteolysis and viable lactic acid bacte-
roqueforti on caseins. J. Dairy Res. 49:487–500. ria in commercial Cheddar cheeses treated with high pressure. J.
Visser, S. 1993. Proteolytic enzymes and their relation to cheese ripen- Dairy Res. 71:107–115.
ing and flavor: An overview. J. Dairy Sci. 76:329–350. Yvon, M., and L. Rijnen. 2001. Cheese flavor formation by amino acid
Voigt, D. D., F. Chevalier, M. C. Qian, and A. L. Kelly. 2010. Effect of catabolism. Int. Dairy J. 11:185–201.
high-pressure treatment on microbiology, proteolysis, lipolysis and
levels of flavor compounds in mature blue-veined cheese. Innov.
Food Sci. Emerg. Technol. 11:68–77.

Capítulo 3.
High-pressure processing decelerates lipolysis and formation
of volatile compounds in ovine milk blue-veined cheese.

103

Fotografía: colonias de Penicillium roqueforti y bacterias lácticas en agar M17.

J. Dairy Sci. 96:7500–7510
http://dx.doi.org/10.3168/jds.2013-7221
© American Dairy Science Association®, 2013.
High-pressure processing decelerates lipolysis and formation

of volatile compounds in ovine milk blue-veined cheese
J. Calzada, A. Del Olmo, A. Picon, P. Gaya, and M. Nuñez1
Departamento de Tecnología de Alimentos, Instituto Nacional de Investigación y Tecnología Agraria y Alimentaria (INIA), 28040 Madrid, Spain
ABSTRACT Key words: high-pressure processing, lipolysis, vola-

tile compound, blue-veined cheese
Enzyme-rich cheeses are prone to over-ripening dur-
ing refrigerated storage. Blue-veined cheeses fall within
this category because of the profuse growth of Penicil- INTRODUCTION
lium roqueforti in their interior, which results in the
production of highly active proteinases, lipases, and Flavor, rheological properties, and visual appear-
other enzymes responsible for the formation of a great ance determine cheese quality (Fox and Wallace, 1997).
number of flavor compounds. To control the exces- Cheese flavor, probably the main trait influencing its
sive formation of free fatty acids (FFA) and volatile quality, is caused by the interaction of many compounds
compounds, blue-veined cheeses were submitted to responsible for taste and aroma. These compounds are
high-pressure processing (HPP) at 400 or 600 MPa produced during manufacture and ripening through the
on d 21, 42, or 63 after manufacture. Cheeses were metabolism of lactose, lactate, and citrate, the libera-
ripened for 30 d at 10°C and 93% relative humidity, tion of FFA, and the degradation of caseins to peptides
followed by 60 d at 5°C, and then held at 3°C until and free amino acids (McSweeney and Sousa, 2000;
d 360. High-pressure processing influenced the concen- Collins et al., 2003). Primary degradation is followed
trations of acetic acid and short-chain, medium-chain, by the secondary catabolism of the resulting products
and long-chain FFA. The effect was dependent on to compounds that, in many cases, have higher flavor
treatment conditions (pressure level and cheese age at impact than their respective precursors. More than 600
the time of treatment). The lowest concentrations of volatile compounds have been identified in cheese, most
acetic acid and FFA were recorded for cheeses treated of which have been associated with particular odor and
at 600 MPa on d 21; these cheeses showed the lowest aroma notes (Molimard and Spinnler, 1996; Curioni
esterase activity values. Acetic acid and all FFA groups and Bosset, 2002).
increased during ripening and refrigerated storage. The Coagulant enzymes, together with lactic starter
102 volatile compounds detected in cheese belonged to cultures and their enzymes, are responsible for the
10 chemical groups (5 aldehydes, 12 ketones, 17 alco- biochemical changes occurring during the manufacture
hols, 12 acids, 35 esters, 9 hydrocarbons, 5 aromatic and ripening of semihard and hard cheeses made from
compounds, 3 nitrogen compounds, 3 terpenes, and 1 pasteurized milk, because most of the microorganisms
sulfur compound). High-pressure processing influenced and enzymes present in raw milk have been inactivated
the levels of 97 individual compounds, whereas 68 in- by the thermal treatment. In the case of blue-veined
dividual compounds varied during refrigerated storage. cheeses, Penicillium roqueforti is an additional major
Total concentrations of all groups of volatile compounds ripening agent responsible for their unique flavor.
were influenced by HPP, but only ketones, acids, esters, Penicillium roqueforti consumes lactic acid, causing an
and sulfur compounds varied during refrigerated stor- increase in cheese pH value favorable for many chemical
age. The lowest total concentrations for most groups of reactions, produces extracellular proteinases and lipases
volatile compounds were recorded for the cheese pres- (Gripon et al., 1977; Lamberet and Menassa, 1983),
surized at 600 MPa on d 21. A principal component and has the ability to form methyl ketones through
analysis combining total concentrations of groups of the β-oxidation of FFA followed by a decarboxylation
FFA and volatile compounds discriminated cheeses by reaction (Kinsella and Hwang, 1976).
age and by the pressure level applied to HPP cheeses. The activity of enzymes and microorganisms per-
sists during the refrigerated storage of ripe cheese at
distribution and retail, which can cause over-ripening
if levels of flavor compounds above the desired bal-
Received July 2, 2013.
Accepted September 5, 2013. anced concentrations for a particular cheese variety
1
Corresponding author: nunez@inia.es are attained. Thus, the cheese purchased by the con-
7500
LIPOLYSIS AND VOLATILE COMPOUNDS IN BLUE CHEESE 7501
sumer may have a stronger or different flavor than the L of milk inoculated with lactic cultures and P. roque-
manufacturer intended (Wick et al., 2004). Blue-veined forti. Cheeses, 18 cm in diameter and 10 cm high, were
cheeses, because of their richness in enzymatic activi- ripened at 10°C and 93% relative humidity until d 30
ties, seem particularly prone to over-ripening defects and then at 5°C from d 30 to d 90. After 90 d, they were
during the refrigerated storage of ripe cheese. An ap- held at 3°C until d 360.
proach to prevent over-ripening and prolong the shelf Cheeses were pressurized at 400 or 600 MPa for 5
life of ripe cheese is frozen storage. Although cheese min, after 21, 42, or 63 d of ripening, as described
flavor remains unchanged at thawing, both texture and by Calzada et al. (2013). Treatments were coded as
visual appearance are negatively affected by freezing 400W3, 600W3, 400W6, 600W6, 400W9, and
(Tejada et al., 2000; Van Hekken et al., 2005). 600W9 according to the pressure level applied (400
High-pressure processing (HPP), with a negligible or 600 MPa) and the age of cheese (3, 6, or 9 wk) at
effect on flavor characteristics, meets the increasing pressurization. Cheeses were unpackaged after HPP,
consumer demand for fresh-tasting, minimally pro- and ripening and storage proceeded under the same
cessed foods. It has been successfully applied to milk conditions as for control cheese.
and cheese for the inactivation of pathogenic and spoil- A different cheese per treatment (1 control and 6
age microorganisms (O’Reilly et al., 2000; Arqués et HPP) was sampled at each of the times of analysis.
al., 2006). In addition, HPP may be a useful tool for Two 100-g pieces per cheese were wrapped in aluminum
the inactivation of enzymes present in cheese such as foil, vacuum-packaged, and frozen at −40°C for chemi-
proteinases (García-Risco et al., 2003; Huppertz et al., cal analyses.
2004), peptidases (Malone et al., 2003; Juan et al.,
2007), and esterases (Ávila et al., 2007). The formation FFA Determination
of volatile compounds in cheese is also influenced by
HPP, at a variable degree that depends on the pres- Acetic acid, propionic acid, and FFA from butyric
sure level applied and the age of cheese at the time (C4:0) to linolenic acid (C18:3) in cheese were determined
of treatment (Ávila et al., 2006; Arqués et al., 2007). by gas chromatography with flame-ionization detec-
Consequently, HPP seems a feasible procedure to pre- tion, as described by Fernández-García et al. (2006),
vent over-ripening during the refrigerated storage of with elution in 8 mL of diethyl ether containing 2%
blue-veined cheese. formic acid. Frozen cheese pieces were thawed overnight
In a previous study, we reported the effect of HPP at 4°C before analysis. At all sampling times, acids were
on the proteolysis and formation of biogenic amines in extracted from cheeses using a solid-phase extraction
blue-veined cheese made from ovine milk (Calzada et technique, with pentanoic, nonanoic, and heptadecanoic
al., 2013). However, the effects of HPP on the lipolysis acids added as internal standards. A Hewlett-Packard
and formation of volatile compounds in blue-veined 6890 gas chromatograph (Agilent Technologies, Las Ro-
cheese are not well known. In the only work published zas, Spain) equipped with an automatic sampler (HP
on the subject (Voigt et al., 2010), the authors did 7683), a split/splitless injector, a FFAP column (Agi-
not find significant differences in the concentrations of lent Technologies, 30 m × 0.32 mm i.d. × 0.25 μm film
FFA and methyl ketones when comparing pressurized thickness) and a flame-ionization detector was used for
and control blue-veined cheeses, a result that could the analysis. Injection (1 μL of sample) was performed
be ascribed to the short refrigerated storage period of in split mode at 1:20 split ratio, at 260°C. Helium was
cheeses after HPP (only 28 d). In the present work, we the carrier gas, with the flow set for maintaining a
investigated the influence of HPP applied to ovine milk constant pressure of 0.80 kg/cm2. For chromatographic
blue-veined cheese at 400 or 600 MPa on d 21, 42, or separation, the temperature was increased from 65 to
63 after manufacture on the lipolysis and formation of 240°C at a rate of 10°C/min, and held at 240°C for 12.5
volatile compounds during a 90-d ripening period and min. Fifteen standard solutions of FA were used for
a further 270-d refrigerated storage period. the calculation of calibration curves. Individual FFA
were separated, identified, and quantified, and their
concentrations expressed in milligrams per gram of
MATERIALS AND METHODS
cheese DM.
Cheese Manufacture and HPP
Determination of Esterase Activity
The manufacturing procedure of blue-veined cheese
from pasteurized ovine milk was described in a previous Esterase activity was determined in duplicate on
work (Calzada et al., 2013). Two batches of blue-veined cheese extracts according to the method described by
cheese were made on consecutive days, each from 1,200 Ávila et al. (2007) with some modifications. Ten grams
7502 CALZADA ET AL.
of cheese was homogenized with 20 mL of phosphate CA), with helium flow at 1.4 mL/min for 1 min fol-
buffer (0.1 M, pH 7.0) in an Ultra-Turrax T8 homog- lowed by 1 mL/min, and the following temperature
enizer (IKA Labortechnik, Staufen, Germany), followed program: 7 min at 40°C, first ramp 2°C/min to 90°C,
by centrifugation at 10,000 × g for 20 min at 4°C and second ramp 3°C/min to 150°C, final ramp 9°C/min to
filtering through Whatman No. 2 paper (Whatman 240°C, and 8 min at 240°C. Detection was performed
International Ltd., Maidstone, UK). The chromogenic with electron impact ionization, with 70 eV ionization
substrate was α-naphthylbutyrate (Sigma-Aldrich, energy operating in the full-scan mode at 1.74 scans/s.
Steinhem, Germany). The assay mixture contained 30 Source and quadrupole temperatures were 230 and
μL of chromogenic substrate, 600 μL of distilled water, 150°C, respectively. Compound identification was car-
and different volumes (100, 200, or 400 μL) of cheese ried out by injection of commercial standards and by
homogenate and phosphate buffer (0.1 M, pH 7.5) to a spectra comparison using the Wiley7Nist05 Library
final volume of 1,230 μL. The assay mixture was incu- (Wiley and Sons Inc., Weinheim, Germany). The sum
bated for 1 h at 37°C in a water bath and centrifuged at of abundances of characteristic ions was used for semi-
12,000 × g for 5 min at room temperature. Finally, 900 quantitation of compounds. The relative abundances
μL of supernatant was mixed with 150 μL of Fast Red of volatile compounds were calculated by multiplying
TR salt (Sigma-Aldrich) aqueous solution (2.7 mg/ the respective peak areas by 103 and dividing by the
mL). After 5 min at room temperature, the absorbance cyclohexanone peak area..
was measured at 537 nm using a DU650 spectropho-
tometer (Beckman Coulter Inc., Brea, CA). Esterase Statistical Treatment
activity was calculated from absorbance values in the
range of 0.1 to 0.9 by means of a α-naphthol standard Data obtained were analyzed by a 2-way ANOVA,
curve. One unit of enzymatic activity was defined as the with treatment (6 HPP treatments and control) and
amount of enzyme that liberated 1 pmol of α-naphthol cheese age as the main effects. Means were compared
per minute and gram of cheese at 37°C and pH 7.5. using Tukey’s test, with significance declared at P ≤
0.05. Principal component analysis was carried out on
Determination of Volatile Compounds the total concentrations of groups of FFA and volatile
compounds of 180-d and 360-d cheeses for the discrimi-
Volatile compounds were extracted from cheese us- nation of samples according to treatment and cheese
ing a solid-phase microextraction method (Mallia et age. The SPSS Win 14.0 software (SPSS Inc., Chicago,
al., 2005). Five grams of cheese was homogenized in IL) was used for the statistical analysis of data.
a mechanical grinder with 35 g of Na2SO4 and 100 μL
of an aqueous solution of 1,058 mg/L cyclohexanone
RESULTS AND DISCUSSION
as internal standard. Two grams of the mixture was
weighed in a 15-mL headspace glass vial sealed with a Acetic and Propionic Acids
polytetrafluoroethylene (PTFE)-faced silicone septum
(Supelco, Bellefonte, PA). Vials were submerged in a Acetic acid derives mostly from metabolism of lac-
thermostat-controlled bath at 30°C (D3 model, Haake, tose, lactate, and citrate by lactic acid bacteria and
Berlin, Germany) for both equilibration (20 min) and other microorganisms (McSweeney and Sousa, 2000).
extraction (30 min) phases. A solid-phase microextrac- Both treatment and cheese age significantly influenced
tion manual holder equipped with a 75-μm StableFlex (P < 0.001) the concentration of acetic acid in blue-
carboxen/polydimethylsiloxane (CAR/PDMS) coated veined ovine milk cheese (Table 1). In control cheese,
fiber (Supelco) was inserted through the PTFE septum it increased during ripening from 1.11 mg/g of DM on
for headspace extraction, after which it was inserted d 1 to 1.62 mg/g of DM on d 90, and afterward de-
into the GC injection port for desorption (270°C/10 clined slightly during refrigerated storage, to 1.39 mg/g
min in splitless mode). Before use, the fiber was con- of DM on d 360 (Table 2). High-pressure processing
ditioned in the injection port of the GC (300°C/ 1 h) of cheeses on d 21 arrested the production of acetic
as recommended by manufacturer. After each run, the acid, independently of the pressure level applied, more
fiber was cleaned up to avoid carryover problems and, markedly than in cheeses pressurized on d 42 or d 63.
periodically, fiber sensitivity was tested with an aque- At the end of ripening, the 400W3 and 600W3 cheeses
ous solution of the internal standard. All analyses were showed the lowest concentrations of acetic acid, differ-
run using the same fiber unit. ences that persisted until d 360. Acetic acid, a major
Chromatography (GC-MS) was carried out in a cap- odorant of Cheddar, Gruyère, and Emmental cheeses
illary column (60 m long, 0.25 mm i.d., 0.5 μm film (Curioni and Bosset, 2002), plays an important role in
thickness; Zebron-WAX plus, Phenomenex, Torrance, cheese flavor and aroma by itself and as a substrate for
Table 1. Levels of significance of the ANOVA main effects high- at low concentrations that accounted for 0.4 to 0.6% of
pressure processing (HPP) treatment and cheese age (T), and their
interaction (HPP × T), on the total concentrations of the main groups
total carboxylic acids (Moio et al., 2000).
of carboxylic acids and volatile compounds in ovine milk blue-veined Branched-chain carboxylic acids, deriving from the
cheeses metabolism of leucine, isoleucine, and valine (Yvon and
Factor Rijnen, 2001), were not detected in control or HPP
Chemical compound cheeses. These compounds have been found in many
or group HPP T HPP × T cheese varieties (Curioni and Bosset, 2002), includ-
Acetic acid *** *** *** ing Gorgonzola cheese, in which they represented up
Short-chain FFA *** *** ** to 4.4% of total carboxylic acids (Moio et al., 2000).
Medium-chain FFA *** *** *** Branched-chain carboxylic acids may be formed by cer-
Long-chain FFA *** *** ***
Aldehydes *** NS *** tain strains of lactic acid bacteria and gram-negative
Ketones *** *** *** bacteria such as Pseudomonas spp. (Morales et al.,
Alcohols *** NS NS 2005). The high microbiological quality of the cheeses
Acids *** *** *
Esters *** ** ** studied in the present work, with contaminating bacte-
Hydrocarbons *** NS NS ria at very low levels (Calzada et al., 2013), probably
Aromatic compounds *** NS NS impeded the formation of these compounds.
Nitrogen compounds *** NS *
Terpenes *** NS NS
Sulfur compounds * ** NS
Free Fatty Acids
***P < 0.001; **P < 0.01; *P < 0.05.
Short-chain (SC, C4:0 to C8:0) FFA originate from
ester formation through esterification reactions. In the esterase- or lipase-mediated hydrolysis of triacylglyc-
present work, the decline in acetic acid concentration erides, but also from the fermentation of lactose and
generally observed during refrigerated storage may be lactate, from the degradation of amino acids, and from
ascribed to ester formation. the oxidation of some ketones, esters, and aldehydes
Propionic acid was found at low concentrations, (Molimard and Spinnler, 1996; Collins et al., 2003). In
ranging from 0.01 to 0.03 mg/g of cheese DM (data the present work, the formation of SC FFA was signifi-
not shown), during ripening and refrigerated storage cantly (P < 0.001) influenced by treatment and cheese
of control and HPP cheeses, without a clear influence age (Table 1). Total SC FFA concentration increased
of treatment or cheese age. Propionic acid, mainly 51.7-fold during ripening of control cheese from d 1 to
formed by lactate-metabolizing microorganisms, is 90, and 2.9-fold during refrigerated storage until d 360
characteristic of Swiss-type cheeses and has also been (Table 2). The respective levels of individual SC FFA
detected in Cheddar and Camembert cheeses (Curioni C4:0, C6:0, and C8:0 were 1.32, 0.85, and 0.93 mg/g of
and Bosset, 2002). Within blue-veined varieties, its DM on d 90, and reached 2.94, 2.39, and 3.56 mg/g of
presence has been reported only in Gorgonzola cheese, DM on d 360.
Table 2. Concentrations of acetic acid and short-chain free fatty acids (SC FFA) during ripening and refrigerated storage of ovine milk blue-
veined control and high-pressure processed (HPP) cheeses pressurized at 400 or 600 MPa after 3, 6, or 9 wk (W3, W6, W9) of ripening1
Chemical compound
or group Day Control 400W3 600W3 400W6 600W6 400W9 600W9
Acetic acid 1 1.11 ± 0.03
21 0.63 ± 0.10b 0.45 ± 0.08b 0.19 ± 0.02a
42 1.37 ± 0.09c 0.52 ± 0.05ab 0.24 ± 0.02a 0.83 ± 0.09b 0.62 ± 0.09b
63 1.72 ± 0.21d 0.61 ± 0.07ab 0.39 ± 0.02a 1.08 ± 0.13bc 0.75 ± 0.08abc 1.21 ± 0.11bcd 1.35 ± 0.10cd
90 1.62 ± 0.15c 0.50 ± 0.06a 0.44 ± 0.07a 1.72 ± 0.18c 1.20 ± 0.14bc 1.51 ± 0.23c 0.79 ± 0.12ab
180 1.67 ± 0.12d 0.46 ± 0.01ab 0.39 ± 0.02a 1.33 ± 0.03cd 0.87 ± 0.12bc 0.92 ± 0.10bc 0.68 ± 0.09ab
270 1.49 ± 0.09d 0.55 ± 0.09ab 0.36 ± 0.02a 1.11 ± 0.09cd 0.70 ± 0.04abc 1.42 ± 0.09d 0.92 ± 0.08bc
360 1.39 ± 0.11d 0.55 ± 0.02ab 0.28 ± 0.01a 1.42 ± 0.07d 0.72 ± 0.06bc 1.07 ± 0.06cd 0.58 ± 0.03ab
SC FFA (C4:0-C8:0) 1 0.06 ± 0.01
21 0.72 ± 0.15b 0.74 ± 0.11b 0.19 ± 0.08a
42 2.82 ± 0.38bc 0.82 ± 0.16a 0.34 ± 0.10a 3.25 ± 0.22c 2.47 ± 0.39b
63 2.37 ± 0.20b 0.91 ± 0.19a 0.88 ± 0.21a 3.95 ± 0.36c 3.50 ± 0.42bc 3.81 ± 0.16c 2.25 ± 0.25ab
90 3.10 ± 0.33ab 1.44 ± 0.10a 1.20 ± 0.41a 5.78 ± 0.58c 5.35 ± 0.90c 4.50 ± 0.91bc 5.66 ± 0.69c
180 5.06 ± 0.49c 1.92 ± 0.08ab 1.70 ± 0.80a 6.38 ± 0.78c 4.40 ± 0.85abc 5.95 ± 0.83c 4.56 ± 0.53bc
270 6.32 ± 0.68b 2.83 ± 0.43a 1.58 ± 0.54a 7.94 ± 0.93b 6.67 ± 1.08b 7.20 ± 1.16b 7.74 ± 0.91b
360 8.88 ± 0.78cd 5.21 ± 0.84ab 2.21 ± 0.42a 11.81 ± 1.29d 7.17 ± 0.16bc 10.02 ± 1.02cd 7.55 ± 0.55bc
a–d
1
Mean ± SE (n = 4) of duplicate determinations in 2 cheese-making experiments. Results are expressed in milligrams per gram of cheese DM.

7504 CALZADA ET AL.
Table 3. Concentrations of medium-chain (MC) and long-chain (LC) FFA during ripening and refrigerated storage of ovine milk blue-veined
control and high-pressure processed (HPP) cheeses pressurized at 400 or 600 MPa after 3, 6, or 9 wk (W3, W6, W9) of ripening1
Chemical group Day Control 400W3 600W3 400W6 600W6 400W9 600W9
MC FFA (C10:0-C14:0) 1 0.27 ± 0.07
21 0.92 ± 0.28b 0.92 ± 0.17b 0.32 ± 0.11a
42 3.86 ± 0.83b 1.10 ± 0.14a 0.78 ± 0.26a 5.67 ± 0.78b 3.61 ± 0.63b
63 3.29 ± 0.39abc 1.38 ± 0.23a 2.03 ± 0.55ab 6.30 ± 1.14c 5.22 ± 0.80bc 5.89 ± 0.32c 3.01 ± 0.41abc
90 5.47 ± 0.54abc 1.79 ± 0.15a 3.60 ± 1.22ab 10.84 ± 1.73bc 11.57 ± 1.61c 7.93 ± 1.47abc 11.53 ± 1.96c
180 12.17 ± 1.40b 2.70 ± 0.51a 4.25 ± 1.01a 14.26 ± 2.26b 9.31 ± 1.66ab 12.27 ± 2.67b 7.96 ± 1.82ab
270 14.26 ± 1.79b 4.33 ± 0.88a 3.84 ± 1.27a 14.64 ± 2.01b 10.95 ± 1.80b 16.67 ± 2.36b 14.40 ± 2.45b
360 18.31 ± 1.72bc 11.84 ± 2.59ab 4.50 ± 0.19a 25.74 ± 3.32c 13.55 ± 0.87ab 18.60 ± 2.07bc 12.92 ± 1.96ab
LC FFA (C16:0-C18:3) 1 1.06 ± 0.13
21 5.62 ± 1.21b 5.44 ± 1.57b 2.74 ± 0.88a
42 16.94 ± 2.67bc 5.52 ± 1.19a 3.85 ± 0.70a 24.44 ± 2.43c 13.21 ± 2.73ab
63 15.16 ± 1.51abc 5.98 ± 1.40a 6.72 ± 1.12a 28.06 ± 3.15c 23.49 ± 3.04bc 19.98 ± 1.32bc 10.38 ± 1.08ab
90 19.21 ± 1.83abc 7.13 ± 0.94a 11.50 ± 2.79ab 34.69 ± 4.02bc 38.77 ± 4.86c 26.70 ± 5.02abc 37.85 ± 6.69c
180 36.14 ± 4.00c 10.55 ± 1.23a 12.19 ± 3.82ab 41.09 ± 6.06c 31.19 ± 7.06abc 39.84 ± 6.82c 32.89 ± 5.02bc
270 44.12 ± 5.62b 12.09 ± 2.29a 13.01 ± 2.55a 45.20 ± 7.50b 32.00 ± 8.50ab 50.28 ± 8.95b 45.01 ± 7.39b
360 56.63 ± 4.54bc 38.30 ± 7.96ab 15.11 ± 1.07a 66.85 ± 9.08c 44.17 ± 3.40bc 54.66 ± 5.87bc 40.23 ± 6.81b
a–c
1
Mean ± SE (n = 4) of duplicate determinations in 2 cheese-making experiments. Results are expressed in milligrams per gram of cheese DM.
Medium-chain (MC, C10:0 to C14:0) FFA are formed half those in 400W3 cheese. Milk lipoprotein lipase was
through the lipase-mediated hydrolysis of triacylglyc- presumably inactivated by milk pasteurization in the
erides. Penicillium roqueforti produces 2 extracellular present work.
lipases, an acidic lipase and an alkaline lipase that re- Lactic acid bacteria esterases may withstand cheese
tains activity at pH 4.5 (Lamberet and Menassa, 1983). pressurization, according to Ávila et al. (2007). To our
The formation of MC FFA was significantly (P < 0.001) knowledge, no information on the barotolerance of P.
influenced by treatment and cheese age (Table 1). To- roqueforti lipolytic enzymes has been reported. The es-
tal MC FFA concentration increased 20.3-fold during terase activity values found in the present work, up to
ripening of control cheese and 3.3-fold during refriger- 38.53 pmol of α-naphthol per min per gram for control
ated storage (Table 3). Individual MC FFA C10:0, C12:0, cheese on d 42 (Table 4), were markedly higher than
and C14:0 levels were 2.32, 1.17, and 1.99 mg/g of DM, those reported for a non-mold-ripened variety, which
respectively, on d 90 and attained 7.89, 3.80, and 6.62 ranged from 0.49 to 1.39 pmol of α-naphthol per min
mg/g of DM on d 360. per gram of cheese (Ávila et al., 2007). The high esterase
Long-chain (LC, C16:0 to C18:3) FFA in blue-veined activity values of blue-veined cheese must be attributed
cheeses are mostly derived from triacylglycerides, by to P. roqueforti lipolytic enzymes. These enzymes ex-
the action of P. roqueforti lipases. The accumulation of hibited a certain baroresistance, according to the data
LC FFA in cheese was significantly (P < 0.001) influ- presented in Table 4. Even in the 600W3 cheese, which
enced by treatment and cheese age (Table 1). Total LC showed the lowest (P < 0.05) esterase activity values
FFA concentration increased 18.1-fold during ripening throughout ripening and refrigerated storage, esterase
of control cheese and 2.9-fold during refrigerated stor- activity reached 9.12 pmol of α-naphthol per min per
age (Table 3). The major LC FFA in control cheese, gram on d 90 of ripening. Total FFA concentrations
C16:0 and C18:1, reached levels of 4.31 and 11.14 mg/g of in 90-d cheeses did not correlate significantly with the
DM, respectively, on d 90, and 19.43 and 27.03 mg/g respective esterase activity values. However, total FFA
of DM on d 360. As previously recorded for Stilton concentrations in 360-d cheeses correlated significantly
cheese by Madkor et al. (1987), LC FFA were present (P < 0.05) with the respective esterase activity values,
at higher concentrations than SC and MC FFA. and even more strongly (P < 0.01) with the esterase
Pressurization of cheeses on d 21 limited the forma- activity values obtained for the respective cheeses on d
tion of SC, MC, and LC FFA (Tables 2 and 3). In 90. Levels of significance for the correlations between
contrast, Voigt et al. (2010) did not observe significant SC, MC, and LC FFA and esterase activity values
differences between the FFA content of pressurized equaled those found for total FFA and esterase activity.
and control blue-veined cheeses. In the present work, Butanoic acid plays an important role in the flavor
the effect was more marked for 600W3 cheese than for of many cheese varieties, although at high concentra-
400W3 cheese. On d 360, the concentrations of total tions (usually in cheeses with the late blowing defect
SC, MC, and LC FFA in 600W3 cheese were less than caused by the butyric acid fermentation of lactate by

Table 4. Esterase activity during ripening and refrigerated storage of ovine milk blue-veined control cheese and high-pressure processed (HPP)
cheeses pressurized at 400 or 600 MPa after 3, 6, or 9 wk (W3, W6, W9) of ripening1
Day Control 400W3 600W3 400W6 600W6 400W9 600W9

c b a
21 16.72 ± 0.40 12.62 ± 0.59 4.66 ± 0.67
42 38.53 ± 1.35d 13.71 ± 0.80b 4.98 ± 0.43a 28.85 ± 0.12c 17.74 ± 2.21b
63 23.78 ± 0.67b 15.19 ± 1.05ab 6.57 ± 1.05a 26.80 ± 2.41b 15.96 ± 1.47ab 13.96 ± 0.61ab 10.17 ± 1.90a
90 22.01 ± 2.47bc 17.82 ± 4.02abc 9.12 ± 2.30a 27.36 ± 3.00c 12.46 ± 0.96ab 19.53 ± 0.91abc 16.17 ± 2.60abc
180 33.56 ± 2.25bc 12.49 ± 0.46a 7.65 ± 0.36a 34.47 ± 4.03c 28.37 ± 5.96bc 19.64 ± 3.13ab 10.71 ± 2.67a
270 31.74 ± 0.33d 10.89 ± 1.43ab 5.52 ± 0.80a 32.12 ± 3.76d 17.02 ± 0.72bc 32.01 ± 1.48d 19.75 ± 0.49c
360 28.16 ± 2.06c 8.41 ± 0.43ab 4.31 ± 0.22a 26.18 ± 0.88c 5.95 ± 0.23ab 11.63 ± 0.10b 8.47 ± 0.71ab
a–d
1
Mean ± SE (n = 4) of duplicate determinations in 2 cheese-making experiments. Results are expressed in picomoles of α-naphthol per minute
and gram of cheese.
clostridia), it becomes undesirable. Hexanoic and octa- on d 180 (Table 5), with acetaldehyde being the main
noic acids are characteristic flavor compounds of aged aldehyde at that time. During cheese manufacture and
Grana Padano and Roncal cheeses (Curioni and Bosset, the first days of ripening, acetaldehyde may be formed
2002). Medium-chain FFA such as C10:0 and C12:0 are through the metabolism of lactose, but later it mostly
key aroma compounds in varieties such as Cheddar, derives from the catabolism of threonine (McSweeney
Roncal, and probably others, because of their relatively and Sousa, 2000). The population of lactic acid bacte-
low perception thresholds. In contrast, LC FFA have ria, which declined on average 0.5 log cfu/g in 400 MPa
high perception thresholds, which limit their contribu- cheeses and 4.2 log cfu/g in 600 MPa cheeses imme-
tion to cheese flavor, in spite of the high concentrations diately after HPP, attained respective mean counts of
commonly reached in many cheese types (Curioni and 8.0, 7.1, and 3.5 log cfu/g in control, 400 MPa, and 600
Bosset, 2002). MPa cheeses on d 180 (Calzada et al., 2013). The lower
acetaldehyde content of 600 MPa cheeses on d 180 may
Volatile Compounds be explained by their low lactic acid bacteria counts and
total aerobic counts, which preclude a metabolic activ-
One hundred two compounds were identified in the ity capable of influencing cheese chemical parameters.
volatile fraction of ovine milk blue-veined cheese by Total aldehydes increased in 600 MPa cheeses from d
solid-phase microextraction followed by GC-MS. Lower 180 to 360 due to the formation of 3-methylbutanal
numbers of volatile compounds have generally been (the main aldehyde on d 360), 2-methylbutanal, and
reported for other blue-veined cheeses made exclusively 2-methylpropanal from leucine, isoleucine, and valine,
from bovine milk, such as Gorgonzola or Stilton (Moio respectively, during the last stages of refrigerated stor-
et al., 2000; Gkatzionis et al., 2009), or including ovine age. The increase in total aldehydes recorded from d
milk in their composition, such as Cabrales and Gam- 180 to 360 in 600 MPa cheeses cannot be attributed
onedo (González de Llano et al., 1990; De Frutos et al., to microbial metabolism, because of the low counts of
1991), although 108 volatile compounds were identi- lactic acid bacteria, P. roqueforti, and other microbial
fied or tentatively identified in ovine milk Roquefort groups in those cheeses. Therefore, abiotic chemical re-
cheese (Gallois and Langlois, 1990). The 102 volatile actions seem the most plausible origin of the branched-
compounds identified in the present work included 5 chain aldehydes produced from d 180 to 360 in 600 MPa
aldehydes, 12 ketones, 17 alcohols, 12 acids, 35 esters, cheeses, which doubled their contents during this pe-
9 hydrocarbons, 5 aromatic compounds, 3 nitrogen riod. Aldehydes are key odorants in cheese, with green,
compounds, 3 terpenes, and 1 sulfur compound. Total sweet, pungent notes for acetaldehyde; green, malty,
concentrations of all groups of volatile compounds were acrid, pungent notes that turn into pleasant fruity at
significantly influenced by HPP, but only ketones, ac- low concentrations for 3-methylbutanal; malty, nutty
ids, esters, and sulfur compounds varied significantly notes for 2-methylbutanal; and malty, floral notes for
with cheese age during refrigerated storage (Table 1). 2-methylpropanal (Curioni and Bosset, 2002).
Out of the 102 individual volatile compounds, 97 were Total ketones reached their maximum concentration
significantly influenced by HPP (86 compounds at P < on d 180 in control cheese (Table 5), with 2-pentanone,
0.001), whereas 68 varied significantly with cheese age 2-heptanone, and 2-propanone as the major ketones.
(52 compounds at P < 0.001). Penicillium roqueforti is a main producer of methyl
Total aldehydes reached higher levels in control ketones via the β-oxidation and decarboxylation of
cheese and 400 MPa cheeses than in 600 MPa cheeses FFA, with enhanced production by germinating spores

7506 CALZADA ET AL.
Table 5. Levels of total volatile aldehydes, ketones, alcohols, acids, and esters during refrigerated storage of ovine milk blue-veined control and
high-pressure processed (HPP) cheeses pressurized at 400 or 600 MPa after 3, 6, or 9 wk (W3, W6, W9) of ripening1
Volatile group Day Control cheese 400W3 600W3 400W6 600W6 400W9 600W9
b b a b a ab
Aldehydes 180 15.26 ± 1.43 14.51 ± 1.39 5.32 ± 0.26 15.34 ± 1.65 6.63 ± 0.38 10.40 ± 1.46 5.40 ± 0.66a
360 10.64 ± 0.88a 10.58 ± 0.66a 8.88 ± 1.14a 7.06 ± 0.78a 11.74 ± 1.74a 9.57 ± 0.72a 11.35 ± 0.79a
Ketones 180 1,285 ± 178b 809 ± 107ab 397 ± 25a 355 ± 39a 847 ± 126ab 649 ± 106a 726 ± 149ab
360 1,364 ± 152cd 1,717 ± 201d 670 ± 18ab 1,837 ± 133d 1,005 ± 91abc 1,280 ± 112bcd 534 ± 53a
Alcohols 180 1,452 ± 105d 806 ± 21bc 209 ± 9a 1,070 ± 21c 512 ± 74ab 881 ± 95bc 332 ± 28a
360 1,496 ± 83c 744 ± 54b 239 ± 7a 1,281 ± 115c 481 ± 26ab 775 ± 46b 242 ± 13a
Acids 180 2,465 ± 107b 1,517 ± 144a 1,441 ± 211a 2,631 ± 55b 2,506 ± 88b 2,804 ± 123b 2,844 ± 88b
360 4,211 ± 166c 2,955 ± 529ab 2,090 ± 212a 4,455 ± 270c 3,611 ± 77bc 4,390 ± 112c 3,971 ± 64bc
Esters 180 868 ± 95d 200 ± 15ab 47.2 ± 6.1a 698 ± 52cd 284 ± 23ab 506 ± 66bc 190 ± 14ab
360 1,493 ± 167c 420 ± 36ab 61.2 ± 4.8a 760 ± 55b 311 ± 30ab 841 ± 24b 173 ± 15a
a–d
1
Mean ± SE (n = 4) of duplicate determinations in 2 cheese-making experiments. Results are expressed as relative abundances with respect to
the internal standard.
(Kinsella and Hwang, 1976). On d 180, P. roqueforti its metabolic activity before being inactivated by HPP.
counts reached 7.1 log cfu/g in control cheese and 5.4 The formation of other alcohols is independent from
log cfu/g on average in 400 MPa cheeses, whereas P. the metabolic activity of P. roqueforti. Branched-chain
roqueforti counts were below the detection level in 600 3-methyl-1-butanol is produced through the reduction
MPa cheeses (Calzada et al., 2013). These differences of the respective aldehyde, derived from leucine (Yvon
in P. roqueforti counts may help to explain the higher and Rijnen, 2001). Ethanol is considered to be mostly
ketone content in control cheese than in HPP cheeses produced in cheese through the fermentation of lactose
on d 180. Reduced formation of ketones during the by lactic acid bacteria (Fox and Wallace, 1997). How-
refrigerated storage of pressurized blue-veined cheeses ever, its content doubled in control cheese from d 180
compared with control cheese has been reported by to 360, a stage at which lactose, glucose, and galactose
Voigt et al. (2010). In the present work, the formation were certainly exhausted, whereas ethanol content did
of methyl ketones in 600 MPa cheeses, with P. roqueforti not vary in HPP cheeses. Reduction of acetaldehyde
counts below detection level following HPP treatments, seems a plausible source of ethanol during the refriger-
needs further explanation. Either sublethally injured P. ated storage of control cheese. Reduction of 1-butanal
roqueforti cells unable to grow on selective media car- may be also the origin of 1-butanol, which increased
ried out the chemical reactions responsible for methyl from d 180 to 360 in control cheese but not in HPP
ketone formation or these reactions took place abioti- cheeses. Primary alcohols show green, alcoholic notes,
cally in the absence of viable P. roqueforti cells. From d and secondary alcohols have fruity, herbaceous notes,
180 to 360, we observed marked increases in the ketone whereas 3-methyl-1-butanol has pleasant, fresh cheese
content of 400 MPa cheeses, whereas ketones remained notes (Curioni and Bosset, 2002). They all contribute
stable in control cheese. Methyl ketones are the char- to cheese flavor, directly and as substrates for ester
acteristic flavor compounds of blue-veined cheeses, in formation.
particular, 2-heptanone, which has a blue cheese note Total volatile acids attained lower levels on d 180
and a low perception threshold, but also 2-pentanone, in 400W3 and 600W3 cheeses (Table 5). The main ac-
with a sweet fruity odor note, and 2-propanone, with ids on d 180 as determined by GC-MS were butanoic,
hay and wood pulp odor notes (Molimard and Spinnler, hexanoic, and acetic acids, which accounted for 54, 29,
1996). and 12%, respectively, of total acids in cheeses, on aver-
Total alcohols were at higher levels in control cheese age. Total acids increased in all cheeses from d 180 to
and 400 MPa cheeses than in 600 MPa cheeses on d 360, more so in control cheese and 400 MPa cheeses
180 (Table 5). The main alcohols at d 180 were 2-pen- than in 600 MPa cheeses, in spite of their contribution
tanol, 2-propanol, and ethanol, followed by 1-butanol, to ester formation, with significant increases of all the
2-heptanol, and 3-methyl-1-butanol. The formation of individual acids. Because acids are formed through mi-
2-alkanols through the reduction of the correspond- crobial metabolism and lipolysis, these results indicate
ing 2-alkanones by both spores and mycelium of P. that both phenomena probably remained active even
roqueforti has been suggested as a cellular detoxifying after HPP of cheese at the higher pressure level.
mechanism (Kinsella and Hwang, 1976). The lowest Total esters were negatively affected by HPP, in par-
content of 2-alkanols was found in 600W3 cheese, in ticular at 600 MPa, with 600W3 cheese showing the
which P. roqueforti had the shortest period to exert lowest ester contents on d 180 and 360, and control

cheese the highest contents (Table 5). Ethyl butano- pungent, fetid, buttery notes have been assigned to
ate, methyl hexanoate, and ethyl hexanoate were the γ-butyrolactone (Molimard and Spinnler, 1996).
major esters on d 180, followed by methyl butanoate, Total hydrocarbons tended to reach higher levels in
ethyl octanoate, and methyl octanoate. On d 360, ethyl control cheese than in HPP cheeses, although the dif-
butanoate, ethyl hexanoate, and methyl hexanoate ferences were significant (P < 0.05) only for 600W3,
were predominant, followed by ethyl octanoate, methyl 400W9, and 600W9 cheeses on d 180 and for 600W9
butyrate, and ethyl acetate. Esters may be formed in cheeses on d 360 (Table 6). We observed no significant
cheese by esterification, from alcohols and carboxylic increase in the hydrocarbon content during refriger-
acids, or by alcoholysis, a transferase reaction in which ated storage of cheeses from d 180 to 360. The main
fatty acyl groups from acylglycerols and acyl-CoA de- hydrocarbons found—octane, hexane, and pentameth-
rivatives are directly transferred to alcohols, the second ylheptane—have solvent, gasoline-like odor notes. The
mechanism being the major one in lactic acid bacteria presence of hydrocarbons, generally derived from lipid
and yeasts (Liu et al., 2004). To our knowledge, the oxidation (Carbonell et al., 2002; Collins et al., 2003),
existence of alcohol acyltransferases in Penicillium has been reported in ovine milk cheeses such as La Ser-
spp. has not been reported. It may be hypothesized ena and Manchego (Carbonell et al., 2002; Fernández-
that ester formation in control and 400 MPa cheeses García et al., 2002) but not in Roquefort, Gorgonzola,
took place by both mechanisms, whereas in 600 MPa or Stilton blue-veined cheeses (Gallois and Langlois,
cheeses, with a negligible microbial population, the 1990; Moio et al., 2000; Gkatzionis et al., 2009).
first mechanism would predominate. Esters generally Total aromatic compounds were generally at lower
show sweet, fruity, and floral notes, have a low percep- levels in 400W3 and 600W3 cheeses than in the oth-
tion threshold, and are key odorants in many cheeses. ers (Table 6). Early HPP treatment, on d 21, which
At low concentrations, esters contribute positively severely impaired growth and metabolism of P. roque-
to the overall flavor balance in cheese, by themselves forti (Calzada et al., 2013), was more crucial than the
and by masking the effect of unclean off-flavors with pressure level applied. The main aromatic compounds
pungent, sharp, cowy, and barny notes. At higher con- found were toluene and 1-methoxy-4-methylbenzene,
centrations, a fruity flavor defect may arise (Curioni which did not vary significantly from d 180 to 360;
and Bosset, 2002; Liu et al., 2004). Two cyclic esters, phenol, which declined; and 4-methyl-phenol, which
γ-butyrolactone and γ-caprolactone, were detected at increased. Aromatic compounds may be formed in
low levels and increased moderately from d 180 to 360. cheese through the microbial catabolism of aromatic
Their highest contents were found in control and 400 amino acids (Yvon and Rijnen, 2001). Toluene and
MPa cheeses. Lactones are considered to be generated phenol have been detected in the volatile fraction of
by the hydrolysis of hydroxy-fatty acid triglycerides fol- cheeses made from ovine raw milk (Carbonell et al.,
lowed by lactonization (Jolly and Kosikowski, 1975). 2002; Fernández-García et al., 2004), but not in blue-
They generally have pronounced fruity flavor notes and veined cheeses (Gallois and Langlois, 1990; Lawlor et
a low perception threshold, and may be of importance al., 2003).
to the aroma of some cheese varieties such as Bleu Total nitrogen compounds were significantly influ-
d’Auvergne (Gallois and Langlois, 1990). However, enced by HPP and did not vary with cheese age from
Table 6. Levels of total volatile hydrocarbons, aromatic compounds, terpenes, nitrogen compounds, and sulfur compounds during refrigerated
storage of ovine milk blue-veined control and high-pressure processed (HPP) cheeses pressurized at 400 or 600 MPa after 3, 6, or 9 wk (W3,
W6, W9) of ripening1
Volatile group Day Control 400W3 600W3 400W6 600W6 400W9 600W9
b ab a ab ab a
Hydrocarbons 180 10.29 ± 0.85 7.84 ± 1.00 6.31 ± 0.90 6.61 ± 0.62 7.27 ± 1.12 6.30 ± 0.59 5.76 ± 0.48a
360 9.86 ± 1.27b 8.14 ± 0.73ab 6.27 ± 0.57ab 6.93 ± 0.70ab 6.87 ± 0.79ab 6.00 ± 0.45ab 4.56 ± 0.28a
Aromatic compounds 180 17.37 ± 1.61c 8.86 ± 0.29ab 5.71 ± 0.63a 14.44 ± 0.91bc 13.67 ± 0.88bc 10.42 ± 0.13ab 14.03 ± 0.68bc
360 15.71 ± 1.43c 8.33 ± 0.38ab 5.42 ± 0.09a 11.94 ± 0.45bc 13.05 ± 0.52bc 11.05 ± 0.45b 10.79 ± 0.42b
Nitrogen compounds 180 3.37 ± 0.17ab 2.26 ± 0.28a 3.14 ± 0.38ab 3.91 ± 0.27b 3.67 ± 0.28b 3.78 ± 0.31b 3.46 ± 0.41ab
360 5.40 ± 0.60b 2.89 ± 0.45a 2.41 ± 0.20a 4.06 ± 0.40ab 3.34 ± 0.40ab 4.38 ± 0.69ab 3.29 ± 0.34ab
Terpenes 180 1.50 ± 0.10b 1.12 ± 0.08ab 0.89 ± 0.02a 1.72 ± 0.22b 1.64 ± 0.13b 1.43 ± 0.16ab 1.47 ± 0.13ab
360 1.90 ± 0.27b 1.05 ± 0.07ab 0.56 ± 0.09a 1.36 ± 0.22ab 1.66 ± 0.10b 1.53 ± 0.06ab 1.43 ± 0.12ab
Sulfur compounds 180 0.95 ± 0.07a 0.80 ± 0.07a 0.72 ± 0.06a 1.00 ± 0.10a 1.01 ± 0.05a 1.04 ± 0.05a 0.90 ± 0.04a
360 1.18 ± 0.12a 0.94 ± 0.10a 0.92 ± 0.07a 1.22 ± 0.13a 1.04 ± 0.12a 1.18 ± 0.10a 1.05 ± 0.10a
a–c
1
Mean ± SE (n = 4) of duplicate determinations in 2 cheese-making experiments. Results are expressed as relative abundances with respect to
the internal standard.

7508 CALZADA ET AL.
d 180 to 360 (Tables 1 and 6). Methanamide, butyr- groups of flavor compounds, which might be unsatisfac-
amide, and 2,6-dimethylpyrazine, the nitrogen com- torily balanced from a sensory point of view.
pounds detected in the present work, were at low levels
in all cheeses (data not shown). The concentrations Principal Component Analysis
of butyramide and 2,6-dimethylpyrazine more than
doubled in control cheese from d 180 to 360, whereas Principal component analysis was carried out on the
they remained largely constant in HPP cheeses during total concentrations of the 4 groups of carboxylic acids
this period. The increases of these compounds in con- (acetic + propionic, SC FFA, MC FFA, and LC FFA)
trol cheese might be associated with the metabolism of determined by GC and the total levels of the 10 groups
cheese microbiota, but their origin could not be traced of volatile compounds (aldehydes, ketones, alcohols,
to a particular microbial group. acids, esters, hydrocarbons, aromatic compounds, ni-
Total terpenes attained lower levels in 400W3 and trogen compounds, terpenes, and sulfur compounds)
600W3 cheeses than in the other cheeses (Table 6). determined by GC-MS. Two functions explained each
Terpenes are secondary plant metabolites, whose pres- over 10% of the variance. Function 1, formed by the
ence in an ovine milk cheese not ripened by molds has 4 groups of carboxylic acids and 8 groups of volatile
been traced to the feed (Fernández-García et al., 2008). compounds (excluding aldehydes and hydrocarbons),
Additionally, P. roqueforti strains cultivated on syn- explained 50.6% of the variance, whereas function
thetic media have been reported to produce terpenes 2, formed by aldehydes and hydrocarbons, explained
(Chalier and Crouzet, 1993). Pressurization of cheeses 15.7% of the variance. Function 1 was associated with
on d 21, which impeded normal growth of P. roqueforti cheese age, and function 2 with absence of (control) or
(Calzada et al., 2013), likely hindered terpene synthesis milder (400 MPa) HPP treatment. When cheeses were
during ripening and refrigerated storage of 400W3 and plotted against functions 1 and 2 (Figure 1), function 1
600W3 cheeses. Terpenes found in the present work correctly separated all the 360-d cheeses, located in the
were endoborneol and 2 unidentified sesquiterpenes, positive semi-plane, from all the 180-d cheeses, located
all common volatile compounds in ovine milk cheeses in the negative semi-plane, with the only exception be-
(Carbonell et al., 2002) but not in blue-veined cheeses ing the 360-d 600W3 cheese, which was located in the
(Lawlor et al., 2003). In contrast to our results, the negative semi-plane. The decelerating effect of early
latter authors reported the presence of α-pinene and HPP (d 21) at the higher pressure level (600 MPa) on
limonene in the 6 blue-veined varieties they investi- the biochemical changes occurring during cheese ripen-
gated. Terpenes confer herbaceous, floral, and fruity ing and refrigerated storage was thus highlighted. The
odor notes, and may be useful as geographical markers contents of FFA and volatile compounds in the 360-d
of cheese origin (Bugaud et al., 2001). 600W3 cheese were closer to those of 180-d cheeses than
Dimethylsulfone, the only sulfur compound detected to the contents of 360-d cheeses.
in blue-veined ovine milk cheese, varied with HPP
and cheese age (Tables 1 and 6). Dimethylsulfone has CONCLUSIONS
been detected in Cheddar cheese at levels that doubled
from medium-age to extra sharp cheese (Burbank and High-pressure processing influenced the total concen-
Qian, 2005). It can be formed through the oxidation of trations of all groups of carboxylic acids and volatile
dimethylsulfide (Al-Attabi et al., 2009) or may derive compounds found in ovine milk blue-veined cheese. The
directly from the pasture, because it is present in cru- greatest reduction in the concentrations of carboxylic
ciferous and bulbaceous plants. More common sulfur acids during ripening and refrigerated storage was re-
compounds such as dimethylsulfide or dimethyldisul- corded for the cheese pressurized at 600 MPa on d 21,
fide (formed through the degradation of methionine), which showed the lowest esterase activity values. Con-
which have been reported in other blue-veined cheese trol cheese was richest in alcohols, esters, and aromatic
varieties (Gallois and Langlois, 1990; Lawlor et al., compounds, whereas cheese pressurized at 600 MPa on
2003), were not found in HPP or control cheeses in the d 21 showed the lowest contents of aldehydes, ketones,
current study. alcohols, acids, esters, aromatic compounds, and ter-
The sensory characteristics of control cheeses and penes. The reduced levels of flavor compounds in the
HPP cheeses except the 600W3 cheese showed negli- 600W3 cheese, which were associated with low counts
gible variations during the refrigerated storage period, of P. roqueforti and other microbial groups, resulted in
according to the results obtained in a previous work lower flavor quality scores during refrigerated storage.
(Calzada et al., 2013). In contrast, the flavor quality of The high concentrations of carboxylic acids and volatile
the 600W3 cheese declined during refrigerated storage, compounds in control cheese and the other HPP cheeses
which can be attributed to the lower contents of some did not impair their sensory characteristics, probably
ity, fatty acid and volatile compound composition of milk. Lait
81:401–414.
Burbank, H. M., and M. C. Qian. 2005. Volatile sulfur compounds in
Cheddar cheese determined by headspace solid-phase microextrac-
tion and gas chromatograph-pulsed flame photometric detection.
J. Chromatogr. A 1066:149–157.
Calzada, J., A. del Olmo, A. Picon, P. Gaya, and M. Nuñez. 2013. Pro-
teolysis and biogenic amine buildup in high-pressure treated ovine
milk blue-veined cheese. J. Dairy Sci. 96:4816–4829.
Carbonell, M., M. Nuñez, and E. Fernández-García. 2002. Evolution of
the volatile components of ewe raw milk La Serena cheese during
ripening. Correlation with flavor characteristics. Lait 82:683–698.
Chalier, P., and J. Crouzet. 1993. Production of volatile components
by Penicillium roqueforti cultivated in the presence of soya bean
oil. Flavour Frag. J. 8:43–49.
Collins, Y. F., P. L. H. McSweeney, and M. G. Wilkinson. 2003. Lipol-
ysis and free fatty acid catabolism in cheese: A review of current
knowledge. Int. Dairy J. 13:841–866.
Curioni, P. M. G., and J. O. Bosset. 2002. Key odorants in various
cheese types as determined by gas chromatography-olfactometry.
Int. Dairy J. 12:959–984.
Figure 1. Distribution of control and high-pressure processed De Frutos, M., J. Sanz, and I. Martinez-Castro. 1991. Characterization
(HPP) blue-veined cheeses pressurized at 400 or 600 MPa on d 21, of artisanal cheeses by GC and GC/MS analysis of their medium
42, or 63 after manufacture (W3, W6, W9) on the plane defined by volatility (SDE) fraction. J. Agric. Food Chem. 39:524–530.
functions 1 and 2 of the principal component analysis. Each symbol Fernández-García, E., M. Carbonell, J. Calzada, and M. Nuñez. 2006.
represents the averaged value of the 2 batches of cheese. Treatments Seasonal variation of the free fatty acids contents of Spanish ovine
are as follows: C = control; = 400W3; = 600W3; Δ = 400W6; milk cheeses protected by a designation of origin: A comparative
= 600W6; □ = 400W9; = 600W9. Cheese age is as follows: 180 d = study. Int. Dairy J. 16:252–261.
6 mo; 360 d = 12 mo. Fernández-García, E., M. Carbonell, P. Gaya, and M. Nuñez. 2004.
Evolution of the volatile components of ewes raw milk Zamorano
cheese. Seasonal variation. Int. Dairy J. 14:701–711.
because an adequate balance of flavor compounds was Fernández-García, E., M. Carbonell, and M. Nuñez. 2002. Volatile
fraction and sensory characteristics of Manchego cheese. 1. Com-
maintained during refrigerated storage. parison of raw and pasteurized milk cheese. J. Dairy Res. 69:579–
593.
Fernández-García, E., M. Imhof, H. Schlichtherle-Cerny, J. O. Bosset,
ACKNOWLEDGMENTS and M. Nuñez. 2008. Terpenoids and benzenoids in La Serena
cheese made at different seasons of the year with a Cynara cardun-
This research was funded by AGL 2009-07801 project culus extract as coagulant. Int. Dairy J. 18:147–157.
from the Ministry of Science and Innovation (MICINN, Fox, P. F., and J. M. Wallace. 1997. Formation of flavor compounds in
cheese. Adv. Appl. Microbiol. 45:17–85.
Spain). J. Calzada is the recipient of a FPI grant (MI- Gallois, A., and D. Langlois. 1990. New results in the volatile odorous
CINN, Spain). The authors thank the valuable help compounds of French cheeses. Lait 70:89–106.
of V. Carbonero (Lácteas Toledo, Guadamur, Spain) García-Risco, M. R., I. Recio, E. Molina, and R. López-Fandiño. 2003.
Plasmin activity in pressurized milk. J. Dairy Sci. 86:728–734.
with cheese supply and of F. Purroy (NC Hyperbaric, Gkatzionis, K., R. S. T. Linforth, and C. E. R. Dodd. 2009. Vola-
Burgos, Spain) with high pressure treatments. tile profile of Stilton cheeses: Differences between zones within a
cheese and dairies. Food Chem. 113:506–512.
González de Llano, D., M. Ramos, C. Polo, J. Sanz, and I. Martinez-
REFERENCES Castro. 1990. Evolution of the volatile components of an artisanal
blue cheese during ripening. J. Dairy Sci. 73:1676–1683.
Al-Attabi, Z., B. R. D’Arcy, and H. C. Deeth. 2009. Volatile sulphur Gripon, J.-C., M. J. Desmazeaud, D. Le Bars, and J. L. Bergère. 1977.
compounds in UHT milk. Crit. Rev. Food Sci. Nutr. 49:28–47. Role of proteolytic enzymes of Streptococcus lactis, Penicillium
Arqués, J. L., S. Garde, E. Fernández-García, P. Gaya, and M. Nuñez. roqueforti, and Penicillium caseicolum during cheese ripening. J.
2007. Volatile compounds, odor, and aroma of La Serena cheese Dairy Sci. 60:1532–1538.
high-pressure treated at two different stages of ripening. J. Dairy Huppertz, T., P. F. Fox, and A. L. Kelly. 2004. Susceptibility of plas-
Sci. 90:3627–3639. min and chymosin in Cheddar cheese to inactivation by high pres-
Arqués, J. L., S. Garde, P. Gaya, M. Medina, and M. Nuñez. 2006. In- sure. J. Dairy Res. 71:496–499.
activation of microbial contaminants in raw milk La Serena cheese Jolly, R. C., and F. V. Kosikowski. 1975. Quantification of lactones in
by high-pressure treatments. J. Dairy Sci. 89:888–891. ripening pasteurized milk blue cheese containing added microbial
Ávila, M., J. Calzada, S. Garde, and M. Nuñez. 2007. Effect of a lipases. J. Agric. Food Chem. 23:1175–1176.
bacteriocin-producing Lactococcus lactis strain and high-pressure Juan, B., V. Ferragut, M. Buffa, B. Guamis, and A. Trujillo. 2007.
treatment on the esterase activity and free fatty acids in Hispánico Effects of high pressure on proteolytic enzymes in cheese: Rela-
cheese. Int. Dairy J. 17:1415–1423. tionship with the proteolysis of ewe milk cheese. J. Dairy Sci.
Ávila, M., S. Garde, E. Fernández-García, M. Medina, and M. Nuñez. 90:2113–2125.
2006. Effect of high-pressure treatment and a bacteriocin-produc- Kinsella, J. E., and D. Hwang. 1976. Biosynthesis of flavors by Penicil-
ing lactic culture on the odor and aroma of Hispánico cheese: Cor- lium roqueforti. Biotechnol. Bioeng. 18:927–938.
relation of volatile compounds and sensory analysis. J. Agric. Lamberet, G., and A. Menassa. 1983. Purification and properties of an
Food Chem. 54:382–389. acid lipase from Penicillium roqueforti. J. Dairy Res. 50:459–468.
Bugaud, C., S. Buchin, J. B. Coulon, A. Hauwuy, and D. Dupont. Lawlor, J. B., C. M. Delahunty, J. Sheehan, and M. G. Wilkinson.
2001. Influence of the nature of alpine pastures on plasmin activ- 2003. Relationships between sensory attributes and the volatile

7510 CALZADA ET AL.
compounds, non-volatile and gross compositional constituents of belonging to six different species. J. Agric. Food Chem. 53:6835–
six blue-type cheese. Int. Dairy J. 13:481–494. 6843.
Liu, S.-Q., R. Holland, and V. L. Crow. 2004. Esters and their bio- O’Reilly, C. E., P. M. O’Connor, A. L. Kelly, T. P. Beresford, and
synthesis in fermented dairy products: A review. Int. Dairy J. P. M. Murphy. 2000. Use of hydrostatic pressure for inactivation
14:923–945. of microbial contaminants in cheese. Appl. Environ. Microbiol.
Madkor, S., P. F. Fox, S. I. Shalabi, and N. H. Metwalli. 1987. Studies 66:4890–4896.
on the ripening of Stilton cheese: Lipolysis. Food Chem. 25:93– Tejada, L., R. Gómez, M. Vioque, E. Sánchez, C. Mata, and J. Fernán-
109. dez-Salguero. 2000. Effect of freezing and frozen storage on the
Mallia, S., E. Fernández-García, and J. O. Bosset. 2005. Comparison sensorial characteristics of Los Pedroches, a Spanish ewe milk
of purge and trap and solid phase microextraction techniques for cheese. J. Sens. Stud. 15:251–262.
studying the volatile aroma compounds of three European PDO Van Hekken, D. L., M. H. Tunick, and Y. W. Park. 2005. Effect of
hard cheeses. Int. Dairy J. 15:741–758. frozen storage on the proteolytic and rheological properties of soft
Malone, A. S., C. Wick, T. H. Shellhammer, and P. D. Courtney. caprine milk cheese. J. Dairy Sci. 88:1966–1972.
2003. High pressure effects on proteolytic and glycolytic enzymes Voigt, D. D., F. Chevalier, M. C. Qian, and A. L. Kelly. 2010. Effect of
involved in cheese manufacturing. J. Dairy Sci. 86:1139–1146. high-pressure treatment on microbiology, proteolysis, lipolysis and
McSweeney, P. L. H., and M. J. Sousa. 2000. Biochemical pathways for levels of flavor compounds in mature blue-veined cheese. Innov.
the production of flavour compounds in cheese during ripening: A Food Sci. Emerg. Technol. 11:68–77.
review. Lait 80:293–324. Wick, C., U. Nienaber, O. Anggraeni, T. H. Shellhammer, and P. D.
Moio, L., L. Piombino, and F. Addeo. 2000. Odour-impact compounds Courtney. 2004. Texture, proteolysis and viable lactic acid bacte-
of Gorgonzola cheese. J. Dairy Res. 67:273–285. ria in commercial Cheddar cheeses treated with high pressure. J.
Molimard, P., and H. E. Spinnler. 1996. Compounds involved in the Dairy Res. 71:107–115.
flavour of surface mold-ripened cheeses: Origins and properties. J. Yvon, M., and L. Rijnen. 2001. Cheese flavour formation by amino
Dairy Sci. 79:169–184. acid catabolism. Int. Dairy J. 11:185–201.
Morales, P., E. Fernández-García, and M. Nuñez. 2005. Volatile com-
pounds produced in cheese by Pseudomonas strains of dairy origin

Capítulo 4.
Using high-pressure processing for reduction of proteolysis
and prevention of over-ripening of raw milk cheese.

117

Fotografía: cortes de queso control (no presurizado) de Torta del Casar a día 60
(superior),120 (centro) y 240 (inferior).

Food Bioprocess Technol (2014) 7:1404–1413
DOI 10.1007/s11947-013-1141-5
ORIGINAL PAPER
Using High-Pressure Processing for Reduction of Proteolysis

and Prevention of Over-ripening of Raw Milk Cheese
Javier Calzada & Ana del Olmo & Antonia Picon &
Pilar Gaya & Manuel Nuñez
Received: 20 February 2013 / Accepted: 29 May 2013 / Published online: 13 June 2013
# Springer Science+Business Media New York 2013
Abstract High-pressure-processing (HPP) at 400 or 600 MPa Introduction

was applied to cheeses made from ewe raw milk, on days 21 or
35 after manufacturing, to reduce proteolysis and prevent over- The breakdown of proteins and lipids, and the metabolism
ripening. The characteristics of HPP and non-pressurized of lactose and citrate, are primary biochemical events
(control) cheeses were compared during ripening at 8 °C until which take place during cheese manufacturing and early
day 60 and further storage at 4 °C until day 240. HPP and ripening. Afterwards, secondary biochemical events such
control cheeses showed similar pH values throughout ripening, as the further hydrolysis of peptides and the catabolism of
but on day 240 pH values remained 0.4–0.6 units lower for amino acids, fatty acids and lactate, give rise to the
HPP cheeses than for the control cheeses. Casein degrada- formation of the compounds responsible for the character-
tion was significantly retarded in the 600 MPa cheeses. istic flavour and aroma of each particular cheese variety
Their α-casein concentration was 48–52 % higher on day (McSweeney and Sousa 2000; Collins et al. 2003; Yvon
60 and 30–33 % higher on day 240 than in the control and Rijnen 2001). Simultaneously, the typical texture and
cheeses while β-casein concentration was 25–26 % higher the microstructure of the product develop (O’Reilly et al.
on day 60 and 100–103 % higher on day 240. No signif- 2003; Picon et al. 2013b). Secondary biochemical events
icant differences in para-κ-casein concentration between occurring during mid- and late ripening pursue during
cheeses were found on day 60, but on day 240, it was refrigerated storage of cheese at the dairy and during its
22–35 % higher in the 600 MPa cheeses than in the control shelf life at retailers and homes. Consequently, over-
cheese. Hydrophilic peptides, hydrophobic peptides and to- ripening defects may surge before consumption, particularly
tal free amino acids evolved similarly in HPP and control in strongly proteolyzed cheese varieties.
cheeses during the 60-day ripening period. However, on day When exceeding a certain threshold, ammonia, amines,
240 hydrophilic peptides were at 34–39 % lower levels in alcohols, aldehydes, carboxylic acids and thiol compounds
the 600 MPa cheeses than in the control cheeses, hydro- formed in cheese through amino acid catabolism (Yvon and
phobic peptides at 7–16 % lower levels and total free amino Rijnen 2001) are among the main causative agents for the
acids at 25–29 % lower levels. Flavour intensity scores flavor and aroma defects associated to over-ripening. Freezing
increased at a slower rate in HPP cheeses than in the of fully ripened cheeses has been assayed to prevent over-
control cheese. Flavour quality declined markedly in the ripening and prolong their shelf life, by stopping or retarding
control cheeses during refrigerated storage while it did not enzyme activity and chemical reactions. Even though flavour
vary significantly in 600 MPa cheeses. characteristics of cheeses remained unchanged during frozen
storage, texture defects at thawing were common (Tejada et al.
Keywords High-pressure processing . Proteolysis . 2000; Van Hekken et al. 2005).
Over-ripening . Cheese . Flavour High-pressure processing (HPP) is a technology that can
achieve the food safety level of heat pasteurization whilst
meeting consumer demand for fresher-tasting minimally
processed foods (Norton and Sun 2008). The inactivation of
J. Calzada : A. del Olmo : A. Picon : P. Gaya : M. Nuñez (*)
pathogenic and spoilage microorganisms has been the main
Departamento de Tecnología de Alimentos, INIA,
Carretera de La Coruña Km 7, Madrid 28040, Spain objective when applying HPP to milk and cheese (O’Reilly
e-mail: nunez@inia.es et al. 2000; Trujillo et al. 2000; Shao and Ramaswamy 2011).
Food Bioprocess Technol (2014) 7:1404–1413 1405
In addition, HPP could be useful to stop or retard some of the of tap water and filtering through a cheese cloth. Curds were
chemical reactions taking place during cheese ripening, since cut into 10-mm cubes, held at 30 °C for 15 min and distrib-
cheese-related enzymes, including the proteinases and pepti- uted into cylindrical moulds. Cheeses, 13 cm in diameter and
dases responsible for the formation of peptides and free amino 6-cm high, were pressed for 3 h in a horizontal press, salted
acids (FAA), are affected by HPP (García-Risco et al. 2003; by rubbing dry salt twice on all the surfaces, and ripened at
Malone et al. 2003; Juan et al. 2007). Pressure level is more 8 °C and 92 % relative humidity.
crucial to enzyme inactivation than temperature or time of
exposure, such as shown for chymosin which loses less than High-Pressure Processing
5 % of its activity in Cheddar cheese at 400 MPa and 93–96 %
at 600 MPa (Huppertz et al. 2004b). HPP has been reported to Cheeses coded as 400W3, 600W3, 400W5 and 600W5 were
hinder the formation of FAA by some authors. Thus, lower vacuum-packaged in CN300 bags (Cryovac Grace S.A.,
FAA concentrations were found in 1-month-old Cheddar Barcelona, Spain) and pressurized at 400 or 600 MPa for
cheese pressurized at 400 MPa on day 1 than in the respective 5 min, after 3 or 5 weeks of ripening, respectively. HPP was
control cheese (O'Reilly et al. 2002) and in 7-month-old performed in a 120-L capacity isostatic press (Hiperbaric,
Cheddar cheeses pressurized on day 30 at 400, 500 or Burgos, Spain). Times to reach 400 and 600 MPa were 1.85
800 MPa than in the respective control cheese (Wick et al. and 2.83 min, respectively and depressurization times, 7 and
2004). Within the current applications of emerging technolo- 8 s. The temperature of the water used as transmitting fluid
gies as alternatives to milk pasteurization in cheese manufac- remained under 14 °C during the whole process. Pressurized
ture, pulsed electric fields have also been assayed, with satis- cheeses were unpackaged after HPP. Control cheeses were
factory results (Yu et al. 2012). vacuum-packaged after 3 weeks of ripening and unpackaged
In the present work, we applied HPP to Casar cheese, a simultaneously with cheeses pressurized at that time.
variety made from ewe raw milk coagulated with an aqueous Cheeses were ripened at 8 °C and 92 % RH until day 60,
extract of Cynara cardunculus L. (cardoon) flowers. This and afterwards held at 4 °C until day 240.
extract contains the aspartic proteinases named cyprosins or
cynarases, with optimum activity around pH 5.1 (Heimgartner Chemical Determinations
et al.1990). The strong proteolytic activity of cyprosins, the
microbial load of raw milk and the high pH value of the cheese Caseins and whey proteins were determined on duplicate sam-
thus produced make it prone to over-ripening. Because of the ples by capillary gel electrophoresis according to a previously
seasonality in ewe milk production, ripe cheeses made during described method (Garde et al. 2002) with some modifications,
the first half of the year may be held at refrigeration temper- on an automated P/ACETM MDQ capillary electrophoresis
atures for several months before marketed. This is also valid apparatus controlled by the 32 Karat Software (Beckman
for similar varieties such as La Serena and Los Pedroches Instruments España S.A., Madrid, Spain). Briefly, 5 g of cheese
cheeses in Spain, and Serra da Estrela and Azeitao cheeses in were mixed with 25 mL of 2 % trisodium citrate solution,
Portugal. With the objective of preventing the undesirable homogenized for 1 min in an Ultra-Turrax T-10 blender
consequences of a prolonged storage period, we applied (IKA, Staufen, Germany) at high speed on ice. Twenty
HPP to cheeses 3 or 5 weeks after manufacture. The break- microlitres of cheese homogenate was mixed with 170 μL of
down of caseins, the formation of peptides and FAA, the 100 mM Tris–HCl buffer (pH 9.0) containing 1 % SDS, 10 μL
texture and the sensory characteristics of HPP cheeses of 2-mercaptoethanol and 4 μL of a 10-kDa internal standard
during a 60-day ripening period and a further 180-day (Beckman), and heated at 95 °C for 10 min before injection at
refrigerated storage period were investigated and compared 5 kV. Electrophoretic separation was performed at 15 kV for
with the characteristics of control cheese. 30 min after a 4-min ramp, in a bare-fused silica capillary
column (Beckman) of 50 μm internal diameter and 30 cm total
length, in SDS-buffer gel (Beckman). To calculate the MW of
Materials and Methods peaks monitored at 214 nm, the coefficient of relative time
mobility to the internal standard was compared with those of a
Cheese Manufacture mixture of 10, 20, 35, 50, 100, 150 and 225 kDa protein
standards (Beckman, SDS-MW protein size standard).
Two batches of Casar cheese were made on consecutive Commercial standards (Sigma, Alcobendas, Spain) of bovine
days, each from 600 L of refrigerated ewe raw milk without α-casein, β-casein, κ-casein, α-lactalbumin, β-lactalbumin,
added starter cultures, at a Protected Designation of Origin serum albumin and lactoferrin were used for the identification
dairy. Milk was coagulated at 30 °C for 60 min in a semi- of proteins. Proteins were quantified with respect to the internal
automated open vat with an aqueous extract obtained by standard area and expressed as milligramme of protein per
macerating 600 g of dry cardoon flowers overnight in 6 L gramme of cheese dry matter (DM).
1406 Food Bioprocess Technol (2014) 7:1404–1413
Hydrophilic and hydrophobic peptides in the water- Wycombe, Bucks, UK), with crosshead and chart speeds of
soluble fraction of cheese were determined on duplicate 50 and 500 mm/min, respectively. Fracturability (force at
samples by RP-HPLC using a Beckman System Gold chro- breaking point, expressed in Newtons), elasticity (apparent
matograph (Beckman Instruments España) equipped with a elastic modulus, expressed in Newtons per square millimetre)
diode array detector module 168, with detection wavelength and firmness (work done on the cheese, expressed in Joules)
at 280 nm, as previously described (Lau et al. 1991). Peaks were calculated from compression curves (Picon et al. 2013a).
with retention times from 5.5 to 14.6 min were considered
to correspond to hydrophilic peptides and those with Sensory Evaluation
retention times from 14.6 to 20.5 min to hydrophobic
peptides. Peptide levels were expressed in arbitrary units, A trained 15-member panel carried out the evaluation of
calculated as units of chromatogram area per milligramme flavour intensity and quality (preference), scoring on a 10-
of cheese DM. point scale as previously described (Nuñez et al. 1991). Five
Free amino acids were extracted from duplicate samples additional flavour attributes (acid, bitter, salty, sweet and
(Krause et al. 1995) and individual FAA determined by RP- umami) were also determined by panelists on a 10-point
HPLC using a Beckman System Gold chromatograph, after scale (Picon et al. 2013a).
derivatization with 6-aminoquinolyl-N-hydroxysuccinimidyl
carbamate. They were expressed as milligrammes per gramme Statistical Analysis
of cheese DM.
Cheese pH was measured in triplicate directly with a Crison Data obtained were analyzed by a two-way analysis of
penetration electrode (model 52–3,2; Crison Instruments, variance, with HPP treatment and days of ripening as main
Barcelona, Spain) by means of a Crison GPL 22 pH-meter. effects. Means were compared using Tukey’s test, with
Dry matter was determined on triplicate samples after drying to p=0.05. The SPSS Win 14.0 software (SPSS, Inc., Chicago,
constant weight in an oven at 102 °C. IL) was employed for the statistical analysis of data.
Cyprosin activity and overall peptidolytic activity in
cheese were determined on duplicate samples. A homoge-
nate of 5 g of cheese, 25 mL of 2 % trisodium citrate solution Results and Discussion
and 2 g of skim milk was acidified with 1 N HCl to pH 5.2,
favourable for cyprosin activity, and incubated for 3 h at 37 ° Protein Breakdown
C. Then, the pH was adjusted with 1 N NaOH to 6.5,
favourable for peptidase activity, and the incubation was Casein concentrations declined significantly (p<0.001) with
prolonged for 3 h. The OPA test (Church et al. 1983), based ripening time in control cheese, according to the analysis of
on the reaction of released α-amino groups with this com- variance, from 38.36, 272.55 and 25.76 mg of α-, β- and
pound and with β-mercaptoethanol to form an adduct that para-κ-casein per gramme of cheese DM on day 1 (data not
absorbs strongly at 340 nm, was run on 0-h, 3-h and 6-h shown) to 5.80, 81.93 and 16.09 mg/g DM, respectively, on
samples. Cyprosin activity was estimated as the increase in day 60. These marked decreases in casein levels agree with
absorbance (OPA method) from 0 to 3 h during the incuba- previous works on cheese varieties made from ewe milk
tion at pH 5.2 and 37 °C. Overall, peptidolytic activity was coagulated with cardoon extract (Garde et al. 2007; Delgado
estimated as the increase in absorbance (OPA method) from et al. 2010). Regarding whey proteins, α-lactalbumin, at a
3 to 6 h during the incubation at pH 6.5, which reduces concentration of 1.54 mg/g DM on day 1, could not be
drastically cyprosin activity, and 37 °C. To discard the detected in control cheese from day 21 onwards (data not
possible interference of milk plasmin with cyprosins or shown) while β-lactoglobulin concentration decreased signif-
enzymes of bacterial origin present in cheese samples, icantly (p<0.01) in control cheese with ripening time, from
the same assay was run on milk samples (taken from the 11.63 mg/g DM on day 1 to 5.13 mg/g DM on day 60.
vat, without added cardoon extract) at pH 6.7, the initial HPP at 400 or 600 MPa did not significantly influence
pH value of milk, and at pH values of 6.2, 5.7 and 5.2, casein levels immediately after treatment. However, higher
after acidification with 1 N HCl. α-casein levels were found for the 600 MPa cheeses during
ripening and refrigerated storage (Fig. 1), with a concentra-
Textural Determinations tion of α-casein 48–52 % higher in the 600 MPa cheeses than
in control cheeses on day 60 and 30–33 % higher on day 240.
Six cylinder-shaped (17-mm height, 17-mm diameter) sam- The concentration of β-casein in pressurized and control
ples from each cheese were compressed to 75 % of their cheeses did not differ significantly during ripening, but it
original height after 10 min at room temperature (20–22 °C) was 100–103 % higher in the 600 MPa cheeses than in the
using an Instron Compression Tester 4301 (Instron, High control cheese on day 240. No significant differences in
12 20 a
b a a a a a
a
b 18 a a a
a a a c
10 a a b b b
a 16 bc
a b a
mg para- -casein/g DM
a ab abc
b 14
8
mg -casein/g DM
ab b a ab
ab
12 a a
a a ab
a b
6 b 10
a ab
ab ab 8
ab
4 a a
6
4
2
2
0 0
35 d 60 d 120 d 180 d 240 d 35 d 60 d 120 d 180 d 240 d
140 8
a
a c
a 7 bc
120 a a a a a a a a
a b a
a b
a
mg -lactoglobulin/g DM
c 6 a a ab
100 a a ab
a a ab a a
b b
mg -casein/g DM
a b a ab
5
a a a
80 ab a
a a 4
60 a
a a 3
a a a
40
2
20 1
0 0
35 d 60 d 120 d 180 d 240 d 35 d 60 d 120 d 180 d 240 d
Fig. 1 Concentrations of caseins and β-lactoglobulin during ripening (horizontally striped bar), 600W5 (dotted bar). Means (bars with
and refrigerated storage of control and HPP cheeses. Control (black SEM) at the same sampling date with the same letter do not differ
bar), 400W3 (obliquely striped bar), 600W3 (white bar), 400W5 significantly (p>0.05)
para-κ-casein concentration between cheeses were found and peptidases from lactic acid bacteria. Apparently, there was
during ripening but on day 240 it was 22–35 % higher in a pressure-induced enhancement of cyprosin activity in
the 600 MPa cheeses than in the control cheese. Similarly, 400 MPa cheeses, but only the cyprosin activity of 400W5
there were no significant differences in β-lactoglobulin con- cheese on day 240 showed a significant (p<0.05) difference
centration between cheeses on day 60, whereas on day 240 it when compared with the respective control cheese (Table 1).
was 27–37 % higher in the 600 MPa cheeses than in the Increases of activity after HPP have been reported for
control cheese. enzymes such as the cell envelope proteinase and the
Protein degradation in the control cheese and in the pres- PepC aminopeptidase of Lactococcus lactis MG1363
surized cheeses until submitted to HPP can be associated to (Malone et al. 2003). Cyprosin activity values in 600 MPa
both plasmin and cyprosins. Cheese pressurization at 400 MPa cheeses did not differ significantly from those of the respec-
has a negligible effect on plasmin activity, while treatment at tive control cheese. A high stability of cyprosins under cheese
600 MPa reduces its activity by less than 10 % (Huppertz et al. ripening conditions has been reported (Picon et al. 1999). In
2004b). Therefore, considerable plasmin activity would the present work, considerable cyprosin activity persisted,
remain in pressurized cheeses. The role of cyprosins in even in 600 MPa cheeses, until day 240. The increases in
the proteolysis of HPP cheeses is difficult to establish absorbance recorded for cheese samples during the 3 h of
since there is no information available on cyprosine incubation at pH 5.2 cannot be attributed to the activity of
barostability. In La Serena cheese, also made from ewe milk plasmin. When the assay was run on milk without added
milk coagulated with cardoon extract, treatments at 300 and cardoon extract, the increases of absorbance in the OPA test
400 MPa on day 2 resulted in higher levels of αs- and β- were as low as 0.022, 0.008, 0.004 and 0.003 for milk samples
caseins on days 30 and 60 of ripening than in the control at pH values of 6.7, 6.2, 5.7 and 5.2, respectively. The overall
cheese (Garde et al. 2007). peptidolytic activity was estimated by further incubation of
To ascertain the role of cyprosins in protein breakdown, cheese samples at pH 6.5 for 3 h, which resulted in additional
cyprosin activity was estimated in HPP and control cheeses by increases in absorbance. The peptidolytic activity in 400 MPa
determining the increase in absorbance by the OPA method cheeses and in control cheese did not differ (Table 1).
after incubation for 3 h at 37 °C and pH 5.2, a pH value However, the peptidolytic activity was 74 % lower in
favourable for cyprosin activity (Heimgartner et al. 1990) but 600W3 cheese and 63 % lower in 600W5 cheese than in
unfavourable for the activity of plasmin and most proteinases control cheese on day 60, while it was 90 % and 89 % lower,
Table 1 Cyprosin activity and peptidolytic activity in control and HPP proteins in the control cheese than in the pressurized cheeses.
cheeses at the end of ripening (day 60) and after refrigerated storage
Regarding cheese DM, no significant differences were found
(day 240)
between HPP and control cheeses during ripening, until day
Days Cheese Cyprosin activitya Peptidolytic activitya 60. At the end of the refrigerated storage period, on day 240,
higher (p < 0.05) DM content was recorded for 600W3,
60 Control 0.148±0.017ab 0.081±0.008b
400W5 and 600W5 cheeses than for the 400W3 cheese and
400W3 0.172±0.023ab 0.074±0.005b
control cheese (Fig. 2).
600W3 0.123±0.016a 0.021±0.016a
400W5 0.207±0.025b 0.107±0.010b
Peptide and FAA Formation
600W5 0.193±0.013ab 0.030±0.007a
240 Control 0.336±0.071a 0.119±0.038a The levels of hydrophilic and hydrophobic peptides and their
400W3 0.439±0.052a 0.125±0.052a ratio were significantly (p < 0.001) influenced by time,
600W3 0.346±0.041a 0.012±0.015a according to the analysis of variance. There were no signifi-
400W5 0.596±0.045b 0.056±0.037a cant changes in peptide levels attributable to HPP immediately
600W5 0.424±0.031a 0.013±0.022a after treatment on days 21 (data not shown) and 35. On day
Results are expressed as mean±SEM of duplicate determinations on
60, at the end of ripening, the 400W3 cheese had a signifi-
two cheese making trials. Means in the same column at the same cantly (p<0.05) higher level of hydrophilic peptides than the
sampling date with the same letters do not differ significantly control cheese (Fig. 3) while the rest of the HPP cheeses did
a
Cyprosin and peptidolytic activities are expressed as increases in not differ from the control cheese. This result may be ascribed
absorbance (OPA method) either to an enhancement of cyprosin activity by HPP at
400 MPa, a pressure level which apparently increased its
respectively, on day 240 (Table 1). These results point to the activity, or to changes in the conformation of proteins caused
inactivation of peptidolytic enzymes of bacterial origin at by HPP (García-Risco et al. 2000; Huppertz et al. 2004a)
600 MPa, in agreement with previous works (Malone et al. which might have favoured the access of cyprosins to
2003; Avila et al. 2006), and may be of practical interest to their substrates. The level of hydrophobic peptides on
control over-ripening at the cheese industry. day 60 in control and HPP cheeses did not differ, with
Cheese pH and DM increased significantly (p<0.001) the only exception of 400W5 cheese which showed a
with time, according to the analysis of variance. They mod- lower level. In 60-day-old La Serena cheese pressurized
ulate enzyme activity and may influence protein breakdown. at 400 MPa on day 2, hydrophilic peptides also attained a
Control cheese and HPP cheeses had similar pH values until higher level than in the control cheese, while hydrophobic
day 60 (Fig. 2) but afterwards the pH rose more rapidly in the peptides were at a lower level (Garde et al. 2007). The
control cheese, which showed significantly (p<0.05) higher hydrophobic peptides/hydrophilic peptides ratio reached on
values than HPP cheeses from day 120 onwards. The higher day 60 values of 1.55–1.68 in HPP cheeses and 1.87 in
pH values of the control cheese during refrigerated storage control cheese, markedly higher than the 0.54 mean value
would have been less favourable for the activity of cyprosins, recorded for ewe raw milk Manchego cheese made using
with optimum pH values around 5.1 (Heimgartner et al. animal rennet (Gaya et al. 2005) but close to the 1.50
1990), but more favourable for plasmin, with maximal ac- value found for La Serena cheese (Garde et al. 2007).
tivity at slightly alkaline pH (Visser 1981). This fact may The level of hydrophilic peptides increased from day 60 to
contribute to explain the lower concentrations of residual day 240 by factors of 3.19 in the control cheese and 1.78–2.56
Fig. 2 Values (means with 6.20 60

SEM) of pH and dry matter pH DM, %
(DM) content during ripening 6.00 58
and refrigerated storage of
control and HPP cheeses.
5.80 56
Control (X), 400W3 (triangle),
600W3 (circle), 400W5
(diamond), 600W5 (square) 5.60 54
5.40 52
5.20 50
5.00 48
35 d 60 d 120 d 180 d 240 d 35 d 60 d 120 d 180 d 240 d
in the HPP cheese. At the end of refrigerated storage, the level the extracellular peptidolytic activity in cheese. Higher FAA
of hydrophilic peptides in the control cheese was significantly concentrations on days 30 and 60 of ripening were also found
(p<0.05) higher than in the rest (Fig. 3), in agreement for La Serena cheese pressurized at 400 MPa on day 2 (Garde
with its low concentrations of residual proteins. On the et al. 2007). In the present work, the FAA concentration was
contrary, the levels of hydrophobic peptides hardly varied 53 % higher in the 400W3 cheese than in the control cheese on
during refrigerated storage and on day 240 there were no day 60, but afterwards the differences in FAA between the
significant differences in the contents of hydrophobic pep- 400W3 cheese and the control cheese were no longer
tides between HPP and control cheeses. The hydrophobic statistically significant. The higher pH values of the control
peptides/hydrophilic peptides ratio fell sharply during cheese during refrigerated storage most probably enhanced
refrigerated storage, to values of 0.64–0.92 on day 240, with the activity of the peptidolytic enzymes of microbial origin,
no significant differences between HPP and control cheeses. increasing the FAA concentration to levels close to those of
To our knowledge, no information is available on the post- the 400 MPa cheeses. At the end of the refrigerated storage
ripening evolution of peptides in the Casar or La Serena period, on day 240, the cheeses pressurized at 600 MPa
cheeses during the refrigerated storage period, which may showed significantly (p<0.05) lower FAA concentrations
span for several months because of the seasonality in milk than the rest (Fig. 3), a result which can be associated to
and cheese production. the observed inactivation of peptidolytic enzymes at the
The concentration of FAA increased significantly (p<0.001) higher pressure level (Table 1).
with time. It did not vary immediately after the HPP of cheeses
on days 21 (data not shown) and 35. However, total FAA were Texture
on day 35 at significantly (p<0.05) higher levels in the 400W3
cheese than in the rest (Fig. 3), as already found for hydrophilic Firmness and elasticity declined significantly (p<0.001) with
peptides, which served as substrates for the activity of time. A marked softening of cheese texture was observed
peptidolytic enzymes. Pressurization of cheeses at 400 MPa during the first weeks of ripening, resulting in the soft creamy
most probably lyses a fraction of the bacterial cells without texture desirable for this type of cheese. Firmness, as deter-
a negative effect on the activity of the enzymes which are mined from the compression curves, reached low values in
released into the medium (Picon et al. 2013a), thus increasing HPP and control cheeses (Fig. 4), consequently with the
45 b 2.50
a a a
40 2.25 a
Hydrophilic peptides, AU/mg DM
a
a a b
35 a 2.00
ab
a a ab
1.75 a a
30 a a
a
Peptide ratio
a
a a 1.50
25 a a
a a 1.25 a a a
a a a
20 b a a
ab 1.00 a a a
ab
a a a a a
15 a a a a
0.75 a
10 0.50
5 0.25
0 0.a00
35 d 60 d 120 d 180 d 240 d 35 d 60 d 120 d 180 d 240 d
30 a 30
a a b a c
a a c a c
b a
Hydrophobic peptides, AU/mg DM
bc ab a
abc ab ab
25 ab ab 25 bc
a a b
a a a
a
Total FAA, mg/g DM
ab ab
20 20
a a
b a a
ab ab
15 15
a a
10 10
b ab
ab
b a a
5 5 a a a a
0 0
35 d 60 d 120 d 180 d 240 d 35 d 60 d 120 d 180 d 240 d
Fig. 3 Levels of hydrophilic peptides, hydrophobic peptides, hydro- bar), 400W5 (horizontally striped bar), 600W5 (dotted bar). Means
phobic/hydrophilic ratio and total free amino acids (FAA) during (bars with SEM) at the same sampling date with the same letter do
ripening and refrigerated storage of control and HPP cheeses. not differ significantly (p>0.05)
Control (black bar), 400W3 (obliquely striped bar), 600W3 (white
Fig. 4 Texture parameters 0.06 b c

(firmness and elasticity) during b
ripening and refrigerated
0.05 b
storage of control and HPP
cheeses. Control (black bar),
b
400W3 (obliquely striped bar), 0.04
600W3 (white bar), 400W5 bc
(horizontally striped bar),
Firmness (J)
0.03 b b
600W5 (dotted bar). Means b b
ab
(bars with SEM) at the same ab ab b
0.02 a b
sampling date with the same
letter do not differ significantly ab ab
a ab ab
(p>0.05) 0.01 a
a
a
a
0.00
35 d 60 d 120 d 180 d 240 d
0.06
c
b
0.05 b
b
Elasticity (N/mm2)
0.04
b
bc
bc c
0.03 b
b
ab b
ab
0.02 abc a
ab a
a
a a a
ab a
0.01
a a
0.00
35 d 60 d 120 d 180 d 240 d
extensive proteolysis due to the use of cardoon extract as milk Elasticity exhibited a pattern similar to that of firmness, with
coagulant. The level of intact caseins, in particular of αs1- higher values generally for the 600 MPa cheeses and lower
casein, influences the stability of the cheese protein network values for the 400 MPa cheeses (Fig. 4). The compression
(Creamer and Olson 1982). In the present work, higher firm- curves showed no breaking point, impeding the determina-
ness values were generally found for the 600 MPa cheeses, tion of the fracturability parameter.
which showed the highest levels of intact caseins. A strong
correlation between residual caseins and firmness had been Sensory Evaluation
recorded for La Serena cheese (Fernández del Pozo et al.
1988). Also, HPP by itself strengthens the texture of cheeses Both flavour intensity and flavour quality were significantly
made from milk coagulated with cardoon extract, with an (p<0.001) influenced by time, according to the analysis of
increase in the values of texture parameters immediately after variance. HPP at 600 MPa applied on day 21 retarded flavour
treatments which is more marked at higher pressure levels development during ripening, with lower (p<0.05) intensity
(Garde et al. 2007). scores on day 60 for the 600W3 cheese than for the respective
Two additional factors influencing cheese texture are pH control cheese (Fig. 5) and no significant differences between
value and DM content. At high pH values the casein mole- the control cheese and the rest of the HPP cheeses. Afterwards,
cules acquire a negative net charge and ionic interactions flavour intensity increased in all cheeses, with significantly
change from attraction to repulsion, weakening cheese tex- (p<0.05) lower scores for the 600 MPa cheeses than for the
ture (Creamer and Olson 1982). In the present work, the control cheese from day 180 onwards (Fig. 5) and no signif-
lower pH values of 600 MPa cheeses during refrigerated icant differences between the 400 MPa cheeses and control
storage probably contributed to their relatively higher firm- cheese. A significant (p<0.01) correlation was found between
ness. Negative correlations between pH values and firmness flavour intensity scores of cheeses during the whole refriger-
of La Serena cheese had been reported (Fernández del Pozo ated storage period and the respective levels of total FAA.
et al. 1988). A higher DM content is known to strengthen the Flavour intensity scores of 60-day-old La Serena cheeses
matrix structure (Nuñez et al. 1991). However, the small pressurized at 300 or 400 MPa on days 2 and 50 after
differences in DM content between cheeses observed in the manufacturing did not differ significantly from that of control
present work precluded a significant effect of DM on texture. cheese (Garde et al. 2007). The results obtained in the present
Fig. 5 Sensory characteristics 10

(flavour intensity and flavour
9
quality) during ripening and
c
refrigerated storage of control 8 b
b b c
b b bc
and HPP cheeses. Control b b
Flavour intensity
7 ab ab ab b a ab
(black bar), 400W3 (obliquely a a
striped bar), 600W3 (white 6 a
a
bar), 400W5 (horizontally
5
striped bar), 600W5 (dotted
bar). Means (bars with SEM) at 4
the same sampling date with the
3
same letter do not differ
significantly (p>0.05) 2
0
60 d 120 d 180 d 240 d
10
8 b
ab ab b
7 ab c
a b b b
b b bc
Flavour quality
6 b b b
b
5
a
4
3 a
2 a
1
0
60 d 120 d 180 d 240 d
work for 400 MPa cheeses are in agreement with those cheese and 400W3 cheese in comparison with 600W3 cheese,
obtained for La Serena cheese treated at 400 MPa in spite of on day 180 for control cheese in comparison with 600W3
the different days of ripening at the time of pressurization. and 600W5 cheeses, and on day 240 for control cheese in
Flavour quality scores at the end of ripening, on day 60, comparison with 600W3 cheese (data not shown). These
showed few significant differences between cheeses. differences could not be associated to the levels of hydro-
However, the flavour quality of the control cheese declined phobic peptides. An opposite effect of HPP was reported
dramatically throughout refrigerated storage reaching signifi- for La Serena cheese, which showed increased bitterness
cantly (p<0.05) lower scores than all the HPP cheeses from when pressurized on day 2 at 300 or 400 MPa and on
day 120 onwards (Fig. 5). The flavour quality score of 400W3 day 50 at 300 MPa in comparison with the control cheese
cheese also declined with time, attaining significantly lower (Garde et al. 2007). The different ripening time of cheeses
values on days 180 and 240 than on day 60. The highest at HPP and the higher pressure level (600 MPa) applied
flavour quality score on day 240 was that of 600W5 cheese, in the present work may be responsible for the variation
which did not vary during refrigerated storage. La Serena in results.
cheese pressurized at 400 MPa on day 2 after manufacture
showed a significantly lower flavour quality score on day 60
than control cheese, but there was no difference in quality Conclusions
if the cheese was pressurized on day 50 (Garde et al.
2007). As the study on La Serena cheese ceased on day Research on HPP application to cheese has been mainly
60, the results of the present work during refrigerated focused on its effects during the ripening period. The objective
storage cannot be compared. of the present work was to preserve cheese flavour quality
Flavour descriptors “acid”, “salty”, “sweet” and “umami” during post-ripening refrigerated storage, mostly of cheese
were not influenced by HPP. Differences were occasionally varieties with seasonal variations in production. According
found during refrigerated storage for “bitter” scores, which to the results obtained, HPP appears as a useful tool to prevent
attained higher (p<0.05) values on day 120 for control over-ripening of raw milk cheeses, in particular of those
varieties which suffer extensive proteolysis because of Garde, S., Arqués, J. L., Gaya, P., Medina, M., & Nuñez, M. (2007).
Effect of high-pressure treatments on proteolysis and texture of
manufacturing procedures, e.g. if cardoon extract is used for
ewes’ raw milk La Serena cheese. International Dairy Journal, 17,
milk coagulation. HPP at 600 MPa was particularly effective 1424–1433.
in retarding the breakdown of proteins and the formation of Gaya, P., Sánchez, C., Nuñez, M., & Fernández-García, E. (2005).
peptides and FAA during prolonged refrigerated storage. Proteolysis during ripening of Manchego cheese made from raw
or pasteurized ewes’ milk. Seasonal variation. Journal of Dairy
Flavour development during the 60-day ripening period was
Research, 72, 287–295.
similar in pressurized and control cheeses. Afterwards, HPP Heimgartner, U., Pietrzak, M., Geertsen, R., Brodelius, P., Da Silva
prevented the dramatic decline in flavour quality recorded for Figueiredo, A. C., & Pais, M. S. S. (1990). Purification and partial
the control cheese throughout the refrigerated storage. HPP characterization of milk clotting proteases from flowers of Cynara
cardunculus. Phytochemistry, 29, 1405–1410.
cheeses, with the only exception of 400W3 cheese, retained
Huppertz, T., Fox, P. F., & Kelly, A. L. (2004a). Dissociation of caseins
until day 240 flavour quality scores not differing from in high pressure-treated bovine milk. International Dairy Journal,
those of the respective 60-day-old cheeses. The HPP of 14, 675–680.
cheeses at 600 MPa may thus be recommended to prevent Huppertz, T., Fox, P. F., & Kelly, A. L. (2004b). Susceptibility of
plasmin and chymosin in Cheddar cheese to inactivation by high
over-ripening and maintain flavour quality during prolonged
pressure. Journal of Dairy Research, 71, 496–499.
refrigerated storage. Juan, B., Ferragut, V., Buffa, M., Guamis, B., & Trujillo, A. (2007).
Effects of high pressure on proteolytic enzymes in cheese: rela-
tionship with the proteolysis of ewe milk cheese. Journal of Dairy
Acknowledgments This research was funded by the AGL 2009– Science, 90, 2113–2125.
07801 project (Ministry of Science and Innovation, Spain). The authors Krause, I., Bockhardt, A., Neckermann, H., Henle, T., & Klostermeyer,
thank the Protected Designation of Origin dairy for providing the H. (1995). Simultaneous determination of amino acids and bio-
cheeses and Hiperbaric for the HPP treatments. J. Calzada is the genic amines by reversed-phase high-performance liquid chroma-
recipient of an FPI grant (Ministry of Science and Innovation, Spain). tography of the dabsyl derivatives. Journal of Chromatography. A,
715, 67–79.
Lau, K. Y., Barbano, D. M., & Rasmussen, R. R. (1991). Influence of
pasteurization of milk on protein breakdown in Cheddar cheese
References during aging. Journal of Dairy Science, 74, 727–740.
Malone, A. S., Wick, C., Shellhammer, T. H., & Courtney, P. D. (2003).
High pressure effects on proteolytic and glycolytic enzymes
Avila, M., Garde, S., Gaya, P., Medina, M., & Nuñez, M. (2006). Effect involved in cheese manufacturing. Journal of Dairy Science,
of high-pressure treatment and a bacteriocin-producing lactic culture 86, 1139–1146.
on the proteolysis, texture, and taste of Hispánico cheese. Journal of McSweeney, P. L. H., & Sousa, M. J. (2000). Biochemical pathways for
Dairy Science, 89, 2882–2893. the production of flavour compounds in cheese during ripening: A
Church, F. C., Swaisgood, H. E., Porter, D. H., & Catignani, G. L. review. Le Lait, 80, 293–324.
(1983). Spectrophotometric assay using o-phtaldialdehyde for Norton, T., & Sun, D.-W. (2008). Recent advances in the use of high
determination of proteolysis in milk and isolated milk proteins. pressure as an effective processing technique in the food industry.
Journal of Dairy Science, 66, 1219–1227. Food and Bioprocess Technology, 1, 2–34.
Collins, Y. F., McSweeney, P. L. H., & Wilkinson, M. G. (2003). Nuñez, M., Guillén, A. M., Rodríguez-Marín, M. A., Marcilla, A. M.,
Lipolysis and free fatty acid catabolism in cheese: a review of Gaya, P., & Medina, M. (1991). Accelerated ripening of ewes' milk
current knowledge. International Dairy Journal, 13, 841–866. Manchego cheese: the effect of neutral proteinases. Journal of
Creamer, L. K., & Olson, N. F. (1982). Rheological evaluation of Dairy Science, 74, 4108–4118.
maturing Cheddar cheese. Journal of Food Science, 47, 631–636. O'Reilly, C. E., O’Connor, P. M., Kelly, A. L., Beresford, T. P., &
Delgado, F.-J., Rodríguez-Pinilla, J., González-Crespo, J., Ramírez, R., Murphy, P. M. (2000). Use of hydrostatic pressure for inactivation
& Roa, I. (2010). Proteolysis and texture changes of a Spanish soft of microbial contaminants in cheese. Applied and Environmental
cheese (Torta del Casar) manufactured with raw ewe milk and Microbiology, 66, 4890–4896.
vegetable rennet during ripening. International Journal of Food O'Reilly, C. E., O'Connor, P. M., Murphy, P. M., Kelly, A. L., &
Science and Technology, 45, 512–519. Beresford, T. P. (2002). Effects of high-pressure treatment on
Fernández del Pozo, B., Gaya, P., Medina, M., Rodríguez-Marín, M. A., viability and autolysis of starter bacteria and proteolysis in Cheddar
& Nuñez, M. (1988). Changes in chemical and rheological charac- cheese. International Dairy Journal, 12, 915–922.
teristics of La Serena ewes’ milk cheese during ripening. Journal of O'Reilly, C. E., Kelly, A. L., Oliveira, J. C., Murphy, P. M., Auty, M. A.
Dairy Research, 55, 457–464. E., & Beresford, T. P. (2003). Effect of varying high-pressure
García-Risco, M. R., Olano, A., Ramos, M., & López-Fandiño, R. treatment conditions on acceleration of ripening of cheddar cheese.
(2000). Micellar changes induced by high pressure. Influence in Innovative Food Science and Emerging Technologies, 4, 277–284.
the proteolytic activity and organoleptic properties of milk. Journal Picon, A., Fernandez, J., Gaya, P., Medina, M., & Nuñez, M. (1999).
of Dairy Science, 83, 2184–2189. Stability of chymosin and cyprosins under milk-coagulation and
García-Risco, M. R., Recio, I., Molina, E., & López-Fandiño, R. cheese-ripening conditions. Journal of Dairy Science, 82, 2331–
(2003). Plasmin activity in pressurized milk. Journal of Dairy 2333.
Science, 86, 728–734. Picon, A., Alonso, R., Gaya, P., & Nuñez, M. (2013a). High-pressure
Garde, S., Tomillo, J., Gaya, P., Medina, M., & Nuñez, M. (2002). treatment and freezing of raw goat milk curd for cheese manufac-
Proteolysis in Hispánico cheese manufactured using a mesophilic ture: effects on cheese characteristics. Food and Bioprocess Tech-
starter, a thermophilic starter and bacteriocin-producing Lactococcus nology. doi:10.1007/s11947-012-0923-5.
lactis subsp. lactis INIA 415 adjunct culture. Journal of Agricultural Picon, A., Alonso, R., van Wely, K. H. M., & Nuñez, M. (2013b).
and Food Chemistry, 50, 3479–3485. Microstructural, textural and colour characteristics during ripening of
Hispánico cheese made using high-pressure-treated ovine milk curd. Van Hekken, D. L., Tunick, M. H., & Park, Y. W. (2005). Effect of
Food and Bioprocess Technology. doi:10.1007/s11947-012-0955-x. frozen storage on the proteolytic and rheological properties of soft
Shao, Y. W., & Ramaswamy, H. S. (2011). Clostridium sporogenes- caprine milk cheese. Journal of Dairy Science, 88, 1966–1972.
ATCC 7955 spore destruction kinetics in milk under high pressure Visser, S. (1981). Proteolytic enzymes and their action on milk proteins.
and elevated temperature treatment conditions. Food and A review. Netherlands Milk and Dairy Journal, 35, 65–88.
Bioprocess Technology, 4, 458–468. Wick, C., Nienaber, U., Anggraeni, O., Shellhammer, T. H., &
Tejada, L., Gómez, R., Vioque, M., Sánchez, E., Mata, C., & Fernández- Courtney, P. D. (2004). Texture, proteolysis and viable lactic acid
Salguero, J. (2000). Effect of freezing and frozen storage on the bacteria in commercial Cheddar cheeses treated with high pressure.
sensorial characteristics of Los Pedroches, a Spanish ewe milk Journal of Dairy Research, 71, 107–115.
cheese. Journal of Sensory Studies, 15, 251–262. Yu, L. J., Ngadi, M., & Raghavan, V. (2012). Proteolysis of cheese
Trujillo, A. J., Capellas, M., Buffa, M., Royo, C., Gervilla, R., Felipe, slurry made from pulsed electric field-treated milk. Food and
X., Sendra, E., Saldo, J., Ferragut, V., & Guamis, B. (2000). Bioprocess Technology, 5, 47–54.
Applications of high pressure treatment for cheese production. Yvon, M., & Rijnen, L. (2001). Cheese flavour formation by amino acid
Food Research International, 33, 311–316. catabolism. International Dairy Journal, 11, 185–201.

Capítulo 5.
Reducing biogenic-amine-producing bacteria, decarboxylase
activity, and biogenic amines in raw milk cheese by high-
pressure treatments.

129

Fotografía: colonias de bacterias aerobias totales en agar PC (superior, izquierda),

bacterias Gram negativas en agar MacConkey (superior, derecha), lactobacilos en agar
Rogosa (centro, izquierda), enterococos en agar KF (centro, derecha), coliformes y
enterobacterias lactosa negativo en agar VRB (inferior, izquierda), y estafilococos
coagulasa positivo y negativo en agar BP (inferior, derecha), obtenidas a partir de
siembras de queso Torta del Casar.

Reducing Biogenic-Amine-Producing Bacteria, Decarboxylase Activity,
and Biogenic Amines in Raw Milk Cheese by High-Pressure
Treatments
Javier Calzada, Ana del Olmo, Antonia Picón, Pilar Gaya, Manuel Nuñez
Departamento de Tecnología de Alimentos, INIA, Madrid, Spain
Biogenic amines may reach concentrations of public health concern in some cheeses. To minimize biogenic amine buildup in
raw milk cheese, high-pressure treatments of 400 or 600 MPa for 5 min were applied on days 21 and 35 of ripening. On day 60,
counts of lactic acid bacteria, enterococci, and lactobacilli were 1 to 2 log units lower in cheeses treated at 400 MPa and 4 to 6 log
units lower in cheeses treated at 600 MPa than in control cheese. At that time, aminopeptidase activity was 16 to 75% lower in
cheeses treated at 400 MPa and 56 to 81% lower in cheeses treated at 600 MPa than in control cheese, while the total free amino
acid concentration was 35 to 53% higher in cheeses treated at 400 MPa and 3 to 15% higher in cheeses treated at 600 MPa, and
decarboxylase activity was 86 to 96% lower in cheeses treated at 400 MPa and 93 to 100% lower in cheeses treated at 600 MPa.
Tyramine, putrescine, and cadaverine were the most abundant amines in control cheese. The total biogenic amine concentration
on day 60, which reached a maximum of 1.089 mg/g dry matter in control cheese, was 27 to 33% lower in cheeses treated at 400
MPa and 40 to 65% lower in cheeses treated at 600 MPa. On day 240, total biogenic amines attained a concentration of 3.690
mg/g dry matter in control cheese and contents 11 to 45% lower in cheeses treated at 400 MPa and 73 to 76% lower in cheeses
treated at 600 MPa. Over 80% of the histidine and 95% of the tyrosine had been converted into histamine and tyramine in con-
trol cheese by day 60. Substrate depletion played an important role in the rate of biogenic amine buildup, becoming a limiting
factor in the case of some amino acids.
B iogenic amines (BA) are low-molecular-weight organic bases

showing biological activity (1). Monoamines, diamines, and
polyamines consist of an aliphatic, aromatic, or heterocyclic struc-
also be formed by “deureation,” an alternative metabolic pathway
(9). Tyramine, phenylethylamine, histamine, tryptamine, cadav-
erine, and putrescine are formed through the decarboxylation of
ture with one, two, or more attached reactive amino groups, re- tyrosine, phenylalanine, histidine, tryptophan, lysine, and orni-
spectively. The main monoamines are tyramine, a potent vaso- thine or arginine (via agmatine), respectively. Enterococci and
constrictor with an effect on healthy individuals usually limited to heterofermentative lactobacilli have been considered the main ty-
headache or migraine (2), and histamine, also vasoactive, which ramine and histamine formers, respectively, but other lactic acid
may cause urticaria, hypotension, headache, flushing, and ab- bacteria (LAB) and some Gram-negative bacteria may also be in-
dominal cramps (3). The diamines putrescine and cadaverine can volved in BA formation in cheese (10, 11).
react with nitrite to form carcinogenic nitrosamines (4). Poly- The main source of decarboxylase-positive bacteria in cheese is
amines and diamines may be converted into stable carcinogenic raw milk. Bacterial reduction procedures for milk, such as bac-
N-nitroso compounds and enhance the growth of chemically in- tofugation, pasteurization, pressurization, or high-pressure ho-
duced aberrant crypt foci in the intestine (5). Accumulation of BA mogenization (12–15), may diminish the levels of decarboxylase-
in cheese and other foods is therefore a matter of public health positive bacteria and BA in cheese. Also, bacteriocinogenic strains
concern. of LAB have been shown to inhibit decarboxylase-positive bacte-
The sensitivity of individuals to BA varies considerably. Con- ria, thus hindering BA formation (16). Even irradiation has been
centrations of histamine above 500 to 1,000 mg/kg of food are investigated to control BA buildup in cheese, decreasing BA con-
regarded as potentially dangerous for human health (6), but tents with respect to a nonirradiated control but showing a dele-
cheeses with lower histamine contents have been involved in out- terious effect on sensory characteristics (17). High-pressure (HP)
breaks (7). Alcohol and other potentiating factors may increase treatments have been successfully applied to inactivate microbial
the effect of biogenic amines. For this reason, acceptable BA levels contaminants in raw milk cheese (18, 19). To impede BA forma-
for foods have not been established as a general rule. Only an tion, HP treatments have been assayed only on pasteurized goat
upper limit for histamine, 100 mg/kg, has been set for certain fish milk cheese, but the low BA concentration in control cheese did
and fish products. Threshold values of 100 to 800 mg/kg for ty- not permit an evaluation of the efficacy of the procedure (20).
ramine and 30 mg/kg for phenylethylamine have been suggested
as the toxic dose (6).
During cheese manufacture and ripening, lactic starter cul- Received 31 October 2012 Accepted 6 December 2012
tures, together with other microorganisms, milk enzymes, and Published ahead of print 14 December 2012
coagulant enzymes, degrade milk proteins into peptides, which Address correspondence to Manuel Nuñez, nunez@inia.es.
are further hydrolyzed to free amino acids (FAA). Bacterial decar- Copyright © 2013, American Society for Microbiology. All Rights Reserved.
boxylases are responsible for the conversion of precursor amino doi:10.1128/AEM.03368-12
acids into monoamines and diamines (8), while polyamines can
February 2013 Volume 79 Number 4 Applied and Environmental Microbiology p. 1277–1283 aem.asmPSH
Calzada et al.
In spite of these and other works, there are aspects of BA for- Analysis of FAA was carried out by RP-HPLC after derivatization with
mation in cheese by decarboxylase-positive bacteria and its con- Waters AccQ Fluor reagent, using a Waters AccQ Tag column. Concen-
trol by HP treatments which need to be elucidated. To our knowl- trations of BA and FAA were expressed as mg per g of dry matter (DM).
edge, the fate of bacterial decarboxylases during ripening and the Cheese DM was determined in triplicate, after cheese grinding with sand,
limiting conditions of substrate depletion or availability for decar- by drying to a constant weight in an oven at 102°C. Cheese pH was mea-
sured in triplicate directly by means of a Crison penetration electrode
boxylation reactions have not been investigated. Also, although
(model 52-3,2; Crison Instruments S.A., Barcelona, Spain) coupled to a
the inactivation of glycolytic and proteolytic enzymes of milk- and Crison GPL 22 pH meter.
cheese-borne bacteria by HP treatments has been studied, there is Enzymatic determinations. Aminopeptidase activity was determined
no information on the effect of high pressure on amino acid de- using an extract obtained by homogenizing 10 g of cheese with 20 ml of
carboxylases of bacterial origin in cheese. In the present work, the 100 mM sodium phosphate buffer (pH 7) for 1 min in an Ultra-Turrax
buildup of BA in raw milk cheeses, pressurized or untreated, T-10 blender (IKA) at high speed on ice, followed by centrifugation (10,000 ⫻
throughout their ripening and refrigerated storage periods was g for 30 min at 4°C) and filtering through Whatman no. 2 paper. Activity on
investigated with regard to the populations of bacterial groups duplicate samples was measured with lysine p-nitroanilide (Lys p-NA) and
potentially able to form BA, the activity of decarboxylating en- leucine p-nitroanilide (Leu p-NA) as substrates and expressed as nmol of
zymes present in cheese, and the concentration of amino acids as p-nitroaniline produced per min per g of cheese DM.
For the determination of tyrosine decarboxylase (TDC) activity in
the substrates required for BA formation.
cheese, a standard curve was obtained by adding up to 32 mAU/ml of the
L-TDC apoenzyme (Sigma, Alcobendas, Spain). (One arbitrary unit [AU]
MATERIALS AND METHODS of tyrosine decarboxylase liberates 1.0 ␮mol of CO2 from tyrosine per min
Cheeses and high-pressure treatments. Raw ewe milk (mean total viable at pH 5.5 and 37°C.) Methods described previously by Børresen et al. (21),
counts, 4.78 log10 CFU/ml) was used for the manufacture of Casar cheese with some modifications, were followed. Briefly, each reaction was initi-
in duplicate trials, carried out on consecutive days. Milk (400 liters) with ated by adding different volumes of L-TDC solution (0.625 mg/ml apoen-
no lactic cultures added was coagulated at 30°C for 60 min with cardoon zyme stock solution in 0.5 M acetate buffer [pH 5.5]) in a total volume of
extract. Curd was cut into 1-cm cubes, held at 30°C for 15 min, and 2.2 ml acetate buffer (0.5 M, pH 5.5) in the presence of 1.5 mM L-tyrosine
distributed into cylindrical molds. Cheeses, 13 cm in diameter and 6 cm (Sigma) and 0.15 mM pyridoxal-5=-phosphate (Sigma). The reaction mix-
high, were pressed for 3 h and salted by rubbing dry salt onto all the ture was incubated at 37°C for 24 h and stopped by adding 0.55 ml of 1 M
surfaces. They were ripened at 8°C and in 92% relative humidity (RH) perchloric acid to the mixture. After centrifugation (10,000 ⫻ g for 15 min at
until day 60 and afterwards held at 4°C until day 240. 25°C), tyramine was quantified by HPLC, as described above, and ty-
Cheeses from each trial were pressurized for 5 min at 400 or 600 MPa, ramine concentrations were plotted against concentrations of the added
after 3 or 5 weeks of ripening, and coded as cheeses 400W3, 600W3, TDC. To determine TDC concentrations in cheese, 10 g of cheese was
400W5, and 600W5. Before high-pressure (HP) treatment, cheeses were mixed with 20 ml of a 2% sodium citrate solution, homogenized for 1 min
vacuum packaged in CN300 bags (Cryovac Grace S.A., Barcelona, Spain). in an Ultra-Turrax T-10 blender (IKA) at high speed on ice, and centri-
A 120-liter-capacity isostatic press (NC Hyperbaric, Burgos, Spain) was fuged at 10,000 ⫻ g for 30 min at 4°C. The supernatant was recovered and
used for HP treatments. Come-up times to reach 400 and 600 MPa were dialyzed against sterile double-distilled water for 24 h at 5°C in a dialysis
1.85 and 2.83 min, respectively, and depressurization times were 7 and 8 s. tube with a nominal molecular mass cutoff of 10 kDa (Medicell Interna-
The temperature of the water used as transmitting fluid remained below tional Ltd., London, United Kingdom). The dialysate was centrifuged at
14°C during the process. After HP treatments, cheeses were unpackaged, 10,000 ⫻ g for 20 min at 5°C, and the supernatant was stored at ⫺20°C
and ripening proceeded under the same conditions as the control cheese. until analysis. TDC in cheese was assayed on 0.3 ml of this supernatant as
Microbiological methods. Representative cheese samples (10 g) were described above, without the addition of the apoenzyme. Reaction mix-
homogenized with 90 ml of a sterile 2% (wt/vol) sodium citrate solution at tures were incubated at 37°C, and the tyramine concentration was deter-
45°C in a Colworth Stomacher 400 (A. J. Seward Ltd., London, United mined by HPLC after 0 h and 24 h of incubation. TDC activity in cheeses
Kingdom). Decimal dilutions of samples were prepared in a sterile 0.1% was calculated from the increase in tyramine concentrations after 24 h by
peptone solution. Total viable counts and counts of LAB, enterococci, and means of the TDC standard curve.
lactobacilli were determined in duplicate on plates of plate count agar Statistical treatment. Analysis of variance with HP treatment (four
(Biolife, Milano, Italy) incubated for 48 h at 30°C, De Man-Rogosa- treatments and control) and cheese age as the main effects was performed
Sharpe (MRS) agar (acidified at pH 5.7 with acetic acid; Biolife) incubated on the analytical variables by means of the SPSS Win 14.0 program. Cal-
for 48 h at 30°C, Kenner Fecal (KF) Streptococcus agar (Oxoid, Basing- culation of correlations and comparison of means by Tukey’s test, with
stoke, United Kingdom) incubated for 48 h at 37°C, and Rogosa agar the significance assigned at a P value of ⬍0.05, were carried out by using
(acidified at pH 5.4 with acetic acid; Biolife) incubated anaerobically for the same program.
48 h at 37°C, respectively. Counts of Micrococcaceae were determined in
duplicate on plates of mannitol salt agar (MSA; Oxoid) incubated for 72 h
at 30°C, counts of coagulase-positive staphylococci were determined on RESULTS
Baird-Parker agar (Oxoid) with rabbit plasma fibrinogen (RPF) supple- Cheese microbiota. Levels of total viable counts and LAB were
ment II (Biolife) incubated at 37°C for 48 h, counts of Gram-negative over 7.5 log10 CFU/g on day 1 (Fig. 1) and had increased to almost
bacteria were determined on MacConkey agar (Biolife) incubated for 24 h 9.5 log10 CFU/g by day 21 (data not shown). HP treatments
at 30°C, and counts of coliforms were determined on violet red bile agar brought about significant (P ⬍ 0.05) decreases in counts of all
(VRBA; Oxoid) incubated for 24 h at 37°C. microbial groups, which were particularly pronounced at 600
Chemical analyses. BA and FAA were simultaneously extracted from MPa. Immediately after pressurization, total viable counts were
duplicate samples according to procedures described previously by
0.88 to 1.33 log units lower and LAB counts were 1.32 to 1.64 log
Krause et al. (12). Quantitative analysis of BA, after derivatization with
dabsyl chloride, was carried out by reverse-phase high-pressure liquid
units lower in cheeses treated at 400 MPa than in control cheese,
chromatography (RP-HPLC), using a System Gold HPLC apparatus while the decreases in cheeses treated at 600 MPa were 3.99 to 4.43
(Beckman Coulter, Palo Alto, CA) equipped with a Nova-pack C18 col- log units for total viable counts and 6.03 to 6.51 log units for LAB
umn (Waters, Milford, MA). A standard mixture of BA (Sigma-Aldrich, counts. Afterwards, a slight recovery of LAB was recorded for
Alcobendas, Spain) was used for their identification and quantification. cheeses treated with 600 MPa, with counts 4.14 to 5.17 log units
1278 aem.asm.org Applied and Environmental Microbiology

Biogenic Amines in High-Pressure-Treated Cheeses
lower in cheeses treated at 600 MPa than in control cheese on day

60. Similar patterns were found for counts of enterococci and
lactobacilli. Micrococcaceae were more baroresistant than LAB,
with decreases immediately after treatment of 1.17 to 1.44 log
units in cheeses treated at 400 MPa and 2.78 to 3.89 log units in
cheeses treated at 600 MPa. However, on day 60, coagulase-posi-
tive staphylococci were below the detection level in all pressurized
cheeses. At that time, counts of Gram-negative bacteria and coli-
forms were more than 3 log units lower in 400W3 cheese than in
control cheese and had declined below the detection level in the
rest of the HP-treated cheeses. Differences in microbial counts
between pressurized and control cheeses persisted throughout re-
frigerated storage, until day 240. Contrarily, minor differences
were recorded for pH values, which ranged from pH 5.33 to 5.42
on day 60 and from pH 5.56 to 5.73 on day 240, and in dry matter
content, which ranged from 50.41% to 53.40% on day 60 and
from 54.51% to 57.81% on day 240 (data not shown).
Aminopeptidase activity. Cheese pressurization on day 21 re-
duced significantly (P ⬍ 0.05) the aminopeptidase activity re-
corded with Leu p-NA as the substrate, by 35% in 400W3 cheese
and by 48% in 600W3 cheese, while the respective decreases were
26% and 37% with Lys p-NA as the substrate (data not shown).
On day 35, the decreases in activity caused by pressurization were
21% in 400W5 cheese and 30% in 600W5 cheese with Leu p-NA as
the substrate and 24% in 400W5 cheese and 28% in 600W5 cheese
with Lys p-NA as the substrate (data not shown). Aminopeptidase
activity increased in all cheeses during ripening, attaining its max-
imum values on day 60 (Table 1). Thereafter, a gradual decline of
aminopeptidase activity was recorded, with few significant differ-
ences between cheeses at the last stages of refrigerated storage.
Free amino acids. Accumulation of FAA during ripening pro-
ceeded at different rates in HP-treated and control cheeses from
the 0.24 mg/g DM found in control cheese on day 1 (data not
shown). Pressurization of cheese at 400 MPa, but not at 600 MPa,
enhanced significantly (P ⬍ 0.05) the formation of FAA during
FIG 1 Bacterial counts (log CFU/g) during ripening and storage in control ripening (Fig. 2). Leucine, valine, glutamic acid, and lysine were
() cheeses and cheeses submitted to treatment at 400 MPa (䊐) or 600 MPa
(p) at 3 weeks and at 400 MPa (z) or 600 MPa at 5 weeks. Bars indicate the most abundant FAA in control cheese on day 60. At the end of
standard errors of the means. refrigerated storage, on day 240, concentrations of FAA totaled
23.31 mg/g DM in control cheese, 25.40 to 26.05 in cheeses treated
at 400 MPa, and only 16.47 to 17.47 mg/g DM in cheeses treated at
600 MPa (Fig. 2). At that time, leucine, valine, lysine, and phenyl-
alanine were the most abundant FAA in control cheese.
TABLE 1 Aminopeptidase activity in control and pressurized cheeses during ripening and storagea
Mean aminopeptidase activity (nmol p-NA/min per g cheese DM) ⫾ SEM
Substrate Cheese Day 60 Day 120 Day 180 Day 240
Leu p-NA Control 80.67 ⫾ 11.11D 30.95 ⫾ 4.64B 9.41 ⫾ 0.28AB 6.42 ⫾ 0.60A
400W3 67.45 ⫾ 4.96CD 25.25 ⫾ 0.83B 9.74 ⫾ 1.34B 5.47 ⫾ 0.31A
600W3 19.05 ⫾ 1.32A 9.70 ⫾ 0.93A 5.88 ⫾ 0.87A 4.45 ⫾ 1.01A
400W5 41.27 ⫾ 3.26BC 27.58 ⫾ 3.81B 8.07 ⫾ 0.40AB 4.43 ⫾ 0.43A
600W5 35.17 ⫾ 3.15AB 21.81 ⫾ 1.59B 7.75 ⫾ 0.12AB 5.78 ⫾ 0.90A
Lys p-NA Control 38.10 ⫾ 8.32C 11.34 ⫾ 1.74B 7.07 ⫾ 0.09A 4.67 ⫾ 0.13AB
400W3 25.69 ⫾ 1.65B 9.04 ⫾ 0.46B 6.76 ⫾ 0.50A 5.04 ⫾ 0.83B
600W3 7.22 ⫾ 0.70A 4.15 ⫾ 0.50A 5.57 ⫾ 1.19A 2.81 ⫾ 0.37A
400W5 9.42 ⫾ 2.64A 8.91 ⫾ 1.02B 6.73 ⫾ 0.38A 4.28 ⫾ 0.49AB
600W5 14.15 ⫾ 0.21AB 8.24 ⫾ 0.69AB 6.64 ⫾ 0.95A 3.82 ⫾ 0.14AB
a
On day 1, aminopeptidase activities in control cheese with Leu p-anilide and Lys p-nitroanilide as substrates were 60.93 ⫾ 1.45 and 44.67 ⫾ 0.83 nmol p-nitroaniline/min per g
cheese DM, respectively. Mean values in the same column for the same substrate bearing the same superscript did not differ significantly (P ⬎ 0.05).
February 2013 Volume 79 Number 4 aem.asm.org 1279

Calzada et al.
FIG 2 Selected free amino acids (mg/g DM) during ripening and storage in FIG 3 Biogenic amines (mg/g DM) during ripening and storage in control
control () cheeses and cheeses submitted to treatment at 400 MPa (䊐) or 600 () cheeses and cheeses submitted to 400 MPa (䊐) or 600 MPa (p) at 3 weeks
MPa (p) at 3 weeks and at 400 MPa (z) or 600 MPa at 5 weeks. Bars and at 400 MPa (z) or 600 MPa at 5 weeks. Bars indicate standard errors
indicate standard errors of the means. of the means.
the same cheeses were significant (P ⬍ 0.05). Significant correla-

Biogenic amines. None of the BA was detected in control
cheese on day 1. However, by day 21, the concentration of total BA tions between histamine concentrations and log counts of LAB or
in control cheese had reached 0.371 mg/g DM, and by day 35, it enterococci were occasionally found, with lower r2 values than for
reached 0.740 mg/g DM (data not shown). On day 60, the most lactobacilli. Log counts of Micrococcaceae did not correlate signif-
abundant BA in control cheese were tyramine, putrescine, cadav- icantly with tyramine or histamine concentrations at any of the
erine, and histamine (Fig. 3). The concentration of total BA in- sampling times. Coagulase-positive staphylococci, Gram-negative
creased up to 1.089 mg/g DM in control cheese on day 60, while it bacteria, and coliforms were below detection levels in pressurized
remained at 0.728 to 0.794 mg/g DM in cheeses treated at 400 MPa cheeses, and their correlations with BA could not be calculated.
and 0.377 to 0.656 mg/g DM in cheeses treated at 600 MPa. For- The correlations of histamine contents to histidine contents (r2 ⫽
mation of BA progressed during refrigerated storage, and on day 0.128) and of tyramine concentrations to tyrosine concentrations
240, the concentration of total BA reached 3.690 mg/g DM in (r2 ⫽ 0.290) in all cheeses, at all times of ripening and storage, were
control cheese, 2.022 to 3.276 mg/g DM in cheeses treated with nonsignificant.
400 MPa, and only 0.896 to 1.011 mg/g DM in cheeses treated with Substrate depletion. The proportion of FAA decarboxylated
600 MPa. The most abundant BA in control cheese on day 240 to the respective BA depended on the amino acid and on the HP
were tyramine, putrescine, cadaverine, and tryptamine (Fig. 3). treatment. Decarboxylated phenylalanine in control cheese up to
Spermidine and spermine were not found in any of the cheeses at day 60 represented 27.82% of the FAA formed during the first 60
any stage of ripening or refrigerated storage. Concentrations of all days of ripening, while in pressurized cheeses, it ranged from
individual BA in control cheese correlated significantly (P ⬍ 0.05) 1.86% to 12.53% of the total free phenylalanine formed (Table 2).
with ripening time. Values of r2 ranged from 0.891 for histamine Similar proportions were recorded for other FAA such as lysine
to 0.984 for tyramine when concentrations of individual BA in and tryptophan (data not shown). In the case of histidine, the
control cheese from day 1 to day 240 were plotted against time. decarboxylated FAA in control cheese up to day 60 was 82.82% of
Tyramine concentrations in control and experimental cheeses the free histidine formed until that time, and in pressurized
on days 60, 120, 180, and 240 correlated significantly (P ⬍ 0.05) cheeses, it accounted for 14.84% to 62.83% of the free histidine
with log counts of enterococci in the same cheeses. Some signifi- formed. Decarboxylated tyrosine in control cheese up to day 60
cant correlations were also found between tyramine concentra- reached 96.01% of the free tyrosine formed during the first 2
tions and log counts of LAB or lactobacilli although with lower r2 months, while in pressurized cheeses, it ranged from 68.98% to
values than for enterococci. Correlations of histamine concentra- 81.20%. On day 180, the differences in the availability of FAA
tions on days 120, 180, and 240 with log counts of lactobacilli in between control and pressurized cheeses persisted (Table 2).

TABLE 2 Decarboxylationa of phenylalanine, histidine, and tyrosine to TABLE 3 Tyrosine decarboxylase activity in control and pressurized
the respective biogenic amine in cheese during ripening and storage cheeses during ripening and storage
Decarboxylation (%) Mean tyrosine decarboxylase activity
(mAU/g cheese DM) ⫾ SEMa
Phenylalanine Histidine Tyrosine
Cheese Day 60 Day 180
Cheese Day 60 Day 180 Day 60 Day 180 Day 60 Day 180
Control 0.738 ⫾ 0.027 B
3.008 ⫾ 0.162C
Control 27.82 15.41 82.82 53.44 96.01 72.99
400W3 0.031 ⫾ 0.023A 0.183 ⫾ 0.071AB
400W3 4.16 13.20 62.83 78.05 81.20 73.75
600W3 0.052 ⫾ 0.031A 0.080 ⫾ 0.053A
600W3 1.86 16.13 14.84 5.76 68.98 61.59
400W5 0.103 ⫾ 0.028A 0.689 ⫾ 0.221B
400W5 10.85 7.26 61.04 70.81 80.15 63.86
600W5 ND 0.061 ⫾ 0.044A
600W5 12.53 7.04 36.21 21.25 69.39 57.69
a
a
Means in the same column bearing the same superscript letter did not differ
Expressed as the percentage of decarboxylated FAA of the total amount of FAA
significantly (P ⬎ 0.05). On day 1, TDC activity of control cheese was below the
formed, which was calculated as the sum of the present amount of FAA plus the
detection level. ND, not determined (below the detection level).
maximum (present or past) amount of the respective biogenic amine plus the
equimolecular CO2 resulting from decarboxylation.
formed in control cheese and less than 15% of that formed in

Decarboxylase activity. The standard curve for the determi- pressurized cheeses had been decarboxylated on day 60, while on
nation of tyrosine decarboxylase activity fit a linear model within day 180, this proportion was below 20% in all cheeses (Table 2).
the concentration range of 0 to 0.2 mAU of TDC/ml. It followed Histidine content might limit histamine formation to a certain
the equation y ⫽ 72.185x ⫹ 0.22219, in which y was ␮g tyramine extent, particularly in control cheese, in which more than 80% of
formed/ml and x was mAU TDC/ml, with an r2 value of 0.999. By the free histidine had been decarboxylated by day 60, and in
means of this standard curve, TDC activity in control and pres- cheeses treated with 400 MPa, with over 60% decarboxylated his-
surized cheeses was determined. On day 60, TDC activity was tidine, while in cheeses treated with 600 MPa, the proportion of
significantly (P ⬍ 0.05) higher in control cheese than in pressur- decarboxylated histidine remained below 40%. The availability of
ized cheeses (Table 3). Activity increased during refrigerated stor- free tyrosine appeared to be strongly limiting for tyramine forma-
age, by approximately 4-fold in control cheese and 6-fold in tion, with more than 95% of the free tyrosine being decarboxy-
cheeses treated at 400 MPa from day 60 to day 180, while a minor lated by day 60 in control cheese, while in cheeses pressurized at
increase was observed for cheeses treated at 600 MPa. Differences 400 and 600 MPa, more than 80% and almost 70% of tyrosine,
between the TDC activity of pressurized and control cheeses per- respectively, had been decarboxylated. The remarkably higher
sisted on day 180 (Table 3). tyrosine decarboxylase activity recorded for control cheese (Ta-
ble 3), in which the enzyme had not been inactivated since high
DISCUSSION pressure had not been applied, than in pressurized cheeses, in
Prerequisites for BA formation are availability of FAA, presence of particular when treated at 600 MPa, appears to be the main cause
decarboxylase-positive microorganisms, and environmental con- for tyrosine becoming practically exhausted in the former cheese.
ditions allowing decarboxylase activity (8, 22). In the present Decarboxylase-positive microorganisms in cheese were nega-
work, HP treatments influenced the first two prerequisites, while tively influenced by HP treatments, as expected. Pressurization
the environmental conditions (pH value and dry matter content) reduced the populations of all the analyzed microbial groups, with
were not substantially affected. pH values for control and pressur- a more marked effect on Gram-negative bacteria. Counts of En-
ized cheeses appeared to be favorable for tyramine formation by terococcus, a genus known to harbor abundant decarboxylase-pos-
LAB, while they might slightly hinder its formation by Enterobac- itive strains (11), suffered important declines in pressurized
teriaceae, according to results reported previously for the decar- cheeses. A similar result was obtained for Lactobacillus, also rich in
boxylase activity of various bacterial sources at different pH values strains with decarboxylating activity (27). Decreases in the levels
(23). of all microbial groups were significantly more pronounced after
According to our results, the overall availability of FAA did not treatments at 600 MPa than at 400 MPa, in agreement with results
appear to be a limiting factor for BA formation. On day 60, the from previous studies regarding the effect of pressurization on the
concentration of total FAA in control cheese was 4.82 mg/g DM microbiota of raw milk cheeses (18, 19, 28).
versus 1.09 mg/g DM of total BA, while on day 240, the respective Tyramine concentrations in control and experimental cheeses
concentrations were 23.31 and 3.69 mg/g DM. These concentra- correlated significantly with the respective log counts of entero-
tions of FAA and BA exceed considerably the respective levels cocci from day 60 onwards, and histamine concentrations corre-
previously reported for other cheese varieties (13, 24, 25). On day lated significantly with the respective log counts of lactobacilli
60, total FAA attained higher concentrations in cheeses treated at from day 120 onwards. In spite of these correlations, the possible
400 MPa than in control cheese, a result in apparent contradiction contribution of microorganisms from other genera to the forma-
with the lower aminopeptidase activity recorded for the former tion of tyramine and histamine in the present work cannot be
cheeses. The enhanced proteolysis of HP-treated cheeses observed excluded. Strains of Lactobacillus, Lactococcus, Leuconostoc, Car-
in the present work can be explained by the conformational nobacterium, and Enterobacteriaceae have been found to produce
changes in the caseins of pressurized cheeses, revealed by confocal tyramine (23, 29), and strains of Enterococcus, Lactococcus, Oeno-
laser scanning microscopy (26), which facilitate the access of en- coccus, Pediococcus, Streptococcus, and Enterobacteriaceae have
zymes to their substrates. Regarding some particular FAA, the been found to form histamine (30–32). On the other hand, bac-
phenylalanine concentration did not seem to restrain phenylethyl- terial strains belonging to genera frequently found in cheese, such
amine formation, since less than 30% of the free phenylalanine as Lactobacillus, Micrococcus, and Pediococcus, have been reported

Calzada et al.
to degrade tyramine and histamine, generally by means of mono- ACKNOWLEDGMENTS

amine oxidases and preferably under aerobic conditions (33). This work was supported by project AGL2009-07801 of the Spanish Min-
Also, the concentration of BA in a minicheese model (34) ripened istry of Science and Innovation (MICINN). Javier Calzada was the recip-
for 4 months was approximately 95% lower when selected Lacto- ient of a MICINN fellowship.
bacillus casei strains capable of degrading tyramine and histamine
in MRS broth were added to milk as adjunct cultures. This phe- REFERENCES
nomenon was probably enhanced by the small size of the 1. Silla-Santos MH. 1996. Biogenic amines: their importance in foods. Int. J.
Food Microbiol. 29:213–231.
minicheeses, favorable for aeration. In the case of real-size cheeses,
2. Til HP, Falke HE, Prinsen MK, Willems MI. 1997. Acute and subacute
the low oxygen concentration present in their interior would toxicity of tyramine, spermidine, spermine, putrescine and cadaverine in
probably preclude a significant contribution of oxidases to BA rats. Food Chem. Toxicol. 35:337–348.
degradation. In another work (16), the use of a nisin-producing 3. Stratton JE, Hutkins RW, Taylor SL. 1991. Biogenic amines in cheese
Lactococcus lactis strain as the main starter or two enterocin-pro- and other fermented foods. J. Food Prot. 54:460 – 470.
4. Hotchkiss JH, Scanlan RA, Libbey LM. 1977. Formation of bis(hydroxy-
ducing Enterococcus faecalis strains as adjunct cultures inhibited a alkyl)-n-nitrosamines as products of nitrosation of spermidine. J. Agric.
histamine-forming Lactobacillus buchneri strain and reduced his- Food Chem. 25:1183–1189.
tamine contents in 4-month-old cheese from 177 to 214 mg/kg to 5. Eliassen KA, Reistad R, Risøen U, Rønning HF. 2002. Dietary poly-
less than 10 mg/kg. amines. Food Chem. 78:273–280.
6. ten Brink B, Damink C, Joosten HMLJ, Huis in’t Veld JHJ. 1990.
In the present work, pressurization had a negative effect on all
Occurrence and formation of biologically-active amines in foods. Int. J.
microbial groups and on TDC activity. Levels of microbial groups Food Microbiol. 11:73– 84.
decreased further during ripening and storage of all cheeses. On 7. Taylor SL. 1986. Histamine food poisoning: toxicology and clinical as-
the contrary, TDC activity increased by approximately 4-fold pects. Crit. Rev. Toxicol. 17:91–128.
from day 60 to day 180 in control cheese and, unexpectedly, even 8. Halász A, Baráthá, Simon-Sarkadi L, Holzapfel W. 1994. Biogenic
amines and their production by microorganisms in food. Trends Food Sci.
more in cheeses treated at 400 MPa, by approximately 6-fold. The Technol. 5:42– 49.
increase of TDC activity in control cheese can be explained by the 9. Komprda P, Burdychová R, Dohnal V, Cwiková O, Sládková P. 2008.
production of the enzyme by intact bacterial cells, followed by its Some factors influencing biogenic amines and polyamines content in
release into the surrounding medium, favored by spontaneous cell Dutch-type semi-hard cheese. Eur. Food Res. Technol. 227:29 –36.
10. Joosten HMLJ, Northolt MD. 1987. Conditions allowing the formation
lysis. In cheeses treated at 400 MPa, counts of bacterial groups
of biogenic amines in cheese. 2. Decarboxylase properties of some non-
after pressurization were lower than in control cheese, but the starter bacteria. Neth. Milk Dairy J. 41:259 –280.
mild-pressure treatment would most probably enhance the lysis 11. Sarantinopoulos P, Andrighetto C, Georgalaki MD, Rea MC, Lombardi
of LAB cells, as shown previously for lactococci (35), without in- A, Cogan TM, Kalantzopoulos G, Tsakalidou E. 2001. Biochemical
activating the enzyme. The minor increase in TDC activity re- properties of enterococci relevant to their technological performance. Int.
Dairy J. 11:621– 647.
corded for cheeses treated at 600 MPa might be due to a double 12. Krause I, Bockhardt A, Klostenmeyer H. 1997. Characterization of
effect of high pressure at this level, on the one hand promoting cell cheese ripening by free amino acids and biogenic amines and influence of
lysis, as in the case of cheeses treated at 400 MPa, and on the other bactofugation and heat-treatment of milk. Lait 77:101–108.
hand inactivating the enzyme released into the medium. 13. Lanciotti R, Patrignani F, Iucci L, Guerzoni ME, Suzzi G, Belletti N,
Gardini F. 2007. Effects of milk pressure homogenization on biogenic
In spite of the well-known need for sufficient amounts of
amine accumulation during ripening of ovine and bovine Italian cheeses.
FAA for BA formation, weak correlations between the concen- Food Chem. 104:693–701.
trations of BA and the concentrations of their precursor FAA in 14. Novella-Rodríguez S, Veciana-Nogués MT, Trujillo-Mesa AJ, Vidal-
control and pressurized cheeses of different ages were ob- Carou MC. 2002. Profile of biogenic amines in goat cheese made from
tained. Pools of FAA and BA in cheese are subject to consider- pasteurized and pressurized milks. J. Food Sci. 67:2940 –2944.
15. Ordóñez AI, Ibañez FC, Torre P, Barcina Y. 1997. Formation of biogenic
able variations during ripening. The FAA generated in cheese amines in Idiazábal ewe’s-milk cheese: effect of ripening, pasteurization
by microbial peptidases, besides being decarboxylated to BA, and starter. J. Food Prot. 60:1371–1375.
may be converted into a wide range of compounds through the 16. Joosten HMLJ, Nuñez M. 1996. Prevention of histamine formation in
catabolism processes initiated by aminotransferases, amino cheese by bacteriocin-producing lactic acid bacteria. Appl. Environ. Mi-
crobiol. 62:1178 –1181.
acid lyases, and deaminases (36). The BA formed in cheese
17. Rabie MA, Siliha HI, El-Saidy SM, El-Badawy AA, Malcata FX. 2011.
through FAA decarboxylation may be degraded by oxidases Effect of ␥-irradiation upon biogenic amine formation in blue cheese dur-
and possibly by other enzymes of microbial origin. From a ing storage. Int. Dairy J. 21:373–376.
public health point of view, it must be taken into account that 18. Arqués JL, Garde S, Gaya P, Medina M, Nuñez M. 2006. Inactivation of
the increase in the concentration of FAA brought about by microbial contaminants in raw milk La Serena cheese by high pressure
treatments. J. Dairy Sci. 89:888 – 891.
some of the manufacturing procedures designed to accelerate 19. O’Reilly CE, O’Connor PM, Kelly AL, Beresford TP, Murphy PM. 2000.
cheese ripening may enhance BA formation (37, 38), in partic- Use of hydrostatic pressure for inactivation of microbial contaminants in
ular in the presence of decarboxylase-positive microbiota. cheese. Appl. Environ. Microbiol. 66:4890 – 4896.
According to the results obtained in the present work, it may be 20. Novella-Rodríguez S, Veciana-Nogués MT, Saldo J, Vidal-Carou MC.
2002. Effects of high hydrostatic pressure treatments on biogenic amine
concluded that HP treatments were capable of reducing not only
contents in goat cheeses during ripening. J. Agric. Food Chem. 50:7288 –
the population of potentially decarboxylating microbiota but also 7292.
the level of decarboxylating enzymes and the concentrations of 21. Børresen T, Klausen NK, Larsen LM, Sørensen H. 1989. Purification and
BA. Contrary to the results found for some FAA in control cheese, characterization of tyrosine decarboxylase and aromatic-L-amino-acid
no substrate depletion occurred in pressurized cheeses. In spite of decarboxylase. Biochim. Biophys. Acta 993:108 –115.
22. Linares D, Del Rio B, Ladero V, Martinez N, Fernandez M, Martin MC,
this fact, cheeses treated at 400 and 600 MPa showed total BA Alvarez MA. 2012. Factors influencing biogenic amines accumulation in
concentrations up to 45% and 76% lower, respectively, than those dairy products. Front. Microbiol. 3:180. doi:10.3389/fmicb.2012.00180.
in control cheese. 23. Pircher A, Bauer F, Paulsen P. 2007. Formation of cadaverine, histamine,

putrescine and tyramine by bacteria isolated from meat, fermented sau- Identification of the Enterococcus faecalis tyrosine decarboxylase operon
sages and cheeses. Eur. Food Res. Technol. 226:225–231. involved in tyramine production. Appl. Environ. Microbiol. 68:3537–
24. Gaya P, Sánchez C, Fernández-García E, Nuñez M. 2005. Proteolysis 3544.
during ripening of Manchego cheese made from raw or pasteurized ewes’ 32. Rossi F, Gardini F, Rizzotti L, La Gioia F, Tabanelli G, Torriani S. 2011.
milk. Seasonal variation. J. Dairy Res. 72:287–295. Quantitative analysis of histidine decarboxylase gene (hdcA) transcription
25. Mayer HK, Fiechter G, Fischer E. 2010. A new ultra-pressure liquid and histamine production by Streptococcus thermophilus PRI60 under
chromatography method for the determination of biogenic amines in conditions relevant to cheese making. Appl. Environ. Microbiol. 77:2817–
cheese. J. Chromatogr. A 1217:3251–3257. 2822.
26. O’Reilly CE, Kelly AL, Oliveira JC, Murphy PM, Auty MAE, Beresford 33. Leuschner RG, Heidel M, Hammes WP. 1998. Histamine and tyramine
TP. 2003. Effect of varying high-pressure treatment conditions on accel- degradation by food fermenting microorganisms. Int. J. Food Microbiol.
eration of ripening of cheddar cheese. Innov. Food Sci. Emerg. Technol. 39:1–10.
4:277–284. 34. Herrero-Fresno A, Martínez N, Sánchez-Llana E, Díaz M, Fernández
27. Joosten HMLJ, Northolt MD. 1989. Detection, growth and amine pro- M, Martin MC, Ladero V, Alvarez MA. 2012. Lactobacillus casei strains
ducing capacity of lactobacilli in cheese. Appl. Environ. Microbiol. 55: isolated from cheese reduce biogenic amine accumulation in an experi-
2356 –2359. mental model. Int. J. Food Microbiol. 157:297–304.
28. Rodríguez E, Arqués JL, Nuñez M, Gaya P, Medina M. 2005. Combined 35. Malone AS, Shellhammer TH, Courtney PD. 2002. Effects of high pres-
effect of high-pressure treatments and bacteriocin-producing lactic acid sure on the viability, morphology, lysis, and cell wall hydrolase activity of
bacteria on the inactivation of Escherichia coli O157:H7 in raw milk cheese. Lactococcus lactis subsp. cremoris. Appl. Environ. Microbiol. 68:4357–
Appl. Environ. Microbiol. 71:3399 –3404. 4363.
29. González de Llano D, Cuesta P, Rodríguez A. 1998. Biogenic amine 36. Yvon M, Rijnen L. 2001. Cheese flavor formation by amino acid catabo-
production by wild lactococcal and leuconostoc strains. Lett. Appl. Micro- lism. Int. Dairy J. 11:185–201.
biol. 26:270 –274. 37. Fernández-García E, Tomillo J, Nuñez M. 1999. Effect of added protein-
30. Chaves-López C, De Angelis M, Martuscelli M, Serio A, Paparella A, ases and level of starter culture on the formation of biogenic amines in raw
Suzzi G. 2006. Characterization of the Enterobacteriaceae isolated from milk Manchego cheese. Int. J. Food Microbiol. 52:189 –196.
an artisanal Italian ewe’s cheese (Pecorino Abruzzese). J. Appl. Microbiol. 38. Leuschner RGK, Kurihara R, Hammes WP. 1998. Effect of enhanced
101:353–360. proteolysis on formation of biogenic amines by lactobacilli during Gouda
31. Connil N, Le Breton Y, Dousset X, Auffray Y, Rincé A, Prévost H. 2002. cheese ripening. Int. J. Food Microbiol. 44:15–20.

Capítulo 6.
High-pressure processing for the control of lipolysis, volatile
compounds and off-odours in raw milk cheese.

139

Fotografía: quesos de Torta del Casar en cámara de maduración.

Food Bioprocess Technol (2014) 7:2207–2217
DOI 10.1007/s11947-013-1206-5
ORIGINAL PAPER
High-Pressure Processing for the Control of Lipolysis, Volatile

Compounds and Off-odours in Raw Milk Cheese
Received: 1 July 2013 / Accepted: 26 September 2013 / Published online: 12 October 2013
Abstract Build-up of flavour compounds throughout ripening Introduction

of raw milk cheeses may result in strong over-ripening notes
during refrigerated storage. In order to control the formation of Cheese quality is determined by flavour, rheological proper-
free fatty acids (FFAs) and volatile compounds, and the appear- ties and visual appearance (Fox and Wallace 1997). Flavour,
ance of off-odours, raw milk cheeses were high-pressure- probably the main trait influencing cheese quality, has been
processed (HPP) 21 or 35 days after manufacture at 400 or the subject of numerous scientific investigations. Compounds
600 MPa. Ripening proceeded at 8 °C until day 60 and, responsible for cheese flavour are produced through glycoly-
afterwards, cheeses were held at 4 °C until day 240. The effect sis, lipolysis and proteolysis, followed by the secondary reac-
of HPP on the formation of FFAs and volatile compounds was tions which occur during ripening (McSweeney and Sousa
dependent on pressure level and cheese age at the time of 2000; Yvon and Rijnen 2001; Collins et al. 2003). More than
treatment. On day 60, acetic and propionic acids, branched- 600 volatile compounds have been identified in different
chain FFAs and short-chain FFAs showed the lowest (p <0.05) cheese varieties and many have been related to particular
concentrations in cheeses treated at 400 or 600 MPa on day 21, odour and aroma notes (Molimard and Spinnler 1996; Sablé
while medium- and long-chain FFAs were at similar levels in and Cottenceau 1999; Curioni and Bosset 2002).
all cheeses. HPP influenced significantly (p <0.05) 84 out of The main agents responsible for the biochemical changes
the 94 volatile compounds found in cheese. On day 60, the taking place during the manufacture and ripening of pasteur-
lowest (p <0.05) concentrations of acids, alcohols and esters ized milk cheeses are coagulant enzymes together with starter
were recorded for cheeses treated at 400 or 600 MPa on day 21, cultures and their enzymes. Pasteurization inactivates most
and the lowest (p <0.05) concentrations of ketones for cheeses enzymes and microorganisms present in raw milk. However,
treated at 400 MPa on days 21 or 35. On day 240, all HPP the enzymes and microorganisms coming from milk are cru-
cheeses showed lower (p <0.05) concentrations of aldehydes, cial for the ripening process of raw milk cheeses (Fernández-
esters and, particularly, sulphur compounds than control García et al. 2002; Gaya et al. 2005). In the case of varieties
cheese, which exhibited putrid and rancid off-odours from not produced, or produced at a lesser amount, during certain
day 120 onwards. Principal component analysis combining periods of the year, generally due to the seasonality in milk
FFAs and volatile compounds discriminated 240-day control production, refrigerated storage of ripe cheese for several
cheese from 120-day control cheese and both from the rest of months is a must. Since the action of enzymes and microor-
cheeses. ganisms persists during refrigerated storage of ripe cheese,
over-ripening may happen before the product is consumed.
Raw milk cheeses, with pronounced enzyme activity and
Keywords High-pressure processing . Lipolysis . Volatile microbial metabolism associated reactions, are particularly
compounds . Cheese prone to it. Freezing of ripe cheeses has been investigated to
prevent over-ripening and prolong their shelf life. However,
even though their flavour characteristics remained unchanged
J. Calzada : A. del Olmo : A. Picon : P. Gaya : M. Nuñez (*)
at thawing, both texture and visual appearance were negative-
Departamento de Tecnología de Alimentos, INIA, Carretera de La
Coruña Km 7, Madrid 28040, Spain ly affected (Tejada et al. 2000; Van Hekken et al. 2005). In the
e-mail: nunez@inia.es particular case of mixed milk (ewe + cow or goat + cow)
cheese, curd made from ewe or goat milk in the period of vacuum-packaged and unpackaged simultaneously with the
maximum production was frozen, and after several months cheeses pressurized after 21 days of ripening. Ripening took
thawed and mixed with fresh cow milk curd for cheese man- place at 8 °C and 92 % RH until day 60, and afterwards
ufacture with satisfactory results (Picon et al. 2013a, b). cheeses were held at 4 °C until day 240.
Because of its negligible effect on flavour characteristics, At each of the sampling times, two 100 g pieces per cheese
high-pressure processing (HPP) meets the increasing consum- were wrapped in aluminum foil, vacuum-packaged and frozen
er demand for fresher tasting minimally processed foods at −40 °C for chemical analyses.
(Norton and Sun 2008). It has been successfully applied to
milk and cheese for the inactivation of pathogenic and spoil- Chemical Determinations
age microorganisms (O’Reilly et al. 2000; Shao and
Ramaswamy 2011). Some cheese-related enzymes such as Acetic and propionic acids, branched-chain carboxylic acids
proteinases and peptidases lose their activity when subjected (BCCAs) and free fatty acids (FFAs) from butyric (C4:0) to
to HPP, totally or partially depending on the pressure level linolenic acid (C18:3) in cheese were extracted in duplicate and
applied (Malone et al. 2003; Huppertz et al. 2004; Juan et al. determined by gas chromatography (GC) with flame ioniza-
2007). The effect of HPP on cheese esterase activity depends tion detection, as described by Fernández-García et al. (2006).
on the pressure level but also on the type of starter culture used Frozen cheese pieces were thawed overnight at 4 °C, prior to
(O’Reilly et al. 2002; Ávila et al. 2007). Formation of volatile analysis. At all sampling times, acids were extracted from
compounds in pressurized cheese is influenced by both the cheeses using a solid-phase extraction technique, with
pressure level and the age of cheese at the time of treatment pentanoic, nonanoic and heptadecanoic acids added as inter-
(Ávila et al. 2006; Arqués et al. 2007). On the basis of those nal standards (IS). Fifteen standard solutions of fatty acids
results, it seemed feasible to apply HPP for preventing over- were used for the calculation of calibration curves. Individual
ripening during refrigerated storage of raw milk cheese. FFAs were separated, identified and quantified, and their
Casar cheese studied in the present work is made from ewe concentrations expressed in milligram per gram cheese dry
raw milk coagulated with an aqueous extract of Cynara matter.
cardunculus L. (cardoon) flowers. The strong activity of Volatile compounds were extracted in duplicate from
cardoon enzymes, the raw milk microbiota and the high cheese using a solid-phase microextraction method (Mallia
cheese pH values may cause over-ripening during refrigerated et al. 2005). Ten grams of cheese was homogenized in a
storage of ripe cheese. In a previous work (Calzada et al. mechanical grinder with 20 g of Na2SO4 and 25 μL of an
2013), we achieved a delay in proteolysis during refrigerated aqueous solution of 1,058 mg/L cyclohexanone as IS. Five
storage of Casar cheese by means of HPP treatments on days grams of the mixture was weighed in a 15-mL headspace glass
21 and 35 after manufacture, at 400 or 600 MPa. In the present vial sealed with a PTFE-faced silicone septum (Supelco,
work, we have investigated the influence of HPP treatments Bellefonte, PA, USA). Vials were submerged in a thermostatic
under the same conditions on the lipolysis, the formation of bath at 30 °C (D3 model, HAAKE, Berlin, Germany) for both
volatile compounds and the appearance of off-odours through- equilibration (20 min) and extraction (30 min) phases. A
out ripening and refrigerated storage of Casar cheese. SPME manual holder equipped with a 2 cm×50/30 μm
StableFlex divinylbenzene/carboxen/polydimethylsiloxane-
coated fibre (Supelco) was inserted through the PTFE septum
Materials and Methods for headspace extraction, after which it was inserted into the
GC injection port for desorption (270 °C/10 min in splitless
Cheese Manufacture and High-Pressure Processing mode). Before use, the fibre was conditioned in the injection
port of the GC (270 °C/1 h) as recommended by the manu-
The manufacturing procedure of Casar cheese from ewe raw facturer. After each run, the fibre was cleaned up to avoid
milk was described by Calzada et al. (2013). The experiment carry-over problems, and periodically, fibre sensitivity was
was carried out in duplicate, at a protected designation of tested with an aqueous solution of the internal standard. All
origin (PDO) dairy, on two batches of cheese made on con- analyses were run using the same fibre unit.
secutive days, using an aqueous extract of dry cardoon flowers Gas chromatography–mass spectrometry was carried out
as milk coagulant. using a HP 6890-MSD HP 5973 apparatus (Agilent, Palo Alto,
Cheeses were pressurized at 400 or 600 MPa for 5 min, CA, USA) with a capillary column (60 m long; 0.25 mm i.d.;
after 21 or 35 days of ripening, as described by Calzada et al. 0.5 μm film thickness; Zebron-WAX plus, Phenomenex, Tor-
(2013). They were coded as 400W3, 600W3, 400W5 and rance, CA, USA) and helium flow at 1.4 mL/min for 1 min
600W5, according to the pressure level applied (400 or followed by 1 mL/min. The temperature program was 7 min at
600 MPa) and the age of cheese (3 or 5 weeks) at pressuriza- 40 °C, first ramp 2 °C/min to 90 °C, second ramp 3 °C/min to
tion. After HPP, they were unpackaged. Control cheeses were 150 °C, final ramp 9 °C/min to 240 °C, and 8 min at 240 °C.
Detection was performed with electron impact ionization, with cheese at day 35. A certain recovery of microbial metabolism
70 eV ionization energy operating in the full-scan mode at in the 400W3 cheese occurred afterwards, as shown by the
1.74 scans/s. Source and quadrupole temperatures were 230 increase in its acetic acid concentration from day 35 to day 60
and 150 °C, respectively. Compound identification was car- and during refrigerated storage. On the contrary, the acetic
ried out by injection of commercial standards and by spectra acid concentration hardly varied in 600W3 cheese from day
comparison using the Wiley7Nist05 Library (Wiley & Sons 35 onwards. The effect of HPP when applied on day 35 was
Inc., Germany). The sum of abundances of characteristic ions less marked, with no significant differences in acetic acid
was used for semi-quantitation of compounds. The areas have concentration between cheeses pressurized on day 35 and
been referred to the IS (compound peak area multiplied by 103 control cheese during the rest of ripening and refrigerated
and divided by the IS peak area). storage. Acetic acid, a major odorant of Cheddar, Gruyère
and Emmental cheeses (Curioni and Bosset 2002) and of other
Sensory Evaluation varieties including most ewes milk cheeses (Fernández-García
et al. 2006), with a typical vinegar odour note, plays an
A trained 15-member panel carried out the evaluation of odour important role in cheese flavour and aroma by itself and as a
intensity and quality (preference) of 60-, 120- and 240-day substrate for ester formation through esterification reactions.
cheeses, scoring on a 0–10 points scale (Nuñez et al. 1991). Concentrations of acetic acid ranging from 3.9 to 6.1 mg/g had
Odour was defined as the olfactory sensation felt directly by been reported for ripe Casar cheese by Delgado et al. (2009).
smelling the cheese (Fernández-García et al. 2002). In addi- Propionic acid, generally derived from the microbial me-
tion, panellists were asked to evaluate putrid and rancid odour tabolism of lactate, was not detected on day 1, but its concen-
notes, also scoring on a 0–10 points scale. Cheese samples of tration increased almost sixfold from day 21 to day 60 in
the same ripening time from the same experiment (four HPP control cheese (Table 1). Formation of propionic acid was
cheeses and one control cheese) were simultaneously presen- influenced by HPP of cheeses on day 21, with significantly
ted to panellists at each of the sensory evaluation dates. (p <0.05) lower concentrations in 400W3 and 600W3 cheeses
than in control cheese not only on day 35 but also during the
Statistical Analysis rest of the ripening period and throughout refrigerated storage
until day 240. The different patterns of acetic and propionic
Data obtained were analyzed by a two-way analysis of variance, acid concentrations from day 35 to day 240 in cheeses pres-
with HPP treatment and days after manufacture as the main surized on day 21 were probably due to a higher barotolerance
effects. Means were compared using Tukey’s test, with p of acetic acid-producing bacteria compared with propionic
assigned at 0.05. Principal component analysis (PCA) was acid-producing bacteria. Propionic acid, characteristic of
carried out on individual volatile compounds and on the total Swiss-type cheeses, also with a vinegar odour note, is also
levels of groups of FFAs and volatile compounds of 60-, 120- present in Camembert and some ewes milk cheeses (Molimard
and 240-day cheeses for the discrimination of samples according and Spinnler 1996; Fernández-García et al. 2004), although it
to the HPP treatment applied and the days after manufacture. was not found in a previous work on Casar cheese (Delgado
The SPSS Win 14.0 software (SPSS Inc., Chicago, IL, USA) et al. 2009). Besides its direct role in cheese odour and aroma,
was used for the statistical analysis of data. propionic acid contributes to ester formation.
BCCAs were detected at very low concentration on day 1
and increased by more than eightfold from day 21 to day 60 of
Results and Discussion ripening in control cheese (Table 2). Their formation was
hindered by HPP of cheeses on day 21, as shown by the lower
Acetic, Propionic and Branched-Chain Carboxylic Acids (p < 0.05) concentrations of 2-methylpropanoic and 3-
methylbutanoic acids in 400W3 and 600W3 cheeses than in
Acetic acid, which derives from the metabolism of lactose, control cheese from day 35 onwards. On the contrary, differ-
lactate and citrate by lactic acid bacteria and other microor- ences in BCCAs concentrations between cheeses pressurized
ganisms, could be detected as early as day 1 (Table 1). Pro- at 400 or 600 MPa on day 35 and control cheese remained
duction of acetic acid proceeded steadily during ripening of non-significant during the rest of ripening and refrigerated
control cheese, reaching a concentration 32 times higher on storage. 2-Methylpropanoic and 3-methylbutanoic acids de-
day 60 than on day 1. Significantly (p <0.05) lower concen- rive, respectively, from the catabolism of valine and leucine,
trations of acetic acid were recorded for 400W3 and 600W3 which starts with a transamination catalysed by aminotrans-
cheeses than for control cheese at day 35, most probably ferases, with α-ketoglutarate or other α-ketoacid as acceptor,
because of the microbial death and injury caused by HPP of followed by an oxidative decarboxylation generating acyl-
cheeses on day 21, while the acetic acid concentrations of CoAs which are further hydrolysed to carboxylic acids
400W5 and 600W5 cheeses did not differ from that of control (Yvon and Rijnen 2001). Lactic acid bacteria are not the only
Table 1 Concentrations of acetic

and propionic acids during ripen- Acid Days Control cheese HPP cheesesa
ing and refrigerated storage of
control and HPP cheeses 400W3 600W3 400W5 600W5
Aceticb 1 0.17±0.06
21 2.39±0.12 a 2.86±0.15 a 2.64±0.39 a
a 35 4.85±0.42 b 3.21±0.21 a 2.86±0.30 a 5.17±0.42 b 4.97±0.24 b
Codes for HPP cheeses are
400W3, HPP at 400MPa on day 60 5.40±0.82 ab 3.92±0.19 ab 3.47±0.23 a 5.35±0.15 ab 6.01±1.00 b
21; 600W3, HPP at 600 MPa on 120 5.65±0.44 b 4.40±0.36 ab 3.27±0.35 a 5.85±0.49 b 5.41±0.47 b
day 21; 400W5, HPP at 400 MPa 180 6.06±1.04 ab 4.83±0.32 ab 3.40±0.21 a 6.35±0.49 b 5.57±0.76 ab
on day 35; 600W5, HPP at
600 MPa on day 35 240 5.82±0.73 b 5.17±0.21 b 3.17±0.24 a 6.23±0.15 b 4.75±0.09 ab
b
Concentrations are expressed in Propionicb 1 ND
milligram per gram cheese dry 21 0.22±0.08 a 0.18±0.09 a 0.22±0.11 a
matter, as mean±SEM of dupli- 35 1.53±0.50 b 0.18±0.06 a 0.23±0.11 a 1.37±0.38 b 1.70±0.51 b
cate determinations on two cheese 60 1.29±0.37 b 0.20±0.08 a 0.27±0.11 a 1.04±0.41 ab 0.96±0.47 ab
making trials. Means in the same
row at the same sampling date 120 2.03±0.26 b 0.39±0.10 a 0.53±0.02 a 0.48±0.15 a 0.68±0.35 a
followed by the same letter do not 180 1.88±0.42 b 0.23±0.07 a 0.26±0.08 a 0.89±0.30 ab 1.54±0.51 ab
differ significantly. ND, below 240 2.99±0.20 b 0.34±0.04 a 0.74±0.31 a 0.82±0.27 a 0.97±0.39 a
detection level
microorganisms capable of BCCAs formation. Gram-negative characteristic of ewes and goat cheeses (Curioni and Bosset
bacteria such as Pseudomonas spp. (Morales et al. 2005) have 2002) and have been found in Casar cheese (Delgado et al.
also been reported to form BCCAs in cheese. In the present 2010) and in La Serena cheese, a similar variety (Carbonell
work, counts of Gram negative bacteria on McConkey agar et al. 2002).
reached high levels in control Casar cheese, 7.43 log cfu/g on
day 21 and 6.85 log cfu/g on day 35 (data not shown), but Free Fatty Acids
declined below detection level after HPP of cheeses at 400 or
600 MPa on days 21 or 35. Early HPP treatment on day 21 Short-chain (SC, C4:0 to C8:0) FFAs originate not only from
was more effective in controlling the formation of BCCAs by the esterase-mediated hydrolysis of triacylglycerides but also
raw milk microorganisms than HPP on day 35. The odour of from the fermentation of lactose and lactate, from the degra-
2-methylpropanoic acid has been described as cheesy, sweaty dation of amino acids and from the oxidation of some ketones,
and sour, and that of 3-methylbutanoic acid as rancid, cheesy, esters and aldehydes (Molimard and Spinnler 1996; Curioni
sweaty and putrid. Both BCCAs are flavour compounds and Bosset 2002; Collins et al. 2003). In the present work, the
Table 2 Concentrations of
branched-chain carboxylic acids Acid Days Control cheese HPP cheesesa
(BCCAs) and short-chain (SC,
C4:0 to C8:0) free fatty acids 400W3 600W3 400W5 600W5
(FFAs) during ripening and re-
frigerated storage of control and BCCAsb 1 0.01±0.00
HPP cheeses 21 0.09±0.02 a 0.14±0.03 a 0.10±0.01 a
35 0.51±0.10 b 0.19±0.05 a 0.12±0.03 a 0.48±0.04 b 0.45±0.09 b
a
Codes for HPP cheeses are 60 0.76±0.08 b 0.33±0.04 a 0.27±0.03 a 0.57±0.16 ab 0.63±0.02 ab
400W3, HPP at 400 MPa on day 120 1.04±0.08 b 0.37±0.07 a 0.32±0.06 a 0.69±0.21 ab 0.64±0.09 ab
21; 600W3, HPP at 600 MPa on 180 0.80±0.02 b 0.32±0.03 a 0.32±0.03 a 0.84±0.12 b 0.81±0.07 b
day 21; 400W5, HPP at 400 MPa
on day 35; 600W5, HPP at 240 1.15±0.18 b 0.40±0.08 a 0.30±0.06 a 0.76±0.06 ab 0.68±0.15 ab
600 MPa on day 35 SC FFAsb 1 0.06±0.01
b 21 0.22±0.07 a 0.26±0.07 a 0.29±0.02 a
Concentrations are expressed in
milligram per gram cheese dry 35 1.55±0.05 b 0.36±0.05 a 0.37±0.07 a 1.67±0.06 b 1.41±0.10 b
matter, as mean±SEM of dupli- 60 1.93±0.13 b 0.55±0.02 a 0.72±0.03 a 1.41±0.46 ab 1.52±0.12 ab
cate determinations on two cheese
making trials. Means in the same 120 5.15±0.49 b 0.92±0.02 a 0.99±0.21 a 0.95±0.09 a 1.01±0.22 a
row at the same sampling date 180 3.42±1.09 b 0.82±0.04 a 0.89±0.04 a 1.95±0.13 ab 2.55±0.17 ab
followed by the same letter do not 240 8.69±1.54 b 1.11±0.03 a 2.34±0.88 a 1.75±0.09 a 1.25±0.37 a
differ significantly
concentration of SC FFAs increased 32-fold from day 1 to day FFAs such as C10:0 and C12:0 contribute to the aroma of Ched-
60 in control cheese (Table 2), with butanoic acid accounting dar, Roncal and other cheese varieties because of their relatively
for 95 % of the total SC FFAs on day 60. The concentrations low perception thresholds (Curioni and Bosset 2002).
of SC FFAs were significantly (p <0.05) lower in 400W3 and Long chain (LC, C16:0 to C18:3) FFAs also increased at a
600W3 cheeses than in control cheese from day 35 onwards, slow rate during ripening of control cheese, by 2.6-fold from
while in 400W5 and 600W5 cheeses, they were significantly day 1 to day 60 (Table 3). On day 60, there were no significant
(p <0.05) lower only on days 120 and 240. Esterases from differences in the concentration of LC FFAs between HPP
lactic acid bacteria seem to be resistant to the pressurization of cheeses and control cheese. Afterwards, significantly
cheese at 400 MPa for 5 min according to Ávila et al. (2007), (p <0.05) higher levels of LC FFAs in control cheese than in
who did not find differences in the concentration of SC FFAs HPP cheeses were recorded at all sampling dates throughout
between HPP cheese and control cheese both made from refrigerated storage. Since yeasts increase the concentration of
pasteurized milk. Voigt et al. (2012) reported lower concen- LC FFAs in cheese (Chen et al. 2012), the high yeast counts
trations of SC FFAs in 21-day cheeses made from HPP (400 or present in Casar cheese (Poullet et al. 1991) might be involved
600 MPa) milk than when made from raw milk, but higher SC in LC FFAs formation. Milk lipoprotein lipase was probably
FFAs concentrations on day 180 for the cheese made from the main agent responsible for LC FFAs formation. In cheeses
milk treated at 600 MPa. In the present work, butanoic acid made from pressurized goat or cow milk, milk lipoprotein
was at significantly lower concentrations in 400W3 and lipase remained active. Thus, the concentration of LC FFAs
600W3 cheeses than in control cheese from day 35 until day in 60-day cheeses made from raw and from HPP (500 MPa)
240 while few significant differences were recorded for C6:0 goat milk did not differ and both were significantly higher
and C8:0 FFAs. This result points to the production of butanoic than in cheese made from pasteurized milk (Buffa et al. 2001).
acid in cheese by microorganisms from raw milk, rather than to When Cheddar cheeses made from HPP milk and from raw
its formation through the esterase-mediated hydrolysis of milk were compared, LC FFAs concentrations were signifi-
triacylglycerides. In a work on pressurized cheese made from cantly higher in raw milk cheese on day 21 and in cheese from
goat raw milk, minor differences in C4:0, C6:0 and C8:0 FFAs milk treated at 600 MPa on day 180 (Voigt et al. 2012). Also,
concentrations during ripening were found between HPP the concentration of LC FFAs in goat raw milk cheese on day
cheeses and control cheese (Delgado et al. 2012). Butanoic 60 was not influenced by HPP treatment of cheese (Delgado
acid, with a rancid cheese-like odour, plays an important role et al. 2012). There is no available information on the
in the flavour of many cheese types made from cow and ewe barotolerance of ewes milk lipoprotein lipase, which is not
milk, although large amounts, usually associated to the butyric necessarily coincident with those of cow or goat milk lipopro-
acid fermentation of lactate, become undesirable. Hexanoic and tein lipases. This might explain the discrepancy between the
octanoic acids are characteristic flavour compounds of aged results obtained in the present work and those reported for
Grana Padano and Roncal cheeses (Curioni and Bosset 2002). cheeses made from cow or goat milk, independently of wheth-
Medium chain (MC, C10:0 to C14:0) FFAs concentration er HPP was applied to the milk or to the cheese. The high
increased at a slow rate during ripening of control cheese, only perception thresholds of LC FFAs limit their contribution to
1.7-fold from day 1 to day 60 (Table 3). By day 60, HPP of cheese aroma, in spite of the high concentrations usually
cheeses had not affected the concentration of MC FFAs. reached in many cheese types (Curioni and Bosset 2002).
Significantly (p <0.05) higher levels of MC FFAs in control
cheese than in HPP cheeses were recorded during refrigerated Volatile Compounds
storage on days 120 and 240, with the only exception of the
400W3 cheese on day 120 which did not differ from the Ninety-four volatile compounds were detected in Casar cheese
respective control cheese. Our results seem to agree with the by SPME followed by GC-MS, namely 5 aldehydes (acetalde-
reported barotolerance of the lipoprotein lipase from goat milk hyde, 3-methylbutanal, 3-ethylbenzaldehyde, 4-ethylben-
(Trujillo et al. 1999). Similar concentrations of C10:0, C12:0 zaldehide, vinylbenzaldehyde), 7 ketones (acetone, 2-
and C14:0 FFAs were found in 60-day cheeses made from raw butanone, 2-pentanone, 2-heptanone, 3-hydroxybutanone, 1-
and from HPP (500 MPa) goat milk, while cheese made from phenyl-2-propanone, tertbutyl-hydroxy-propiophenone), 12 al-
pasteurized milk showed significantly lower concentrations of cohols (2-propanol, ethanol, 2-butanol, 1-propanol, 2-methyl-1-
the three MC FFAs (Buffa et al. 2001). Differences in MC propanol, 1-butanol, methyl-1-butanol, 3-methyl-3-buten-1-ol,
FFAs concentrations on day 60 between HPP cheeses and 2-heptanol, 3-methyl-2-buten-1-ol, 1-hexanol, 2,5-dimethyl-3-
control cheese made from goat raw milk were non-significant hexanol), 12 acids (acetic, propanoic, 2-methylpropanoic,
(Delgado et al. 2012), and this also occurred when the MC butanoic, 3-methylbutanoic, 2-methylbutanoic, pentanoic, 2-
FFAs concentrations in Cheddar cheeses made from HPP and butenoic, hexanoic, 4-hexenoic, heptanoic, octanoic), 33 esters
raw milk were compared, with the only exception of C14:0 (ethylacetate, ethylpropanoate, ethyl-2-methylpropanoate,
which behaved similarly to SC FFAs (Voigt et al. 2012). MC propylacetate, methylbutyrate, 1-methylpropylacetate, 2-
Table 3 Concentrations of medi-

um-chain (MC, C10:0 to C14:0) and Acid Days Control cheese HPP cheesesa
long-chain (LC, C16:0 to C18:3) free
fatty acids (FFAs) during 400W3 600W3 400W5 600W5
ripening and refrigerated storage
of control and HPP cheeses MC FFAsb 1 0.23±0.03
21 0.30±0.03 a 0.28±0.02 a 0.29±0.03 a
35 0.32±0.03 a 0.30±0.02 a 0.29±0.04 a 0.30±0.01 a 0.28±0.01 a
a
Codes for HPP cheeses are 60 0.38±0.01 a 0.37±0.02 a 0.35±0.01 a 0.34±0.01 a 0.36±0.02 a
400W3, HPP at 400 MPa on day 120 0.66±0.08 b 0.52±0.06 ab 0.42±0.02 a 0.44±0.03 a 0.42±0.02 a
21; 600W3, HPP at 600 MPa on 180 0.69±0.11 a 0.55±0.03 a 0.56±0.03 a 0.52±0.01 a 0.48±0.03 a
day 21; 400W5, HPP at 400 MPa
on day 35; 600W5, HPP at 240 0.88±0.08 b 0.63±0.05 a 0.52±0.03 a 0.55±0.03 a 0.54±0.07 a
600 MPa on day 35 LC FFAsb 1 0.54±0.11
b 21 1.09±0.07 a 1.04±0.05 a 1.03±0.04 a
Concentrations are expressed in
milligram per gram cheese dry 35 1.14±0.08 a 1.07±0.09 a 1.00±0.06 a 1.11±0.02 a 1.00±0.06 a
matter, as mean±SEM of dupli- 60 1.40±0.03 a 1.39±0.06 a 1.23±0.02 a 1.25±0.04 a 1.36±0.06 a
cate determinations on two cheese
making trials. Means in the same 120 2.61±0.05 c 2.01±0.03 b 1.50±0.10 a 1.65±0.04 a 1.65±0.01 a
row at the same sampling date 180 2.95±0.14 b 2.03±0.10 a 1.89±0.06 a 1.86±0.13 a 1.62±0.09 a
followed by the same letter do not 240 3.02±0.09 b 2.14±0.10 a 1.70±0.08 a 1.76±0.05 a 1.74±0.18 a
differ significantly
methylpropylacetate, ethylbutyrate, propylpropanoate, 1- 3-Methylbutanal, formed from leucine, increased in control,

methylpropylpropanoate, ethyl-2-methylbutyrate, ethyl-3- 400W5 and 600W5 cheeses during refrigerated storage but
methylbutyrate, butylacetate, isoamylacetate, propylbutyrate, 1- declined in the rest. In a previous study on Casar cheese, it was
methylpropylbutyrate, ethylpentanoate, butylpropanoate, propyl- detected only until day 30 (Delgado et al. 2010). It is a potent
3-methylbutyrate, 2-methylpropylbutyrate, amylpropionate, odorant in Camembert, Cheddar, Emmental and Gruyère
butylbutyrate, ethylhexanoate, butyl-3-methylbutyrate, cheeses and has a pleasant fruity odour which turns into green
ethyltrans-4-hexenoate, propylhexanoate, isobutylhexanoate, malty at high concentrations (Curioni and Bosset 2002). The
ethylheptanoate, ethyloctanoate, pentylhexanoate, ethyl-3- three aromatic aldehydes found, 3-ethylbenzaldehyde,
hydroxybutyrate, ethyldecanoate, butyrolactone), 9 sulphur com- 4-ethylbenzaldehyde and vinylbenzaldehyde, probably derived
pounds (methanethiol, dimethylsulphide, sulphur dioxide, S- from the catabolism of aromatic amino acids by Gram negative
methyl-thioacetate, dimethyldisulphide, S-methylthiobutyrate, bacteria, increased during refrigerated storage. The last one
dimethyltrisulphide, 2-hydroxyethylmethylsulphide, 3- became the dominant aldehyde on day 240. Aromatic alde-
methylthio-1-propanol), 8 aromatic compounds (toluene, sty- hydes are not common compounds in cheese, but 4-
rene, ethylstyrene, 1,3-ethenylbenzene, benzenemethanol, ethylbenzaldehyde has been found in ready-to-use lettuce
benzeneethanol, phenol, 4-methylphenol), 5 hydrocarbons (hex- (Lonchamp et al. 2009).
ane, 1,4-pentadiene, heptane, octane, 3-octene) and 3 ethers (2- Total ketones were at significantly (p <0.05) higher levels
butoxyethanol, 2-(2-ethoxyethoxy)ethanol, 3-phenoxyethanol). in control, 600W3 and 600W5 cheeses than in 400W3 and
According to the analysis of variance, 84 compounds were 400W5 cheeses on day 60, but the differences were no longer
significantly (p <0.05) affected by the HPP treatment applied significant on day 240 (Table 4). 2-Butanone, with a sweet
and 71 compounds by the days elapsed after manufacture. The odor reminiscent of butterscotch (Curioni and Bosset 2002),
number of volatile compounds found in the present work was was the major ketone in control cheese, accounting for ap-
higher than the 46 volatile compounds reported by Delgado proximately 90 % of total ketones on day 60, in agreement
et al. (2010) for this cheese variety and closer to the 112 with a previous work (Delgado et al. 2010). Its levels rose in
volatile compounds found for La Serena cheese, a similar HPP and control cheeses during refrigerated storage, becom-
variety, by Carbonell et al. (2002). ing dominant on day 240 in all the cheeses. 3-Hydroxy-2-
Total aldehydes were at similar levels in HPP cheeses and in butanone (acetoin), with a sour milk odour note, was the
control cheese on day 60 (Table 4). During refrigerated storage, dominant ketone in 600W3 and 600W5 cheeses at the end
they increased at a lesser rate in HPP cheeses than in control of ripening, accounting for 50–80 % of total ketones, but
cheese, remaining on day 240 at significantly (p <0.05) lower afterwards declined. 2-Propanone, with hay and wood pulp
levels in all the HPP cheeses than in control cheese. Acetalde- odour notes, 2-pentanone, with a sweet fruity odour note, and
hyde, the major aldehyde during ripening, mostly formed 2-heptanone, with a Blue cheese note and a low perception
through the metabolism of lactose but also derived from gly- threshold (Molimard and Spinnler 1996), also reached rela-
cine, kept fairly stable during refrigerated storage. tively high levels in HPP and control cheeses.
Table 4 Levels of total volatile aldehydes, ketones, alcohols, acids and esters during ripening and refrigerated storage of control and HPP cheeses
Chemical group Days Control cheese HPP cheesesa
400W3 600W3 400W5 600W5
Aldehydesb 60 10.7±0.7 a 9.2±0.9 a 10.8±0.5 a 9.7±0.4 a 9.3±0.3 a

120 15.8±1.5 b 12.6±1.5 ab 13.3±0.8 ab 11.2±0.4 a 11.0±0.7 a
240 17.1±0.8 b 12.5±1.0 a 11.9±0.4 a 13.3±0.5 a 12.3±0.8 a
Ketonesb 60 1,429±119 b 152±25 a 1,857±135 b 387±113 a 1,523±121 b
120 1,778±266 b 1,836±254 b 1,411±263 ab 1,439±39 ab 1,095±132 a
240 2,147±179 a 1,796±55 a 2,475±478 a 2,117±172 a 1,625±254 a
Alcoholsb 60 9,065±248 b 5,191±276 a 4,866±119 a 9,071±180 b 7,879±876 b
120 13,490±971 c 10,363±498 bc 6,707±418 a 11,375±528 bc 10,067±810 b
240 15,914±1,379 b 14,701±414 b 7,450±547 a 14,065±557 b 9,052±113 a
Acidsb 60 12,275±629 b 5,239±127 a 5,246±302 a 9,878±1,344 b 10,962±529 b
120 10,444±420 b 6,176±260 a 5,866±306 a 7,571±852 ab 8,662±1,279 ab
240 9,483±344 a 6,680±358 a 7,201±865 a 8,332±381 a 8,537±1,341 a
Estersb 60 3,651±430 b 1,411±154 a 862±45 a 2,753±157 b 1,502±34 a
120 3,984±372 c 1,394±69 a 1,197±68 a 3,093±408 bc 2,058±81 ab
240 5,536±361 c 1,533±78 a 1,488±188 a 3,522±256 b 2,403±403 ab
a
Codes for HPP cheeses are 400W3, HPP at 400 MPa on day 21; 600W3, HPP at 600 MPa on day 21; 400W5, HPP at 400 MPa on day 35; 600W5, HPP
at 600 MPa on day 35
b
Levels are expressed in relative abundances to the internal standard, as mean±SEM of duplicate determinations on two cheese making trials. Means in
the same row at the same sampling date followed by the same letter do not differ significantly
Total alcohols were at significantly (p <0.05) higher levels metabolism of lactose, lactate, citrate and amino acids (Yvon
in control cheese and 400W5 and 600W5 cheeses than in and Rijnen 2001), but butanoic acid may also derive from the
400W3 and 600W3 cheeses on day 60 (Table 4). Main alco- hydrolysis of triglycerides (Collins et al. 2003). Their concen-
hols at the end of ripening were ethanol, 2-butanol and 3- tration may decrease in cheese because of ester formation,
methyl-1-butanol, followed by 2-methyl-1-propanol and 1- mainly at late ripening stages and during refrigerated storage.
butanol. The secondary alcohol 2-butanol is formed by the Total esters reached significantly (p <0.05) higher levels in
enzymatic reduction of 2-butanone (Molimard and Spinnler control and 400W5 cheeses than in the other treated cheeses on
1996), while branched-chain primary alcohols derive from the day 60 (Table 4). Ethyl butanoate was the main ester, account-
reduction of the respective aldehydes produced from leucine, ing for more than 60 % of the total esters in control cheese,
isoleucine and valine (Yvon and Rijnen 2001). 3-Methyl-1- followed by ethyl acetate, ethyl hexanoate, ethyl propanoate,
butanol has pleasant fresh cheese notes, while primary alco- 3-methylbutyl acetate and ethyl 3-methylbutanoate. Esters gen-
hols have green, alcoholic notes and secondary alcohols fruity, erally show sweet, fruity and floral notes, have a low percep-
herbaceous notes (Curioni and Bosset 2002). Ethanol declined tion threshold and, by masking the sharpness and bitterness of
slightly during refrigerated storage, while 2-butanol increased other compounds, contribute to a pleasant cheese flavour. In
markedly. On day 240, control cheese and 400W3 and 400W5 particular, ethyl butanoate has been identified as a potent
cheeses showed the maximum (p <0.05) levels of alcohols odorant in many cow and ewes milk cheese varieties
due to their high contents of 2-butanol, which accounted for (Curioni and Bosset 2002). During refrigerated storage, total
more than 60 % of total alcohols. esters increased, the highest (p <0.05) level on day 240 being
Total volatile acids reached their highest (p <0.05) levels at that of control cheese followed by 400W5 cheese. The in-
the end of the ripening period in control cheese and in 400W5 crease in esters levels during refrigerated storage was accom-
and 600W5 cheeses (Table 4). Butanoic, acetic, 2-methyl- panied by a decline in acids levels in control cheese and
propanoic, propanoic and 3-methylbutanoic were the main vol- 400W5 and 600W5 cheeses. However, alcohols did not de-
atile acids on day 60, while acetic, 3-methylbutanoic, butanoic, cline during refrigerated storage in spite of their contribution
decanoic and octanoic had been reported as the major volatile to ester formation. The only cyclic ester detected in cheeses,
acids in 60-day Casar cheese by Delgado et al. (2010). On day γ-butyrolactone, showed the highest (p <0.05) levels in con-
240, butanoic, acetic, 2-methylpropanoic, 3-methylbutanoic trol cheese on day 60, and increased afterwards in all cheeses,
and propanoic remained as the main volatile acids in the present with no significant differences between them on day 240. On
work. They are mostly produced through the microbial the contrary, γ-butyrolactone was not found in Casar cheese
by Delgado et al. (2010), who reported δ-decalactone as the day 240 (Calzada et al. 2013). The characteristic odour notes of
only lactone present. sulphur compounds at low concentrations (cowy, feedy, garlic,
Total sulphur compounds were at low levels at the end of onion, cooked cabbage and cauliflower, mashed potato) make
ripening in HPP and control cheeses, with no significant them essential contributors to the aroma of varieties such as
differences between cheeses (Table 5). During refrigerated Limburger, Camembert, Cheddar, Blue and ewes milk cheeses
storage, their level increased dramatically in control cheese, (Bonnarme et al. 2000; Fernández-García et al. 2004). Howev-
by factors of 82 on day 120 and 467 on day 240, and at a lesser er, because of to their very low perception thresholds, from 0.2
degree in 400W3 and 600W3 cheeses, by factors of 9 and 10, to 20 ppb (Weimer et al. 1999), high concentrations of sulphur
respectively, from day 60 to day 240. The number of sulphur compounds in cheese may result unpleasant.
compounds also increased, from 4 on day 60 to 9 on day 240. Total aromatic compounds were at similar levels in HPP
Dimethyldisulphide, methanethiol and dimethyltrisulphide and control cheeses (Table 5). The main aromatic compounds
were below detection level on day 60 and accounted for were 2-phenylethanol, toluene, styrene, ethylstyrene and phe-
76.6, 7.6 and 4.1 % of total sulphur compounds in 240-day nol. They generally increased during refrigerated storage in all
control cheese, while S -methyl-thioacetate accounted for cheeses. Aromatic compounds probably derive from the ca-
11.3 %, after a 20-fold increase from day 60 to day 240. tabolism of aromatic amino acids by cheese microbiota other
Carbon disulphide, dimethylsulphide and dimethyldisulphide than lactic acid bacteria. The concentration of aromatic amino
were the sulphur compounds detected, at low levels, in a study acids in cheeses ranged from 0.27 to 0.63 mg/g DM on day 60
on La Serena cheese (Carbonell et al. 2002), while 3- and from 1.26 to 2.63 mg/g DM on day 240 (Calzada et al.
methylthio-1-propanol was the only sulphur compound found 2013). Only styrene and phenol had been previously detected
in Casar cheese, at levels which declined during ripening in Casar cheese (Delgado et al. 2010), while most of the
(Delgado et al. 2010). Sulphur compounds may be produced aromatic compounds found in the present work had been
in cheese by Lactococcus, Lactobacillus , Brevibacterium , previously reported for other ewe raw milk cheeses (Carbonell
Micrococcus, Corynebacterium, Pseudomonas and probably et al. 2002; Fernández-García et al. 2002, 2004).
other genera of Gram negative bacteria. Most of them derive Total hydrocarbons levels increased slightly during refriger-
from methanethiol, a compound of putrid and faecal-like ated storage in HPP and control cheeses, with no significant
aroma which is formed from methionine by the action of differences at any of the sampling times (Table 5). Octane,
cystathionine or methionine lyases (Weimer et al. 1999). The hexane and 1,4-pentadiene were the main linear hydrocarbons,
concentration of methionine in cheeses ranged from 0.15 to which are formed through the oxidation of FFAs. Hydrocarbons
0.27 mg/g DM on day 60 and from 0.73 to 1.32 mg/g DM on found in the present work have been found in other ewe raw
Table 5 Levels of total volatile sulphur compounds, aromatic compounds, hydrocarbons and ethers during ripening and refrigerated storage of control
and HPP cheeses
Chemical group Days Control cheese HPP cheesesa
400W3 600W3 400W5 600W5
Sulphur compoundsb 60 14.1±1.6 a 7.3±0.7 a 7.5±1.2 a 10.2±2.6 a 14.6±2.3 a

120 1,158±219 b 50.5±17.3 a 68.8±19.7 a 10.1±3.0 a 12.7±3.0 a
240 6,583±1,005 b 66.6±27.1 a 78.1±21.7 a 30.6±11.0 a 8.0±1.5 a
Aromatic compoundsb 60 48.0±2.8 a 40.8±1.1 a 36.7±0.9 a 42.7±8.7 a 42.8±4.0 a
120 55.8±3.7 a 49.0±1.8 a 41.0±1.0 a 43.2±4.5 a 48.2±5.8 a
240 69.8±9.9 a 63.8±8.1 a 50.7±1.5 a 62.8±4.4 a 54.8±6.0 a
Hydrocarbonsb 60 12.6±0.5 a 18.9±0.8 a 19.1±1.4 a 19.0±2.8 a 19.6±1.8 a
120 16.6±3.2 a 16.3±2.4 a 18.4±1.7 a 24.4±3.3 a 21.0±1.8 a
240 21.8±3.8 a 18.6±3.7 a 17.5±1.8 a 20.8±4.4 a 21.9±1.8 a
Ethersb 60 14.4±2.6 a 11.3±0.4 a 12.2±0.9 a 17.1±3.1 a 17.4±0.9 a
120 20.1±1.2 b 15.5±2.1 ab 11.9±1.7 a 16.1±0.7 ab 21.2±2.6 b
240 23.7±4.2 a 18.1±1.6 a 16.2±0.8 a 24.3±3.3 a 24.6±4.1 a
a
b
Levels are expressed in relative abundances to the internal standard, as mean±SEM of duplicate determinations on two cheese making trials. Means in
the same row at the same sampling date followed by the same letter do not differ significantly
Table 6 Odour characteristics and off-odours during ripening and refrigerated storage of control and HPP cheeses
Characteristic Days Control cheese HPP cheesesa
400W3 600W3 400W5 600W5
Odour intensityb 60 5.92±0.29 a 5.81±0.37 a 5.60±0.28 a 5.88±0.34 a 5.74±0.41 a

120 6.65±0.38 a 5.66±0.42 a 5.79±0.35 a 6.07±0.29 a 6.45±0.36 a
240 7.89±0.33 b 6.42±0.26 a 6.14±0.40 a 6.71±0.35 a 6.53±0.32 a
Odour qualityb 60 6.42±0.38 a 6.23±0.30 a 5.96±0.28 a 6.45±0.34 a 6.18±0.41 a
120 4.59±0.44 a 5.36±0.29 ab 5.77±0.35 ab 6.11±0.42 b 5.94±0.30 ab
240 1.87±0.20 a 4.78±0.47 b 4.90±0.36 b 5.32±0.29 b 5.56±0.43 b
Putrid odourb 60 1.02±0.19 a 1.46±0.15 a 0.91±0.23 a 0.98±0.09 a 1.32±0.26 a
120 2.59±0.31b 1.70±0.28 ab 1.14±0.15 a 1.36±0.23 a 1.19±0.30 a
240 6.82±0.38 b 2.85±0.26 a 2.40±0.37 a 1.97±0.29 a 2.28±0.33 a
Rancid odourb 60 0.53±0.12 a 0.76±0.20 a 0.49±0.08 a 0.38±0.19 a 0.62±0.21 a
120 1.85±0.23 b 1.20±0.17 a 1.06±0.15 a 1.32±0.10 a 0.91±0.22 a
240 3.28±0.34 b 1.59±0.26 a 1.80±0.32 a 1.47±0.29 a 1.38±0.16 a
a
b
Results are expressed as mean±SEM of the scores from a 15-member panel using a 10-point scale on two cheese making trials. Means in the same row
at the same sampling date followed by the same letter do not differ significantly
milk cheeses (Carbonell et al. 2002; Fernández-García et al. such as dimethyldisulphide, S-methyl-thioacetate, methanethiol
2002, 2004), while 3-methylpentane was the only hydrocarbon and dimethyltrisulphide seem to be associated to the appearance
detected in Casar cheese (Delgado et al. 2010). of putrid odour and the loss of odour quality during the refrig-
Total ethers were at low levels on day 60 in HPP and control erated storage of control cheese.
cheeses, with no significant differences between them at that Odour characteristics of 60-day La Serena cheeses were not
time (Table 5). They increased slightly in all cheeses during significantly affected by HPP at 300 or 400 MPa on days 2 or
refrigerated storage, with 600W3 cheese generally exhibiting 50 after manufacture (Arqués et al. 2007). On the contrary, the
the lowest values. Main ethers were 2-butoxyethanol and results obtained in the present work show significant benefi-
2-(2-ethoxy-ethoxy)ethanol. The presence of ethers in smoked cial effects of HPP at 400 or 600 MPa on the odour
goat raw milk cheese was reported by Guillén et al. (2004), who
attributed their origin to pine needle smoke. The possible
sources of ethers in non-smoked cheeses are unknown. 2.0
2m
Odour Characteristics 1.5 C
2m
2m
Function 2 (14.1%)
1.0 4m
Odour intensity of HPP and control cheeses on day 60 did not C
differ (Table 6). During refrigerated storage, odour intensity 0.5 8m 4m 8m
scores rose slightly in HPP cheeses and more markedly in 4m C

8m
0.0
control cheese, which showed a significantly (p <0.05) higher
value than all the HPP cheeses on day 240. The increase in -0.5 2m
8m
odour intensity scores from day 60 to day 240 may be associ- 8m 4m
-1.0 2m 4m
ated to the build-up of volatile compounds such as aldehydes,
alcohols, esters, ketones and probably sulphur compounds too. -1.5
Regarding odour quality, HPP and control cheeses did not differ -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
on day 60. Control cheese received the lowest (p <0.05) score Function 1 (49.9%)
from the panellists on day 240, while HPP cheeses did not differ Fig. 1 Distribution of HPP and control cheeses on the plane defined by
between them. The increasing concentrations of some volatile functions 1 and 2 of the principal component analysis. Each symbol
represents the averaged value of the two batches of cheese. Treatments
compounds in control cheese during refrigerated storage might are as follows: control, C; 400W3, open circle; 600W3, black circle;
be responsible for the observed changes in odour quality. In 400W5, open triangle; 600W5, black triangle. Days after manufacture
particular, the outstanding increases of sulphur compounds are as follows: 60 days, 2m; 120 days, 4m; 240 days, 8m
characteristics of Casar cheese when held under refrigeration showed considerably higher concentrations of propionic acid,
for a prolonged period. BCCAs and SC FFAs than HPP cheeses. Also, major changes
in the volatile fraction of cheeses were recorded during ripen-
Principal Component Analysis ing and refrigerated storage, which were more marked in the
case of control cheese. The excessive formation of some
Principal component analysis was firstly carried out on 57 impact flavour compounds affected negatively the sensory
volatile compounds, selected by the high statistical signifi- characteristics of control cheese, which suffered a dramatic
cance assigned to them in the analysis of variance. The objec- loss of odour quality during refrigerated storage. Sulphur
tive was to discriminate HPP and control cheeses on days 60, compounds appeared as the main causative agents of this
120 and 240 after manufacture, according to pressurization phenomenon. HPP was an effective procedure to reduce the
treatment and cheese age, on the basis of their volatile fraction. formation of carboxylic acids and volatile compounds and to
Function 1, formed by 25 esters, 7 sulphur compounds, 4 control off-odours, thus maintaining cheese sensory charac-
alcohols, 2 aldehydes, 2 ketones, 3 aromatic compounds, 1 teristics throughout a prolonged refrigerated storage period.
acid and 1 hydrocarbon, explained 48.2 % of the variance,
while function 2, formed by four alcohols, four acids, one Acknowledgments This research was funded by AGL 2009-07801
ester and one ketone, explained 13.5 % and function 3, formed project (Ministry of Science and Innovation, Spain). The authors thank
solely by one alcohol and one acid, explained 8.8 %. When the PDO dairy for providing the cheeses and Hyperbaric for HPP treat-
control and HPP cheeses were plotted against functions 1 and ments. J. Calzada is the recipient of a FPI grant (Ministry of Science and
Innovation, Spain).
2, two groups of cheeses were obtained, the 240-day control
cheese and the rest, which included the 60- and 120-day
control cheeses and all the HPP cheeses (data not shown).
A second PCA was performed on the total concentrations References
of the five groups of acids as determined by GC and the total
levels of the nine groups of volatile compounds as determined Arqués, J. A., Garde, S., Fernández-García, E., Gaya, P., & Nuñez, M.
by GC-MS. The objective was to achieve a better discrimina- (2007). Volatile compounds, odor, and aroma of La Serena cheese
tion of cheeses on the basis of their FFAs contents in addition high-pressure treated at two different stages of ripening. Journal of
Dairy Science, 90(8), 3627–3639.
to their volatile fraction. Function 1, formed by all groups of Ávila, M., Garde, S., Fernández-García, E., Medina, M., & Nuñez, M.
acids (LC FFAs, SC FFAs, MC FFAs, BCCAs, acetic + (2006). Effect of high-pressure treatment and a bacteriocin-
propionic, in decreasing order of significance) and seven producing lactic culture on the odor and aroma of Hispánico cheese:
groups of volatile compounds (alcohols, esters, aromatic com- correlation of volatile compounds and sensory analysis. Journal of
Agricultural and Food Chemistry, 54(2), 382–389.
pounds, sulphur compounds, aldehydes, ethers and ketones, in Ávila, M., Calzada, J., Garde, S., & Nuñez, M. (2007). Effect of a
decreasing order of significance), explained 49.9 % of the bacteriocin-producing Lactococcus lactis strain and high-pressure
variance. Function 2 (formed by hydrocarbons and, with a treatment on the esterase activity and free fatty acids in Hispánico
negative coefficient, by volatile acids) explained 14.1 % of the cheese. International Dairy Journal, 17(12), 1415–1423.
Bonnarme, P., Psoni, L., & Spinnler, H. E. (2000). Diversity of L-
variance. Function 3 (formed by volatile acids and, with a methionine catabolism pathways in cheese-ripening bacteria.
negative coefficient, by ketones) explained 12.6 % of the Applied and Environmental Microbiology, 66(12), 5514–5517.
variance. Control and HPP cheeses were plotted against func- Buffa, M., Guamis, B., Pavia, M., & Trujillo, A. J. (2001). Lipolysis in
tions 1 and 2, which may be respectively associated to over- cheese made from raw, pasteurized or high-pressure-treated goats’
milk. International Dairy Journal, 11(3), 175–179.
ripening and to cheese age at the time of HPP treatment. Four Calzada, J., del Olmo, A., Picon, A., Gaya, P., & Nuñez, M. (2013). Using
groups were thus obtained: (1) 240-day control cheese, (2) high-pressure-processing for reduction of proteolysis and prevention
120-day control cheese, (3) 400W3 and 600W3 cheeses at all of over-ripening of raw milk cheese. Food and Bioprocess
sampling times, and (4) 60-day control cheese together with Technology. doi:10.1007/s11947-013-1141-5.
Carbonell, M., Nuñez, M., & Fernández-García, E. (2002). Evolution of the
400W5 and 600W5 cheeses at all sampling times (Fig. 1). volatile components of ewe raw milk La Serena cheese during ripen-
According to the results of this second PCA, cheeses pressur- ing. Correlation with flavor characteristics. Lait, 82(6), 683–698.
ized at 400 or 600 MPa on day 35 after manufacture Chen, L., Cui, J., Ding, Q., Ma, Y., Chen, L., Dong, J., Jiang, T., &
maintained during a prolonged refrigerated storage period Maubois, J. (2012). The effect of yeast species from raw milk in
China on proteolysis and aroma compound formation in
characteristics similar to those of 60-day control cheese. Camembert-type cheese. Food and Bioprocess Technology, 5(6),
2548–2556.
Collins, Y. F., McSweeney, P. L. H., & Wilkinson, M. G. (2003).
Conclusions Lipolysis and free fatty acid catabolism in cheese: a review of
current knowledge. International Dairy Journal, 13(11), 841–866.
Curioni, P. M. G., & Bosset, J. O. (2002). Key odorants in various cheese
Carboxylic acids increased during ripening and refrigerated types as determined by gas chromatography-olfactometry.
storage in all cheeses, in particular in control cheese which International Dairy Journal, 12(12), 959–984.
Delgado, F. J., González-Crespo, J., Ladero, L., Cava, R., & Ramírez, R. Morales, P., Fernández-García, E., & Nuñez, M. (2005). Volatile com-
(2009). Free fatty acids and oxidative changes of a Spanish soft pounds produced in cheese by Pseudomonas strains of dairy origin
cheese (PDO ‘Torta del Casar’) during ripening. International belonging to six different species. Journal of Agricultural and Food
Journal of Food Science and Technology, 44(9), 1721–1728. Chemistry, 53(17), 6835–6843.
Delgado, F. J., González-Crespo, J., Cava, R., García-Parra, J., & Norton, T., & Sun, D.-W. (2008). Recent advances in the use of high
Ramírez, R. (2010). Characterisation by SPME-GC-MS of the pressure as an effective processing technique in the food industry.
volatile profile of a Spanish soft cheese P.D.O. Torta del Casar Food and Bioprocess Technology, 1(1), 2–34.
during ripening. Food Chemistry, 118(1), 182–189. Nuñez, M., Guillén, A. M., Rodríguez-Marín, M. A., Marcilla, A. M.,
Delgado, F. J., González-Crespo, J., Cava, R., & Ramírez, R. (2012). High- Gaya, P., & Medina, M. (1991). Accelerated ripening of ewes’ milk
pressure treatment applied throughout ripening of a goat cheese Manchego cheese: the effect of neutral proteinases. Journal of Dairy
caused minimal changes on free fatty acids content and oxidation in Science, 74(12), 4108–4118.
mature cheese. Dairy Science and Technology, 92(3), 237–248. O’Reilly, C. E., O’Connor, P. M., Kelly, A. L., Beresford, T. P., &
Fernández-García, E., Carbonell, M., & Nuñez, M. (2002). Volatile Murphy, P. M. (2000). Use of hydrostatic pressure for inactivation
fraction and sensory characteristics of Manchego cheese. 1. of microbial contaminants in cheese. Applied and Environmental
Comparison of raw and pasteurized milk cheese. Journal of Dairy Microbiology, 66(11), 4890–4896.
Research, 69(4), 579–593. O’Reilly, C. E., O’Connor, P. M., Murphy, P. M., Kelly, A. L., &
Fernández-García, E., Carbonell, M., Gaya, P., & Nuñez, M. (2004). Beresford, T. P. (2002). Effects of high-pressure treatment on via-
Evolution of the volatile components of ewes raw milk Zamorano bility and autolysis of starter bacteria and proteolysis in Cheddar
cheese. Seasonal variation. International Dairy Journal, 14(8), cheese. International Dairy Journal, 12(11), 915–922.
701–711. Picon, A., Alonso, R., Gaya, P., & Nuñez, M. (2013a). High-pressure
Fernández-García, E., Carbonell, M., Calzada, J., & Nuñez, M. (2006). treatment and freezing of raw goat milk curd for cheese manufac-
Seasonal variation of the free fatty acids contents of Spanish ovine ture: effects on cheese characteristics. Food and Bioprocess
milk cheeses protected by a designation of origin: a comparative Technology, 6(10), 2820–2830.
study. International Dairy Journal, 16(3), 252–261. Picon, A., Alonso, R., van Wely, K. H. M., & Nuñez, M. (2013b).
Fox, P. F., & Wallace, J. M. (1997). Formation of flavor compounds in Microstructural, textural and colour characteristics during ripening
cheese. Advances in Applied Microbiology, 45, 17–85. of Hispánico cheese made using high-pressure-treated ovine milk
Gaya, P., Sánchez, C., Nuñez, M., & Fernández-García, E. (2005). curd. Food and Bioprocess Technology, 6(11), 3056–3067
Proteolysis during ripening of Manchego cheese made from raw Poullet, B., Huertas, M., Sánchez, A., Cáceres, P., & Larriba, G. (1991).
or pasteurized ewes’ milk. Seasonal variation. Journal of Dairy Microbial study of Casar de Cáceres cheese throughout ripening.
Research, 72(3), 287–295. Journal of Dairy Research, 58(2), 231–238.
Guillén, M. D., Ibargoitia, M. L., Sopelana, P., Palencia, G., & Fresno, M. Sablé, S., & Cottenceau, G. (1999). Current knowledge of soft cheeses
(2004). Components detected by means of solid-phase micro- flavor and related compounds. Journal of Agricultural and Food
extraction and gas chromatography/mass spectrometry in the head- Chemistry, 47(12), 4825–4836.
space of artisan fresh goat cheese smoked by traditional methods. Shao, Y. W., & Ramaswamy, H. S. (2011). Clostridium sporogenes-
Journal of Dairy Science, 87(2), 284–299. ATCC 7955 spore destruction kinetics in milk under high pressure
Huppertz, T., Fox, P. F., & Kelly, A. L. (2004). Susceptibility of plasmin and elevated temperature treatment conditions. Food and
and chymosin in Cheddar cheese to inactivation by high pressure. Bioprocess Technology, 4(3), 458–468.
Journal of Dairy Research, 71(4), 496–499. Tejada, L., Gómez, R., Vioque, M., Sánchez, E., Mata, C., & Fernández-
Juan, B., Ferragut, V., Buffa, M., Guamis, B., & Trujillo, A. (2007). Salguero, J. (2000). Effect of freezing and frozen storage on the
Effects of high pressure on proteolytic enzymes in cheese: relation- sensorial characteristics of Los Pedroches, a Spanish ewe milk
ship with the proteolysis of ewe milk cheese. Journal of Dairy cheese. Journal of Sensory Studies, 15(3), 251–262.
Science, 90(5), 2113–2125. Trujillo, A. J., Royo, C., Ferragut, V., & Guamis, B. (1999). Ripening
Lonchamp, J., Barry-Ryan, C., & Devereux, M. (2009). Identification of profiles of goat cheese produced from milk treated with high pres-
volatile quality markers of ready-to-use lettuce and cabbage. Food sure. Journal of Food Science, 64(5), 833–837.
Research International, 42(8), 1077–1086. Van Hekken, D. L., Tunick, M. H., & Park, Y. W. (2005). Effect
Mallia, S., Fernández-García, E., & Bosset, J. O. (2005). Comparison of of frozen storage on the proteolytic and rheological properties
purge and trap and solid phase microextraction techniques for of soft caprine milk cheese. Journal of Dairy Science, 88 (6),
studying the volatile aroma compounds of three European PDO 1966–1972.
hard cheeses. International Dairy Journal, 15(6–9), 741–758. Voigt, D. D., Chevalier, F., Donaghy, J. A., Patterson, M. F., Qian, M. C.,
Malone, A. S., Wick, C., Shellhammer, T. H., & Courtney, P. D. (2003). & Kelly, A. L. (2012). Effect of high-pressure treatment of milk for
High pressure effects on proteolytic and glycolytic enzymes involved cheese manufacture on proteolysis, lipolysis, texture and function-
in cheese manufacturing. Journal of Dairy Science, 86(4), 1139–1146. ality of Cheddar cheese during ripening. Innovative Food Science
McSweeney, P. L. H., & Sousa, M. J. (2000). Biochemical pathways for and Emerging Technologies, 13, 23–30.
the production of flavour compounds in cheese during ripening: a Weimer, B., Seefeldt, K., & Dias, B. (1999). Sulfur metabolism in
review. Lait, 80(3), 293–324. bacteria associated with cheese. Antonie van Leeuwenhoek,
Molimard, P., & Spinnler, H. E. (1996). Compounds involved in the 76(1–4), 247–261.
flavour of surface mold-ripened cheeses: origins and properties. Yvon, M., & Rijnen, L. (2001). Cheese flavour formation by amino acid
Journal of Dairy Science, 79(2), 169–184. catabolism. International Dairy Journal, 11(4–7), 185–201.

Capítulo 7.
Effect of high-pressure-processing on the microbiology,
proteolysis, texture and flavour of Brie cheese during ripening
and refrigerated storage.

153

Fotografía: cuñas de queso Brie control (no presurizado), de arriba abajo: a día 21,30, 60
y 120.
International Dairy Journal 37 (2014) 64e73
Contents lists available at ScienceDirect
International Dairy Journal

journal homepage: www.elsevier.com/locate/idairyj
Effect of high-pressure-processing on the microbiology, proteolysis,

texture and flavour of Brie cheese during ripening and refrigerated
storage
Javier Calzada, Ana del Olmo, Antonia Picon, Pilar Gaya, Manuel Nuñez*
Departamento de Tecnología de Alimentos, INIA, Carretera de La Coruña Km 7, Madrid 28040, Spain
a r t i c l e i n f o a b s t r a c t
Article history: Brie cheeses were high pressure (HP)-treated at 400 or 600 MPa on days 14 or 21 after manufacture to
Received 20 January 2014 prevent over-ripening. Lactic acid bacteria and Penicillium camemberti numbers declined markedly after
Received in revised form HP treatment. In control cheese pH increased 2.0 units from day 21 to day 60, but less than 0.3 units in
3 March 2014
HP-treated cheeses. Cheeses treated at 600 MPa showed the maximum concentrations of residual caseins
Accepted 12 March 2014
during refrigerated storage and control cheese the minimum concentrations. A 7.6-fold increase in hy-
drophobic peptides was recorded from day 21 to day 60 in control cheese and 0.8e1.6-fold increases in
HP-treated cheeses. The maximum aminopeptidase activity was detected in control cheese, the highest
free amino acid concentrations in cheeses treated at 400 MPa. The firmest texture was recorded for
cheeses treated on day 14 at 400 or 600 MPa. HP-treated cheeses showed higher flavour quality scores
than control cheese from day 60 onwards.
Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction P. camemberti, are responsible for the primary proteolysis taking

place during the manufacture and ripening of Camembert cheese
Brie cheese, a surface mould-ripened variety, is usually manu- (Leclercq-Perlat et al., 2004; Trieu-Cuot & Gripon, 1982). The same
factured from pasteurised milk inoculated with a mesophilic lactic causative agents are considered to be involved in primary prote-
starter culture and Penicillium camemberti spores, with Geotrichum olysis of Brie cheese. Afterwards, peptidases from lactic acid bac-
candidum and Kluyveromyces lactis as optional adjunct cultures. The teria (Gripon, Desmazeaud, Le Bars, & Bergère, 1977; Lane & Fox,
colonization of cheese surfaces by P. camemberti mycelium reduces 1997) and Geotrichum candidum (Auberger, Lenoir, & Bergère,
the development of spoilage microorganisms and modulates the 1997) carry out further hydrolysis of the products resulting from
external appearance (Lessard, Bélanger, St-Gelais, & Labrie, 2012). primary proteolysis to free amino acids. Finally, the catabolism of
Most of the scientific knowledge on the ripening process of surface free amino acids by lactic acid bacteria and other microorganisms
mould-ripened cheeses has been obtained on Camembert cheese (Yvon & Rijnen, 2001) results in the formation of the volatile potent
(Spinnler & Gripon, 2004), more suitable for experimental work than odorants (Kubícková & Grosch, 1997) and the non-volatile flavour
Brie cheese because of its smaller size. Growth of P. camemberti on the compounds (Kubícková & Grosch, 1998) present in surface mould-
surface of Camembert cheese is accompanied by the production of ripened cheeses.
extracellular enzymes and the consumption of lactic acid, with a A characteristic phenomenon taking place in surface mould-
concomitant rise in pH value (Desmazeaud, Gripon, Le Bars, & ripened cheeses is the migration of calcium and phosphate from
Bergère, 1976). Although phenomena occurring during the ripening the cheese interior to the surface, associated with the pH gradient
process of Brie and Camembert cheeses are similar, the lesser surface created by lactic acid consumption by moulds (Le Graet, Lepienne,
to volume ratio of the former variety may result in some differences. Brule, & Ducruet, 1983). The rise in pH value, firstly at the cheese
Milk native plasmin, clotting enzymes and proteinases pro- surface and then at the interior, the reduction of the concentration
duced by microorganisms, mostly lactic acid bacteria and of calcium phosphate, and the proteolysis, bring about the desirable
changes in the texture of surface mould-ripened cheeses (Abraham,
Cachon, Colas, Feron, & De Coninck, 2007; Vassal, Monnet, Le Bars,
* Corresponding author. Tel.: þ34 913476799. Roux, & Gripon, 1986), which becomes softer and mature to the
E-mail addresses: nunez@inia.es, mng1950@gmail.com (M. Nuñez). core during ripening. However, the pH increase throughout
http://dx.doi.org/10.1016/j.idairyj.2014.03.002
0958-6946/Ó 2014 Elsevier Ltd. All rights reserved.
J. Calzada et al. / International Dairy Journal 37 (2014) 64e73 65
ripening of Brie and Camembert cheeses creates conditions ripened at 12 C and 90% RH. On day 14 after manufacture, cheeses
favourable for the growth of undesirable bacteria such as Escher- were cut into 200 g wedges that were wrapped in food-grade
ichia coli, Yersinia enterocolitica, Staphylococcus aureus, Bacillus ce- metalised paper. Ripening proceeded under the same conditions
reus and Listeria monocytogenes (Maisnier-Patin, Deschamps, Tatini, until day 21 and afterwards wedges were held at 4 C until day 150.
& Richard, 1992; Nooitgedagt & Hartog, 1988). Wedges of cheeses from both trials were pressurised for 5 min at
Chemical reactions associated with cheese ripening continue 400 or 600 MPa on days 14 or 21 after manufacture. They were
during refrigerated storage at retail and at homes. Therefore, cheese respectively coded as 400W2, 600W2, 400W3, and 600W3. Before
at the time of consumption may be of a stronger or different flavour HPP treatments, cheese wedges were vacuum packed in BB325
than the manufacturer intended (Wick, Nienaber, Anggraeni, plastic film bags (Cryovac, Barcelona, Spain). A 120-L capacity
Shellhammer, & Courtney, 2004). Brie cheese is particularly prone isostatic press (Hiperbaric, Burgos, Spain) was used for HPP treat-
to over-ripening not long after the optimal date of consumption, ments. Times to reach 400 and 600 MPa were 1.79 and 2.61 min,
because of its pH value close to neutrality and the presence of respectively, and depressurisation times, 6 and 8 s. The initial
potent enzymes of fungal origin. Another undesirable event which temperature of the water used as transmitting fluid was 9 C and
may occur in cheese if ripening or refrigerated storage are pro- remained under 14 C during the process. After treatments, cheese
longed in excess is the formation of biogenic amines, a group of wedges were taken out of the plastic film bags. Ripening and
compounds of notable public health significance (Silla-Santos, refrigerated storage conditions of HPP cheese wedges were the
1996). In surface mould-ripened varieties such as Brie cheese, the same as for control cheese.
build-up of biogenic amines would be favoured by their high con-
centrations of free amino acids, resulting from extensive 2.2. Microbiological analysis and chemical determinations
proteolysis.
High pressure processing (HPP) efficiently inactivates microbial Representative 10 g samples of cheese wedges, including the
contaminants in cheese (O’Reilly, O’Connor, Kelly, Beresford, & rind, were used for microbiological analysis. Total viable counts,
Murphy, 2000). It also affects the activity of cheese enzymes, lactic acid bacteria, staphylococci, L. monocytogenes, Gram-negative
which may be increased at pressure levels of 400 MPa or lower and bacteria, and coliforms were determined as previously described
is generally decreased if pressures above 500 MPa are applied (Arqués, Garde, Gaya, Medina, & Nuñez, 2006). P. camemberti and
(Huppertz, Fox, & Kelly, 2004; O’Reilly, O’Connor, Murphy, Kelly, & other fungi were enumerated in duplicate on plates of chloram-
Beresford, 2002; Wick et al., 2004). In addition, HPP has shown to phenicol glucose agar (Scharlab, Barcelona, Spain), incubated for 3e
be useful in preventing the formation of biogenic amines during 7 days at 25 C.
ripening and refrigerated storage of cheese (Calzada, del Olmo, Representative samples for chemical determinations were pre-
Picon, Gaya, & Nuñez, 2013a). pared by removing a 2-mm thick layer of rind from all the surfaces of
The effect of HPP on the characteristics of surface mould- cheese wedges at 4 C and mincing the rest of the wedge. Dry matter
ripened cheeses has been studied on Camembert and Paillardin, a (DM) content and pH value were determined in triplicate as previ-
Belgian variety. Enhanced proteolysis was recorded for Camembert ously described (Garde, Tomillo, Gaya, Medina, & Nuñez, 2002).
cheese treated at 50 MPa for 4 h, accompanied by an increase in pH Proteins were analysed in triplicate by capillary gel electropho-
value up to 0.50 units (Kolakowski, Reps, & Babuchowski, 1998). resis on an automated PACE/MDQ CE controlled by 32 Karat Soft-
Also, Paillardin cheese pressurised at 50 MPa for 8 h exhibited ware (Beckman Instruments España S.A., Madrid, Spain), according
higher soluble N contents and slightly increased pH values than to a previously described method (Calzada, del Olmo, Picon, Gaya, &
control cheese (Messens, Foubert, Dewettinck, & Huyghebaert, Nuñez, 2013b). Hydrophilic and hydrophobic peptides in the water-
2001). However, there is no information on the effect of HPP on soluble fraction were extracted in duplicate and determined by
the ripening of surface mould-ripened cheeses under the condi- reverse phase-high pressure liquid chromatography (RP-HPLC) at
tions currently used in the food industry, with higher pressures and 214 and 280 nm as described by Lau, Barbano, and Rasmussen
shorter process times. The objective of the present work was to (1991). Free amino acids (FAAs) and biogenic amines, simulta-
investigate the effect of HPP at 400 or 600 MPa for 5 min, applied on neously extracted in duplicate as described by Krause, Bockhardt,
days 14 or 21 after manufacture, as a procedure to control over- Neckermann, Henle, and Klostermeyer (1995), were determined
ripening of Brie cheese. The microbiology, proteolysis, formation by RP-HPLC and quantified as previously described (Calzada et al.,
of biogenic amines, texture and flavour of high pressure (HP)- 2013b). Overall proteolysis was determined by the o-phthaldialde-
treated Brie cheeses during ripening and refrigerated storage were hyde (OPA) method in duplicate and expressed as the absorbance at
compared with the characteristics of control cheese. 340 nm (Church, Swaisgood, Porter, & Catignani, 1983).
Aminopeptidase activity released into the cheese was deter-
2. Materials and methods mined in duplicate with leucine p-nitroanilide (Leu-p-NA) and
lysine p-nitroanilide (Lys-p-NA) as substrates and expressed in
2.1. Manufacture and high pressure processing of Brie cheese nmol p-nitroaniline per min per g (Garde et al., 2002).
Brie cheese was made from 2100 L pasteurised (73 C for 20 s) 2.3. Texture determinations, sensory analysis and statistical
milk, in duplicate trials carried out on consecutive days at a dairy analysis
factory in Central Spain. Mesophilic lactic culture (Flora Danica, Chr.
Hansen S.L., Tres Cantos, Spain), P. camemberti PC V5 (Sacco, Fracturability (force at fracture, N), elasticity (Young’s modulus,
Cadorago, Italy), and 0.02% CaCl2 were added to milk at 37 C, which N mm2), and firmness (energy, area under the force-distance
was coagulated with equal amounts of animal rennet (Ha-Bo, Chr. curve up to 75% compression, N m) were determined by uniaxial
Hansen) and microbial rennet (Rennilase, Novo Nordisk, Bagsvaerd, compression testing on six representative samples per cheese
Denmark) in 20 min. The curd was cut into 15-mm cubes, stirred for wedge after removing the rind, using an Instron Compression
15 min, and transferred into circular moulds 32 cm in diameter. Tester 4301 (Instron, High Wycombe, UK), equipped with Instron
Cheeses (2.7 kg in weight, approximately 120 per vat) were held at Bluehill 2.3 software.
28 C for 22 h, and turned over after 1, 2 and 8 h in the moulds. They A trained 16-member panel carried out the evaluation of flavour
were salted by immersion in brine on day 1 after manufacture, and intensity, flavour quality (preference) and five flavour attributes
66 J. Calzada et al. / International Dairy Journal 37 (2014) 64e73
(acid, bitter, salty, sweet and umami) scoring on a 10-point scale as

previously described (Picon et al., 2010). Panellists were also asked
to report on the presence of off-flavours.
Statistical analysis of results was carried out by means of SPSS
Win 14.0 program (SPSS Inc. Chicago, IL). Two-way analysis of
variance included the main effects treatment (four experimental
cheeses and one control cheese) and cheese age, as well as the
“treatment age” interaction. Means were compared using Tukey’s
test, with the significance assigned at P < 0.05.
3. Results and discussion
3.1. Microbial groups
Lactic acid bacteria counts in Brie cheese were 8.99 log cfu g1 on
day 1, a similar population to that of total viable counts. They
decreased by 1.65, 3.71, 0.96 and 3.45 log cfu g1 respectively in
400W2, 600W2, 400W3 and 600W3 cheeses immediately after HPP
(Fig. 1). Further decrease in counts during ripening and refrigerated
storage occurred only in 600W2 cheese, which might be associated
with the particular physiological status of sub-lethally injured cells
of lactic acid bacteria in this cheese, which are unable to recover and
grow on a selective medium. The recorded inactivation of lactic acid
bacteria was in agreement with the results obtained for other cheese
varieties submitted to HPP (Calzada et al., 2013a; Voigt, Chevalier,
Qian, & Kelly, 2010; Wick et al., 2004).
Counts of P. camemberti declined significantly (P < 0.05) imme-
diately after HPP (Fig. 1), by 2.25 and 3.29 log cfu g1 in 400W2 and
400W3 cheeses and by at least 5 log cfu g1, to counts below the
detection limit, in 600W2 and 600W3 cheeses. This does not
exclude the presence of sublethally injured cells of P. camemberti on
the cheese rind. There is no information on the resistance of
P. camemberti to high pressure. Treatment at 400 MPa lowered
counts of Penicillium roqueforti in cheese slurry by more than 2
log cfu g1 at 10 C and by 6 log cfu g1 at 20 C (O’Reilly et al., 2000),
and in blue-veined cheese by 0.74e5.33 log cfu g1 depending on
cheese age at treatment (Calzada et al., 2013b; Voigt et al., 2010).
Fig. 1. Counts of the main microbial groups (A, lactic acid bacteria; B, P. camemberti)
Those results point to the influence of the strain, its physiological
during ripening and refrigerated storage of Brie control cheese (-) and cheeses HP-
status, growth substrate and HPP temperature on its baroresistance. treated at 400 MPa on day 14 (,), 600 MPa on day 14 ( ), 400 MPa on day 21 ( )
Non-Penicillium fungi, mostly yeasts, were not detected until day 90, or 600 MPa on day 21 ( ). Bars indicate standard error of the means.
at levels ranging from 6.52 to 6.92 log cfu g1 in control cheese and
from 2.30 to 3.79 log cfu g1 in cheeses treated at 400 MPa (data not
shown). Prior to day 90, the detection of other fungi was most 400 MPa than in control cheese, with a difference in DM content of
probably hindered by the higher counts of P. camemberti, colonies of 6.1% on day 28 (Saldo, McSweeney, Sendra, Kelly, & Guamis, 2002),
which spread on chloramphenicol glucose agar plates. and in ovine milk cheese treated at 500 MPa than in control cheese,
Microbial contaminants tested for such as staphylococci, with a difference in DM content of 3.3% on day 30 (Juan, Ferragut,
L. monocytogenes, Gram-negative bacteria and coliforms were not Buffa, Guamis, & Trujillo, 2007). These authors ascribed such dif-
detected throughout ripening or refrigerated storage of Brie cheese, ferences in moisture loss to changes in the structure of the para-
indicating satisfactory hygiene conditions during manufacture. caseinate network caused by HPP, which would have increased
water retention in the cheese. Blue-veined cheeses are more
3.2. Dry matter and pH demineralised than hard and semi-hard cheeses, which might in-
fluence their water retention ability and DM content after HPP.
Dry matter (DM) content of Brie cheese was 44.31% on day 1 Thus, the DM content of blue-veined cheese treated at 600 MPa was
(data not shown) and did not vary with HPP immediately after 2.7% higher than that of control cheese on day 28 after HPP (Voigt
treatments (Fig. 2). On day 21, at the end of ripening, there were no et al., 2010), and the DM contents of HP-treated ovine milk blue-
significant (P < 0.05) differences in DM content between cheeses veined cheeses was similar to that of control cheese during
but, during refrigerated storage, DM decreased in control cheese ripening and refrigerated storage (Calzada et al., 2013b). In the case
and increased in HPP cheeses. The intense proteolysis of control of Brie cheese, curd demineralisation may also affect water reten-
cheese would render it highly hygroscopic as a consequence of the tion ability and DM content after HP treatment.
increase in the number of water-binding sites, thus promoting The pH value declined to 4.81 on day 1 (data not shown), and
absorption of moisture from the atmosphere of the storage cham- afterwards rose to 5.35 by day 14 in control cheese (Fig. 2). It was
ber, with a resultant decrease in DM content. The increase in the not significantly affected by HPP immediately after treatment,
DM content of HPP cheeses may be directly associated with mois- contrarily to the 0.6 pH units increase recorded for caprine milk
ture loss during refrigerated storage. In contrast, previous studies cheese after treatment on day 4 at 400 MPa (Saldo et al., 2002),
have reported less moisture loss in caprine milk cheese treated at which was explained by the release of colloidal calcium phosphate
were 55, 60, 64 and 83% of those on day 1 (Table 1). During
refrigerated storage, there was a drastic hydrolysis in aS-, b- and k-
caseins, which resulted in concentrations on day 120 of only 1, 3
and 4% of those on day 1, while para-k-casein concentration was
45%. In addition to the activity of rennet enzymes and proteinases
from the lactic cultures, activity of the heat-tolerant protease milk
plasmin persists in pasteurised milk cheeses (Lane & Fox, 1997).
Pepsin and chymosin from rennet show acidic optimum pH values,
gradually losing their activity at pH values close to and above 6.0.
They would be more active during ripening of control Brie cheese,
with pH values ranging from 4.81 to 5.55, than during refrigerated
storage. Plasmin hydrolyses aS- and b-caseins but not k-casein,
which is resistant to the enzyme (Eigel, 1977), and has a slightly
alkaline optimum pH value (Visser, 1981). Therefore, plasmin
would be more active during refrigerated storage, with pH values
ranging from 6.30 to 7.87. P. camemberti aspartyl proteinase (Trieu-
Cuot, Archieri-Haze, & Gripon, 1982a) and metalloproteinase
(Trieu-Cuot, Archieri-Haze, & Gripon, 1982b) are both active on aS-
and b-caseins; values of pH close to and above 6.0 enhance the
activity of the metalloproteinase and reduce the activity of the
aspartyl proteinase. The considerable decline in the concentrations
of aS- and b-caseins from day 30 onwards (Table 1) as the pH rose
from 5.55 on day 21 to 7.55 on day 60 (Fig. 2) can be explained by
the enhancement of the activity of plasmin and the P. camemberti
metalloproteinase at the higher pH values. The effect on k- and
para-k-caseins was less marked (Table 1), since plasmin is not
active on these proteins and the metalloproteinase probably has a
higher affinity for aS- and b-caseins as substrates, which might
retard its action on k- and para-k-caseins. Concentrations of aS-, b-,
k- and para-k-caseins were considerably higher in control Brie
cheese on day 90 than in control blue-veined cheese of the same
age, in which b- and para-k-caseins were 1.22 and 5.94 mg per g of
cheese DM, respectively, while aS- and k-caseins were below the
detection level (Calzada et al., 2013b).
In HPP cheeses, the concentrations of aS-, b-, k- and para-k-ca-
seins at the end of ripening, on day 21, were similar to those of
control cheese (Table 1). Afterwards, a slower rate of proteolysis
Fig. 2. Dry matter content (A) and pH value (B) during ripening and refrigerated was generally recorded for HPP cheeses than for control cheese and,
storage of Brie control cheese (-) and cheeses HP-treated at 400 MPa on day 14 (,), among HPP cheeses, for 600 MPa cheeses than for 400 MPa
600 MPa on day 14 ( ), 400 MPa on day 21 ( ) or 600 MPa on day 21 ( ). Bars indicate cheeses. The lower pH values of HPP cheeses were less favourable
standard error of the means. for the activity of plasmin and the P. camemberti metalloproteinase
during refrigerated storage than the pH values of control cheese.
The effect of chymosin and lactic acid bacteria proteinases would be
into the aqueous phase. The higher demineralisation of the curd considerably diminished in HPP cheeses in comparison with con-
during the manufacture of Brie and other soft cheeses in compar- trol cheese, since these enzymes are gradually inactivated at
ison with hard and semi-hard varieties may limit the release of pressures above 400 MPa (Huppertz et al., 2004; Malone, Wick,
colloidal calcium phosphate, thus reducing the effect of HPP on Shellhammer, & Courtney, 2003). However, plasmin activity is
their pH value. In control cheese, the pH rose sharply from day 21 to stable at 600 MPa (Huppertz et al., 2004; Malone et al., 2003) and
day 60, by 2.0 pH units, while in HPP cheeses it increased during the would remain active in HPP cheeses. The action of pepsin and
same period by less than 0.3 pH units. The markedly lower P. camemberti proteinases on the caseins in HPP cheeses is more
P. camemberti counts in HPP cheeses than in control cheese (Fig. 1), difficult to ascertain since there is no available information, to our
which may be associated with lower lactic acid consumption, seem knowledge, on the baroresistance of these enzymes. The proteol-
to be responsible for the more moderate increase in their pH values. ysis of k- and para-k-caseins in 600 MPa cheeses cannot be ascribed
Significant (P < 0.05) differences in pH between HPP and control to chymosin and lactic acid bacteria proteinases, which are inacti-
cheeses persisted until the end of the refrigerated storage. A similar vated at this pressure level, or to plasmin, which is not active on
phenomenon was recorded for blue-veined cheeses, although the these casein fractions. Therefore, the proteolysis of k- and para-k-
differences between the pH values of control and HPP cheeses did caseins in 600 MPa cheeses must be attributed to pepsin and/or the
not exceed 0.8 pH units at any of the sampling dates (Calzada et al., P. camemberti metalloproteinase, which implies residual activity of
2013b; Voigt et al., 2010). at least one of these enzymes after HPP at 600 MPa. The hydrolysis
of aS- and b-caseins in 600 MPa cheeses can be related to the action
3.3. Hydrolysis of proteins of plasmin together with pepsin and/or the P. camemberti metal-
loproteinase. These results on proteolysis in Brie cheese were
Concentrations of aS-, b-, k- and para-k-caseins in control Brie difficult to compare with those of Kolakowski et al. (1998) for
cheese were 170.48, 176.28, 43.94 and 32.94 mg per g of cheese DM, Camembert cheese, due to the lower pressures (50e200 MPa) and
respectively, on day 1 (data not shown). On day 21, concentrations longer treatment time (4 h) applied by these authors.
Table 1
Levels of caseins during ripening and refrigerated storage of control Brie cheese and cheeses HP-treated at 400 or 600 MPa after 14 days (W2) or 21 days (W3) of ripening.a
Casein Days Control cheese 400W2 600W2 400W3 600W3
aS-Casein 21 92.88 5.03a 95.31 3.68a 106.76 4.32a 98.29 4.67a 99.84 3.37a
30 87.22 3.32ab 87.04 4.60ab 98.46 2.42b 85.17 2.20a 84.69 1.53a
60 68.30 5.12ab 54.49 1.61a 84.78 7.48b 79.81 5.79b 80.87 6.80b
90 21.43 4.00a 21.27 4.32a 56.70 2.81b 15.86 1.63a 58.79 4.17b
120 1.73 0.27a 12.09 2.02ab 25.40 6.45b 5.24 0.38a 17.16 0.82ab
b-Casein 21 105.60 5.39a 108.62 3.18a 112.05 6.27a 118.80 3.89a 113.70 8.41a
30 101.64 3.44a 104.08 4.31a 108.20 3.66a 95.35 2.85a 99.71 1.10a
60 62.38 5.53a 72.66 2.82ab 100.99 4.36c 88.32 3.60bc 92.05 10.14bc
90 37.38 6.19a 41.71 5.24a 75.58 7.25b 26.04 3.77a 84.89 2.83b
120 5.05 0.78a 36.42 6.65bc 54.51 9.82c 19.47 0.88ab 33.68 2.05bc
k-Casein 21 27.89 1.42a 28.87 3.29a 31.19 2.83a 30.03 1.51a 29.96 0.82a
30 28.27 1.02a 27.90 1.31a 28.35 0.95a 25.82 0.75a 25.65 0.61a
60 21.39 1.96a 21.00 1.75a 26.69 2.24a 24.51 1.53a 22.60 2.27a
90 18.15 2.84ab 14.89 3.04ab 19.94 1.48b 9.94 1.61a 21.03 1.52b
120 1.94 0.29a 9.60 1.58b 13.29 2.73b 2.92 0.46a 3.40 0.60a
para-k-Casein 21 27.26 1.57a 28.45 2.17a 30.20 2.46a 29.63 1.35a 28.95 0.93a
30 28.18 0.72ab 29.50 1.17b 27.71 0.94ab 28.04 0.69ab 25.57 0.42a
60 27.91 0.77a 24.85 1.07a 28.03 0.78a 27.06 1.25a 27.11 2.34a
90 28.63 1.44b 23.48 1.60ab 27.54 0.53b 21.43 4.30a 25.99 1.04ab
120 14.91 2.21a 19.58 1.85a 20.99 3.14a 14.74 1.09a 18.61 2.25a
a
Caseins are expressed in mg per g of cheese DM and presented as mean SE (n ¼ 6) of triplicate determinations in two cheese-making experiments.
Means on the same row followed by different superscript letters differ significantly (P < 0.05).
Concentrations of a-lactalbumin, b-lactoglobulin, serum albu- in control Brie cheese on day 90 than in control blue-veined cheese
min and g-caseins in control Brie cheese were 3.53, 15.37, 0.22 and of the same age (Calzada et al., 2013b).
13.44 mg per g of cheese DM, respectively, on day 1 (data not In HPP cheeses, the breakdown of a-lactalbumin and b-lacto-
shown). By the end of the ripening period, a-lactalbumin had not globulin evolved similarly to control cheese, with no significant
varied in control cheese, b-lactoglobulin had declined by 32%, differences during ripening and few significant differences during
serum albumin was no longer detected and g-caseins had increased refrigerated storage (Table 2). The concentration of g-caseins did
by 71% in comparison with control cheese concentrations on day 1 not differ significantly (P < 0.05) between HPP and control cheeses
(Table 2). During refrigerated storage, a-lactalbumin and b-lacto- on day 21, and showed few significant differences between cheeses
globulin concentrations in control cheese markedly decreased from from day 30 to day 90. However, from day 90 onwards, there was a
day 60 onwards and from day 90 onwards, respectively, which can less pronounced decline in the concentration of g-caseins in HPP
be related to the high pH values of cheese (Fig. 2), which are cheeses than in control cheese (Table 2), probably due to the lower
favourable for proteolysis. There was a strong increase in g-caseins pH values of the former cheeses, which are less favourable for the
from day 30 to day 60 (Table 2), which can be associated to the activity of plasmin and P. camemberti metalloproteinase.
decline in b-casein concentration (Table 1) during the same period,
favoured by the high pH values of control cheese. Afterwards, there 3.4. Levels of peptides, free amino acids and biogenic amines
was a considerable decrease of g-caseins from day 90 to day 120, as
they were broken down to smaller size peptides. Concentrations of The formation of hydrophilic peptides was enhanced in Brie
a-lactalbumin, b-lactoglobulin and g-caseins reached higher levels cheeses HP-treated on day 14, which showed significantly
Table 2
Levels of whey proteins and g-caseins during ripening and refrigerated storage of control Brie cheese and cheeses HP-treated at 400 or 600 MPa after 14 days (W2) or 21 days
(W3) of ripening.a
Protein Days Control cheese 400W2 600W2 400W3 600W3
a-Lactalbumin 21 4.25 0.26a 4.38 0.32a 3.78 0.23a 4.57 0.24a 4.28 0.20a
30 5.12 0.41b 4.89 0.17b 3.46 0.37a 4.84 0.33b 4.13 0.11ab
60 5.73 0.84a 5.72 0.97a 4.66 0.92a 4.89 0.18a 4.28 0.51a
90 1.33 0.27b 0.66 0.12ab 0.52 0.08a 0.52 0.15a 1.03 0.28ab
120 0.87 0.15a 0.69 0.11a 0.53 0.16a 0.66 0.07a 0.80 0.24a
b-Lactoglobulin 21 10.38 0.70a 11.67 0.57a 12.30 0.40a 11.02 0.73a 11.26 0.54a
30 10.95 0.51a 14.42 1.00b 13.16 0.44ab 14.49 0.40b 11.83 0.38a
60 11.37 0.37a 13.22 0.78a 14.03 1.31a 14.04 0.57a 12.16 0.80a
90 10.25 0.60a 10.07 0.43a 13.04 0.53a 10.10 1.18a 10.77 0.49a
120 3.82 0.43a 8.32 1.62a 6.80 1.16a 7.47 0.56a 5.09 1.23a
g-Caseins 21 23.05 1.48a 24.90 1.09a 24.38 1.32a 24.58 0.81a 22.76 0.93a
30 28.43 1.31ab 31.65 1.19b 27.02 1.23ab 31.89 1.17b 24.97 0.67a
60 64.80 3.31b 48.74 3.48a 56.13 4.02ab 51.36 2.73ab 50.52 3.49ab
90 63.49 4.78a 53.91 2.41a 63.54 1.38a 49.39 4.37a 60.07 3.85a
120 20.91 1.31a 47.60 2.01b 44.35 5.12b 38.80 1.46b 36.31 2.62b
a
Proteins are expressed in mg per g of cheese DM and presented as mean SE (n ¼ 6) of triplicate determinations in two cheese-making experiments.
Table 3
Levels of hydrophilic peptides, hydrophobic peptides and the ratio of hydrophobic peptides/hydrophilic peptides during ripening and refrigerated storage of control Brie
cheese and cheeses HP-treated at 400 or 600 MPa after 14 days (W2) or 21 days (W3) of ripening.a
Peptides Days Control cheese 400W2 600W2 400W3 600W3

a b b a
Hydrophilic 21 11.40 0.17 18.56 0.72 17.15 0.68 13.31 0.61 12.64 0.49a
30 15.70 0.24a 27.78 2.61b 22.57 0.19ab 17.56 0.34a 17.79 0.16a
60 23.99 1.36a 37.09 1.55bc 32.25 0.98b 44.48 1.35d 40.12 2.68cd
90 43.32 3.56a 47.42 0.96a 48.67 3.35a 42.56 4.14a 42.46 0.96a
120 47.35 1.00b 33.56 1.73a 37.22 3.14ab 33.94 0.41a 37.45 3.07ab
Hydrophobic 21 4.02 0.19a 4.99 0.45a 4.73 0.51a 3.80 0.22a 3.77 0.30a
30 6.28 0.05a 6.16 0.88a 4.66 0.10a 4.57 0.19a 4.62 0.11a
60 30.34 3.16b 6.04 0.45a 3.72 0.58a 5.66 0.52a 5.88 1.17a
90 72.41 3.75b 7.81 0.75a 7.56 0.94a 6.14 0.75a 5.93 0.24a
120 36.90 1.13b 5.81 0.53a 6.66 1.50a 5.02 0.08a 5.61 0.20a
Ratio 21 0.352 0.011b 0.269 0.014a 0.276 0.019a 0.287 0.008a 0.297 0.012a
30 0.401 0.009c 0.221 0.005a 0.207 0.006a 0.260 0.006b 0.260 0.008b
60 1.252 0.077b 0.162 0.006a 0.114 0.015a 0.126 0.008a 0.143 0.020a
90 1.684 0.052b 0.166 0.019a 0.154 0.006a 0.144 0.003a 0.140 0.004a
120 0.779 0.007b 0.172 0.008a 0.172 0.021a 0.148 0.004a 0.152 0.009a
a
Peptides (determined at 280 nm) are expressed as chromatogram area units mg1 of cheese DM and presented as mean SE (n ¼ 4) of duplicate determinations in two
cheese-making experiments.
(P < 0.05) higher levels than control cheese on day 21 (Table 3). HP- responsible for the patterns of hydrophobic peptides in the two
treatment of cheeses on day 21 had no significant effect on the level cheese varieties.
of hydrophilic peptides immediately after HPP. However, all HPP The ratio of hydrophobic peptides/hydrophilic peptides was
cheeses at day 60 had significantly (P < 0.05) higher levels of hy- significantly (P < 0.05) higher in control cheese than in HPP cheeses
drophilic peptides than control cheese. From day 60 onwards, there at all sampling times (Table 3). This ratio has been related to cheese
was a build-up of hydrophilic peptides in control Brie cheese, bitterness (Gómez, Garde, Gaya, Medina, & Nuñez, 1997). In control
probably enhanced by its high pH values, while these tended to cheese, it increased dramatically from 0.352 on day 21 to 1.684 on
decrease in HPP cheeses. A much less marked increase in hydro- day 90, while in HPP cheeses it declined gradually from 0.269 e
philic peptides was recorded during ripening of HPP and control 0.297 on day 21 to 0.140e0.166 on day 90. This ratio varied to a
blue-veined cheeses (Calzada et al., 2013b). The different pH optima lesser extent during ripening of blue-veined cheese, from 0.31 on
of the proteolytic systems of P. camemberti and P. roqueforti (Gripon day 21 to 0.20 on day 90 in control cheese and from 0.29 on day 21
et al., 1977) may partly explain the differences in the formation of to 0.20e0.32 on day 90 in HPP cheeses (Calzada et al., 2013b).
hydrophilic peptides in these two cheese varieties. Aminopeptidase activity on Leu-p-NA as substrate was
The levels of hydrophobic peptides did not show significant 1.79 nmol p-nitroaniline min1 g1 on day 1 (data not shown) and
differences between cheeses at the end of ripening, on day 21 did not vary immediately after HPP of Brie cheeses. Higher enzy-
(Table 3). During refrigerated storage, the level of hydrophobic matic activity was generally recorded for control, 400W3 and
peptides remained fairly unchanged in HPP cheeses. However, a 600W3 cheeses than for 400W2 and 600W2 cheeses from day 21 to
drastic increase occurred in control cheese from day 30 onwards, day 90 (Table 4). Aminopeptidase activity on Lys-p-NA as substrate
up to a value of 72.41 on day 90, approximately 10-fold the level in was 2.74 nmol p-nitroaniline min1 g1 on day 1 (data not shown)
the respective HPP cheeses. This increase can be associated to the and declined immediately after HPP in 400W2 and 600W2 cheeses,
decline in the concentration of aS- and b-caseins (Table 1). In which showed generally lower activity than the rest until day 90
contrast, hydrophobic peptides hardly varied during ripening of (Table 4). Aminopeptidase activities were not correlated with the
HPP and control blue-veined cheese (Calzada et al., 2013b). As for populations of viable lactic acid bacteria or P. camemberti in HPP
the hydrophilic peptides, differences in the pH optima of the pro- and control cheeses (Fig. 1). They did not correlate either with the
teolytic systems of P. camemberti and P. roqueforti might be declines in the counts of these microbial groups caused by HPP,
Table 4
Aminopeptidase activity on Leu-p-NA and Lys-p-NA during ripening and refrigerated storage of control Brie cheese and cheeses HP-treated at 400 or 600 MPa after 14 days
(W2) or 21 days (W3) of ripening.a
Substrate Days Control cheese 400W2 600W2 400W3 600W3
Leu-p-NA 21 3.43 0.09b 2.19 0.15a 2.33 0.26a 4.29 0.28b 3.53 0.32b
30 4.16 0.31c 2.28 0.15a 2.06 0.05a 3.97 0.41bc 3.05 0.15ab
60 7.36 1.06b 3.56 0.34a 3.46 0.26a 5.42 0.28ab 3.89 0.49a
90 5.05 0.10c 2.07 0.34a 1.09 0.16a 4.39 0.08bc 3.43 0.31b
120 7.88 1.06b 1.62 0.27a 1.56 0.34a 1.50 0.05a 1.74 0.18a
Lys-p-NA 21 4.34 0.12b 2.83 0.29a 3.04 0.30a 4.26 0.34b 4.07 0.17b
30 5.13 0.80b 3.17 0.42a 3.06 0.09a 4.03 0.15ab 4.15 0.06ab
60 8.26 1.33b 4.08 0.06a 3.75 0.34a 6.43 0.19ab 4.10 0.49a
90 5.71 0.53c 1.76 0.04a 1.31 0.18a 4.89 0.32c 3.21 0.16b
120 7.93 0.64b 2.04 0.51a 1.81 0.39a 1.68 0.02a 1.48 0.14a
a
Aminopeptidase activities are expressed in nmol p-nitroaniline min1 g1 and presented as mean SE (n ¼ 4) of duplicate determinations in two cheese-making ex-
periments.
Table 5
Levels of free amino acids (FAAs) and overall proteolysis during ripening and refrigerated storage of control cheese and cheeses HP-treated at 400 or 600 MPa after 14 days
(W2) or 21 days (W3) of ripening.a
Variable Days Control cheese 400W2 600W2 400W3 600W3

a b b b
FAAs 21 3.88 0.15 9.47 1.09 7.85 0.66 7.27 1.28 5.02 0.19a
30 10.96 1.69a 27.63 0.39c 17.04 0.35b 18.32 1.94b 18.67 0.80b
60 28.64 0.90a 71.44 1.12c 49.30 0.37b 74.82 2.03c 47.90 1.92b
90 53.76 1.65a 89.32 1.35c 68.64 4.87b 110.66 1.89d 73.24 2.10b
120 78.24 4.78a 101.84 1.82b 78.03 1.75a 123.36 1.33c 94.63 4.79b
Proteolysis 21 0.43 0.01a 0.77 0.06c 0.54 0.05b 0.47 0.00ab 0.47 0.01a
30 0.55 0.02a 1.76 0.11d 1.30 0.02c 1.39 0.10c 0.96 0.01b
60 1.74 0.03a 4.42 0.09c 3.27 0.14b 4.45 0.13c 3.63 0.01b
90 3.46 0.14a 5.43 0.26c 4.25 0.07ab 6.81 0.36d 5.06 0.06bc
120 5.73 0.53a 7.47 0.12a 5.80 0.46a 9.59 0.50b 7.22 0.19a
a
Total free amino acids are expressed in mg g1 cheese DM and overall proteolysis (determined by the OPA method) as the absorbance at 340 nm.
Results are presented as mean SE (n ¼ 4) of duplicate determinations in two cheese-making experiments.
which enhances the release of intracellular enzymes to the medium 3.5. Texture
after the death and lysis of microbial cells. In blue-veined cheese,
aminopeptidase activity was significantly higher in control cheese The softening of texture recorded during ripening and refrig-
than in all the HPP cheeses during refrigerated storage, indepen- erated storage of control Brie cheese (Fig. 3) is characteristic of this
dently of the pressure applied or the age of cheeses at treatment cheese variety. Fracturability of control cheese decreased gradually
(Calzada et al., 2013b). from 10.18 N on day 1 (data not shown) to 0 N on day 60. Elasticity
The concentration of total FAA reached 1.94 mg g1 DM in con- of control cheese was 0.122 N mm2 on day 1 (data not shown),
trol cheese on day 1 (data not shown). From day 21 onwards, higher increased to 0.171 N mm2 on day 14, and declined afterwards to
FAA concentrations were found in HPP cheeses than in control 0.002 N mm2 on day 60. Firmness of control cheese was 0.075 N m
cheese, and in 400 MPa cheeses than in 600 MPa cheeses (Table 5). on day 1 (data not shown), remained at that level until day 14, and
Differences in FAA between HPP cheeses and control cheese can afterwards declined to 0.002 N m on day 60. Texture parameters of
tentatively be attributed to a higher activity of peptidolytic enzymes control cheese showed close to zero values from day 60 onwards,
at the lower pH value of HPP cheeses, acting preferentially on the typical of mature to the core Brie cheese (Abraham et al., 2007;
hydrophobic peptides. This would explain why the levels of hy- Vassal et al., 1986).
drophobic peptides did not increase during the refrigerated storage Brie cheeses HP-treated on day 14 exhibited a firmer texture
of HPP cheeses, in spite of the sharp decrease in their a- and b-casein than control cheese, with significantly (P < 0.05) higher values of
concentrations, while hydrophobic peptides rose considerably in texture parameters (Fig. 3). Cheeses HP-treated on day 21 showed
control cheese (Table 3). The concentration of total FAA in 90-day- different patterns from those of 400W2 and 600W2 cheeses, with
old control Brie cheese was close to the 61.20 mg g1 DM reported marked declines in texture parameters immediately after HPP,
for control blue-veined cheese of the same age by Calzada et al. generally followed by further declines until day 60, and a final re-
(2013b), but total FAA concentrations of HPP Brie cheeses on day covery of texture parameters at the end of the refrigerated storage
90 were markedly higher than the level of 25.21e64.68 mg g1 DM period, in particular in the 400W3 cheese, which might be associ-
reported for HPP blue-veined cheeses of the same age. ated with the increase in DM content (Fig. 2).
Overall proteolysis, as determined by the OPA method, attained The concentrations of residual caseins could explain the differ-
an absorbance value of 0.19 on day 1 (data not shown). As in the ences in texture between control cheese and HP-treated cheeses,
case of FAA, higher absorbance values were recorded from day 21 but they do not justify the differences between cheeses HP-treated
onwards for HPP cheeses than for control cheese, and for 400 MPa on days 14 and 21. To ascertain the factors responsible for these
cheeses than for 600 MPa cheeses (Table 5). HPP treatment at the differences in texture, the correlations of texture parameters with
lower pressure level might have favoured the release of intracel- residual caseins, DM content and pH value were calculated for
lular enzymes without causing their inactivation, thus enhancing control and HP-treated cheeses from days 21e90. It was found that
secondary proteolysis reactions in 400 MPa cheeses. the fracturability was significantly correlated with residual b-casein
Biogenic amines were not detected in any sample of control, level (r ¼ 0.447; P < 0.05) and pH (r ¼ 0.459; P < 0.05), the
400W2 and 600W2 Brie cheeses. Cadaverine, found in 400W3 and elasticity also with residual b-casein level (r ¼ 0.493; P < 0.05) and
600W3 cheeses (data not shown), increased from 19 mg kg1 DM on pH (r ¼ 0.648; P < 0.01), and the firmness with residual aS-casein
day 30 to 117 mg kg1 DM on day 120 in 400W3 cheese and from level (r ¼ 0.488; P < 0.05), residual b-casein level (r ¼ 0.541;
6 mg kg1 DM on day 60 to 131 mg kg1 DM on day 120 in 600W3 P < 0.05), pH value (r ¼ 0.686; P < 0.001) and DM content
cheese. Putrescine was found only in 600W3 cheese on day 120, at (r ¼ 0.468; P < 0.05). These results emphasise the important role of
20 mg kg1 DM (data not shown). The absence or low concentrations the increase in pH value during ripening and storage on the soft-
of biogenic amines in control and HPP cheeses are in agreement with ening of the texture of Brie cheese and similar varieties, as previ-
the above mentioned low levels of bacterial contaminants during ously reported by Abraham et al. (2007) and Vassal et al. (1986)
ripening and refrigerated storage. Cadaverine and putrescine have who also found significant negative correlations between firm-
been reported to be the most abundant biogenic amines in ness and pH in Camembert cheeses.
Camembert and Brie cheeses (Stratton, Hutkins, & Taylor, 1991).
Concentrations of 564 mg kg1 cadaverine, 127 mg kg1 putrescine, 3.6. Sensory characteristics
185 mg kg1 tyramine and 6 mg kg1 histamine were found in a
commercial raw milk Camembert cheese (Kubí cková & Grosch, Flavour quality scores of control and HPP Brie cheeses did not
1998), and were partly responsible for its bitter taste. differ significantly (P < 0.05) during the first month (Fig. 4).
Fig. 3. Texture characteristics (A, fracturability; B, elasticity; C, firmness) during Fig. 4. Sensory scores (A, flavour quality; B, flavour intensity; C, bitterness) during
ripening and refrigerated storage of Brie control cheese (-) and cheeses HP-treated at ripening and refrigerated storage of Brie control cheese (-) and cheeses HP-treated at
400 MPa on day 14 (,), 600 MPa on day 14 ( ), 400 MPa on day 21 ( ) or 600 MPa on 400 MPa on day 14 (,), 600 MPa on day 14 ( ), 400 MPa on day 21 ( ) or 600 MPa on
day 21 ( ). Bars indicate standard error of the means. day 21 ( ). Bars indicate standard error of the means.
However, a drastic decline in flavour quality was recorded for acceptable flavour quality scores, above 5.5, for control and HPP
control cheese from day 30 to day 60, while flavour quality scores of blue-veined cheeses until day 360 and no significant differences in
HPP cheeses did not vary during this period. The significantly bitterness scores between cheeses. In the present work, bitterness
(P < 0.05) higher flavour quality scores of HPP Brie cheeses per- evolved more rapidly in control cheese, in which it attained
sisted until day 120, a finding in contrast to the results obtained for significantly (P < 0.05) higher scores than in all HPP cheeses from
HPP blue-veined cheeses by Calzada et al. (2013b), who observed day 60 onwards (Fig. 4), which most probably impaired flavour
quality. A highly significant (r ¼ 0.799; P < 0.001) negative cor- Arqués, J. L., Garde, S., Gaya, P., Medina, M., & Nuñez, M. (2006). Inactivation of
microbial contaminants in raw milk La Serena cheese by high-pressure treat-
relation was found between flavour quality and bitterness,
ments. Journal of Dairy Science, 89, 888e891.
comparing control and HPP cheeses from day 21 to day 90. The level Auberger, B., Lenoir, J., & Bergère, J. L. (1997). Partial characterization of exo-
of hydrophobic peptides in those cheeses correlated significantly peptidases produced by a strain of Geotrichum candidum. Sciences des Aliments,
with bitterness scores (r ¼ 0.791; P < 0.001) and, negatively, with 17, 655e670.
Calzada, J., Del Olmo, A., Picon, A., Gaya, P., & Nuñez, M. (2013a). Reducing biogenic-
flavour quality scores (r ¼ 0.920; P < 0.001). According to Engel, amine-producing bacteria, decarboxylase activity, and biogenic amines in raw
Tournier, Salles, and Le Quéré (2001), bitterness of Camembert milk cheese by high-pressure treatments. Applied and Environmental Microbi-
cheese is due mainly to the taste of small peptides released during ology, 79, 1277e1283.
Calzada, J., del Olmo, A., Picon, A., Gaya, P., & Nuñez, M. (2013b). Proteolysis and
cheese ripening, modulated by changes in the cheese matrix, which biogenic amine buildup in high-pressure treated ovine milk blue-veined
would increase the relative availability of bitter peptides and/or cheese. Journal of Dairy Science, 96, 4816e4829.
modify texture-taste interactions. Flavour quality of control cheese Church, F. C., Swaisgood, H. E., Porter, D. H., & Catignani, G. L. (1983). Spectropho-
tometric assay using o-phtaldialdehyde for determination of proteolysis in milk
was most probably also impaired by compounds other than bitter and isolated milk proteins. Journal of Dairy Science, 66, 1219e1227.
peptides resulting from the secondary degradation of proteolysis Desmazeaud, M. J., Gripon, J.-C., Le Bars, D., & Bergère, J. L. (1976). Study of the role
and lipolysis products, with unpleasant taste and aroma notes, of microorganisms and enzymes during cheese ripening. III. Influence of mi-
croorganisms. Lait, 56, 379e396.
which conferred over-ripening flavour characteristics such as pu- Eigel, W. N. (1977). Effect of bovine plasmin on aS1-B and k-A caseins. Journal of
trid and rancid to 90- and 120-day control cheeses, according to the Dairy Science, 60, 1399e1403.
remarks from panellists. Engel, E., Tournier, C., Salles, C., & Le Quéré, J. L. (2001). Evolution of the composition
of a selected bitter Camembert cheese during ripening: release and migration of
Flavour intensity scores showed less marked differences than
taste-active compounds. Journal of Agricultural and Food Chemistry, 49, 2940e
quality scores between control and HP-treated cheeses (Fig. 4). 2947.
Some of the HPP cheeses showed higher intensity scores than Garde, S., Tomillo, J., Gaya, P., Medina, M., & Nuñez, M. (2002). Proteolysis in His-
control cheese on day 30, but intensity scores of control cheese pánico cheese manufactured using a mesophilic starter, a thermophilic starter,
and bacteriocin-producing Lactococcus lactis subsp. lactis INIA 415 adjunct
generally exceeded those of HPP cheeses on days 90 and 120. culture. Journal of Agricultural and Food Chemistry, 50, 3479e3485.
Flavour intensity of control and HP-treated cheeses from days 21 to Gómez, M. J., Garde, S., Gaya, P., Nuñez, M., & Medina, M. (1997). Relationship be-
90 was highly correlated with levels of hydrophilic peptides tween level of hydrophobic peptides and bitterness in cheese made from
pasteurized and raw milk. Journal of Dairy Research, 64, 289e297.
(r ¼ 0.881; P < 0.001) and total FAA (r ¼ 0.837; P < 0.001), less with Gripon, J.-C., Desmazeaud, M. J., Le Bars, D., & Bergère, J. L. (1977). Role of proteolytic
levels of hydrophobic peptides (r ¼ 0.498; P < 0.05), and negatively enzymes of Streptococcus lactis, Penicillium roqueforti, and Penicillium caseicolum
correlated (r ¼ 0.491; P < 0.05) with flavour quality. A significant during cheese ripening. Journal of Dairy Science, 60, 1532e1538.
Huppertz, T., Fox, P. F., & Kelly, A. L. (2004). Susceptibility of plasmin and chymosin
correlation between flavour intensity and total FAA was also found in Cheddar cheese to inactivation by high pressure. Journal of Dairy Research, 71,
for ovine-milk blue-veined cheeses (Calzada et al., 2013b). 496e499.
At the end of Brie cheese refrigerated storage period, on day 150, Juan, B., Ferragut, V., Buffa, M., Guamis, B., & Trujillo, A. J. (2007). Effects of high
pressure on proteolytic enzymes in cheese: relationship with the proteolysis of
extreme over-ripening of control cheese and over-ripening of all ewe milk cheese. Journal of Dairy Science, 90, 2113e2125.
the HPP cheeses was observed. For this reason, the results corre- Kolakowski, P., Reps, A., & Babuchowski, A. (1998). Characteristics of pressurized
sponding to 150-day cheeses are not presented. ripened cheeses. Polish Journal of Food and Nutrition Sciences, 48, 473e483.
Krause, I., Bockhardt, A., Neckermann, H., Henle, T., & Klostermeyer, H. (1995).
Simultaneous determination of amino acids and biogenic amines by reversed-
phase high-performance liquid chromatography of the dabsyl derivatives.
4. Conclusions Journal of Chromatography A, 715, 67e79.
Kubícková, J., & Grosch, W. (1997). Evaluation of potent odorants of Camembert cheese
High-pressure-processing of Brie cheese lowered counts of by dilution and concentration techniques. International Dairy Journal, 7, 65e70.
Kubícková, J., & Grosch, W. (1998). Evaluation of flavour compounds of Camembert
P. camemberti by more than 2 log units at 400 MPa, and by more than cheese. International Dairy Journal, 8, 11e16.
5 log units at 600 MPa. The increase in the pH value of HPP cheeses Lane, C. N., & Fox, P. F. (1997). Role of starter enzymes during ripening of Cheddar
was retarded in comparison with control cheese, most probably cheese made from pasteurised milk under controlled microbiological condi-
tions. International Dairy Journal, 7, 55e63.
because the reduced fungal population metabolised lactic acid at a Lau, K. Y., Barbano, D. M., & Rasmussen, R. R. (1991). Influence of pasteurization of
considerably slower rate. The lower pH values of HPP cheeses milk on protein breakdown in Cheddar cheese during aging. Journal of Dairy
together with the inactivation of proteolytic enzymes by high Science, 74, 727e740.
Leclercq-Perlat, M.-N., Buono, F., Lambert, D., Latrille, E., Spinnler, H.-E., & Corrieu, G.
pressure resulted in higher concentrations of residual caseins, in
(2004). Controlled production of Camembert-type cheeses. Part I. Microbio-
particular in 600 MPa cheeses, and in markedly lower concentra- logical and physicochemical evolutions. Journal of Dairy Research, 71, 346e354.
tions of hydrophobic peptides. Firmer texture, higher flavour quality Le Graet, Y., Lepienne, A., Brule, G., & Ducruet, P. (1983). Mineral migration in soft
and reduced bitterness of Brie cheese were achieved by means of cheese during ripening. Lait, 63, 317e332.
Lessard, M. H., Bélanger, G., St-Gelais, D., & Labrie, S. (2012). The composition of
HPP, a procedure which may be applied to prevent over-ripening. Camembert cheese-ripening cultures modulates both mycelium growth and
appearance. Applied and Environmental Microbiology, 78, 1813e1819.
Maisnier-Patin, S., Deschamps, N., Tatini, S. R., & Richard, J. (1992). Inhibition of
Acknowledgements Listeria monocytogenes in Camembert cheese made with a nisin-producing
starter. Lait, 72, 249e263.
Malone, A. S., Wick, C., Shellhammer, T. H., & Courtney, P. D. (2003). High pressure
Funding from project AGL2009-07801 from the Ministry of Sci- effects on proteolytic and glycolytic enzymes involved in cheese manufacturing.
ence and Innovation (MICINN, Madrid, Spain, BES-2010-030444) is Journal of Dairy Science, 86, 1139e1146.
Messens, W., Foubert, I., Dewettinck, K., & Huyghebaert, A. (2001). Proteolysis of a
acknowledged by the authors. J. Calzada was the recipient of a
high-pressure-treated mould-ripened cheese. Milchwissenschaft, 56, 201e204.
MICINN fellowship. The authors are grateful to ILAS S.A. (Madrid, Nooitgedagt, A. J., & Hartog, B. J. (1988). A survey of the microbiological quality of
Spain) for providing Brie cheeses and to Hiperbaric (Burgos, Spain) Brie and Camembert cheese. Netherlands Milk and Dairy Journal, 42, 57e72.
for valuable help with HPP. O’Reilly, C. E., O’Connor, P. M., Kelly, A. L., Beresford, T. P., & Murphy, P. M. (2000).
Use of hydrostatic pressure for inactivation of microbial contaminants in
cheese. Applied and Environmental Microbiology, 66, 4890e4896.
O’Reilly, C. E., O’Connor, P. M., Murphy, P. M., Kelly, A. L., & Beresford, T. P. (2002).
References Effect of high-pressure treatment on viability and autolysis of starter bacteria
and proteolysis of Cheddar cheese. International Dairy Journal, 12, 915e922.
Abraham, S., Cachon, R., Colas, B., Feron, G., & De Coninck, J. (2007). Eh and pH Picon, A., Alonso, R., Gaya, P., Fernández-García, E., Rodríguez, B., De Paz, M., et al.
gradients in Camembert cheese during ripening: measurements using (2010). Microbiological, chemical, textural and sensory characteristics of His-
microelectrodes and correlations with texture. International Dairy Journal, 17, pánico cheese manufactured using frozen ovine milk curds scalded at different
954e960. temperatures. International Dairy Journal, 20, 344e351.
Saldo, J., McSweeney, P. L. H., Sendra, E., Kelly, A. L., & Guamis, B. (2002). Proteolysis Trieu-Cuot, P., & Gripon, J.-C. (1982). A study of proteolysis during Camembert
in caprine milk cheese treated by high pressure to accelerate cheese ripening. cheese ripening using isoelectric focusing and two-dimensional electropho-
International Dairy Journal, 13, 35e44. resis. Journal of Dairy Research, 49, 501e510.
Silla-Santos, M. H. (1996). Biogenic amines: their importance in foods. International Vassal, L., Monnet, V., Le Bars, D., Roux, C., & Gripon, J.-C. (1986). Relation be-
Journal of Food Microbiology, 29, 213e231. tween pH, chemical composition and texture of Camembert cheese. Lait, 66,
Spinnler, H.-E., & Gripon, J.-C. (2004). Surface mould-ripened cheeses. In P. F. Fox, 341e351.
P. L. H. McSweeney, T. M. Cogan, & T. P. Guinee (Eds.) (3rd ed.)Major cheese Visser, S. (1981). Proteolytic enzymes and their action on milk proteins. A review.
groups: Vol. 2. Cheese. Chemistry, physics and microbiology (pp. 157e174). Lon- Netherlands Milk and Dairy Journal, 35, 65e88.
don, UK: Elsevier Academic Press. Voigt, D. D., Chevalier, F., Qian, M. C., & Kelly, A. L. (2010). Effect of high-pressure
Stratton, J. E., Hutkins, R. W., & Taylor, S. L. (1991). Biogenic amines in cheese and treatment on microbiology, proteolysis, lipolysis and levels of flavor com-
other fermented foods: a review. Journal of Food Protection, 54, 460e470. pounds in mature blue-veined cheese. Innovative Food Science and Emerging
Trieu-Cuot, P., Archieri-Haze, M.-J., & Gripon, J.-C. (1982a). Effect of aspartyl pro- Technologies, 11, 68e77.
teinases of Penicillium caseicolum and Penicillium roqueforti on caseins. Journal Wick, C., Nienaber, U., Anggraeni, O., Shellhammer, T. H., & Courtney, P. D. (2004).
of Dairy Research, 49, 487e500. Texture, proteolysis and viable lactic acid bacteria in commercial Cheddar
Trieu-Cuot, P., Archieri-Haze, M.-J., & Gripon, J.-C. (1982b). Comparative study of the cheeses treated with high pressure. Journal of Dairy Research, 71, 107e115.
action of metalloproteinases of Penicillium caseicolum and Penicillium roqueforti Yvon, M., & Rijnen, L. (2001). Cheese flavour formation by amino acid catabolism.
on aS1- and b-caseins. Lait, 62, 234e249. International Dairy Journal, 11, 185e201.

Capítulo 8.
Effect of high-pressure-processing on the lipolysis and
volatile compounds of Brie cheese during ripening and
refrigerated storage.

165

Fotografía: colonias de Penicillium camemberti en agar CG.

International Dairy Journal 39 (2014) 232e239
Contents lists available at ScienceDirect
International Dairy Journal

journal homepage: www.elsevier.com/locate/idairyj
Effect of high-pressure-processing on lipolysis and volatile

compounds of Brie cheese during ripening and refrigerated storage
~ ez*
Javier Calzada, Ana del Olmo, Antonia Picon, Manuel Nun
~ a Km 7, Madrid 28040, Spain
Departamento de Tecnología de Alimentos, INIA, Carretera de La Corun
a r t i c l e i n f o a b s t r a c t
Article history: Brie cheeses were high pressure (HP)-treated at 400 or 600 MPa, on days 14 or 21 after manufacture, to
Received 19 March 2014 prevent over-ripening. HP-treatment reduced total free fatty acid content of 120-day-old cheese by up to
Received in revised form 88.5%. On day 120, HP-treated cheeses had levels of alcohols, aldehydes, acids, esters and ethers up to 3.4,
25 June 2014
1.9, 43.8, 18.7 and 5.6 times higher, respectively, than control cheese, while ketones and hydrocarbons
Accepted 1 July 2014
levels were up to 88.6% and 48.9% lower, respectively. Levels of sulphur compounds, pyrazines and
Available online 29 July 2014
amines increased drastically in control cheese from day 60 onwards, resulting in lower odour quality
scores. On day 120, HP-treated cheeses had levels of sulphur compounds, pyrazines and amines up to
96.9%, 99.3% and 99.4% lower, respectively, than control cheese. Cheese appearance was impaired by HP-
treatment, resulting in lower lightness and slightly more reddish and yellowish colour. These changes
might diminish consumer acceptance of HP-treated Brie cheese.
© 2014 Elsevier Ltd. All rights reserved.
1. Introduction & Cottenceau, 1999). Some of the non-volatile compounds interact

with volatile compounds in Camembert cheese, influencing the
Soft cheeses ripened by Penicillium camemberti, such as Brie and release of aroma compounds and consumer perception (Pionnier,
Camembert, are characterised by a thin layer of white mould col- Engel, Salles, & Le Quere
, 2002). The potent enzymes of fungal origin
onising their rind that metabolises lactic acid with a concomitant present in soft cheeses ripened by P. camemberti and their high pH
increase in cheese pH to values close to neutrality. Extracellular values make these cheeses particularly prone to the development of
enzymes produced by P. camemberti, namely aspartyl proteinase unbalanced flavour profiles, which lead to over-ripening in spite of
(Trieu-Cuot, Archieri-Haze, & Gripon, 1982a), metalloproteinase adequate refrigerated storage conditions.
(Trieu-Cuot, Archieri-Haze, & Gripon, 1982b) and lipase (Lamberet High pressure processing (HPP) of cheese at sufficiently high
& Lenoir, 1976), are main contributors to the biochemical changes pressures, generally above 500 MPa, has been proven to eliminate
involved in the ripening process of these cheese varieties. The ac- pathogenic microorganisms (O'Reilly, O'Connor, Kelly, Beresford, &
tivity of P. camemberti enzymes is generally enhanced at the high Murphy, 2000), inactivate enzymes (Huppertz, Fox, & Kelly, 2004)
pH values of Brie and Camembert cheeses. and prevent biogenic amine formation (Calzada, del Olmo, Picon,
Products resulting from the primary hydrolysis of proteins and Gaya, & Nun ~ ez, 2013a). It may also affect other cheese character-
triglycerides are further degraded, giving rise to the formation of non- istics such as flavour and texture (Martínez-Rodríguez et al., 2012).
volatile flavour compounds, as well as volatile potent odorants The effect of HPP on the proteolysis, texture and flavour of Brie
responsible for the typical flavour and aroma notes of Brie and cheese during ripening and refrigerated storage was reported in a
Camembert cheeses (Adda & Dumont, 1974; Kubíckova & Grosch, previous paper (Calzada, del Olmo, Picon, Gaya, & Nun ~ ez, 2014b).
1997, 1998). The ripening microbiota influences the formation of free However, to our knowledge, there is no information available about
fatty acids (FFAs) and volatile compounds in Camembert-type cheeses the influence of HPP on the lipolysis and the formation of volatile
(Leclercq-Perlat, Corrieu, & Spinnler, 2007; Leclercq-Perlat, Latrille, compounds in Brie cheese.
Corrieu, & Spinnler, 2004). Soft white mould cheeses are rich in FFAs, In the present study, the formation of FFAs and volatile com-
ketones, esters and sulphur compounds (Adda & Dumont, 1974; Sable pounds in the control Brie cheese and in the cheeses HP-treated at
400 or 600 MPa on days 14 and 21 after manufacture (Calzada et al.,
2014b), to prevent over-ripening during refrigerated storage, was
* Corresponding author. Tel.: þ34 913476799.
investigated. The odour and the colour of control and HP-treated
~ ez).
E-mail addresses: nunez@inia.es, mng1950@gmail.com (M. Nun Brie cheeses were also compared.
http://dx.doi.org/10.1016/j.idairyj.2014.07.007
0958-6946/© 2014 Elsevier Ltd. All rights reserved.
2. Materials and methods
2.1. Brie cheeses and chemical determinations
The manufacture of Brie cheeses studied in the present work

was described previously (Calzada et al., 2014b). In each of two
trials, carried out on consecutive days, approximately 120 cheeses,
2.7 kg in weight, were made from 2100 L pasteurised milk. High-
pressure (HP) processing of cheeses was performed on days 14 or
21 after manufacture, at 400 or 600 MPa for 5 min. Cheeses HP-
treated on day 14 were coded as 400W2 and 600W2, and those
HP-treated on day 21 as 400W3 and 600W3. Control and HP-
treated cheeses were ripened at 12 C until day 21 and then held
at 4 C until day 120.
One cheese wedge per trial was used for enzymatic and chem-
ical analyses, after discarding the rind, and another wedge per trial
for odour and colour analyses. Esterase activity was determined in
duplicate as described by Calzada, del Olmo, Picon, Gaya, and Nun~ ez
(2013b). Free fatty acids were analysed in duplicate by gas-
chromatography (GC) according to Ferna ndez-García, Carbonell,
Calzada, and Nun ~ ez (2006). Volatile compounds were determined
in triplicate by GC coupled to mass-spectrometry (MS), after solid-
phase microextraction (SPME), as described by Calzada, del Olmo,
Picon, Gaya, and Nun ~ ez (2014a).
2.2. Odour, colour and statistical analysis
Quality (preference) and intensity of odour, defined as the ol-

factory sensation perceived by smelling the cheese, were evaluated
by a trained 16-member panel scoring on a 0e10 point scale as
described by Calzada et al. (2014a). Colour parameters of lightness
(L*), redness (a*) and yellowness (b*) were determined at the core
and the upper rind of cheeses, on each site at four different points,
using a CM-2600d spectrocolorimeter (Minolta, Osaka, Japan) with
a D65 illuminator at 10 observer angle and specular component
included (SCI), at room temperature, according to Picon, Alonso, van
Wely, and Nun ~ ez (2013). Statistical analysis of results, including
Fig. 1. Levels of (A) ethanoic and (B) propanoic acids during ripening and refrigerated
two-way analysis of variance (ANOVA), comparison of means by storage of control Brie cheese (black) and cheeses HP-treated at 400 MPa on day 14
Tukey's test with the significance assigned at P < 0.05 and principal (white), 600 MPa on day 14 (horizontally striped), 400 MPa on day 21 (dotted) or
component analysis, was carried out using SPSS Win 14.0 program 600 MPa on day 21 (obliquely striped). Bars indicate standard error of the means.
(SPSS Inc., Chicago, IL, USA), as described by Calzada et al. (2013b).
3. Results and discussion Esterase activity increased markedly during ripening and
refrigerated storage of control cheese, from 0.83 pmol a-naphthol
3.1. Ethanoic acid, propanoic acid and free fatty acids min1 g1 on day 1 to 9.01 on day 60 and 18.14 on day 120
(Supplementary Fig. S1). HPP of Brie cheese, in particular at
The concentration of ethanoic acid, a product of microbial 400 MPa, tended to increase esterase activity, probably by
origin, was 1.305 mg g1 cheese dry matter (DM) on day 1 (Fig. 1); it enhancing the release of enzymes from dead or injured cells of
decreased in control cheese to a minimum of 0.08 mg g1 on day lactic acid bacteria and P. camemberti. Esterase activity in cheeses
30, most probably because of acid consumption by moulds, and treated at 400 MPa increased from 15.91e18.37 pmol a-naphthol
then increased gradually. In 400W2 and 400W3 cheeses, the con- min1 g1 on day 60 to 33.25e44.07 on day 120, while in cheeses
centration of ethanoic acid was generally higher than in control treated at 600 MPa it rose from 8.14e8.92 pmol a-naphthol
cheese until day 120, most probably because of lower acid con- min1 g1 on day 60 to 22.12e27.61 on day 120. Although treat-
sumption by the smaller population of moulds, whereas in 600W2 ment at 600 MPa would cause more intense damage to lactic acid
and 600W3 cheeses it remained fairly unchanged, which could be bacteria and P. camemberti than that at 400 MPa (Calzada et al.,
associated with slow production and slow consumption by HPP- 2014a), favouring the release of intracellular enzymes from
injured microorganisms. Propanoic acid, a product of microbial injured cells, the secreted enzymes outside the microbial cells
origin too, was present at 0.053 mg g1 cheese DM on day 1 (Fig. 1); would presumably be more susceptible to inactivation by HP than
it declined in control cheese to a minimum on day 60 and after- those in the interior of the microbial cells.
wards rose to 0.055 mg g1 cheese DM on day 120. In HP-treated Short-chain (SC) FFAs were at higher concentrations in 400W3
cheeses, it also declined to a minimum on day 60, but there was cheese than in the rest of cheeses from day 21 to day 30 (Fig. 2),
no further recovery probably because of the pressure-induced which may be associated with its higher esterase activity. However,
changes in their microbiota. Ethanoic and propanoic acids, with a from day 60 onwards, SC-FFA concentrations were higher in control
typical vinegar odour, are key odorants of Cheddar, Gruye re and cheese than in all the HP-treated cheeses. Lactic acid bacteria es-
Emmental cheeses (Curioni & Bosset, 2002). terases are responsible for the generation of SC-FFAs in many
4.0 (Calzada et al., 2014b) would imply less injured cells and less
SC-FFA
A released esterase in control cheese, which seems contradictory in
3.5 terms of its higher SC-FFA concentrations. Some of the SC-FFAs, in
g cheese DM)
particular branched-chain FFAs, may originate from the breakdown

3.0 of free amino acids (Urbach, 1993), but there was no correlation
between the concentrations of free amino acids in control and HP-
mg g cheese DM
2.5
treated Brie cheeses (Calzada et al., 2014b) and their SC-FFA content
-1
(Fig. 2). SC-FFAs can also derive from ketones, esters and aldehydes
2.0
by oxidation (Molimard & Spinnler, 1996). However, the more
SC-FFAs (mg
-1
1.5 plausible explanation for differences in SC-FFA contents may be the

enhancement of P. camemberti lipase activity in control cheese. This
1.0 enzyme, which is capable of hydrolysing SC-fatty acid triglycerides,
has optimum activity at alkaline pH (Lamberet & Lenoir, 1976).
0.5 Consequently, more pronounced lipolysis should be expected to
occur during refrigerated storage in control Brie cheese, in which
0.0 the pH value increased from 5.54 on day 21 to 7.55 on day 60, than
14 d 21 d 30 d 60 d 90 d 120 d in HP-treated cheeses, with pH values which only rose from
5.26e5.70 on day 21 to 5.47e5.82 on day 60 (Calzada et al., 2014b).
6.0
MC-FFA On day 60, the concentrations of individual SC-FFAs in control Brie
B
cheese were 1.299, 0.050, 0.103, 0.483 and 0.186 mg g1 cheese DM
5.0 for C4:0, iso-C4:0, iso-C5:0, C6:0 and C8:0, respectively (data not
g cheese DM)
shown), markedly higher than the levels of these FFAs in com-

mercial Brie cheese reported by Woo, Kollodge, and Lindsay (1984);
4.0
mg g cheese DM
those authors did not indicate ripening time, which influences FFA
concentration.
-1
3.0 Medium-chain (MC) FFAs followed a pattern of accumulation

MC-FFAs (mg
similar to that of SC-FFAs, with higher concentrations in control

-1
cheese than in HP-treated cheeses from day 90 onwards (Fig. 2). As

2.0
for SC-FFAs, a lower contribution of lactic acid bacteria esterases
than of P. camemberti lipase to MC-FFA formation would be ex-
1.0 pected. Although control cheese had lower esterase activity than
400W2 and 400W3 cheeses from day 60 onwards, its higher pH
0.0
value (Calzada et al., 2014b) most probably enhanced MC-FFA for-
14 d 21 d 30 d 60 d 90 d 120 d mation. Concentrations of individual MC-FFAs in 60-day-old con-
trol cheese were 0.369, 0.340, 0.014 and 1.051 mg g1 cheese DM,
LC-FFA respectively, for C10:0, C12:0, C13:0 and C14:0 (data not shown), higher
25.0
C than in commercial Brie cheese (Woo et al., 1984).
22.5 Treatment of Brie cheese at 400 MPa on day 21 had a pro-
nounced favourable effect on long-chain (LC) FFA concentrations
g cheese DM)
20.0
from day 21 to day 60 (Fig. 2). This can be attributed to the lipase of
17.5 P. camemberti, which is more active than lactic acid bacteria ester-
mg g cheese DM
15.0 ases on the triglycerides containing LC-FFAs. Individual LC-FFA

concentrations in 60-day-old control cheese were 2.136, 1.038,
-1
12.5 6.942, 0.905 and 0.138 mg g1 cheese DM for C16:0, C18:0, C18:1, C18:2
LC-FFAs (mg
and C18:3, higher than in commercial Brie cheese (Woo et al., 1984).
-1
10.0
From day 90 onwards, LC-FFA concentrations were higher in control
7.5 cheese than in HP-treated cheeses (Fig. 2), most probably because
5.0 of the enhancement of the activity of P. camemberti lipase at high
pH values. Concentrations of LC-FFAs in Camembert-type cheeses
2.5 were correlated with the levels of P. camemberti spores (Leclercq-
0.0 Perlat et al., 2007).
14 d 21 d 30 d 60 d 90 d 120 d Previous studies on the lipolysis of HP-treated mould-ripened
Storage time cheeses dealt only with blue-veined cheeses. No significant effect of
HPP at 400 or 600 MPa on the lipolysis in mature blue-veined
Fig. 2. Levels of (A) short-chain free fatty acids (SC-FFAs, C4:0 to C8:0), (B) medium-
cheese was recorded after 28 days at 4 C, probably because of
chain FFAs (MC-FFAs, C10:0 to C14:0) and (C) long-chain FFAs (LC-FFAs, C16:0 to C18:3)
during ripening and refrigerated storage of control Brie cheese (black) and cheeses HP- the short storage period (Voigt, Chevalier, Qian, & Kelly, 2010). In
treated at 400 MPa on day 14 (white), 600 MPa on day 14 (horizontally striped), contrast, total FFA concentration in 360-day-old ovine milk blue-
400 MPa on day 21 (dotted) or 600 MPa on day 21 (obliquely striped). Bars indicate veined cheese declined by 41% on average when treated at
standard error of the means. 600 MPa at different stages of ripening, while HPP at 400 MPa had a
negligible effect on total FFAs (Calzada et al., 2013b).
FFAs are important contributors to the aroma of surface mould-
cheese varieties, lipolysis in cheese being related to the autolytic ripened cheeses such as Camembert and Brie, both directly and as
release of intracellular enzymes from lactic acid bacteria (Collins, precursors of methyl ketones, alcohols, lactones and esters. They
McSweeney, & Wilkinson, 2003). The higher counts of viable lac- may represent up to 10% of total fatty acids in Camembert cheese,
tic acid bacteria in control Brie cheese than in HP-treated cheeses which contains all the even-numbered FFAs, together with iso-C4:0
and iso-C5:0 acids (Molimard & Spinnler, 1996). Although LC-FFAs Some branched-chain aldehydes have been identified as potent
are usually the most abundant FFAs in cheeses, they play a minor cheese odorants; their green, malty, sweet, floral, fruity, nutty,
role in flavour, because of higher perception thresholds than those acrid, pungent notes contribute to cheese aroma, generally
of SC- and MC-FFAs (Curioni & Bosset, 2002). conferring more pleasant odours at lower levels (Curioni & Bosset,
2002).
3.2. Volatile compounds The pattern of accumulation of total ketones in Brie cheese
differed depending on HPP treatment (Table 1). On day 30, 3-
One hundred and sixteen compounds were detected in the hydroxybutanone was the main ketone in all cheeses
volatile fraction of Brie cheese by GCeMS, after SPME extraction, (Supplementary Table S2). It is generated from pyruvate by lactic
including 9 aldehydes, 19 ketones, 20 alcohols, 10 acids, 11 esters, 9 acid bacteria, but P. camemberti and P. caseifulvum strains do not
sulphur compounds, 9 pyrazines, 2 amines, 8 ethers, 14 hydrocar- produce it (Jollivet et al., 1993; Larsen, 1998). On day 120, 3-
bons, and 5 benzenic compounds. This high number of volatile hydroxybutanone remained as the major ketone in HP-treated
compounds agrees with the results of previous works (Kubí ckova& cheeses, while 2-pentanone and 2-propanone predominated in
Grosch, 1997; Molimard & Spinnler, 1996). In the present study, 4- control cheese. Formation of methyl ketones by P. camemberti ATCC
methyl-2-pentanone, 2-tridecanone, trimethylamine, 2-ethyl-3,5- 4845 in a model system was reported by Okumura and Kinsella
dimethylpyrazine and 3-ethyl-2,5-dimethylpyrazine were found (1985) and confirmed afterwards for other P. camemberti and
only in control cheese, while 2-propenal, 2-propen-1-ol, octanoic P. caseifulvum strains (Jollivet et al., 1993; Larsen, 1998). In the
acid, methyl acetate and 3-methyl-1-butanol acetate were detected present study, methyl ketone formation in HP-treated cheeses was
only in HP-treated cheeses. Terpenes, compounds commonly pre- most probably hindered by their low P. camemberti counts (Calzada
sent in the volatile fraction of cheeses, come from the animal diet et al., 2014b). Methyl ketones contribute to cheese flavour with
(Toso, Procida, & Stefanon, 2002) or are produced by P. camemberti floral, fruity, musty, earthy, blue-cheese notes, whereas 3-
and Penicillium caseifulvum strains used in cheese manufacture hydroxybutanone has a buttery note (Molimard & Spinnler, 1996;
(Jollivet, Belin, & Vayssier, 1993; Larsen, 1998); they were not found Sable & Cottenceau, 1999).
in the present study. Total alcohol levels were significantly (P < 0.05) higher in HP-
The concentration of total aldehydes increased consistently treated cheeses than in control cheese (Table 1). Alcohols
during refrigerated storage in all cheeses, independently of HPP contribute to cheese flavour directly and as substrates for ester
treatment (Table 1). Branched-chain aldehydes, particularly 3- formation. The main individual alcohols in control cheese were
methylbutanal (Supplementary Table S1), predominated in Brie ethanol, 2-propanol and 2-pentanol, whereas ethanol clearly pre-
cheese, in agreement with previous findings for Camembert cheese dominated in HP-treated cheeses, followed by 3-methyl-1-butanol
(Leclercq-Perlat et al., 2004; Sable & Cottenceau, 1999). Ethanal, (Supplementary Table S3). The higher ethanol levels in HP-treated
another abundant aldehyde, can be produced through lactose cheeses cannot be associated with microbial fermentation since the
fermentation by lactic acid bacteria during the early stages of increase persisted after day 30, when carbohydrates were no longer
ripening, and is also formed by some P. caseifulvum strains (Larsen, available. In this regard, reduction of ethanal as a mechanism for
1998) but not by P. camemberti strains (Jollivet et al., 1993). ethanol formation in blue-veined cheese was suggested by Calzada
Branched-chain aldehydes in cheese generally derive from free et al. (2013b). Some P. caseifulvum strains are capable of producing
amino acids by transamination or by Strecker degradation (Yvon & ethanol in carbohydrate-containing media (Larsen, 1998), but this
Rijnen, 2001). Differences in the concentration of precursor amino alcohol was not detected in P. camemberti cultures (Jollivet et al.,
acids or in lactic acid bacteria counts between HP-treated and 1993). 2-Alkanols are derived from the reduction of the respec-
control Brie cheeses (Calzada et al., 2014b) do not explain the dif- tive methyl ketones (Molimard & Spinnler, 1996). Branched-chain
ferences in the concentration of total aldehydes. Resting cells or alcohols may originate from the respective amino acids through
cellular extracts are capable of degrading amino acids to aroma sequential transamination, decarboxylation and reduction (Yvon &
compounds in cheese models (Yvon & Rijnen, 2001) and abiotic Rijnen, 2001). Their formation by P. camemberti and P. caseifulvum
chemical reactions have been suggested as a mechanism of for- strains has been reported (Jollivet et al., 1993; Larsen, 1998). A key
mation of branched-chain aldehydes in blue-veined cheese odorant of Brie and Camembert cheeses, 1-octen-3-ol, was found at
(Calzada et al., 2013b), what might be also valid for Brie cheese. higher levels in control cheese than in HP-treated cheeses from day
Table 1
Levels of total aldehydes, total ketones, total alcohols, total acids, and total esters in the volatile fraction during refrigerated storage of Brie control cheese and cheeses HP-
treated at 400 or 600 MPa after 14 days (W2) or 21 days (W3) of ripening.
Compoundsa Days Control cheese 400W2 600W2 400W3 600W3

A C B A
Aldehydes 30 5.75 ± 0.41 17.02 ± 1.02 11.05 ± 0.44 7.45 ± 0.45 6.69 ± 0.56A
60 8.74 ± 0.29A 18.69 ± 1.29B 18.85 ± 1.05B 19.37 ± 1.09B 30.74 ± 2.38C
120 39.13 ± 3.20A 43.72 ± 2.62A 35.03 ± 3.10A 34.55 ± 3.00A 73.72 ± 5.34B
Ketones 30 724.68 ± 84.19B 1458.41 ± 117.46C 632.37 ± 50.90AB 344.47 ± 28.78A 533.13 ± 27.77AB
60 321.55 ± 22.75A 249.41 ± 26.73A 688.42 ± 27.92B 168.36 ± 8.59A 1252.16 ± 84.64C
120 2674.07 ± 130.72B 304.85 ± 6.42A 561.79 ± 47.94A 326.77 ± 36.06A 669.40 ± 24.73A
Alcohols 30 181.65 ± 7.17A 656.39 ± 36.98BC 715.22 ± 34.56C 744.83 ± 48.99C 529.25 ± 43.29B
60 310.19 ± 11.39A 993.30 ± 59.28C 915.22 ± 32.39C 1252.44 ± 29.02D 483.38 ± 18.15B
120 358.39 ± 24.05A 1119.69 ± 32.42D 809.11 ± 28.41C 1226.36 ± 36.88D 552.77 ± 57.40B
Acids 30 795.51 ± 51.80A 1688.07 ± 82.47C 1271.88 ± 79.12B 956.89 ± 89.88A 833.65 ± 30.99A
60 70.01 ± 6.12A 2599.83 ± 253.00B 1889.55 ± 27.52B 1909.83 ± 74.37B 1570.19 ± 93.04B
120 85.04 ± 7.65A 3294.02 ± 161.69BC 2773.47 ± 74.93B 3722.64 ± 180.89C 3206.43 ± 280.71BC
Esters 30 14.37 ± 1.12A 112.39 ± 16.69B 273.80 ± 28.27C 67.11 ± 5.08AB 80.19 ± 5.27B
60 14.30 ± 0.76A 165.18 ± 10.84C 302.74 ± 17.28D 75.78 ± 3.48B 176.10 ± 8.32C
120 12.52 ± 0.69A 172.66 ± 5.01C 234.29 ± 16.84D 128.41 ± 5.24B 144.86 ± 5.56BC
a
Volatile compounds are expressed as relative abundances on the internal standard (103); values are means ± SE (n ¼ 6) of triplicate determinations in two cheese-making
experiments. Means on the same row followed by different superscripts differ significantly (P < 0.05).
Table 2
Levels of total sulphur compounds, total pyrazines, total amines, total ethers, total hydrocarbons, and total benzenic compounds in the volatile fraction during refrigerated
storage of Brie control cheese and cheeses HP-treated at 400 or 600 MPa after 14 days (W2) or 21 days (W3) of ripening.
Compoundsa Days Control cheese 400W2 600W2 400W3 600W3

A A A A
Sulphur compounds 30 8.20 ± 0.36 9.12 ± 0.54 8.18 ± 0.87 8.55 ± 0.48 8.11 ± 0.44A
60 9.06 ± 0.72A 7.29 ± 0.55A 9.06 ± 0.94A 7.84 ± 0.56A 7.32 ± 0.49A
120 232.89 ± 39.33B 8.78 ± 0.52A 8.79 ± 0.90A 8.69 ± 0.53A 7.26 ± 0.48A
Pyrazines 30 1.58 ± 0.18A 2.49 ± 0.15C 2.24 ± 0.12BC 2.05 ± 0.12ABC 1.75 ± 0.14AB
60 2.74 ± 0.21A 3.02 ± 0.29A 2.27 ± 0.14A 4.89 ± 0.18B 4.68 ± 0.18B
120 408.84 ± 27.32B 4.98 ± 0.23A 2.79 ± 0.20A 5.59 ± 0.55A 7.33 ± 0.97A
Amines 30 1.60 ± 0.23B 0.97 ± 0.05A 0.71 ± 0.13A 0.93 ± 0.11A 0.67 ± 0.12A
60 1.18 ± 0.23C 1.14 ± 0.20BC 0.54 ± 0.05AB 0.63 ± 0.07ABC 0.52 ± 0.06A
120 13.26 ± 1.13B 0.70 ± 0.06A 0.14 ± 0.02A 0.16 ± 0.02A 0.08 ± 0.01A
Ethers 30 31.65 ± 3.79A 138.91 ± 12.23C 132.43 ± 10.91C 91.59 ± 7.14B 89.55 ± 4.36B
60 24.80 ± 3.55A 173.71 ± 5.48C 183.28 ± 6.29C 136.48 ± 8.60B 172.28 ± 2.54C
120 33.57 ± 2.64A 188.53 ± 7.01C 176.25 ± 3.66BC 161.61 ± 5.60B 159.42 ± 3.63B
Hydrocarbons 30 24.60 ± 1.58A 24.77 ± 1.89A 25.00 ± 1.40A 28.20 ± 1.86A 26.25 ± 1.58A
60 19.79 ± 1.69A 26.23 ± 2.17AB 28.79 ± 1.76B 27.14 ± 1.82B 24.65 ± 1.27AB
120 21.56 ± 1.19A 27.58 ± 1.43A 27.82 ± 1.78A 26.57 ± 1.19A 25.74 ± 1.89A
Benzenic compounds 30 18.74 ± 2.01A 16.72 ± 1.53A 16.91 ± 0.98A 15.84 ± 0.91A 15.96 ± 1.50A
60 16.58 ± 1.45A 15.98 ± 1.44A 13.83 ± 2.13A 17.99 ± 1.93A 17.17 ± 0.87A
120 31.51 ± 2.26B 16.11 ± 1.54A 16.88 ± 2.56A 16.67 ± 1.72A 21.18 ± 0.69A
a
Volatile compounds are expressed as relative abundances on the internal standard (103); values are means ± SE (n ¼ 6) of triplicate determinations in two cheese-making
experiments (ND, not detected). Means on the same row followed by different superscripts differ significantly (P < 0.05).
30 to day 120 (data not shown). It derives from linoleic and lino- butyrolactone (Supplementary Table S5), did not vary with HPP or
lenic acids and has green and mushroom-like notes and a low with time of storage, and was more abundant in Brie cheese than in
perception threshold (Curioni & Bosset, 2002). Phenylethanol, blue-veined cheese (Calzada et al., 2013b). Depending on their
another key odorant of Brie and Camembert cheeses, with floral concentration in cheese, g-butyrolactone and other lactones exhibit
notes, generated from phenylalanine (Sable & Cottenceau, 1999), fruity notes at low levels, or pungent, fetid, buttery notes at higher
reached higher levels in control cheese than in HP-treated cheeses levels (Molimard & Spinnler, 1996).
on day 120 (data not shown). Nine out of 10 assayed P. camemberti Total sulphur compounds remained fairly unchanged from day
strains produced phenylethanol, while only 3 strains formed 1- 30 to day 60, with no significant differences between control and
octen-3-ol (Jollivet et al., 1993). HP-treated cheeses. From day 60 to day 120, they increased 26-fold
Total acids in the volatile fraction of 30-day-old Brie cheeses in control cheese while their levels did not vary in HP-treated
were present at higher levels in 400W2 and 600W2 cheeses than in cheeses (Table 2). On days 30 and 60, dimethyldisulphide and
the rest (Table 1). Afterwards, they declined markedly in control dimethylsulphone were the principal sulphur compounds in all
cheese, because of the sharp decrease in butanoic and hexanoic cheeses (Supplementary Table S6). On day 120, dimethyldisulphide
acids (Supplementary Table S4), and increased gradually in HP- predominated markedly in control cheese, followed by dime-
treated cheeses. The pattern of accumulation of volatile acids thylsulphone, methanethiol and dimethyltrisulphide, while dime-
determined by SPMEeGCeMS differs from that observed for SC- thylsulphone and dimethylsulphide were the most abundant in HP-
FFAs (Fig. 2). This discrepancy might be explained by competition treated cheeses. Methionine g-elimination produces methanethiol,
among volatile compounds for adsorption sites in SPME fibres with which can be further chemically oxidised to dimethyldisulphide or
solid coatings (Rivas-Can ~ edo, Juez-Ojeda, Nun ~ ez, & Fernandez- dimethyltrisulphide (Yvon & Rijnen, 2001). Brevibacterium linens
García, 2012), which would result in the underestimation of buta- and other microorganisms, including Lactococcus lactis, are capable
noic and other acids in 60- and 120-day-old cheeses. Butanoic was of producing methanethiol from methionine. The contribution of
the major acid in the volatile fraction of all Brie cheeses, followed by P. camemberti and P. caseifulvum strains to the formation of sulphur
hexanoic and ethanoic acids (Supplementary Table S4). Some compounds seems less important (Jollivet et al., 1993; Larsen,
P. camemberti and P. caseifulvum strains are capable of producing 1998). The origin of dimethylsulphone, a compound found in
ethanoic, butanoic and hexanoic acids, as well as branched-chain milk and in some Italian cheese varieties (Mallia et al., 2005a; Toso
carboxylic acids (Jollivet et al., 1993; Larsen, 1998). et al., 2002), has not been elucidated, although it seems to be
Total ester levels were significantly (P < 0.05) higher in all the influenced by the diet (Urbach, 1990). Sulphur compounds are
HP-treated cheeses than in control cheese (Table 1). The major ester important key odorants in cheeses, because of their strong garlic,
in HP-treated cheeses throughout refrigerated storage was ethyl onion, cabbage-like and very ripe cheese odours, together with
ethanoate (Supplementary Table S5), which correlates well with their low perception thresholds (Engels, Dekker, de Jong, Neeter, &
their higher ethanol and ethanoic acid contents on day 30. Ethyl & Cottenceau, 1999).
Visser, 1997; Sable
ester formation in control cheese was most probably limited by its Total pyrazines followed a pattern of accumulation similar to
low ethanol content. Esters may be formed by the esterification of that of sulphur compounds (Table 2). Their levels rose moderately
alcohols with carboxylic acids or through alcoholysis, the reaction from day 30 to day 60 in control cheese and in the HP-treated
by which fatty acyl groups are transferred to alcohols (Liu, Holland, cheeses, excepting the 600W2 cheese. From day 60 to day 120,
& Crow, 2004). Most P. camemberti and P. caseifulvum strains pro- they increased 149-fold in control cheese and remained relatively
duce ethyl ethanoate, and some may also produce higher ethyl constant in the HP-treated cheeses. The major pyrazine in all Brie
esters (Jollivet et al., 1993; Larsen, 1998). Esters are key cheese cheeses until day 60 was 2,3-dimethylpyrazine (Supplementary
odorants, which contribute directly to flavour with their fruity, Table S7). However, 2,5-dimethylpyrazine and 2,6-
floral and sweet notes of low perception thresholds, and indirectly dimethylpyrazine, followed by trimethylpyrazine and methylpyr-
by masking unclean off-flavours which might be present (Curioni & azine, predominated on day 120 in control cheese, while 2,3-
Bosset, 2002). The only cyclic ester found in the present work, g- dimethylpyrazine and 2,6-dimethylpyrazine were the most
abundant pyrazines in HP-treated cheeses. Seven different pyr- ethanoate, 3-methylbutanal, 3-methylbutanol, dimethyldisulphide,
azines were detected in Parmigiano-Reggiano cheese, which styrene, ethyl butanoate, ethyl hexanoate, 2-pentanone and 2-
contributed to its nutty, roasted aroma (Qian & Reineccius, 2002). octanone) were associated by the authors with P. camemberti,
Pyrazines, some of which exhibit earthy, musty, soil odour notes, while ethyl ethanoate and butyl ethanoate were associated with
have also been found in Camembert, Cheddar, Gruye re and Man- Kluyveromyces lactis and 2-heptanone with Geotrichum candidum.
chego cheeses (Curioni & Bosset, 2002; Ferna ndez-García, Ten of those 12 compounds were found in the present study; ex-
Carbonell, & Nun ~ ez, 2002; Mallia, Fernandez-García, & Bosset, ceptions were butyl ethanoate and styrene.
2005b). Principal component analysis was performed on 80 individual
Total amines were present on day 30 at significantly (P < 0.05) volatile compounds, selected on the basis of their high statistical
lower levels in all the HP-treated cheeses than in control cheese, while significance in the ANOVA. Function 1 (PC1), formed by 14 ketones,
on day 60 only the cheeses treated at 600 MPa had lower levels than 8 pyrazines, 7 alcohols, 5 sulphur compounds, propanoic acid,
control cheese (Table 2). From day 60 to day 120, total amines hexanal, trimethylamine and toluene, and with a negative coeffi-
increased 11-fold in control cheese and tended to decline in HP- cient by 3-methylhexane, explained 39.4% of the variance. Function
treated cheeses. Up to 15 amines have been reported in Camembert 2 (PC2), formed by 8 alcohols, 6 volatile acids, 5 esters, 4 hydro-
cheeses of different origins (Molimard & Spinnler, 1996), but only two carbons, 3 aldehydes, 3-methylthio-1-propanol and 1-ethoxy-2-
amines were detected in the present study. They were 4-purinamine, propanol, and with a negative coefficient by 3 ethers and naph-
present in all cheeses from day 30, and trimethylamine, found only in thalene, explained 19.0% of the variance.
120-day-old control cheese, in which it represented more than 90% of Fig. 3 depicts the groups obtained when control and HP-treated
total amines (data not shown). Trimethylamine has a strong fishy cheeses of different ages were plotted against functions 1 and 2.
odour (Molimard & Spinnler, 1996), but no characteristic odour notes Control cheese at days 60 and 120 and 400W3 cheese at day 120
have been reported for 4-purinamine. were clearly separated between them and from the rest of cheeses.
Total ethers remained stable in control cheese from day 30 to The 120-day-old control cheese, located at the right side of x-axis,
day 120, while they increased in HP-treated cheeses by 33e78% was characterised by high amounts of sulphur compounds, amines,
(Table 2). Their relative abundances were at significantly (P < 0.05) pyrazines, ketones and some alcohols, compounds with sulphu-
higher levels in all the HP-treated cheeses than in control cheese. rous, boiled cabbage, garlic, pungent, fruity, fishy and ammonia
Eight alkylealkyl ethers, with 1-methoxy-2-propanol and 1- unpleasant flavour notes. The 60-day-old control cheese, located at
ethoxy-2-propanol being the major ones (data not shown), were the left side of x-axis, was characterised by its low content of al-
detected in the present study, a higher number than the 3 ethers cohols, volatile acids and esters. The 120-day-old 400W3 cheese,
found in smoked fresh goat cheese (Guille n, Ibargoitia, Sopelana, located at the upper side of y axis, was characterised by its high
Palencia, & Fresno, 2004) or in raw ewes milk cheese (Calzada levels of volatile acids, alcohols, esters, aldehydes and hydrocar-
et al., 2014a). Ethers are components of smoke, but their origin in bons, compounds which exhibit vinegar, pungent, sour, musty,
non-smoked cheeses is not known. Although mild, pleasant, sweet green, alcohol, sweet, solvent and ethereal flavour notes. The rest of
and pungent odour notes have been ascribed to ethers, they are not cheeses, located near the origin, showed a balanced flavour profile
considered to be key odorants of mould-ripened cheeses (Curioni & with cheesy notes, and most of them were grouped with the 30-
Bosset, 2002; Sable & Cottenceau, 1999). day-old control cheese.
Total hydrocarbons hardly varied with HPP or with time of
refrigerated storage (Table 2). Major hydrocarbons in control and 3.3. Odour and colour
HP-treated Brie cheeses were hexane, octane and undecane (data
not shown). Some hydrocarbons are already present in milk (Toso Odour quality (preference) scores did not show significant dif-
et al., 2002) but hydrocarbons found in the volatile fraction of ferences between HP-treated and control cheeses from day 21 to
cheese mostly originate from lipid oxidation (Carbonell, Nun ~ ez, &
Ferna ndez-García, 2002). They are not considered to be key odor-
ants in cheese (Thierry, Maillard, & Le Que re
, 1999).
Total benzenic compounds did not vary from day 30 to day 60,
and increased only in control cheese from day 60 to day 120
(Table 2). Main benzenic compounds were nitronaphthalene and
toluene, which levels doubled in control cheese from day 30 to day
120, followed by xylene (data not shown). Some benzenic com-
pounds are formed through the microbial catabolism of aromatic
amino acids (Yvon & Rijnen, 2001). P. caseifulvum strains produce
higher amounts of benzenic compounds than P. camemberti strains
(Jollivet et al., 1993; Larsen, 1998), but none of the strains tested by
those authors was found to produce nitronaphthalene, toluene or
xylene. Naphthalene has been detected in Emmental, La Serena and
Zamorano cheeses (Carbonell et al., 2002; Ferna ndez-García,
Carbonell, Gaya, & Nun ~ ez, 2004; Thierry et al., 1999) but, to our
knowledge, the presence of nitronaphthalene in cheese had not
been reported. Toluene and xylene, frequently found in the volatile
fraction of some cheese varieties (Ferna ndez-García et al., 2004),
are not recognised as important contributors to cheese aroma.
Leclercq-Perlat et al. (2007) selected 12 volatile compounds as Fig. 3. Distribution of HPP and control Brie cheeses on the plane defined by functions 1
markers of the degradation of pyruvate, leucine, phenylalanine, and and 2 of principal component analysis. Each symbol represents the averaged value of
triplicate determinations on two batches of cheese. Treatments are as follows: control
methionine, the terminal oxidation of FFAs, and the intra-chain (C), 400W2 (open circle), 600W2 (black circle), 400W3 (open triangle), 600W3 (black
oxidation of unsaturated FFAs during ripening of Camembert- triangle). Days after manufacture are as follows: 30 days, 1 m; 60 days, 2 m; 120 days,
type cheeses. Nine of these compounds (3-methylbutyl 4 m.
day 60 (Supplementary Fig. S2). On day 90, the odour quality scores impaired by HPP, which brought about less lightness and a slightly
of all the HP-treated cheeses, ranging from 5.73 to 6.11, were more reddish and yellowish colour, changes that might diminish
significantly (P < 0.05) higher than the score 4.02 of control cheese, consumer acceptance.
and by day 120 the difference had increased, with scores of
4.25e4.89 for HP-treated cheeses versus a score of 1.36 for control Acknowledgements
cheese. There were no significant differences in odour intensity
scores (Supplementary Fig. S2) between HP-treated and control Funding from project AGL2009-07801 from the Ministry of
cheeses from day 21 to day 90. However, on day 120, control cheese Science and Innovation (MICINN, Madrid, Spain) is acknowledged
showed a higher (P < 0.05) odour intensity score, 7.90, than all the by the authors. J. Calzada was the recipient of MICINN fellowship
HP-treated cheeses, with scores in the range 6.42e6.81. Differences BES-2010-030444. The authors are grateful to ILAS S.A. (Madrid,
in the odour characteristics of control and HP-treated cheeses on Spain) for providing Brie cheeses and to Hiperbaric (Burgos, Spain)
day 120 can be associated with the marked increase in the levels of for valuable help with HPP.
volatile compounds with unpleasant aroma notes, in particular
sulphurous compounds, pyrazines and amines, in control cheese.
Appendix A. Supplementary data
At the core of Brie cheeses, colour parameters L*, a* and b* values
(data not shown) showed minor differences between cheeses from
Supplementary data related to this article can be found at http://
day 30 to day 90. On day 120, L* value reached a significantly
dx.doi.org/10.1016/j.idairyj.2014.07.007.
(P < 0.05) lower value, 75.92, in control cheese (less lightness) than
in HP-treated cheeses, in which it ranged from 82.11 to 84.60. The a*
value of control cheese, 2.73, was significantly (P < 0.05) higher References
(more reddish) than those of HP-treated cheeses, which ranged
Adda, J., & Dumont, J. P. (1974). Les substances responsables de l'aro ^me des from-
from 0.64 to 1.65. Similarly, the b* value of control cheese, 22.63, ages a pa^te molle. Lait, 74, 1e21.
was significantly (P < 0.05) higher (more yellowish) than those of Calzada, J., del Olmo, A., Picon, A., Gaya, P., & Nun~ ez, M. (2013a). Reducing biogenic-
HP-treated cheeses, which ranged from 19.00 to 20.96. amine-producing bacteria, decarboxylase activity, and biogenic amines in raw
milk cheese by high-pressure treatments. Applied and Environmental Microbi-
Colour parameters at the rind of Brie cheese (data not shown) did ology, 79, 1277e1283.
not evolve as in the core; significant (P < 0.05) differences between Calzada, J., del Olmo, A., Picon, A., Gaya, P., & Nun ~ ez, M. (2013b). High-pressure
the colour parameters of control and HP-treated cheeses were found processing decelerates lipolysis and formation of volatile compounds in ovine
in most cases. On day 30, L*, a* and b* values were 91.34, 0.12 and 6.80 milk blue-veined cheese. Journal of Dairy Science, 96, 7500e7510.
~ ez, M. (2014a). High-pressure
Calzada, J., del Olmo, A., Picon, A., Gaya, P., & Nun
in control cheese, respectively, while L* values ranged from 77.10 to processing for the control of lipolysis, volatile compounds and off-odours in
80.42, a* values from 1.63 to 3.44, and b* values from 18.44 to 21.26 in raw milk cheese. Food and Bioprocess Technology, 7, 2207e2217.
Calzada, J., del Olmo, A., Picon, A., Gaya, P., & Nun ~ ez, M. (2014b). Effect of high-
HP-treated cheeses. At that time, the visual appearance of the rind of
pressure-processing on the microbiology, proteolysis, texture and flavour of
HP-treated cheeses was of lower lightness, together with a slightly Brie cheese during ripening and refrigerated storage. International Dairy Journal,
more reddish and yellowish colour, than that of control cheese. These 37, 64e73.
Carbonell, M., Nun ~ ez, M., & Fernandez-García, E. (2002). Evolution of the volatile
changes were associated with impregnation of the mould mycelium
components of ewe raw milk La Serena cheese during ripening. Correlation
at the rind of HP-treated cheeses by water and fat from the interior, with flavor characteristics. Lait, 82, 683e698.
resulting in the loss of its typical white mossy aspect, which might Collins, Y. F., McSweeney, P. L. H., & Wilkinson, M. G. (2003). Lipolysis and free fatty
diminish consumer acceptance. During the refrigerated storage acid catabolism in cheese: a review of current knowledge. International Dairy
Journal, 13, 841e866.
period, instrumental colour parameters and visual appearance var- Curioni, P. M. G., & Bosset, J. O. (2002). Key odorants in various cheese types as
ied on the surface of control cheese, most probably influenced by the determined by gas chromatographyeolfactometry. International Dairy Journal,
growth of red-orange pigmented microorganisms. By day 120, L* 12, 959e984.
value had decreased to 77.66, a* value had increased to 2.68 and b* Engels, W. J. M., Dekker, R., de Jong, C., Neeter, R., & Visser, S. (1997). A comparative
study of volatile compounds in the water-soluble fraction of various types of
value had increased to 12.01 at the rind of control cheese. Changes in ripened cheese. International Dairy Journal, 7, 255e263.
the colour parameters and visual appearance of the surface of HP- Fern andez-García, E., Carbonell, M., Calzada, J., & Nun ~ ez, M. (2006). Seasonal vari-
treated cheeses were less marked than in control cheese. Thus, L* ation of the free fatty acids contents of Spanish ovine milk cheeses protected by
a designation of origin: a comparative study. International Dairy Journal, 16,
values ranged from 76.68 to 80.52 on day 120, and a* values from 1.26 252e261.
to 3.08, close to the corresponding values of HP-treated cheeses on Fern andez-García, E., Carbonell, M., Gaya, P., & Nun ~ ez, M. (2004). Evolution of the
day 30. In contrast, b* values of HP-treated cheeses increased to levels volatile components of ewes raw milk Zamorano cheese. Seasonal variation.
International Dairy Journal, 14, 701e711.
ranging from 25.12 to 29.36 on day 120, a change which was asso- Fern andez-García, E., Carbonell, M., & Nun ~ ez, M. (2002). Volatile fraction and sen-
ciated with the growth of yellow-pigmented microorganisms. sory characteristics of Manchego cheese. 1. Comparison of raw and pasteurized
milk cheese. Journal of Dairy Research, 69, 579e593.
Guillen, M. D., Ibargoitia, M. L., Sopelana, P., Palencia, G., & Fresno, M. (2004).
4. Conclusions Components detected by means of solid-phase microextraction and gas chro-
matography/mass spectrometry in the headspace of artisan fresh goat cheese
High-pressure processing modified the accumulation pattern of smoked by traditional methods. Journal of Dairy Science, 87, 284e299.
Huppertz, Y., Fox, P. F., & Kelly, A. L. (2004). Susceptibility of plasmin and chymosin
FFAs and volatile compounds in Brie cheese during ripening and in Cheddar cheese to inactivation by high pressure. Journal of Dairy Research, 71,
refrigerated storage. Lipolysis was retarded by HPP, although long- 496e499.
chain FFAs reached higher concentrations in some of the HP- Jollivet, N., Belin, J.-M., & Vayssier, Y. (1993). Comparison of volatile flavor com-
pounds produced by ten strains of Penicillium camemberti Thom. Journal of
treated cheeses than in control cheese until day 60. Alcohols, al-
Dairy Science, 76, 1837e1844.
dehydes, acids, esters and ethers attained higher levels in HP- Kubí ckova, J., & Grosch, W. (1997). Evaluation of potent odorants of Camembert
treated cheeses than in control cheese. Other groups of volatile cheese by dilution and concentration techniques. International Dairy Journal, 7,
65e70.
compounds, such as sulphur compounds, pyrazines and amines,
Kubí ckova, J., & Grosch, W. (1998). Evaluation of flavor compounds of Camembert
increased drastically in control cheese from day 60 to day 120, cheese. International Dairy Journal, 8, 11e16.
which resulted in strong unpleasant odour notes. The odour quality Lamberet, G., & Lenoir, J. (1976). Les caracte res du syste me lipolytique de l'espece
Penicillium caseicolum. Purification et proprie tes de la lipase majeure. Lait, 56,
of Brie cheese was improved by HPP, which yielded a more
622e644.
balanced and pleasant aroma profile and delayed over-ripening. Larsen, T. O. (1998). Volatile flavour production by Penicillium caseifulvum. Inter-
However, the external appearance of Brie cheese became national Dairy Journal, 8, 883e887.
Leclercq-Perlat, M.-N., Corrieu, G., & Spinnler, H.-E. (2007). Controlled production of Qian, M., & Reineccius, G. (2002). Identification of aroma compounds in
Camembert-type cheeses: part III role of the ripening microflora on free fatty Parmigiano-Reggiano cheese by gas chromatography/olfactometry. Journal of
acid concentrations. Journal of Dairy Research, 74, 218e225. Dairy Science, 85, 1362e1369.
Leclercq-Perlat, M.-N., Latrille, E., Corrieu, G., & Spinnler, H.-E. (2004). Controlled Rivas-Can ~ edo, A., Juez-Ojeda, C., Nun
~ ez, M., & Ferna ndez-García, E. (2012). Volatile
production of Camembert-type cheese. Part II. Changes in the concentration of compounds in low-acid fermented sausage “espetec” and sliced cooked pork
the more volatile compounds. Journal of Dairy Research, 71, 355e366. shoulder subjected to high pressure processing. A comparison of dynamic
Liu, S.-Q., Holland, R., & Crow, V. L. (2004). Esters and their biosynthesis in fer- headspace and solid-phase microextraction. Food Chemistry, 132, 18e26.
mented dairy products: a review. International Dairy Journal, 14, 923e945. , S., & Cottenceau, G. (1999). Current knowledge of soft cheeses flavor and
Sable
Mallia, S., Carpino, S., Corallo, L., Tuminello, L., Gelsomino, R., & Licitra, G. (2005a). related compounds. Journal of Agricultural and Food Chemistry, 47, 4825e4836.
Effects on aroma profile of Piacentinu and Ricotta cheese using different tool Thierry, A., Maillard, M.-B., & Le Que re
, J.-L. (1999). Dynamic headspace analysis of
materials during cheesemaking. Royal Society of Chemistry Special Publications, Emmental aqueous phase as a method to quantify changes in volatile flavour
300, 23e34. compounds during ripening. International Dairy Journal, 9, 453e463.
Mallia, S., Fernandez-García, E., & Bosset, J. O. (2005b). Comparison of purge and Toso, B., Procida, G., & Stefanon, B. (2002). Determination of volatile compounds in
trap and solid phase microextraction techniques for studying the volatile aroma cows' milk using headspace GCeMS. Journal of Dairy Research, 69, 569e577.
compounds of three European PDO hard cheeses. International Dairy Journal, 15, Trieu-Cuot, P., Archieri-Haze, M.-J., & Gripon, J.-C. (1982a). Effect of aspartyl pro-
741e758. teinases of Penicillium caseicolum and Penicillium roqueforti on caseins. Journal
Martínez-Rodríguez, Y., Acosta-Mun ~ iz, C., Olivas, G. I., Guerrero-Beltran, J., Rodrigo- of Dairy Research, 49, 487e500.
Aliaga, D., & Sepúlveda, D. R. (2012). High hydrostatic pressure processing of Trieu-Cuot, P., Archieri-Haze, M.-J., & Gripon, J.-C. (1982b). Comparative study of the
cheese. Comprehensive Reviews in Food Science and Food Safety, 11, 399e416. action of metalloproteinases of Penicillium caseicolum and Penicillium roqueforti
Molimard, P., & Spinnler, H. E. (1996). Review: compounds involved in the flavor of on as1- and b-caseins. Lait, 62, 234e249.
surface mold-ripened cheeses: origins and properties. Journal of Dairy Science, Urbach, G. (1990). Effect of feed on flavor in dairy foods. Journal of Dairy Science, 73,
79, 169e184. 3639e3650.
Okumura, J., & Kinsella, J. E. (1985). Methyl ketone formation by Penicillium cam- Urbach, G. (1993). Relations between cheese flavour and chemical composition.
emberti in model systems. Journal of Dairy Science, 68, 11e15. International Dairy Journal, 3, 389e422.
O'Reilly, C. E., O'Connor, P. M., Kelly, A. L., Beresford, T. P., & Murphy, P. M. (2000). Voigt, D. D., Chevalier, F., Qian, M. C., & Kelly, A. L. (2010). Effect of high-pressure
Use of hydrostatic pressure for inactivation of microbial contaminants in treatment on microbiology, proteolysis, lipolysis and levels of flavour com-
cheese. Applied and Environmental Microbiology, 66, 4890e4896. pounds in mature blue-veined cheese. Innovative Food Science and Emerging
Picon, A., Alonso, R., van Wely, K. H. M., & Nun ~ ez, M. (2013). Microstructural, Technologies, 11, 68e77.
textural and colour characteristics during ripening of Hispa nico cheese made Woo, A. H., Kollodge, S., & Lindsay, R. C. (1984). Quantification of major free fatty
using high-pressure-treated ovine milk curd. Food and Bioprocess Technology, 6, acids in several cheese varieties. Journal of Dairy Science, 67, 874e878.
3056e3067. Yvon, M., & Rijnen, L. (2001). Cheese flavour formation by amino acid catabolism.
Pionnier, E., Engel, E., Salles, C., & Le Que re, J. L. (2002). Interactions between non- International Dairy Journal, 11, 185e201.
volatile water-soluble molecules and aroma compounds in Camembert cheese.
Food Chemistry, 76, 13e20.
50
45
pmol D -naphthol min-1 g-1 cheese
40
35
30
25
20
15
10
0
1d 14 d 21 d 30 d 60 d 90 d 120 d
Storage time
Storage time
Fig. S1. Esterase activity during ripening and refrigerated storage of control Brie cheese (black)
and cheeses HP-treated at 400 MPa on day 14 (white), 600 MPa on day 14 (horizontally striped),
400 MPa on day 21 (dotted) or 600 MPa on day 21 (obliquely striped). Bars indicate standard
error of the means.
10
A
9
7
Odour quality
0
21 d 30 d 60 d 90 d 120 d
10
B
9
7
Odour intensity
0
21 d 30 d 60 d 90 d 120 d
Fig. S2. Odour characteristics (A, quality; B, intensity) (scores from 16 panellists on a 0-10
point scale) during ripening and refrigerated storage of control Brie cheese (black) and cheeses
HP-treated at 400 MPa on day 14 (white), 600 MPa on day 14 (horizontally striped), 400 MPa
on day 21 (dotted) or 600 MPa on day 21 (obliquely striped). Bars indicate standard error of the
means.
Table S1
Principal aldehydes in the volatile fraction during refrigerated storage of Brie control cheese and
cheeses HP-treated at 400 or 600 MPa after 14 days (W2) or 21 days (W3) of ripening.
Compound a Days Control cheese 400W2 600W2 400W3 600W3
Ethanal 30 1.13±0.17a 1.59±0.14ab 2.15±0.14b 1.61±0.13ab 1.34±0.11a

60 2.28±0.13bc 1.32±0.10a 1.70±0.15ab 1.65±0.07a 2.31±0.23c
120 2.16±0.19a 1.43±0.10a 1.72±0.06a 1.85±0.18a 3.41±0.43b
2-Methylpropanal 30 0.44±0.04 a 1.29±0.17c 1.06±0.16bc 0.87±0.10abc 0.59±0.12ab

60 0.60±0.04a 1.83±0.13b 1.87±0.13b 1.92±0.08b 3.27±0.34c
120 4.17±0.55a 3.21±0.28a 2.90±0.22a 3.10±0.30a 5.94±0.45b
2-Methylbutanal 30 0.28±0.06a 1.89±0.20c 0.97±0.21b 0.49±0.05ab 0.50±0.06ab

60 1.45±0.10a 3.01±0.42b 1.81±0.11ab 2.99±0.24b 4.57±0.42c
120 5.33±0.32a 4.24±0.44a 5.05±0.62a 4.78±0.48a 11.99±1.83b
3-Methylbutanal 30 2.77±0.24a 11.37±0.99c 6.15±0.41b 3.90±0.41ab 3.61±0.46a

60 2.62±0.17a 11.80±0.85b 12.67±0.94b 11.98±0.80b 19.82±1.51c
120 25.16±3.26a 31.96±2.32a 24.32±2.28a 23.75±2.90a 51.23±3.29b
Heptanal 30 0.54±0.04b 0.39±0.03a 0.39±0.03a 0.28±0.02a 0.30±0.04a

60 1.04±0.19b 0.36±0.03a 0.40±0.03a 0.38±0.03a 0.37±0.04a
120 1.22±0.10b 0.50±0.03a 0.50±0.06a 0.49±0.03a 0.58±0.05a
a
Volatile compounds are expressed as relative abundances on the internal standard (× 103); values are means ± SE
(n = 6) of triplicate determinations in two cheese-making experiments. Means on the same row followed by
different superscripts differ significantly (P <0.05).
Table S2
Principal ketones in the volatile fraction during refrigerated storage of Brie control cheese and cheeses
HP-treated at 400 or 600 MPa after 14 days (W2) or 21 days (W3) of ripening.
2-Propanone 30 61.25±9.26b 73.75±7.79b 66.63±9.88b 28.54±4.40a 70.01±6.51b

60 192.54±21.33c 27.58±4.51ab 78.64±7.61b 13.68±1.52a 136.69±20.84c
120 780.47±82.88b 28.43±1.11a 62.59±5.80a 14.74±2.94a 82.24±6.20a
3-Hydroxybutanone 30 305.36±50.96ab 1123.51±87.20c 424.48±37.54b 158.37±16.70a 296.70±11.07ab

60 7.26±0.37a 110.02±19.49b 442.42±24.43c 93.07±4.90b 710.37±27.84d
120 31.38±3.66a 158.32±5.72b 322.49±33.57c 241.85±26.93bc
304.32±27.91c
2-Pentanone 30 218.45±33.86b 185.30±22.10b 71.43±4.34a 96.72±6.91a 87.00±5.01a

60 60.98±5.53ab 61.98±2.76ab 84.08±4.74b 32.51±1.54a 248.29±26.26c
120 1311.73±107.50b 49.87±3.11a 77.65±6.82a 28.97±3.04a 140.28±9.62a
2-Heptanone 30 104.14±10.46b 31.46±4.19a 24.29±3.24a 38.15±3.66a 47.26±3.77a

60 35.80±4.44ab 31.34±2.32ab 39.26±5.92b 15.81±1.63a 102.59±9.79c
120 330.22±50.80b 39.51±4.05a 58.43±4.80a 25.75±4.36a 100.83±9.62a
2-Nonanone 30 18.08±2.32b 2.46±0.30a 1.01±0.12a 4.79±0.39a 2.70±0.31a

60 9.19±0.35c 2.27±0.36a 2.41±0.30a 1.16±0.22a 6.73±0.82b
120 104.44±13.36b 4.47±0.42a 4.23±0.39a 3.18±0.49a 6.44±0.52a
a
Table S3
Principal alcohols in the volatile fraction during refrigerated storage of Brie control cheese and cheeses
Ethanol 30 73.09±5.48a 433.03±25.74b 508.48±27.42b 507.99±48.02b 388.45±36.24b

60 112.89±7.01a 574.18±55.09c 687.52±21.17cd 734.91±43.66d 266.33±24.84b
120 57.65±7.25a 672.25±24.38c 623.03±26.95c 759.53±49.59c 404.99±56.86b
2-Propanol 30 23.22±2.52a 21.72±2.25a 30.32±2.69a 48.98±5.37b 23.17±2.14a

60 140.35±6.24c 69.25±5.50b 54.26±3.30b 71.42±10.7b 18.97±2.60a
120 116.15±14.60c 60.52±4.58b 23.19±3.61a 69.28±3.02b 12.32±1.17a
2-Butanol 30 5.54±0.91a 18.66±1.12bc 27.49±3.84c 42.61±3.46d 16.59±0.83b

60 9.86±0.31a 99.42±6.67c 38.92±4.41b 49.20±2.26b 9.85±0.49a
120 20.77±2.79ab 162.11±16.38c 23.66±2.07ab 48.00±4.73b 9.07±1.13a
3-Methyl-1-butanol 30 28.11±3.66a 112.67±10.54d 88.73±4.24cd 66.71±6.67bc 51.79±4.81ab

60 1.45±0.12a 134.72±8.88c 73.68±4.27b 233.72±21.71d 111.88±17.20bc
120 16.48±1.44a 96.93±6.38b 77.12±7.76b 181.61±21.91c 69.87±3.50b
2-Pentanol 30 22.55±1.43b 3.27±0.46a 3.96±0.31a 24.32±3.63b 9.21±0.83a

60 20.68±1.94b 21.71±2.20b 6.12±0.51a 43.39±2.44c 5.40±0.46a
120 77.03±5.11d 29.09±2.15b 5.93±0.65a 56.61±3.81c 6.82±0.31a
2-Heptanol 30 6.43±0.81a 4.96±0.41a 6.67±0.49a 17.53±1.95b 6.03±0.66a

60 7.73±1.52a 15.78±2.15b 7.16±0.66a 19.74±1.12b 2.45±0.19a
120 15.92±1.35b 15.39±0.67b 4.97±0.67a 30.51±4.47c 2.78±0.43a
a
Table S4
Principal acids in the volatile fraction during refrigerated storage of Brie control cheese and cheeses
Ethanoic 30 29.81±3.42a 378.10±25.82d 163.46±14.63c 108.53±10.98bc 74.60±10.07ab

60 15.36±1.19a 462.90±38.25b 215.91±15.69b 367.32±21.72b 76.42±4.11a
120 24.86±2.98a 466.96±50.90c 197.46±15.24b 576.72±28.39c 93.08±7.31ab
Butanoic 30 621.38±46.96ab 982.32±35.37c 791.07±44.08bc 641.40±71.22ab 545.21±24.43a

60 31.55±3.94a 1458.99±95.26d 1280.18±24.67cd 1099.37±24.48bc 1057.54±15.97b
120 23.47±2.88a 1945.51±57.46b 1838.89±51.45b 1969.89±79.99b 2062.24±217.36b
3-Methyl butanoic 30 14.88±2.71a 53.14±2.41c 38.64±3.65bc 43.91±5.91bc 31.90±5.24ab

60 0.73±0.08a 155.01±23.24ab 43.72±3.61ab 104.90±13.62ab 168.55±25.41b
120 15.92±2.99a 78.14±10.67a 82.43±8.17a 273.24±52.17b 103.31±29.79a
Pentanoic 30 8.16±0.60ab 14.55±0.84c 10.91±1.16b 8.78±0.99ab 7.16±0.67a

60 1.12±0.19a 22.41±2.01c 18.70±0.26ab 15.89±0.59b 16.26±0.47b
120 0.74±0.08a 30.12±1.02b 31.48±2.68b 35.25±2.02b 45.99±3.26c
Hexanoic 30 110.60±9.64a 236.65±52.84a 249.84±63.42a 135.08±15.84a 158.99±13.37a

60 15.99±2.20a 477.82±75.15c 308.82±31.61bc 290.46±42.54b 218.84±32.24b
120 10.23±1.68a 743.38±67.81bc 597.77±40.06b 815.14±73.61bc 880.67±49.45c
a
Table S5
Principal esters in the volatile fraction during refrigerated storage of Brie control cheese and cheeses

Ethyl ethanoate 30 3.63±0.41a 91.80±16.11b 248.33±27.08c 53.14±5.26ab 66.74±5.84b
60 4.38±0.47a 125.88±8.91c 262.12±15.51d 53.32±2.93b 145.94±8.44c
120 4.09±0.47a 109.19±6.17c 180.40±14.17d 61.16±5.19b 105.53±5.47c
Ethyl butanoate 30 1.32±0.11a 3.49±0.27c 3.05±0.38c 2.49±0.38bc 1.37±0.06ab

60 1.59±0.20a 8.58±0.97c 6.97±0.21c 4.58±0.40b 2.18±0.08a
120 1.28±0.14a 15.66±0.92cd 14.18±0.85c 18.12±1.45d 6.49±0.76b
Ethyl hexanoate 30 0.52±0.05a 1.89±0.20c 1.49±0.12bc 1.26±0.16b 1.27±0.11b

60 0.45±0.06a 4.32±0.49d 3.94±0.64cd 2.75±0.21bc 2.13±0.02b
120 0.47±0.03a 13.47±1.44c 6.41±0.54b 18.24±1.20d 5.68±0.54b
Methylpropyl propanoate 30 4.26±0.34a 4.37±0.31a 4.32±0.25a 3.74±0.34a 3.37±0.43a

60 4.32±0.34b 3.21±0.22ab 2.52±0.18a 3.52±0.38ab 3.25±0.27ab
120 3.12±0.35a 2.58±0.26a 2.46±0.42a 2.99±0.40a 3.17±0.17a
Ȗ-Butyrolactone 30 4.45±0.63a 4.27±0.54a 3.63±0.52a 3.52±0.48a 3.26±0.27a

60 3.53±0.34a 3.24±0.27a 3.55±0.32a 3.43±0.10a 2.96±0.27a
120 3.17±0.07a 2.77±0.19a 3.02±0.26a 2.89±0.25a 2.78±0.23a
a
Table S6
Principal sulphur compounds in the volatile fraction during refrigerated storage of Brie control cheese
and cheeses HP-treated at 400 or 600 MPa after 14 days (W2) or 21 days (W3) of ripening.
Compounds a Days Control cheese 400W2 600W2 400W3 600W3
Dimethylsulphide 30 0.40±0.08a 0.70±0.04a 0.68±0.08a 0.55±0.05a 0.64±0.10a

60 1.75±0.26b 0.92±0.12a 0.87±0.12a 0.63±0.07a 0.61±0.07a
120 2.69±0.42b 0.93±0.13a 0.82±0.11a 0.53±0.07a 0.66±0.10a
Dimethyldisulphide 30 2.57±0.46a 2.23±0.26a 2.04±0.27a 2.31±0.13a 2.07±0.17a

60 2.07±0.32a 1.89±0.22a 2.29±0.34a 1.81±0.28a 1.78±0.19a
120 212.99±37.94b 1.87±0.25a 3.15±0.55a 2.26±0.33a 1.63±0.29a
Dimethyltrisulphide 30 0.12±0.01 ND ND ND ND
60 0.22±0.03a 0.16±0.03a ND ND ND
120 4.00±0.28b 0.23±0.03a 0.05±0.02a 0.06±0.03a 0.07±0.03a
3-Penthanethiol 30 1.01±0.06a 1.35±0.12ab 1.13±0.06a 1.76±0.15b 1.24±0.17a

60 1.12±0.15a 1.16±0.15a 1.12±0.08a 1.55±0.19a 1.18±0.13a
120 1.23±0.16a 1.34±0.08a 1.04±0.13a 1.21±0.07a 1.28±0.18a
Dimethylsulphone 30 4.09±0.27a 4.83±0.50a 4.32±0.62a 3.92±0.37a 4.17±0.48a

60 3.89±0.33a 2.77±0.38a 4.25±0.47a 3.09±0.24a 3.34±0.37a
120 5.91±0.78b 2.59±0.35a 2.10±0.38a 2.53±0.34a 2.12±0.35a
*
(n = 6) of triplicate determinations in two cheese-making experiments (ND, not detected). Means on the same row
followed by different superscripts differ significantly (P <0.05).
Table S7
Principal pyrazines in the volatile fraction during refrigerated storage of Brie control cheese and cheeses
Compounds a Days Control cheese 400W2 600W2 400W3 600W3

Methylpyrazine 30 ND ND ND ND ND
60 0.26±0.04 ND ND ND ND
120 21.29±2.61b 0.06±0.01a 0.12±0.01a 0.22±0.03a 0.33±0.07a
2,3-Dimethylpyrazine 30 1.31±0.15a 1.55±0.22a 1.57±0.12a 1.41±0.09a 1.24±0.13a

60 1.10±0.08a 1.60±0.15ab 1.95±0.12b 1.97±0.13b 1.63±0.22ab
120 1.08±0.08a 2.49±0.32c 1.48±0.11ab 2.02±0.15bc 1.30±0.12ab
2,5-Dimethylpyrazine 30 ND ND ND ND ND
60 ND ND ND ND ND
120 193.13±16.57 b ND ND ND 0.16±0.02a
2,6-Dimethylpyrazine 30 ND 0.41±0.04b 0.36±0.03ab 0.32±0.04ab 0.23±0.03a

60 0.69±0.07a 0.61±0.05a ND 2.04±0.15b 2.18±0.14b
120 144.98±9.47b 1.18±0.10a 0.84±0.10a 2.50±0.29a 4.13±0.67a
Trimethylpyrazine 30 0.06±0.02a 0.21±0.03b ND ND ND

60 0.10±0.01a 0.36±0.03b ND 0.21±0.02a 0.44±0.06b
120 40.11±5.10b 0.41±0.03a ND 0.50±0.04a 0.87±0.27a
a
(n = 6) of triplicate determinations in two cheese-making experiments (ND, not detected). Means on the same row
followed by different superscripts differ significantly (P <0.05).

Capítulo 9.
Effect of high-pressure processing on the microbiology,
proteolysis, biogenic amines and flavour of cheese made from
unpasteurized milk.

185

Fotografía: cortes de queso Arzúa-Ulloa control (no presurizado) a día 1 (superior,

izquierda), 14 (superior derecha), 21 (centro, izquierda), 60 (centro, derecha), 120
(inferior, izquierda) y 240 (inferior, derecha).
Food Bioprocess Technol
DOI 10.1007/s11947-014-1406-7
ORIGINAL PAPER
Effect of High-Pressure Processing on the Microbiology,

Proteolysis, Biogenic Amines and Flavour of Cheese Made
from Unpasteurized Milk
Received: 4 June 2014 / Accepted: 4 September 2014

Abstract Cheese varieties with long ripening periods are Introduction

prone to form biogenic amines and develop off-flavours.
High-pressure processing (HPP), which inactivates microor- Cheese ripening is a complex phenomenon in which indige-
ganisms and enzymes, may be useful in preventing those nous milk enzymes, enzymes from the coagulant, other added
defects. On this aim, cheeses made from unpasteurized milk or in situ produced enzymes, lactic acid bacteria (LAB) from
were treated at 400 or 600 MPa, after 14 or 21 days of starter cultures and other indigenous, added or contaminating
ripening, and their characteristics were compared to those of microorganisms are involved. The importance of their respec-
untreated control cheese throughout a 240-day period. Lactic tive roles in the degradation of milk constituents retained in
acid bacteria declined by 2 log units in 400 MPa cheeses and the curd, mostly proteins and fat, varies among the different
by 6 log units in 600 MPa cheeses after HPP, while Gram- cheese varieties, depending on factors such as milk heat
negative bacteria were below detection level in all the HPP- treatment, type of starter or manufacturing and ripening
treated cheeses. Aminopeptidase activity was significantly parameters.
lower in HPP cheeses than in control cheese from day 21 Milk proteins, mostly caseins, are hydrolyzed by the pro-
onwards. Hydrolysis of αs-casein was enhanced in 400 MPa teolytic enzymes present in cheese. Weakening of the casein
cheeses, but not in 600 MPa cheeses, while a more pro- network during manufacture and ripening contributes to the
nounced hydrolysis of β-, κ- and para-κ-caseins was recorded development of cheese texture. A wide range of peptides are
in all the HPP cheeses from day 60 onwards. Levels of generated through the hydrolysis of caseins, most of which
hydrophilic and hydrophobic peptides were higher in HPP positively influence cheese flavour, although some of the β-
cheeses than in control cheese on day 60 and thereafter. casein-derived peptides are known to be bitter (Sousa et al.
Total free amino acids were at lower concentrations in 2001). Free amino acids (FAA) contribute to cheese flavour
600 MPa cheeses than in the rest from day 60 onwards. The by themselves and indirectly through the compounds resulting
concentration of total biogenic amines was lower in all the from their catabolism, many of which have higher flavour and
HPP cheeses than in control cheese from day 60 onwards. aroma impact than their respective precursors (Yvon and
Flavour quality and flavour intensity of cheese made from Rijnen 2001). Cheese ripening continues during distribution
unpasteurized milk were not significantly affected by HPP. and retail, even at adequate refrigeration temperatures. At the
time of purchase, cheese may have developed a stronger or
different flavour than that intended by the manufacturer (Wick
Keywords High pressure . Cheese . Microbiology . et al. 2004). Overripening and development of off-flavours
Proteolysis . Biogenic amines . Flavour limit the shelf-life of cheese.
FAA also serve as substrates for bacterial decarboxylases,
giving rise to the formation of biogenic amines (BA), a group
J. Calzada : A. del Olmo : A. Picon : P. Gaya : M. Nuñez (*) of public health significance compounds. Monoamines tyra-
Departamento de Tecnología de Alimentos, Instituto Nacional de
mine, phenylethylamine, histamine and tryptamine are formed
Investigación y Tecnología Agraria y Alimentaria (INIA),
28040 Madrid, Spain through the decarboxylation of tyrosine, phenylalanine, histi-
e-mail: nunez@inia.es dine and tryptophan, respectively, while diamines cadaverine
and putrescine are derived from lysine, and ornithine or argi- ripening on the microbiology, proteolysis, biogenic amines,
nine via agmatine (Linares et al. 2011). Enterococci were texture and sensory characteristics of cheese made from un-
traditionally regarded as the main tyramine formers and pasteurized cow milk throughout a 240-day ripening and
heterofermentative lactobacilli as the main histamine pro- storage period, on the aim of improving its safety while
ducers, but other genera of bacteria are also capable of maintaining its flavour traits. Simultaneously, the characteris-
forming BA in cheese (Joosten and Northolt 1987; Pircher tics of untreated control cheese were studied and compared
et al., 2007). Although the levels of decarboxylase-positive with those of HPP cheeses.
bacteria in cheese can be reduced by means of milk
hygienization procedures such as bactofugation,
microfiltration and pasteurization, contamination of pasteur- Materials and Methods
ized milk and curd by adventitious Gram-positive and Gram-
negative bacteria harbouring amino acid decarboxylases Cheese Manufacture
(Linares et al. 2011) may occur.
High-pressure processing (HPP) is a treatment which Cheese was manufactured from 750 L of raw cow milk in each
achieves the safety level of heat pasteurization while meeting of two trials, carried out on consecutive days at a dairy in NW
the consumer demand for fresher-tasting minimally processed Spain. Characteristics of the cow milk used (average for the
foods (Norton and Sun 2008). It may be applied to cheese two trials) were 3.14 % protein, 3.65 % fat, 12.49 % dry
once the manufacturing process has ended, what offers the matter, and 4.10 log cfu/mL viable bacterial counts. Freeze-
advantage that the contamination of the cheese interior is no dried mesophilic starter (25 units of CHN-19, Chr. Hansen
longer possible. Significant reductions of decarboxylase- S.L., Tres Cantos, Spain), composed of Lactococcus lactis
positive bacteria counts and biogenic amine concentrations subsp. lactis, L. lactis subsp. cremoris, L. lactis subsp. lactis
were achieved when HPP was applied to cheese made from biovar diacetylactis and Leuconostoc, was added to the milk
ovine raw milk coagulated with vegetable rennet (Calzada previously warmed to 32.5 °C. After 15 min, animal rennet
et al. 2013a). The procedure was also useful in controlling (150 mL of Naturen®, 80 % chymosin, Chr. Hansen S.L.) was
cheese overripening by lowering microbial counts, proteolytic added to the milk, which was left to coagulate at 32.5 °C for
activity and peptidolytic activity (Calzada et al. 2014a) and 50 min. The curd was cut into 2-cm cubes, stirred in the vat in
had beneficial effects on its volatile profile and odour charac- its own whey, washed with tap water at 35 °C, and finally
teristics (Calzada et al. 2014b). The stage of ripening at which salted in the vat by addition of 9 kg dry salt. The curd was
HPP is applied is a relevant parameter in order to retain cheese dispensed into cylindrical moulds and pressed for 3 h at 20 °C.
sensory characteristics. When HPP treatments at 400 MPa on Total weight of the cheeses out of the press was 95.58 kg
days 2 and 50 of ripening were compared, cheeses treated on (average for the two trials). Cheeses (13.4-cm diameter, 7.0-
day 2 received the lowest scores for taste quality on day 60, cm height, 1.15-kg average weight out of the press) were
while cheeses treated on day 50 did not differ from control ripened at 8 °C and 72 % RH until day 60 and at 5 °C and
cheese (Garde et al. 2007). 75 % RH afterwards.
The economic importance of cheeses made from cow milk
considerably surpasses that of cheeses made from milk of High-Pressure Processing and Sampling
other animal species. In spite of this, the effect of HPP has
been studied only on two major cow milk cheese varieties, Before HPP, cheeses were vacuum-packaged in CN300 bags
Gouda (Messens et al. 1999) and Cheddar (O’Reilly et al. (Cryovac Grace S.A., Barcelona, Spain). Cheeses, respective-
2003; Wick et al. 2004; Rynne et al. 2008), both made from ly coded as 400W2, 600W2, 400W3 and 600W3, were HPP-
pasteurized milk. Some of those studies were carried out on treated for 5 min at 400 or 600 MPa after ripening for 14 or
small cheese samples and pressure levels above 400 MPa 21 days. A 120-L capacity isostatic press (Hiperbaric, Burgos,
were only applied by Wick et al. (2004). To our knowledge, Spain) was used for HPP. Come-up times to reach 400 and
there is no published information on the effect of HPP on cow 600 MPa were 1.76 and 2.58 min, respectively, and depres-
milk cheeses made from unpasteurized milk, more prone to surization times, 5 and 7 s. Initial temperature of the water
formation of biogenic amines and generation of off-flavours used as transmitting fluid was 8 °C; it remained below 14 °C
than cheeses made from pasteurized milk. Tyramine concen- during the process. After treatments, HPP cheeses were
trations ranging from 453.7 to 957.6 mg/kg were reported for unpackaged and ripened under the same conditions of control
artisanal raw milk cheeses, while they ranged from 21.7 to cheese.
216.8 mg/kg in artisanal pasteurized milk cheeses (Ladero From each of the two cheese-making trials, one cheese was
et al. 2010). analyzed on day 1, three cheeses (control, 400W2, 600W2) on
The objective of the present work was to investigate the day 14 and five cheeses (control, 400W2, 600W2, 400W3,
effect of HPP at 400 or 600 MPa applied on days 14 or 21 of 600W3) on each of the sampling dates day 21, day 60, day
120, day 180 and day 240, resulting in 29 cheeses analyzed (Sigma, Alcobendas, Spain) were used for their identification
per trial and 58 cheeses analyzed in the overall experiment. and quantification.
Aminopeptidase activity released into the cheese was mea-
Microbiological Analyses sured on duplicate samples using lysine p-nitroanilide (Lys-p-
NA) and leucine p-nitroanilide (Leu-p-NA) as substrates ac-
Representative cheese samples (10 g) were homogenized with cording to Garde et al. (2002). One activity unit corresponds
90 mL of a sterile 2 % (w/v) sodium citrate solution at 45 °C in to the activity of enzyme(s) producing 1 nmol of p-nitroaniline
a Colworth Stomacher 400 (A. J. Seward Ltd., London, UK). per minute per gram of cheese.
Decimal dilutions of milk and cheese homogenate were pre- Tyrosine decarboxylase (TDC) activity in cheese was de-
pared in sterile 0.1 % peptone solution. Total viable counts, termined in duplicate samples as previously described
LAB, lactobacilli, enterococci, Micrococcaceae, coagulase- (Calzada et al. 2013a). Briefly, a cheese homogenate prepared
positive staphylococci, Gram-negative bacteria, coliforms, by mixing 10 g of cheese with 20 mL of 2 % sodium citrate
moulds and yeasts were determined in duplicate using the solution was centrifuged and the supernatant dialysed (molec-
culture media and incubation conditions previously described ular mass cut-off, 10 kDa). The dialysate was centrifuged and
(Calzada et al. 2013b). Analysis of Listeria monocytogenes in the supernatant stored at −20 °C until analysis. ATDC activity
milk and cheese was performed as indicated by Arqués et al. standard curve was obtained by means of reaction mixtures
(2005). containing 1.5 mM L-tyrosine (Sigma), 0.15 mM pyridoxal-
5′-phosphate (Sigma) and up to 32 mIU/mL of the L-TDC
Chemical and Enzymatic Determinations apoenzyme (Sigma). TDC activity in the supernatants of
cheese homogenates was assayed on reaction mixtures with-
Caseins and whey proteins were analyzed by capillary gel out added apoenzyme. TDC activity in cheese was calculated
electrophoresis on triplicate samples, using an automated from the increase in tyramine concentration (determined by
P/ACE MDQ capillary electrophoresis apparatus controlled HPLC after 0 and 24 h of incubation of the reaction mixtures
by a 32 Karat Software (Beckman Instruments España S.A., at 37 °C) against the TDC activity standard curve.
Madrid, Spain) according to Calzada et al. (2013b).
Commercial standards (Sigma, Alcobendas, Spain) of bovine
Textural Determinations
α-CN, β-CN, κ-CN, α-LA, β-LG, serum albumin and
lactoferrin were used for the identification of proteins.
Eight cylinder-shaped (17-mm height, 17-mm diameter) sam-
Protein peaks were quantified with respect to the standards
ples from each cheese were compressed to 75 % of their
and expressed as milligrams of protein per gram of cheese dry
original height after 10 min at room temperature (20–22 °C)
matter (DM). DM content and pH value of cheeses were
using an Instron Compression Tester 4301 (Instron, High
determined on triplicate samples according to Garde et al.
Wycombe, Bucks, UK), with crosshead and chart speeds of
(2002).
50 and 500 mm/min, respectively. Firmness (work done on the
Hydrophilic and hydrophobic peptides in the water-soluble
cheese up to 75 % compression, expressed in Newton metre
fraction of cheese were determined on duplicate samples by
(N m)), elasticity (apparent elastic modulus, expressed in
reverse-phase high-performance liquid chromatography
Newton per square millimetre (N/mm2)) and fracturability
(HPLC) as previously described (Garde et al. 2002), using a
(force at breaking point, expressed in Newton (N)) were
Beckman System Gold chromatograph (Beckman
determined from the compression curves according to
Instruments España SA, Madrid, Spain) equipped with a
Calzada et al. (2014a).
diode array detector module 168, with detection wavelength
at 280 nm. Peaks with retention times from 5.5 to 14.6 min
were considered to correspond to hydrophilic peptides and Sensory Evaluation
those with retention times from 14.6 to 20.5 min to hydropho-
bic peptides. Results were expressed in arbitrary units (AU), Seventeen trained panellists evaluated the cheeses after 21, 60,
calculated as units of chromatogram area per milligram of 120, 180 and 240 days of ripening for flavour quality (overall
cheese DM. acceptance), flavour intensity (overall intensity) and flavour
FAA and BA were simultaneously extracted from duplicate attributes “acid”, “bitter”, “salty”, “sweet” and “umami” on a
samples according to Krause et al. (1995). Analysis of indi- 0–10-point scale, using a horizontal line anchored in the
vidual FAA after derivatization with Waters AccQ Fluor middle and at both ends, as previously described (Calzada
Reagent (Waters S.A., Barcelona, Spain) and analysis of indi- et al. 2014a). Cheeses were cut into representative cubes (10 to
vidual BA after derivatization with dabsyl chloride were car- 12 g), which were held for 2 h at 20–22 °C prior to sensory
ried out by reverse-phase HPLC as previously described evaluation. Five cheeses per session (one control and four
(Calzada et al. 2013b). Standard mixtures of FAA and BA HPP cheeses, manufactured on the same day), coded with
three-digit numbers, were randomly presented to panellists. 0.7 log units on day 14 and by 0.4 log units on day 21,
Bread and water were used as rinsing agents between cheeses. whereas the respective declines at 600 MPa were 2.9 and 3.2
log units. As for LAB, the use of selective culture media for
Statistical Treatment the enumeration of lactobacilli and enterococci after HPP
might have underestimated their counts and overestimated
Analysis of variance with HPP treatment (four HPP treatments the lethality of 600 MPa treatments. In spite of this,
and control) and ripening time as main effects was performed lactobacilli and enterococci counts decreased less than after
on the analytical variables by means of SPSS Win 14.0 HPP of ovine milk cheese at 600 MPa, in which lactobacilli
program (SPSS Inc., Chicago, USA). Calculation of correla- declined by 5.8–6.1 log units and enterococci by 4.9–5.3 log
tions and comparison of means by Tukey’s test, with the units (Calzada et al. 2013a). A decrease of lactobacilli counts
significance assigned at p<0.05, were carried out using the by more than 4 log units in 1-month-old Cheddar cheese after
same program. treatment at 500–800 MPa was reported by Wick et al. (2004).
Micrococcaceae counts (data not shown) reached levels of
6.92 log cfu/g on day 1 in control cheese and decreased to 4.96
log cfu/g on day 60 and 4.08 log cfu/g on day 240. HPP at
Results and Discussion 400 MPa lowered their counts by 2.1 log units on days 14 and
21, whereas treatment at 600 MPa caused decreases of 3.4 log
Microbial Groups units on day 14 and 3.3 log units on day 21, a lethal effect
similar to the declines of 2.8–3.9 log units after HPP of ovine
Total viable counts declined gradually during ripening of milk cheese at 600 MPa reported by Calzada et al. (2013a). In
control cheese, from 9.2 log cfu/g on day 1 (data not shown) the present work, there was no recovery of Micrococcaceae
to 8.0 log cfu/g on day 240, while LAB counts declined from after HPP with counts remaining below 3.5 log cfu/g in
8.8 log cfu/g on day 1 to 7.9 log cfu/g on day 240 (Fig. 1). 400 MPa cheeses and below 2.5 log cfu/g in 600 MPa cheeses
HPP at 400 MPa lowered total viable counts by 2.4 log units until day 240. Coagulase-positive staphylococci counts (data
on day 14 and by 1.1 log units on day 21, while at 600 MPa not shown) attained levels of 4.83 log cfu/g on day 1, declined
the respective declines were 4.0 and 3.3 log units. Decreases in control cheese to 2.66 log cfu/g on day 14 and to popula-
of LAB counts after HPP at 400 MPa were 2.3 log units on tions below detection level (<1 log cfu/g) during the rest of the
day 14 and 2.0 log units on day 21 whereas the respective ripening period, while they were not found in any of the HPP
decreases at 600 MPa were 6.0 and 5.8 log cfu/g. The decrease cheeses after treatments. L. monocytogenes was not detected
in total viable counts and LAB counts after HPP at 600 MPa (<1 log cfu/g) in control or HPP cheeses at any stage of
should have been similar since LAB accounted for most of the manufacture or ripening.
total viable counts before HPP. The higher decline of LAB Gram-negative bacteria counts (data not shown) were at
counts than of total viable counts after HPP at 600 MPa may 7.52 log cfu/g on day 1 and declined to 6.50, 6.18, 3.90 and
be due to the use of a selective culture medium for LAB 2.04 log cfu/g, respectively, on days 14, 21, 60 and 240 in
enumeration, which probably hindered the recovery of HPP- control cheese. Their population in HPP cheeses was below
injured LAB cells while they were capable of forming colo- the detection level (<1.4 log cfu/g) immediately after treat-
nies on a non-selective medium. The only previously reported ments at 400 or 600 MPa, and they were not found afterwards
result on the decrease of total viable counts and LAB counts in in any of the HPP cheeses, in agreement with the results
cow milk cheeses treated at 600 MPa was a decrease of obtained by Calzada et al. (2013a) for ovine milk cheese.
L. lactis counts by more than 5 log units after HPP of 1- Similarly, coliform counts declined in the present work from
month-old Cheddar cheese at 500–800 MPa (Wick et al. 7.50 log cfu/g on day 1 to 6.08, 5.75, 3.81 and 1.46 log cfu/g,
2004). The decreases of total viable counts and LAB counts respectively, on days 14, 21, 60 and 240 in control cheese
after HPP at 600 MPa in the present study were similar to (data not shown). Coliforms were not detected in any of the
those recorded in ovine milk cheese treated at the same HPP cheeses during ripening. Yeast counts reached 3.62 log
pressure (Calzada et al. 2013a). cfu/g on day 1 and declined afterwards in control cheese to
Lactobacilli counts increased in control cheese from 4.07 3.18 log cfu/g on day 60 and 2.49 log cfu/g on day 240 (data
log cfu/g on day 1 (data not shown) to 8.03 log cfu/g on day not shown). Yeasts were not detected in any of the HPP
240, while enterococci counts decreased from 6.31 log cfu/g cheeses after treatments.
on day 1 (data not shown) to 4.71 log cfu/g on day 240
(Fig. 1). HPP at 400 MPa lowered lactobacilli counts by 1.0 Dry Matter and pH Value
log units on day 14 and by 1.1 log units on day 21, while the
respective declines at 600 MPa were 4.3 and 4.9 log units. In DM of control cheese increased gradually from 47.78 % on
the case of enterococci, HPP at 400 MPa lowered counts by day 1 (data not shown) to 53.13 % on day 60 and 65.43 % on
a b
9 c Total viable counts 9 Lactic acid bacteria counts
c c b b b b c d b b b
8 c c b b b b b 8 b b bb b b b b b
b b
b
7 7 c
a b a
6 a b a a 6 b a a
a a a a
a
Log cfu/g
Log cfu/g
5 a a 5 a a a
a
4 4 a
a a
3 3
2 2
1 1
0 0
14 d 21 d 60 d 120 d 180 d 240 d 14 d 21 d 60 d 120 d 180 d 240 d
c d
9 Lactobacilli counts 9 Enterococci counts
b bb b bb b b
b b
8 b b 8
c
7 c a 7
b
a a b
6 b b 6 b a
a a c b c
a c b
a
Log cfu/g
Log cfu/g
5 a 5 b a
b
a b
4 a 4
a a a
a a a
3 3 a a
2 2
1 1
0 0
14 d 21 d 60 d 120 d 180 d 240 d 14 d 21 d 60 d 120 d 180 d 240 d
Fig. 1 Counts of the main bacterial groups (a total viable bacteria, b 400 MPa on day 21 (dotted) or 600 MPa on day 21 (obliquely striped).
lactic acid bacteria, c lactobacilli, d enterococci) during ripening and Means (bars with SEM) at the same sampling date with the same letter do
refrigerated storage of control cheese (black) and cheeses HP-treated at not differ significantly (p>0.05)
400 MPa on day 14 (white), 600 MPa on day 14 (horizontally striped),
day 240 (Fig. 2). DM contents of HPP cheeses ranged changes in pH values during ripening of control or
from 52.84 to 54.05 % on day 60 and from 65.63 to HPP cheeses were recorded by Messens et al. (1999).
66.64 % on day 240, with no significant (p>0.05) dif-
ferences between control and HPP cheeses at any stage Hydrolysis of Proteins
of ripening. No clear effects of HPP on DM during
ripening were recorded in previous works on cow milk Soluble compounds, including whey proteins, were lost in
cheeses (Messens et al. 1999; Rynne et al. 2008). whey during cheese manufacture. Also, proteolysis took place
Control cheese pH value declined to 5.26 on day 1 (data not in the vat and during pressing of cheese. On day 1, casein
shown) and rose gradually afterwards, to 5.36 on day 60 and concentrations were 88.18, 89.61, 23.25, 16.17 and 3.65 mg/g
5.68 on day 240 (Fig. 2). HPP cheeses had slightly higher pH DM for α-casein, β-casein, κ-casein, para-κ-casein and γ-
values than control cheese immediately after treatment, al- caseins, respectively (data not shown). Concentrations of
though the differences were below 0.1 pH units. A higher whey proteins in 1-day cheese were 0.98 mg/g DM for α-
pH increase, with differences of almost 0.2 pH units, had lactalbumin and 1.37 mg/g DM for β-lactoglobulin, while
been reported in HPP-treated Gouda cheese (Messens bovine serum albumin was not found (data not shown).
et al. 1999). During ripening, the pH increase in HPP From day 1 to day 240, the concentration of caseins in control
cheeses lagged behind that in control cheese, with values cheese declined by 46, 64, 65 and 35 % for αs-, β-, κ- and
ranging from 5.26 to 5.32 on day 60 and from 5.39 to para-κ-caseins, respectively, while a 9.3-fold increase of γ-
5.59 on day 240. These results are in agreement with caseins, products of β-casein hydrolysis, was recorded
those obtained by Calzada et al. (2014a), while no (Fig. 3). The only whey protein detected during ripening of
a enhance the susceptibility of caseins to proteolysis (García-

70
Risco et al. 2000). This might explain the faster degradation of
68 a a a
a a caseins recorded in the present work for HPP cheeses, by the
66 a a a
aa
action of chymosin and plasmin in 400 MPa cheeses and
64
mostly by the action of plasmin in 600 MPa cheeses. Bovine
62
pepsin from rennet is also involved in cheese proteolysis, but its
Dry matter (%)
aa a
60 aa
58
contribution to the proteolysis of HPP cheeses is difficult to
56 a a a
estimate since, to our knowledge, no information about the
a
54
a baroresistance of this enzyme is available. The concentration
52 a a a of γ-caseins increased by 13.8- to 14.1-fold in HPP cheeses
a a a a
50
a from day 1 to day 240 (data not shown), more than in control
48 cheese, in agreement with the more pronounced degradation of
46 β-casein. As for control cheese, the only whey protein detected
14 d 21 d 60 d 120 d 180 d 240 d during ripening of HPP cheeses was α-lactalbumin, at 0.07–
b 0.08 mg/g DM on day 14 (data not shown). Negligible differ-
5.7 c ences in SDS-PAGE profiles were observed between control
c
c b b
and HPP cheeses when Gouda or Cheddar cheeses were treated
5.6 at 400 MPa (Messens et al. 1999; O’Reilly et al. 2003).
b
ab
b The levels of hydrophilic peptides increased gradually
5.5
b b
b ab ab
a
during ripening of control cheese, from 3.53 AU on day 1
a
a a (data not shown) to 7.56 AU on day 60 and 8.72 AU on day
pH value
ab a
a
5.4
b a 240, while the levels of hydrophobic peptides increased from
a
ab aa a 2.20 AU on day 1 to 7.69 AU on day 60 and then declined to
5.3 4.37 AU on day 240 (Fig. 4). All the HPP cheeses showed
a
significantly (p<0.05) higher levels of hydrophilic peptides
5.2
and hydrophobic peptides than control cheese from day 60
onwards, in agreement with the enhanced hydrolysis of ca-
5.1
14 d 21 d 60 d 120 d 180 d 240 d
seins (Fig. 3). Contrarily, a decrease in the levels of hydro-
Fig. 2 Dry matter content (a) and pH value (b) during ripening and philic and hydrophobic peptides with respect to control cheese
refrigerated storage of control cheese (black) and cheeses HP-treated at was reported for pasteurized ewes’ milk cheeses treated at 400
400 MPa on day 14 (white), 600 MPa on day 14 (horizontally striped), or 500 MPa (Juan et al. 2007). The hydrophobic peptides/
400 MPa on day 21 (dotted) or 600 MPa on day 21 (obliquely striped). hydrophilic peptides ratio of control cheese rose from 0.624
Means (bars with SEM) at the same sampling date with the same letter do
not differ significantly (p>0.05) on day 1 (data not shown) to a maximum value of 1.125 on
day 14 in the present work, and then declined to 0.497 on day
240 (Fig. 4). In HPP cheeses, the ratio tended to be lower than
control cheese, α-lactalbumin, reached 0.10 mg/g DM on day in control cheese until day 60, with few significant differences
14 and was not found afterwards (data not shown). between cheeses, but from day 120 onwards, it reached sig-
Proteolysis during ripening of HPP cheeses markedly dif- nificantly (p<0.05) higher values in the 600 MPa cheeses than
fered from that of control cheese (Fig. 3). Degradation of αs- in control cheese.
casein was enhanced only in 400 MPa cheeses, with signifi- Aminopeptidase activity on Leu-p-NA as substrate (data
cantly (p<0.05) lower concentrations than in control cheese or not shown) increased in control cheese from 1.36 activity
in 600 MPa cheeses from day 120 onwards. In the case of β-, units on day 1 to a maximum of 5.87 activity units on day
κ- and para-κ-caseins, concentrations were significantly 14 and remained fairly unchanged afterwards, still reaching
(p<0.05) lower than in control cheese not only in 400 MPa 5.03 activity units on day 240. With Lys-p-NA as substrate,
cheeses but also in 600 MPa cheeses from day 60 onwards. aminopeptidase activity of control cheese rose from 2.25 ac-
From day 1 to day 240, the concentrations of αs-, β-, κ- and tivity units on day 1 (data not shown) to a maximum of
para-κ-caseins in HPP cheeses declined by 41–66, 84–92, 75– 11.39 activity units on day 21 and declined gradually to
81 and 69–88 %, respectively. Chymosin is partially 6.85 activity units on day 240 (Fig. 5a). Aminopeptidase
inactivated by HPP at 400 MPa or higher pressures, but activity values during early ripening were considerably lower
plasmin, a more baroresistant enzyme, maintains full activity than those found for ovine milk cheese of the same age by
after pressurization at 500 or 600 MPa (Malone et al. 2003; Calzada et al. (2014a), but on day 240, similar activity values
Huppertz et al. 2004). HPP causes the disruption of linkages were obtained. In addition to the significant (p<0.05) decrease
between caseins and inorganic constituents of milk, what may in the aminopeptidase activity of HPP cheeses immediately
a b
c d
Fig. 3 Levels of αs-casein (a), β-casein (b), κ-casein (c) and para-κ- on day 14 (horizontally striped), 400 MPa on day 21 (dotted) or 600 MPa
casein (d) during ripening and refrigerated storage of control cheese on day 21 (obliquely striped). Means (bars with SEM) at the same
(black) and cheeses HP-treated at 400 MPa on day 14 (white), 600 MPa sampling date with the same letter do not differ significantly (p>0.05)
after treatment, further decreases were recorded during the rest aminopeptidases, to the surrounding medium. As the released
of the ripening period for most of the HPP cheeses. This aminopeptidases were no longer protected by cell structures,
pattern is in contradiction with the results obtained by Juan they were probably more susceptible to HPP inactivation than
et al. (2007), who observed an increase of the aminopeptidase when inside the cells. This might explain why aminopeptidase
activity in control cheese and in cheeses treated at 400 or inactivation immediately after HPP was higher on day 21,
500 MPa during a 60-day ripening period. The inactivation when more intracellular enzymes had been released from the
of bacterial peptidases by HPP depends not only on process cells, than on day 14.
parameters, but also on the bacterial species and assay sub- The concentration of total FAA increased in control cheese
strates, and the effect may vary for different peptidolytic from 1.06 mg/g DM on day 1 (data not shown) to 8.79 mg/g
enzymes within the same bacterial strain (Malone et al. DM on day 60 and to 31.68 mg/g DM on day 240 (Fig. 5b).
2003). In the present work, aminopeptidase inactivation im- Total FAA concentrations did not vary in HPP cheeses imme-
mediately after HPP was more dependent on the day of diately after treatment, and no significant differences in total
treatment (68–73 % inactivation on day 14 versus 87–89 % FAA were found between control and HPP cheeses on day 21.
on day 21) than on the pressure level (68–88 % at 400 MPa Afterwards, the concentration of total FAA increased in con-
versus 73–89 % at 600 MPa) or the substrate used (68–88 % trol cheese, which showed 28–31 % higher levels than
on Leu-p-NA versus 71–89 % on Lys-p-NA). Approximately 400 MPa cheeses and 63–64 % higher levels than 600 MPa
25 % of the total viable bacteria present in 1-day control cheeses on day 240. The dominant FAA in 240-day control
cheese remained viable on day 14 while only 6 % were still cheese were glutamic acid, leucine, lysine, valine and serine
alive on day 21. Death of bacteria was most probably followed while in the respective HPP cheeses the main FAA were
by cell lysis and release of intracellular enzymes, including glutamic acid, leucine, lysine, valine and phenylalanine (data
a a
14 12 c
b b b
b b
b b b b 11 b
b b bb b bb
12 b 10
Hydrophilic peptides (AU/mg DM)
nmol p -nitroaniline / min g

9 b
10 b bb a
a
ab a 8 b b
a
b
a 7
8 ab
a 6
6 5
a
4 a
4
3
b a a
ab
2 ab a a a a a
2 a a a a
1 a a a a a a
0 0
14 d 21 d 60 d 120 d 180 d 240 d 14 d 21 d 60 d 120 d 180 d 240 d
b b
35 c
14
30
12
Hydrophobic peptides (AU/mg DM)
b
bb b b b c
b b b b
b b 25 b
Total FAA (mg/g DM)
10 b
b b b
a aa b b b
a
a a a
20
8 a a b
a a b b
a 15 a a
6 a
b a a
b
10 ab a
4
a a
aa a
5 aa a
2 a a
0 0
14 d 21 d 60 d 120 d 180 d 240 d 14 d 21 d 60 d 120 d 180 d 240 d
Fig. 5 Aminopeptidase activity on Lys-p-NA (a) and levels of total free

c amino acids (b) during ripening and refrigerated storage of control cheese
1.3 a (black) and cheeses HP-treated at 400 MPa on day 14 (white), 600 MPa
1.2
on day 14 (horizontally striped), 400 MPa on day 21 (dotted) or 600 MPa
b
a a
on day 21 (obliquely striped). Means (bars with SEM) at the same
1.1 sampling date with the same letter do not differ significantly (p>0.05)
1.0 ab aa
a aa a a a b d
ab b b cheese at pressure levels of 200 or 300 MPa had no effect on
Peptide ratio
0.9 a cd
a a a
0.8 a bc its FAA content, while pressure levels of 400 to 800 MPa
a b retarded the evolution of FAA (Wick et al. 2004). Increases of
0.7
total FAA concentrations by 9–12 % were reported for ovine
0.6 a
milk cheese treated at 400 MPa, after a 240-day ripening and
0.5
storage period, while HPP at 600 MPa resulted in decreases of
0.4 25–29 % (Calzada et al. 2014a). In the present study, total
0.3 FAA concentrations in 180- and 240-day cheeses were signif-
14 d 21 d 60 d 120 d 180 d 240 d
icantly correlated with the respective aminopeptidase activity
Fig. 4 Levels of hydrophilic peptides (a), hydrophobic peptides (b) and values, both with Leu-p-NA (r=0.739, p<0.05) and Lys-p-
the ratio of hydrophobic/hydrophilic peptides (c) during ripening and
NA (r=0.784, p<0.01) as substrates.
refrigerated storage of control cheese (black) and cheeses HP-treated at
400 MPa on day 14 (white), 600 MPa on day 14 (horizontally striped),
400 MPa on day 21 (dotted) or 600 MPa on day 21 (obliquely striped).
Means (bars with SEM) at the same sampling date with the same letter do Biogenic Amines
not differ significantly (p>0.05)
Tyrosine decarboxylase was significantly (p < 0.05)
not shown). The concentration of total FAA in Cheddar cheese inactivated by HPP, with lower activity levels in all the HPP
treated at 400 MPa decreased or increased depending on the cheeses than in control cheese on day 21 (Fig. 6). During the
starter culture used (O’Reilly et al. 2002). HPP of Cheddar rest of the ripening period, TDC activity increased gradually
a b
3.0 b 0.45
c
b 0.40
2.5
ab 0.35
b b
TDC activity (mAU/g DM)
Tyramine (mg/g DM)

2.0 0.30 ab
c
a 0.25
1.5 b b b
b a
0.20 b
b
a
1.0 0.15 c
b 0.10 b
0.5
b a
a 0.05 a
aa a a a a a
a a a
0.0 0.00
14 d 21 d 60 d 120 d 180 d 240 d 14 d 21 d 60 d 120 d 180 d 240 d
c d
0.250 0.150
a a
0.225 a c
a 0.125
0.200
b b
Cadaverine (mg/g DM)
0.175
Putrescine (mg/g DM)
a b b
aa a 0.100
ab b b
0.150 a
b a a b
a ab
0.125 0.075 a ab
c a ab
0.100 aa a ab a
a
a a
0.050 a a
0.075 a
0.050 b
0.025
0.025 b
a a
a
0.000 0.000
14 d 21 d 60 d 120 d 180 d 240 d 14 d 21 d 60 d 120 d 180 d 240 d
Fig. 6 Tyrosine decarboxylase activity (a) and levels of tyramine (b), (dotted) or 600 MPa on day 21 (obliquely striped). Means (bars with
putrescine (c) and cadaverine (d) during ripening and refrigerated storage SEM) at the same sampling date with the same letter do not differ
of control cheese (black) and cheeses HP-treated at 400 MPa on day 14 significantly (p>0.05)
(white), 600 MPa on day 14 (horizontally striped), 400 MPa on day 21
in control and 400 MPa cheeses while no activity was detected was not a limiting factor for tyramine formation, since it
in any of the 600 MPa cheeses. On day 240, it reached reached 0.348 mg/g DM in control cheese, 0.214–0.218 mg/
2.619 mAU/g DM in control cheese and 1.148–1.875 mAU/ g DM in 400 MPa cheeses and 0.344–0.416 mg/g DM in
g DM in 400 MPa cheeses. In the present study, TDC activity 600 MPa cheeses on day 240. The low enterococci counts,
values during ripening of control cheese were slightly lower always below 6 log cfu/g throughout ripening of control
than the respective values of control ovine milk cheese while cheese and considerably lower in HPP cheeses, in particular
those of 400 MPa cheeses were higher than the respective in 600 MPa cheeses (Fig. 1), preclude an important contribu-
values of ovine milk cheese treated at 400 MPa (Calzada et al. tion of this genus to tyramine formation. In contrast, the
2013a), results which can be attributed to differences in cheese population of lactobacilli (Fig. 1), at 8 log cfu/g in control
microbiota. and 400 MPa cheeses from day 60 to day 240, was high
Tyramine concentration on day 21 was under 0.010 mg/g enough to actively participate in tyramine formation
DM in control and 400 MPa cheeses and below detection level provided that some of the strains harboured tyrosine
in 600 MPa cheeses. Afterwards, tyramine accumulated at a decarboxylases. Production of tyramine by lactobacilli
faster rate in control cheese than in 400 MPa cheeses while it isolated from Dutch cheese belonging to different species
was only occasionally detected, at levels under 0.010 mg/g was reported by Joosten and Northolt (1987) and by
DM, in some of the 600 MPa cheeses. On day 240, tyramine lactobacilli from various foods including cheeses by Pircher
concentration attained 0.372 mg/g DM in control cheese, et al. (2007). Also, decarboxylation of tyrosine by some
0.170–0.176 mg/g DM in 400 MPa cheeses and was not L. lactis strains has been observed (Curtin and McSweeney
detected in 600 MPa cheeses (Fig. 6). Tyrosine concentration 2003). In the present work, tyrosine decarboxylase-positive
bacteria were not determined, and no population threshold for and pressure-induced in HPP cheeses, cadaverine accumula-
tyramine formation could be established. Thresholds for tyra- tion ceased. In our study, the production of cadaverine may be
mine production in cheese as low as 4 log cfu/g of strains with associated with the presence of Gram-negative bacteria at
decarboxylase activity have been suggested (Ladero et al. sufficiently high levels. Concentrations of cadaverine in 60-
2010). Highly significant (p<0.001) correlations of tyramine to 240-day-old cheeses were not significantly correlated with
concentrations in 60- to 240-day-old cheeses with LAB counts counts of lactic acid bacteria (r=0.049) or lactobacilli (r=
(r=0.757), lactobacilli counts (r=0.781) and TDC activity 0.017). The concentrations of arginine and lysine, precursors
(r=0.915) were found in the present work. Histamine was for putrescine and cadaverine, were not limiting factors for
detected only on days 180 and 240, at 0.009–0.026 mg/g DM biogenic amine formation.
in control cheese and 0.004–0.005 mg/g DM in some of the Other biogenic amines such as phenylethylamine, trypt-
400 MPa cheeses (data not shown). Histidine concentrations, amine, spermine and spermidine were not detected in control
2.121 mg/g DM in control cheese, 1.265–1.269 mg/g DM in or HPP cheeses. Strains belonging to some of the bacterial
400 MPa cheeses and 0.630–0.639 mg/g DM in 600 MPa genera commonly found in cheese form biogenic amines
cheeses on day 240, were sufficient for not limiting histamine through the decarboxylation of FAA, a mechanism which
formation. constitutes an alternative energy source in the absence of
On day 21, putrescine was only found in control and fermentable carbohydrates (Fernández-García et al. 1999).
400 MPa cheeses, at concentrations under 0.010 mg/g DM. Some strains of Lactobacillus, Pediococcus and
Afterwards, putrescine increased more rapidly in control Micrococcus are capable of degrading biogenic amines such
cheese than in 400 MPa cheeses whereas it was not detected as tyramine and histamine by means of monoamine oxidases,
or was at levels of 0.001–0.002 mg/g DM in 600 MPa cheeses preferably under aerobic conditions (Leuschner et al. 1998).
(Fig. 6). On day 240, it reached 0.207 mg/g DM in control The rate of biogenic amine build-up during cheese ripening is
cheese, 0.110–0.143 mg/g DM in 400 MPa cheeses and at the resultant of their formation and degradation in cheese. On
most 0.001 mg/g DM in 600 MPa cheeses. Gram-negative day 240, total biogenic amines reached 0.686 mg/g DM in
bacteria are considered to be the main putrescine and cadav- control cheese, while they only attained 0.316–0.351 mg/g
erine formers (Pircher et al. 2007). They reached counts above DM in 400 MPa cheeses and 0.044–0.081 mg/g DM in
6 log cfu/g until day 21 and below 4 log cfu/g from day 60 600 MPa cheeses. These levels are considerably lower than
onwards in control cheese and were below the detection level in 240-day ovine milk cheeses, in which total biogenic amines
in all samples of HPP cheeses. Low counts of Gram-negative attained 3.690 mg/g DM in control cheese, 2.022–3.276 mg/g
bacteria in control cheese from day 60 onwards precluded a DM in 400 MPa cheeses and 0.896–1.011 mg/g DM in
significant contribution of these microorganisms to the bio- 600 MPa cheeses (Calzada et al. 2013a). The concentrations
synthesis of putrescine, which more than doubled from day 60 of total FAA in control cheese in the present work were 8.79
to day 240. Moreover, putrescine production in 400W2 and and 31.68 mg/g DM on days 60 and 240, respectively, higher
400W3 cheeses took place from day 21 onwards, when Gram- than the concentrations reported by Calzada et al. (2014a),
negative bacteria were no longer detected. Therefore, putres- 4.82 mg/g DM on day 60 and 23.31 mg/g DM on day 240.
cine production in control and HPP cheeses must be ascribed Since FAA concentrations did not limit BA formation in the
to other microorganisms, in particular to lactic acid bacteria. present study, the main reason for the differences in BA
Putrescine concentrations in 60- to 240-day-old cheeses were concentrations between both works seems to be the
highly correlated with counts of lactic acid bacteria (r=0.786; differences in the BA formation ability of cheese microbiota.
p<0.001) and lactobacilli (r=0.808; p<0.001) and at a lower Calzada et al. (2013a) reported Gram-negative bacteria counts
level of significance also with enterococci (r=0.648; p<0.01). from day 60 to day 180 above 6 log cfu/g in control cheese
Production of putrescine via the agmatine deamination path- and below detection level in HPP cheeses. In their work,
way seems to be a species-level trait of Enterococcus faecalis tyramine, histamine, phenylethylamine, tryptamine and pu-
(Ladero et al. 2012). Cadaverine formation was less affected trescine increased in control and HPP cheeses from day 60
by HPP and time of ripening than tyramine or putrescine onwards, even though Gram-negative bacteria were not de-
(Fig. 6). Its pattern of accumulation was markedly different tected at that time in HPP cheeses. In contrast,
from that of putrescine. In control cheese, it reached 0.064 mg/ phenylethylamine and tryptamine were not detected in
g DM on day 14 and a maximum level of 0.113 mg/g DM on control or HPP cheeses in the present work. Also, Calzada
day 21. In HPP cheeses, the maximum concentrations were et al. (2013a) reported a significant increase in cadaverine
generally reached on day 21, with values ranging from 0.067 concentration in control cheese from day 60 onwards while
to 0.083 mg/g DM. On day 240, it attained 0.080 mg/g DM in this biogenic amine did not increase in control cheese from
control cheese, 0.032–0.034 mg/g DM in 400 MPa cheeses day 60 onwards in the present work. These facts point out
and 0.042–0.081 mg/g DM in 600 MPa cheeses. As counts of differences in the BA formation ability of cheese microbiota
Gram-negative bacteria declined, naturally in control cheese between both works.
Textural Characteristics a
0.25
a
Firmness of control cheese declined from 0.075 N m on day 1 a
c
a a
(data not shown) to 0.026 N m on day 21 and afterwards 0.20
a bc
ab
increased gradually up to 0.141 N m on day 240. HPP cheeses ab
a
Firmness (N m)
exhibited lower firmness than control cheese on day 21, with 0.15
values ranging from 0.013 to 0.020 N m, and higher firmness
b b
on day 240, with values ranging from 0.151 to 0.200 N m
0.10 a
(Fig. 7). Elasticity of control cheese decreased from 0.081 N/ aa
mm2 on day 1 down to 0.027 N/mm2 on day 21 and rose a a a aa
afterwards to 0.226 N/mm2 on day 240. HPP cheeses had 0.05 b b b
ab ab
a
elasticity values close to those of control cheese on day 21, a a
ranging from 0.016 to 0.027 N/mm2, and higher values on day 0.00
240, ranging from 0.299 to 0.411 N/mm 2 (Fig. 7). 14 d 21 d 60 d 120 d 180 d 240 d
Fracturability could not be quantified in control cheese until b

day 240, time at which it attained 2.43 N, while in HPP c
0.45
b
cheeses, it could be determined on day 180, with values of bc
bc
0.40
12.05–17.20 N, and on day 240, with values of 18.97– ab ab
24.03 N, markedly higher than those of control cheese 0.35 ab
ab
(Fig. 7). The three texture parameters studied were at consid- a
Elasticity (N/mm )
2 0.30
erably lower levels in 60-day cheeses of the present work than a
0.25
in 50-day HPP and control Hispánico cheeses (Ávila et al.
0.20
2006), in which firmness ranged from 0.21 to 0.34 J, elasticity b
from 0.14 to 0.72 N/mm2 and fracturability from 13.58 to 0.15 ab
aa a
29.28 N. The minor differences in texture parameters caused 0.10
aa a a a
by HPP observed in the present work should not influence b b b
0.05 a ab ab
consumer acceptance. Rheological properties of Cheddar a a
cheese were not adversely affected by pressure, according to 0.00

14 d 21 d 60 d 120 d 180 d 240 d
Wick et al. (2004), while significant increases of fracture strain
and fracture stress values, but not of firmness, were reported c
by Rynne et al. (2008). In the present study, firmness was 26
c
significantly correlated with DM (r=0.950, p<0.001) and pH 24 bc

value (r=0.524, p<0.01) and negatively correlated with α- 22 b
b
casein (r = −0.717, p < 0.001) and β-casein (r = −0.702, 20 b
18
p<0.001), considering control and HPP cheeses from day 21
Fracturability (N)
16
ab
to day 240. Elasticity was also significantly correlated with a
14 a
DM (r=0.941, p<0.001) and pH value (r=0.518, p<0.01) 12
and negatively correlated with α-casein (r=−0.723, p<0.001) 10
and β-casein (r=−0.691, p<0.001), considering control and 8
HPP cheeses from day 21 to day 240. Fracturability was 6 a
significantly correlated only with β-casein (r = −0.762, 4
2
p<0.05), considering control and HPP cheeses from day 180
0
to day 240. 14 d 21 d 60 d 120 d 180 d 240 d
The texture of cheese at any specific stage of ripening is Fig. 7 Texture characteristics (a firmness, b elasticity, c, fracturability)
determined primarily by its pH and the ratio of moisture to during ripening and refrigerated storage of control cheese (black) and
intact casein (Lawrence et al. 1987), although other factors cheeses HP-treated at 400 MPa on day 14 (white), 600 MPa on day 14
such as the fat content also influence cheese textural charac- (horizontally striped), 400 MPa on day 21 (dotted) or 600 MPa on day 21
(obliquely striped). Means (bars with SEM) at the same sampling date
teristics (Oliveira et al. 2011). On the one hand, cheese texture with the same letter do not differ significantly (p>0.05)
tends to soften when the pH rises from 5.0 to 6.0, as proven for
Camembert-type cheeses (Vassal et al. 1986). On the other
hand, the protein network weakens when caseins are hydro- water in the system, which becomes tied up. The lower the
lyzed during ripening, but as each peptide bond in caseins is moisture to casein ratio, the firmer the cheese protein matrix
cleaved the two new ionic groups compete for the available will be. These facts would explain why texture strengthened in
a b
10 10
9 9
b
8 8 a a ab a
a a a a a
a
a a a
a a a a a ab
ab
a a
7 a a a a 7 a a a
a a a a a a a
Flavour intensity
a a
Flavour quality
a a a a
6 a 6
a
5 5
a a a a
a
4 4
3 3
2 2
1 1
0 0
21 d 60 d 120 d 180 d 240 d 21 d 60 d 120 d 180 d 240 d
c d
10 10
9 9
8 8
7 7
Umami flavour
Bitter flavour
6 b b 6
b
ab a
5 ab ab 5
b b a a
b ab ab a a a
4 ab ab 4 a a a
a a a
ab ab ab a a a a a a a
3 3 a a a
a a
a a a a
2 a aa 2 a a
a
1 1
0 0
21 d 60 d 120 d 180 d 240 d 21 d 60 d 120 d 180 d 240 d
Fig. 8 Sensory characteristics (a flavour intensity, b flavour quality, c (dotted) or 600 MPa on day 21 (obliquely striped). Means (bars with
bitter flavour, d umami flavour) during ripening and refrigerated storage SEM) at the same sampling date with the same letter do not differ
of control cheese (black) and cheeses HP-treated at 400 MPa on day 14 significantly (p>0.05)
(white), 600 MPa on day 14 (horizontally striped), 400 MPa on day 21
the present study during cheese ripening, in spite of casein cheese. Increases in flavour intensity during ripening of con-
degradation. The firmer texture of HPP cheeses can be asso- trol and HPP cheeses in the present study were more marked
ciated with the microstructural changes induced by pressuri- than those recorded for ovine milk cheeses (Calzada et al.
zation. HPP cheeses had a more continuous protein matrix 2014a), while flavour quality of control cheese hardly de-
than control cheese (O’Reilly et al. 2003) and denser and more clined during ripening contrarily to the drastic decline ob-
compact protein network (Picon et al. 2013b), as shown by served for ovine milk cheese, probably caused by the abun-
confocal scanning laser microscopy. dance of undesirable microorganisms in the latter cheese
(Calzada et al. 2013a) and the excessive proteolysis brought
Sensory Characteristics about by the cardoon coagulant (Calzada et al. 2014a). In the
present study, acid scores (data not shown) did not vary
Flavour intensity scores increased gradually from day 21 to significantly with HPP or time of ripening. Bitter scores were
day 240 in control and HPP cheeses, with no significant significantly (p<0.05) influenced by HPP and time of ripen-
differences between cheeses at any stage of ripening ing, with a gradual increase during ripening in all cheeses and
(Fig. 8). Flavour quality of control cheese remained fairly with higher values for HPP cheeses than for control cheese
unchanged from day 21 to day 180 and then declined from from day 60 onwards (Fig. 8), but the increase in bitterness did
day 180 to day 240 (Fig. 8). Quality scores of HPP cheeses did not suffice to affect flavour quality. Sweet and salty scores
not differ from those of control cheese throughout ripening, were not influenced by HPP or time of ripening (data not
except for 400W2 cheese on day 120 which had a significant- shown). Umami scores increased gradually in control and
ly (p<0.05) lower quality score than the respective control HPP cheeses from day 21 to day 240, with no significant
differences between cheeses (Fig. 8). HP-treated Cheddar than in control cheese on day 240. Total biogenic amines were
cheese showed significantly less strong flavour than control lowered even more, by up to 94 % in 600 MPa cheeses on day
cheese, and lower pungent, onion-like, salty, acidic, and bitter 240. A firmer texture, determined by instrumental methods,
flavour scores (Rynne et al. 2008). was recorded for HPP cheeses than for control cheese at the
Significant correlations were found for flavour intensity end of the storage period. In spite of the higher bitterness
scores with total FAA (r=0.821, p<0.001) and hydrophilic scores recorded for HPP cheeses from day 60 onwards, fla-
peptides (r=0.554, p<0.01), umami scores with total FAA vour intensity and flavour quality were not significantly influ-
(r=0.881, p<0.001) and hydrophilic peptides (r=0.465, enced by treatments. Minor changes in cheese texture and
p<0.05), and flavour intensity scores with umami scores (r= flavour caused by HPP should not affect consumer accep-
0.943, p<0.001), considering cheeses from day 21 to day 240. tance. HPP is a feasible procedure to improve the safety of
Negative correlations were found for flavour quality scores of cheese made from unpasteurized cow milk, due to its
those cheeses with total FAA (r=−0.505, p<0.01), hydrophil- favourable effect on the reduction of biogenic amines without
ic peptides (r=−0.648, p<0.001), flavour intensity scores (r= compromising sensory characteristics during a prolonged rip-
−0.561, p<0.01), umami scores (r=−0.704, p<0.001) and ening and storage period.
bitterness scores (r=−0.796, p<0.001). Some of these corre-
lations were to be expected since FAA and, at a considerably Acknowledgments This work was supported by project AGL2009-
lesser degree, hydrophilic peptides contribute to the develop- 07801 from the Spanish Ministry of Science and Innovation (MICINN).
ment of cheese flavour and particularly to the umami notes J. Calzada was the recipient of a MICINN fellowship. The authors thank
the valuable help of A. Vence with the manufacture and supply of cheeses
(Sousa et al. 2001). Also, the negative correlation between and of Hiperbaric (Burgos, Spain) with high-pressure treatments.
flavour quality and bitterness has been previously reported
(Picon et al. 2013a). The negative correlations of flavour
quality with total FAA and hydrophilic peptides cannot be
considered as cause-effect relationships but are rather explain- References
able by their opposite evolution during cheese ripening, with a
decrease in flavour quality scores and increases in the levels of Arqués, J. L., Rodriguez, E., Gaya, P., Medina, M., & Nuñez, M. (2005).
FAA and hydrophilic peptides. In fact, the correlations of Effect of combinations of high-pressure treatment and bacteriocin-
producing lactic acid bacteria on the survival of Listeria
flavour quality scores with total FAA at a particular stage of monocytogenes in raw milk cheese. International Dairy Journal,
ripening (i.e. 120, 180 or 240 days) were all of positive sign 15, 893–900.
although non-significant, with r values under 0.30, while the Ávila, M., Garde, S., Gaya, P., Medina, M., & Nuñez, M. (2006). Effect
correlations of flavour intensity scores with total FAA at each of high-pressure treatment and a bacteriocin-producing lactic culture
on the proteolysis, texture, and taste of Hispánico cheese. Journal of
of those stages of ripening showed r values above 0.75. Dairy Science, 89, 2882–2893.
Calzada, J., del Olmo, A., Picón, A., Gaya, P., & Nuñez, M. (2013a).
Reducing biogenic-amine-producing bacteria, decarboxylase activ-
ity, and biogenic amines in raw milk cheese by high-pressure treat-
ments. Applied and Environmental Microbiology, 79, 1277–1283.
Conclusions Calzada, J., del Olmo, A., Picon, A., Gaya, P., & Nuñez, M. (2013b).
Proteolysis and biogenic amine buildup in high-pressure treated
HPP of cheese made from unpasteurized cow milk inactivated ovine milk blue-veined cheese. Journal of Dairy Science, 96,
microorganisms and enzymes. Death rates of microorganisms 4816–4829.
Calzada, J., del Olmo, A., Picon, A., Gaya, P., & Nuñez, M. (2014a).
mostly depended on the pressure level applied, with reduc- Using high-pressure processing for reduction of proteolysis and
tions of lactic acid bacteria counts by 2.0–2.3 log units at prevention of over-ripening of raw milk cheese. Food and
400 MPa and by 5.8–6.0 log units at 600 MPa. Populations Bioprocess Technology, 7, 1404–1413.
of coagulase-positive staphylococci and Gram-negative bac- Calzada, J., del Olmo, A., Picon, A., Gaya, P., & Nuñez, M. (2014b).
High-pressure processing for the control of lipolysis, volatile com-
teria were lowered to counts below the detection level in all pounds and off-odours in raw milk cheese. Food and Bioprocess
HPP cheeses. Aminopeptidase inactivation was more depen- Technology, 7, 2207–2217.
dent on the day cheeses were treated, with activity losses of Curtin, A. C., & McSweeney, P. L. H. (2003). Catabolism of aromatic
68–73 % when HPP was applied on day 14 and 87–89 % on amino acids in cheese-related bacteria: aminotransferase and decar-
boxylase activities. Journal of Dairy Research, 70, 249–252.
day 21. The hydrolysis of αs-casein was enhanced in cheeses Fernández-García, E., Tomillo, J., & Nuñez, M. (1999). Effect of added
treated at 400 MPa and that of β-, κ- and para-κ-caseins in 400 proteinases and level of starter culture on the formation of biogenic
and 600 MPa cheeses. Concomitantly, a more pronounced amines in raw milk Manchego cheese. International Journal of
increase in γ-caseins, hydrophilic peptides and hydrophobic Food Microbiology, 52, 189–196.
García-Risco, M. R., Olano, A., Ramos, M., & López-Fandiño, R.
peptides was recorded for all the HPP cheeses than for control (2000). Micellar changes induced by high pressure. Influence in
cheese. However, the formation of FAA was retarded in HPP the proteolytic activity and organoleptic properties of milk.
cheeses, with up to 64 % less total FAA in 600 MPa cheeses Journal of Dairy Science, 83, 2184–2189.
Garde, S., Arqués, J. L., Gaya, P., Medina, M., & Nuñez, M. (2007). Messens, W., Estepar-Garcia, J., Dewettinck, K., & Huyghebaert, A.
Effect of high-pressure treatments on proteolysis and texture of (1999). Proteolysis of high-pressure-treated gouda cheese.
ewes’ raw milk La Serena cheese. International Dairy Journal, International Dairy Journal, 9, 775–782.
17, 1424–1433. Norton, T., & Sun, D.-W. (2008). Recent advances in the use of high
Garde, S., Tomillo, J., Gaya, P., Medina, M., & Nuñez, M. (2002). pressure as an effective processing technique in the food industry.
Proteolysis in Hispánico cheese manufactured using a mesophilic Food and Bioprocess Technology, 1, 2–34.
starter, a thermophilic starter and bacteriocin-producing Oliveira, N. M., Dourado, F. Q., Peres, A. M., Silva, M. V., Maia, J. M., &
Lactococcus lactis subsp. lactis INIA 415 adjunct culture. Journal Teixeira, J. A. (2011). Effect of guar gum on the physicochemical,
of Agricultural and Food Chemistry, 50, 3479–3485. thermal, rheological and textural properties of green edam cheese.
Huppertz, T., Fox, P. F., & Kelly, A. L. (2004). Susceptibility of plasmin Food and Bioprocess Technology, 4, 1414–1421.
and chymosin in cheddar cheese to inactivation by high pressure. O’Reilly, C. E., O’Connor, P. M., Murphy, P. M., Kelly, A. L., &
Journal of Dairy Research, 71, 496–499. Beresford, T. P. (2002). Effects of high-pressure treatment on via-
Joosten, H. M. L. J., & Northolt, M. D. (1987). Conditions allowing the bility and autolysis of starter bacteria and proteolysis in cheddar
formation of biogenic-amines in cheese. 2. Decarboxylase proper- cheese. International Dairy Journal, 12, 915–922.
ties of some nonstarter bacteria. Netherlands Milk and Dairy O’Reilly, C. E., Kelly, A. L., Oliveira, J. C., Murphy, P. M., Auty, M. A.
Journal, 41, 259–280. E., & Beresford, T. P. (2003). Effect of varying high-pressure
Juan, B., Ferragut, V., Buffa, M., Guamis, B., & Trujillo, A. J. (2007). treatment conditions on acceleration of ripening of cheddar cheese.
Effects of high pressure on proteolytic enzymes in cheese: relation- Innovative Food Science and Emerging Technologies, 4, 277–284.
ship with the proteolysis of ewe milk cheese. Journal of Dairy Picon, A., Alonso, R., Gaya, P., & Nuñez, M. (2013a). High-pressure
Science, 90, 2113–2125. treatment and freezing of raw goat milk curd for cheese manufac-
Krause, I., Bockhardt, A., Neckermann, H., Henle, T., & Klostermeyer, ture: effects on cheese characteristics. Food and Bioprocess
H. (1995). Simultaneous determination of amino acids and biogenic Technology, 6, 2820–2830.
amines by reversed-phase high-performance liquid chromatography Picon, A., Alonso, R., van Wely, K. H. M., & Nuñez, M. (2013b).
of the dabsyl derivatives. Journal of Chromatography A, 715, 67– Microstructural, textural and colour characteristics during ripening
79. of Hispánico cheese made using high-pressure-treated ovine milk
Ladero, V., Fernández, M., Cuesta, I., & Alvarez, M. A. (2010). curd. Food and Bioprocess Technology, 6, 3056–3067.
Quantitative detection and identification of tyramine-producing en- Pircher, A., Bauer, F., & Paulsen, P. (2007). Formation of cadaverine,
terococci and lactobacilli in cheese by multiplex qPCR. Food histamine, putrescine and tyramine by bacteria isolated from meat,
Microbiology, 27, 933–939. fermented sausages and cheeses. European Food Research and
Ladero, V., Fernández, M., Calles-Enríquez, M., Sánchez-Llana, E., Technology, 226, 225–231.
Cañedo, E., Martín, M. C., & Alvarez, M. A. (2012). Is the produc- Rynne, N. M., Beresford, T. P., Guinee, T. P., Sheehan, E., Delahunty, C.
tion of the biogenic amines tyramine and putrescine a species-level M., & Kelly, A. L. (2008). Effect of high-pressure treatment of 1
trait in enterococci? Food Microbiology, 30, 132–138. day-old full-fat Cheddar cheese on subsequent quality and ripening.
Lawrence, R. C., Creamer, L. K., & Gilles, J. (1987). Texture develop- Innovative Food Science and Emerging Technologies, 9, 429–440.
ment during cheese ripening. Journal of Dairy Science, 70, 1748– Sousa, M. J., Ardö, Y., & McSweeney, P. L. H. (2001). Advances in the
1760. study of proteolysis during cheese ripening. International Dairy
Leuschner, R. G., Heidel, M., & Hammes, W. P. (1998). Histamine and Journal, 11, 327–345.
tyramine degradation by food fermenting microorganisms. Vassal, L., Monnet, V., Le Bars, D., Roux, C., & Gripon, J. C. (1986).
International Journal of Food Microbiology, 39, 1–10. Relation entre le pH, la composition chimique et la texture des
Linares, D., Martín, M. C., Ladero, V., Alvarez, M. A., & Fernández, M. fromages de type camembert. Le Lait, 66, 341–351.
(2011). Biogenic amines in dairy products. Critical Reviews in Food Wick, C., Nienaber, U., Anggraeni, O., Shellhammer, T. H., & Courtney,
Science and Nutrition, 51, 691–703. P. D. (2004). Texture, proteolysis and viable lactic acid bacteria in
Malone, A. S., Wick, C., Shellhammer, T. H., & Courtney, P. D. (2003). commercial cheddar cheeses treated with high pressure. Journal of
High pressure effects on proteolytic and glycolytic enzymes in- Dairy Research, 71, 107–115.
volved in cheese manufacturing. Journal of Dairy Science, 86, Yvon, M., & Rijnen, L. (2001). Cheese flavour formation by amino acid
1139–1146. catabolism. International Dairy Journal, 11, 185–201.

Capítulo 10.
Effect of high-pressure-processing on the lipolysis, volatile
compounds, odour and colour of cheese made from
unpasteurized milk.

201

Fotografía: colonias de bacterias aerobias totales en agar PC (superior, izquierda),

bacterias lácticas en agar MRS (superior, derecha), bacterias Gram negativas en agar
MacConkey (centro, izquierda), coliformes y enterobacterias lactosa negativo en agar
VRB (centro, derecha), estafilococos coagulasa positivo y negativo en agar BP (inferior,
izquierda), y Micrococcaceae en agar MS (inferior, derecha), obtenidas a partir de
siembras de queso Arzúa-Ulloa.
Food and Bioprocess Technology: An International Journal
Effect of High-Pressure-Processing on the Lipolysis, Volatile Compounds, Odour and
Colour of Cheese Made from Unpasteurized Milk
--Manuscript Draft--
Manuscript Number:
Full Title: Effect of High-Pressure-Processing on the Lipolysis, Volatile Compounds, Odour and
Colour of Cheese Made from Unpasteurized Milk
Article Type: Original Research
Keywords: High-pressure-processing; Lipolysis; Volatile compounds; Odour; Cheese
Corresponding Author: Manuel Nunez

INIA
Madrid, SPAIN
Corresponding Author Secondary

Information:
Corresponding Author's Institution: INIA
Corresponding Author's Secondary

Institution:
First Author: Javier Calzada, MSc
First Author Secondary Information:

Order of Authors: Javier Calzada, MSc
del Olmo Ana, PhD
Antonia Picon, PhD
Manuel Nunez
Order of Authors Secondary Information:

Abstract: Cheeses made from unpasteurized milk were high-pressure-processed (HPP) 14 or 21
days after manufacture at 400 or 600 MPa, in order to prevent excessive accumulation
of flavour compounds, in particular free fatty acids (FFA) and volatile compounds (VC).
They were compared with untreated control cheese throughout a 240-day period. On
day 60, cheeses treated at 400 MPa had esterase activity values 57.1-58.0% lower
than control cheese while cheeses treated at 600 MPa had values 82.3-82.8% lower,
with no significant differences between total FFA concentrations of cheeses. However,
on day 240 total FFA concentrations were 6.5-9.0% lower in 400 MPa cheeses than in
control cheese, and 15.0-16.9% lower in 600 MPa cheeses, with more marked
differences for short-chain FFA and less marked for hexadecanoic and octadecanoic
acids. Fifty-two of the 70 VC found in cheese were significantly influenced by HPP. On
day 240, total alcohols reached the highest levels in control and 400 MPa cheeses,
total aldehydes, ketones and hydrocarbons in 600 MPa cheeses, and total sulphur
compounds in 400 MPa and 600 MPa cheeses. Levels of total volatile acids, esters,
ethers, benzenic compounds and terpenoids in 240-day cheese were not affected by
HPP. Odour quality and intensity were not significantly influenced by HPP. Higher L*
values were recorded for 600 MPa cheeses, and higher a* values for control and 400
MPa cheeses. Principal component analysis on the groups of volatile compounds
highlighted the different evolvement of volatiles in control, 400 MPa and 600 MPa
cheeses during ripening.
Powered by Editorial Manager® and ProduXion Manager® from Aries Systems Corporation
Manuscript
Click here to download Manuscript: CALZADA et al. FABT.doc
Click here to view linked References
Effect of High-Pressure-Processing on the Lipolysis, Volatile Compounds,
Odour and Colour of Cheese Made from Unpasteurized Milk
Javier Calzada · Ana del Olmo · Antonia Picon · Manuel Nuñez
Departamento de Tecnología de Alimentos, INIA, Carretera de La Coruña Km 7, Madrid, 28040
Spain
Short running head: Lipolysis and volatiles in HPP cheese
Corresponding author:
M. Nuñez. Phone # 34-913476799; Fax # 34-913572293. E-mail: nunez@inia.es
1
Abstract Cheeses made from unpasteurized milk were high-pressure-processed (HPP) 14 or 21
days after manufacture at 400 or 600 MPa, in order to prevent excessive accumulation of flavour
compounds, in particular free fatty acids (FFA) and volatile compounds (VC). They were compared
with untreated control cheese throughout a 240-day period. On day 60, cheeses treated at 400 MPa
had esterase activity values 57.1-58.0% lower than control cheese while cheeses treated at 600 MPa
had values 82.3-82.8% lower, with no significant differences between total FFA concentrations of
cheeses. However, on day 240 total FFA concentrations were 6.5-9.0% lower in 400 MPa cheeses
than in control cheese, and 15.0-16.9% lower in 600 MPa cheeses, with more marked differences
for short-chain FFA and less marked for hexadecanoic and octadecanoic acids. Fifty-two of the 70
VC found in cheese were significantly influenced by HPP. On day 240, total alcohols reached the
highest levels in control and 400 MPa cheeses, total aldehydes, ketones and hydrocarbons in 600
MPa cheeses, and total sulphur compounds in 400 MPa and 600 MPa cheeses. Levels of total
volatile acids, esters, ethers, benzenic compounds and terpenoids in 240-day cheese were not
affected by HPP. Odour quality and intensity were not significantly influenced by HPP. Higher L*
values were recorded for 600 MPa cheeses, and higher a* values for control and 400 MPa cheeses.
Principal component analysis on the groups of volatile compounds highlighted the different
evolvement of volatiles in control, 400 MPa and 600 MPa cheeses during ripening.
Keywords High-pressure-processing · Lipolysis · Volatile compounds · Odour · Cheese
2
Introduction
During the manufacture and ripening of cheese, a myriad of compounds responsible for cheese
flavour are formed by the microorganisms and enzymes involved in glycolysis, lipolysis,
proteolysis and the subsequent secondary reactions (McSweeney and Sousa 2000; Yvon and Rijnen
2001; Collins et al. 2003). The numerous products resulting from carbohydrate metabolism,
triglyceride hydrolysis and protein breakdown are further degraded, giving rise to non-volatile
flavour compounds and volatile potent odorants responsible for the typical flavour, aroma and
odour notes of each cheese variety. More than 600 volatile compounds have been identified in
different cheese varieties and many of those compounds have been related to particular odour and
aroma notes (Curioni and Bosset 2002).
Besides the coagulant enzymes and starter cultures used as a rule in cheese manufacture, the
enzymes and microorganisms coming from milk are main agents in the ripening process of cheeses
made from unpasteurized milk (Fernández-García et al. 2002; Gaya et al. 2005). Long ripening and
refrigerated storage periods are characteristic of some cheese varieties, forced by regulatory issues
or market conditions. The action of enzymes and microorganisms, which persists even at
refrigeration temperatures, may result in cheese over-ripening before consumption. Unpasteurized
milk cheeses are more prone to over-ripening, because of their more marked enzyme activity and
their complex microbial metabolic reactions. Although freezing of ripe cheese considerably retards
biochemical changes and maintains flavour characteristics, the texture and the visual appearance of
cheese may become negatively affected (Tejada et al. 2000; Van Hekken et al. 2005).
High-pressure-processing (HPP) meets the consumer demand for minimally processed fresher-
tasting foods, since its effect on flavour characteristics is negligible (Norton and Sun 2008). HPP
reduces the populations of pathogenic and spoilage microorganisms in milk and cheese (O’Reilly et
al. 2000; Shao and Ramaswamy 2011) and inactivates proteinases, peptidases and esterases present
3
in cheese when sufficiently high pressure levels are applied (Malone et al. 2003; Huppertz et al.
2004; Juan et al. 2007; Ávila et al. 2007). Free fatty acids (FFA) and volatile compounds (VC) of
cheese are also affected by HPP (Ávila et al. 2006; Ávila et al. 2007).
The application of HPP to prevent the excessive accumulation of flavour compounds in cheese
made from unpasteurized milk seems therefore a feasible strategy, on the basis of the results of
previous studies. The effects of HPP are dependent on the pressure level applied and the length of
treatment. In the particular case of cheese, the stage of ripening at which HPP is applied is also a
crucial parameter for the retention of sensory characteristics. Garde et al. (2007) reported the lowest
flavour quality scores for ripe cheese treated at 400 MPa on day 2 after manufacture while HPP did
not impair the sensory characteristics of cheese treated at 400 MPa on day 50.
In a previous study, the proteolysis of unpasteurized milk cheese was decelerated and the levels
of biogenic amines lowered by applying HPP at 400 or 600 MPa, on days 14 and 21 after
manufacture (Calzada et al. 2014b). The objective of the present work was to investigate the
suitability of HPP at 400 or 600 MPa, applied on days 14 or 21 after manufacture, in order to
control the excessive build-up of flavour compounds in unpasteurized milk cheese. Lipolysis,
volatile compounds, odour and colour of HPP-treated cheeses were compared with those of
untreated control cheese throughout a 240-day period.
Materials and Methods
Cheese Manufacture and High-pressure-processing
Cheese was manufactured from 750 L of unpasteurized cow milk in each of two trials, carried out
on consecutive days as previously described (Calzada et al. 2014b). Cheeses (1.15 kg average
weight out of the press) were ripened at 8 ºC and 72% RH until day 60 and at 5 ºC and 75% RH
4
afterwards. Part of the cheeses, coded as 400W2, 600W2, 400W3 and 600W3, were HPP-treated at
400 or 600 MPa after ripening for 14 or 21 days, respectively, as previously described (Calzada et
al. 2014b). After treatments, HPP cheeses were unpackaged and ripened under the same conditions
of control cheese.
From each of the two cheese-making trials, one cheese was analyzed on day 1, three cheeses
(control, 400W2, 600W2) on day 14 and five cheeses (control, 400W2, 600W2, 400W3, 600W3) on
each of the sampling dates day 21, day 60, day 120, day 180 and day 240. At each of these sampling
dates, two 100 g pieces per cheese were wrapped in aluminum foil, vacuum-packaged, and frozen at
-40 ºC until chemical analyses.
Enzymatic and Chemical Determinations
Esterase activity was determined in duplicate as described by Ávila et al. (2007), with some
modifications. Ten grams of cheese were homogenized with 20 mL of 0.1 M, pH 7.0 phosphate
buffer in an Ultra-Turrax T8 homogenizer (IKA Labortechnik, Staufen, Germany), centrifuged at
10,000 g for 20 min at 4 ºC and filtered through Whatman No. 2 paper (Whatman Int. Ltd.,
Maidstone, UK). The chromogenic substrate was α-naphthylbutyrate (Sigma-Aldrich, Steinhem,
Germany). The assay mixture (30 μL of chromogenic substrate, 600 μL of distilled water, 100, 200
or 400 μL of cheese homogenate and 0.1 M, pH 7.5 phosphate buffer up to a final volume of 1230
μL) was incubated for 1 h at 37 ºC in a water bath and centrifuged at 12,000 g for 5 min at room
temperature. Finally, 900 μL of supernatant were mixed with 150 μL of 2.7 mg/mL Fast Red TR
salt (Sigma-Aldrich) aqueous solution. After 5 min at room temperature, the absorbance was
measured at 537 nm using a DU650 spectrophotometer (Beckman Coulter Inc., Brea, CA). Esterase
activity was calculated from absorbance values in the range of 0.1 to 0.9 by means of a α-naphthol
5
standard curve. One unit of enzymatic activity was defined as the amount of enzyme that liberated 1
pmol of α-naphthol per min and g of cheese at 37 ºC and pH 7.5.
Free fatty acids from butyric acid (C4:0) to linolenic acid (C18:3), together with acetic, propionic
and benzoic acids, were determined in duplicate by gas chromatography (GC) as described by
Fernández-García et al. (2006). Prior to analysis, frozen cheese pieces were thawed overnight at 4
ºC. A solid-phase extraction technique was used for the extraction of carboxylic acids from cheese,
with pentanoic, nonanoic and heptadecanoic acids added as internal standards. An HP 6890 gas
chromatograph (Agilent Technologies, Las Rozas, Spain) equipped with an automatic sampler (HP
7683), a split/splitless injector, a FFAP column (Agilent Technologies, 30 m x 0.32 mm i.d. x 0.25
μm film thickness) and a flame ionization detector, was used for the analysis. Injection (1 μL of
sample) was performed in split mode at 1:20 split ratio, at 260 ºC. Helium was the carrier gas, with
the flow set for maintaining a constant pressure of 0.80 kg/cm2. For chromatographic separation the
temperature was increased from 65 to 240 ºC, at a rate of 10 ºC/min, and held at 240 ºC for 12.5
min. Standard solutions of carboxylic acids were used for the calculation of calibration curves.
Individual acids were separated, identified and quantified, and their concentrations expressed in mg
per gram of cheese dry matter (DM).
Volatile compounds were determined in triplicate by gas chromatography-mass spectrometry
(GC-MS) as described by Calzada et al. (2014a). They were extracted from cheese using a solid-
phase microextraction method (SPME). Ten grams of cheese were homogenized with 20 g of
Na2SO4 and 25μL of an aqueous solution of 1,058 mg/L cyclohexanone in a mechanical grinder.
Five grams of the mixture were weighed in a 15 mL headspace glass vial sealed with a PTFE faced
silicone septum (Supelco, Bellefonte, PA, USA) and the vials were submerged in a thermostatic
bath at 30 °C. A SPME manual holder equipped with a 2 cm x 50/30 μm StableFlex
divinylbenzene/carboxen/polydimethylsiloxane (DVB/CAR/PDMS) coated fibre (Supelco) was
inserted through the PTFE septum for headspace extraction, after which it was inserted into the GC
6
injection port for desorption (270 °C/10 min in splitless mode). An HP 6890-MSD HP 5973
apparatus (Agilent, Palo Alto, CA, USA) with a capillary column (60 m long; 0.25 mm i.d.; 0.5 μm
film thickness; Innowax, Agilent Technologies) was used for GC-MS, with helium flow at 1.4
mL/min for 1 min followed by 1 mL/min. The temperature program was 7 min at 40 ºC, first ramp
2 ºC/min to 90 ºC, second ramp 3 ºC/min to 150 ºC, final ramp 9 ºC/min to 240 ºC, and 8 min at
240 ºC. Detection was performed with electron impact ionization, with 70 eV ionization energy
operating in the full-scan mode at 1.74 scans/s. Source and quadrupole temperatures were 230 and
150 ºC, respectively. Compound identification was carried out by injection of commercial standards
and by spectra comparison using the Wiley7Nist05 Library (Wiley & Sons Inc., Germany). The
sum of abundances of characteristic ions was used for the semi-quantitation of compounds. Relative
abundances of compounds were calculated by dividing the peak area of each compounds by the
peak area of the internal standard and multiplying the quotient by 102.
Odour Evaluation
Seventeen trained panellists carried out the evaluation of odour intensity and quality (preference) of
21-, 60-, 120- and 240-day cheeses, scoring on a 0- to 10-point scale. Five cheeses per session (one
control and four HPP cheeses, manufactured on the same day), coded with 3-digit numbers, were
randomly presented to panellists. Odour was defined as the olfactory sensation felt directly by
smelling the cheese (Fernández-García et al. 2002). In addition, panellists were asked to evaluate
putrid, rancid and acid odour notes, also scoring on a 0- to 10-point scale.
Colour Determination
7
Colour parameters in the CIE-LAB colour space L* (lightness), a* (redness) and b* (yellowness)
were determined at six points of the cheese interior, as previously described (Picon et al. 2013). A
CM-2600d spectrocolorimeter equipped with a SpectraMagic 3.6 software (Minolta Camera Co.,
Osaka, Japan), with iluminant D65 (standard daylight) and a 10º observer angle, with specular
component included (SCI), was used.
Statistical Analysis
Two-way analyses of variance (ANOVA) were carried out on the levels of FFA and VC, with HPP
treatment and time after manufacture as main effects. Means were compared using Tukey’s test,
with p assigned at 0.05. Principal component analyses (PCA) were carried out on individual FFA
concentrations, total levels of VC groups and individual VC levels, for the discrimination of
samples according to HPP treatment and time after manufacture. The SPSS Win 14.0 software
(SPSS Inc., Chicago, IL, USA) was used for the statistical analysis of data.
Results and Discussion
Acetic, Propionic and Benzoic Acids
The concentrations of acetic, propionic and benzoic acids in unpasteurized milk cheese were all
significantly (p<0.001) influenced by HPP treatment and time after manufacture, according to the
ANOVA. Acetic acid increased steadily from 0.707 mg/g DM on day 1 (data not shown) to a
maximum level of 2.102 mg/g DM in 240-day control cheese (Table 1). It is produced by lactic acid
bacteria and other microorganisms coming from raw milk, principally through the metabolism of
lactose, lactate and citrate. HPP influenced negatively the formation of acetic acid, in particular
8
when 600 MPa were applied to cheese. Thus, the concentrations of acetic acid on day 240 were
35.9-39.5% lower in 400 MPa cheeses, and 56.4-56.6% lower in 600 MPa cheeses, than in control
cheese (Table 1). Acetic acid, with a typical vinegar odour note, is a major odorant of cheese
varieties such as Cheddar, Gruyère and Emmental (Curioni and Bosset 2002). It contributes to
cheese flavour and aroma by itself and by forming esters with fruity notes.
Propionic acid followed a different accumulation pattern. It sharply increased from 0.031 mg/g
DM on day 1 (data not shown) to a maximum level of 0.395 mg/g DM in 14-day control cheese
(Table 1). Thereafter, its concentration generally declined, with slight differences between HPP and
control cheeses. Propionic acid, which also shows a vinegar odour note, mostly derives from the
microbial metabolism of lactate. It is found in Swiss-type cheeses, Camembert and some ewe milk
cheeses (Molimard and Spinnler 1996; Fernández-García et al. 2004). Similarly to acetic acid,
propionic acid has a direct role in cheese odour and aroma and also contributes to ester formation.
Benzoic acid increased from 0.027 mg/g DM on day 1 (data not shown) to 0.234 mg/g DM on
day 14 (Table 1), similarly to propionic acid. Following this initial increase, its accumulation
persisted in control cheese, in which it reached 0.398 mg/g DM on day 180, and in most of the HPP
cheeses (Table 1). Concentrations found from day 14 onwards (Table 1) clearly exceeded the
maximum concentrations of benzoic acid reported for pasteurized cow milk cheeses (0.0245 mg/g)
or raw cow milk cheeses (0.0287 mg/g) by Iammarino et al. (2011), although up to 0.137 mg/g were
found in Vacherin cheeses from different manufacturers (Sieber et al. 1995). According to these
authors, benzoic acid present in cheese may derive from the microbial enzymatic conversion of
hippuric acid, from phenylalanine breakdown or from benzaldehyde auto-oxidation.
Branched-chain carboxylic acids such as 2-methylpropanoic, 2-methylbutanoic and 3-
methylbutanoic acids, respectively derived from the catabolism of valine, isoleucine and leucine
(Yvon and Rijnen 2001), were not detected in any of the analyzed samples of control and HPP
cheeses, probably due to the absence or low levels of bacteria with the required enzymatic activity.
9
Esterase Activity and Free Fatty Acids
Esterase activity of control cheese increased from 4.63 pmol α-naphthol min-1 g-1 on day 1 to 10.75
pmol α-naphthol min-1 g-1 on day 120 (Table 2), a result which can be associated with the lysis of
starter lactic acid bacteria during ripening followed by the release of their intracellular esterases to
the medium. HPP was responsible for a decline in esterase activity immediately after treatments, by
30.0, 49.1, 45.4 and 70.8% in 400W2, 600W2, 400W3 and 600W3 cheeses, respectively (Table 2).
The more marked declines in esterase activity recorded for 600W2 than for 400W2 cheese and for
600W3 than for 400W3 cheese can be ascribed to the effect of the higher pressure level applied.
The more marked declines in 400W3 than in 400W2 cheese and in 600W3 than in 600W2 cheese
may be explained by the fact that a higher proportion of intracellular esterase from lactic acid
bacteria and other microorganisms had been spontaneously released outside the cells on day 21 than
on day 14, as shown by the increase in the esterase activity values of control cheese. Enzymes
outside the bacterial cells would presumably be more susceptible to inactivation by HPP than in the
interior of the cells. During the rest of the ripening period, a gradual decrease in esterase activity
values occurred in all the HPP cheeses, in particular in those treated at 600 MPa (Table 2).
The concentration of total short-chain (SC, C4:0 to C8:0) FFA increased in control cheese from
0.017 mg/g DM on day 1 (data not shown) to 0.190 mg/g DM on day 240 (Table 3). SC FFA,
mostly produced through the esterase-mediated hydrolysis of triacylglycerides, also derive from the
fermentation of lactose and lactate, from the degradation of amino acids and from the oxidation of
some ketones, esters and aldehydes (Collins et al. 2003). Concentrations of SC FFA were
significantly (p<0.05) lower in 400 MPa cheeses than in control cheese only on day 240 while in
600 MPa cheeses they were lower from day 60 onwards. In the present work, butanoic acid was the
major SC FFA throughout ripening in control and HPP cheeses, followed by hexanoic and octanoic
acids, all three with the same accumulation pattern as total SC FFA. Butanoic acid has a rancid
10
cheese-like odour and is a key odorant of many cheese varieties although at high levels, usually due
to butyric fermentation, may cause flavour defects. Hexanoic and octanoic acids are characteristic
flavour compounds of hard cheese varieties such as Grana Padano and Roncal (Curioni and Bosset
2002).
The concentration of total medium chain (MC, C10:0 to C14:0) FFA increased from 0.178 mg/g
DM on day 1 (data not shown) to 0.400 mg/g DM on day 240 (Table 3), just a 2.25-fold increase
compared to the 11.24-fold increase of SC FFA during the same period. HPP had a lower influence
on MC FFA than on SC FFA. Significant (p<0.05) differences between cheeses were not recorded
until day 240, when control cheese showed the highest MC FFA levels. Similar results on the non-
significant effect of cheese HPP on total MC FFA concentrations were reported for goat milk
cheese (Delgado et al. 2012). MC FFA have relatively low perception thresholds, what makes them
important contributors to the aroma of cheese varieties such as Camembert, Cheddar, Grana Padano
and Roncal (Curioni and Bosset 2002).
The concentration of total long chain (LC, C16:0 to C18:3) FFA increased from 0.669 mg/g DM on
day 1 (data not shown) to 1.307 mg/g DM on day 240 (Table 3). This 1.95-fold increase was even
lower than the increase of MC FFA during the same period. As in the case of MC FFA, significant
(p<0.05) differences between cheeses were not recorded until day 240, time at which control cheese
had higher LC FFA levels. Formation of LC FFA in control and HPP cheeses can be ascribed to the
action of milk lipoprotein lipase, which does not seem to lose activity in cheeses made from HPP-
treated goat milk (Buffa et al. 2001) and in cheese made from goat raw milk HPP-treated at 400 or
600 MPa (Delgado et al. 2012), as proven by the unaltered accumulation of LC FFA during
ripening. LC FFA usually reach high concentrations in many cheese varieties, particularly in blue-
veined cheeses in which the levels frequently exceed 30 mg/g (Calzada et al. 2013), but their
contribution to cheese aroma is lowered by the high perception thresholds (Curioni and Bosset
2002).
11
Volatile Compounds
Seventy individual VC were found in control and HPP cheeses analyzed by SPME followed by GC-
MS. According to the analysis of variance, 52 VC were significantly (p<0.05) influenced by the
HPP treatment and 63 VC by the time elapsed after manufacture. The 70 VC were grouped into 19
alcohols, 9 esters, 9 hydrocarbons, 7 ketones, 6 acids, 6 ethers, 4 aldehydes, 3 benzenic compounds,
2 terpenoids, 2 sulphur compounds and 3 miscellaneous compounds. During early ripening of
control cheese, ketones, acids and alcohols were the predominant VC groups, accounting for 52.6%,
25.6% and 18.3%, respectively, of the overall abundance of VC (pooled data from control cheese
on days 14 and 21), followed by benzenic compounds with 1.16% and esters with 1.12%, while
ethers, terpenoids, hydrocarbons, aldehydes and sulphur compounds accounted for 0.48%, 0.30%,
0.20%, 0.12% and 0.11%, respectively, of the overall abundance (Table 4). The proportions of VC
groups varied with HPP treatment and time of ripening (Tables 5 and 6). During late ripening the
predominant VC groups were ketones, alcohols and acids, accounting for 38.6%, 31.1% and 25.4%
of the overall abundance, respectively (pooled data of control and HPP cheeses on days 60, 120
and 240), followed by benzenic compounds with 1.68% and esters with 1.13%, while ethers,
terpenoids, hydrocarbons, aldehydes and sulphur compounds accounted for 0.89%, 0.58%, 0.42%,
0.16% and 0.06% of the overall abundance, respectively. It must be taken into account that SPME is
more effective in extracting VC of medium and high boiling point in comparison with the purge and
trap method which extracts better the VC of low boiling point (Mallia et al. 2005). According to
these authors, there are exceptions such as acetaldehyde which is better extracted by SPME, in spite
of its low boiling point.
Total ketones of control cheese reached their maximum concentration on day 14 and gradually
declined afterwards (Tables 4 and 5). In contrast, total ketones of HPP cheeses increased markedly
12
from day 21 to day 60, time at which they reached significantly (p<0.05) higher levels than in
control cheese. Afterwards, their levels fell sharply in 400 MPa cheeses while they remained in 600
MPa cheeses at significantly (p<0.05) higher levels than in control cheese (Table 5). Total ketones
represented only 16.5% of the overall abundance of VC in control cheeses while they accounted for
44.5% of the overall abundance in HPP cheeses (pooled data of cheeses on days 60, 120 and 240).
The main ketones (data not shown) were 3-hydroxy-2-butanone (acetoin), 2-butanone, and 2,3-
butanedione (diacetyl), which accounted for 67.7%, 22.4% and 7.0% of the overall abundance of
total ketones, respectively. 3-Hydroxy-2-butanone has a sour milk odour note, 2,3-butanedione a
buttery odour note and 2-butanone a butterscotch odour note (Curioni and Bosset 2002). The
accumulation pattern of 3-hydroxy-2-butanone and 2,3-butanedione in control and HPP cheeses was
similar to that of total ketones. The pattern of 2-butanone differed, with a strong increase in control
and 400 MPA cheeses from day 14 to day 21 which continued until day 120, and a less marked
increase in 600 MPa cheeses which ceased on day 60. Microbial metabolism of lactose, the main
agent responsible for the formation of 2,3-butanedione and 3-hydroxy-2-butanone (Fox and Wallace
1997), is responsible for the accumulation of these ketones until day 14 in control cheese but does
not explain the increase in the levels of these ketones in HPP cheeses afterwards, once lactose is
exhausted. The increase in 2-butanone levels from day 14 to day 60 can be ascribed to the microbial
reduction of 2,3-butanedione and 3-hydroxy-2-butanone. Minor ketones were 2-propanone, with
wood pulp and hay notes, 2-heptanone, with animal and blue cheese notes, and 3-hydroxy-2-
pentanone and 2-hydroxy-3-pentanone, with herbaceous and truffle notes (Curioni and Bosset
2002).
Total volatile acids of control cheese increased until day 21, then kept stable until day 120 and
afterwards declined (Tables 4 and 5). In 400 MPa cheeses their increase persisted until day 60 and
declined afterwards, while in 600 MPa cheeses their accumulation pattern was similar to that of
control cheese, although differences between HPP and control cheeses were not significant. In fact,
13
total acids accounted for 21.1% of the overall abundance of VC in control cheeses and 26.5% in
HPP cheeses (pooled data of days 60, 120 and 240). The major volatile acids (data not shown) were
ethanoic, propanoic and butanoic, accounting for 62.7%, 25.8% and 11.2%, respectively, of the
overall abundance of this chemical group. The first two acids show vinegar odour notes, and the
latter a rancid cheesy note, all being mostly formed through microbial metabolism. Ethanoic and
propanoic acids increased in control cheese until day 21, remained fairly constant until day 120 and
then declined, probably due to ester formation, while butanoic acid increased until day 120 and did
not vary afterwards, what can be associated with the different mechanisms of formation (Yvon and
Rijnen 2001; Collins et al. 2003). Branched-chain carboxylic acids were not detected by SPME
followed by GC-MS, what can be ascribed to the absence or the low levels of bacteria with the
ability to catabolise branched-chain amino acids. Minor volatile acids, at levels 100 times lower or
less than the major volatile acids, were hexanoic, 4-hexenoic and octanoic, presumably derived
from the hydrolysis of triglycerides.
Total alcohols of control cheese increased consistently until day 120 and declined afterwards
(Tables 4 and 5). Their pattern of accumulation in HPP cheeses differed depending on the pressure
level applied, with maximum levels on days 120 or 240 in 400 MPa cheeses and decreases starting
on day 21 in 600 MPa cheeses (Tables 4 and 5). Total alcohols represented 58.2% of the overall
abundance of VC in control cheeses and only 23.9% in HPP cheeses (pooled data of days 60, 120
and 240). The major alcohols (data not shown) were 2-butanol, ethanol and 1,3-butanediol,
accounting for 54.7%, 21.2% and 15.2% of the overall abundance of alcohols, respectively. 2-
Butanol shows chemical and floral notes, ethanol chemical and dry notes, and 1,3-butanediol milk
creamy buttery notes (Curioni and Bosset 2002; Mallia et al. 2005). Biosynthesis of alcohols may
take place in cheese through different mechanisms (Molimard and Spinnler 1996). 2-Butanol is
mostly formed by the enzymatic reduction of 2-butanone and ethanol is derived from lactose
metabolism, while 1,3-butanediol comes from the metabolism of lactose and citrate. Other alcohols
14
found were 2,3-butanediol, 1-propanol, 2-propanol, 2-propen-1-ol, 2-methyl-1-propanol, 1,3-
propanediol, 1-butanol, 1-pentanol, 2-pentanol, 3-methyl-2-buten-1-ol, 3-methyl-3-buten-1-ol, 1-
hexanol, 2,3-hexanediol, 2-ethyl-1-hexanol, cyclohexanol and 2-heptanol, all at levels considerably
lower than those of the major alcohols. A branched-chain alcohol, 2-methyl-1-propanol, was
detected in cheese in spite of the absence of the precursor branched-chain acid and aldehyde.
Total benzenic compounds followed the same accumulation pattern in control and HPP cheeses,
with a slight increase until day 60 and a slight decrease afterwards (Tables 4 and 5). Benzenic
compounds accounted for 1.44% and 1.75% of the overall abundance of VC in control cheeses and
HPP cheeses, respectively, with no significant differences due to treatments. The principal benzenic
compound was naphthalene, accounting for 89.8% of the overall abundance of this group of VC
compounds, followed by benzenethanol and toluene which represented 5.7% and 4.5%,
respectively. Naphtalene, with moth-ball tar-like odour, has been found in cow, ewe and goat milk
smoked cheeses (Guillén and Sopelana 2004). Benzenethanol, with floral, rose notes, has been
detected in raw milk cheeses (Sulejmani et al. 2014). Toluene, with nutty, bitter almond notes,
contributes to Cheddar cheese odour (Arora et al. 1995). Benzenic compounds probably derive from
the catabolism of aromatic amino acids by cheese microbiota other than lactic acid bacteria.
Total esters also followed the same accumulation pattern in control and HPP cheeses, with a
slight increase until day 60 and a slight decrease afterwards (Tables 4 and 5). Esters accounted for
1.16% of the overall abundance of VC in control cheeses and for 1.12% of the overall abundance in
HPP cheeses, with no significant differences due to treatments. The predominant ester was a cyclic
ester, γ-butyrolactone, which accounted for 66.8% of the overall abundance of esters, followed by
ethyl acetate and ethyl butyrate, accounting for 13.0% and 6.4% of the overall abundance. Lactones
are formed from hydroxylated fatty acids by lactonization, with the closing of the ring occurring by
the action of pH, microorganisms or both (Molimard and Spinnler 1996); they have peach, apricot
and coconut fruity notes and low perception thresholds. Linear esters are formed in dairy systems
15
through two enzymatic mechanisms, esterification, a reaction in which esters are formed from
alcohols and carboxylic acids, and alcoholysis, a transferase reaction in which fatty acids from
acylglycerols and acyl-CoA derivatives are directly transferred to alcohols (Liu et al. 2004). In the
present work, the increase in esters levels was not accompanied by a decline in the levels of acids or
alcohols in control or HPP cheeses. Esters generally show sweet, fruity and floral notes, have a low
perception threshold and mask the sharpness and bitterness imparted by fatty acids and amines, thus
contributing to a pleasant cheese flavour. Ethyl acetate, with fruity pineapple notes, and ethyl
butyrate, with green fruit and apple notes, have been identified as potent odorants in cow and ewe
milk cheese varieties (Curioni and Bosset 2002).
Total ethers increased in control cheese and 400 MPa cheeses until day 60 and declined
afterwards, while in 600 MPa cheeses the increase persisted until days 120 or 240 (Tables 4 and 6).
Higher values were reached in HPP cheeses than in control cheese from day 60 onwards, although
the differences were not significant in most cases. Ethers accounted for 0.53% of the overall
abundance of VC in control cheeses and 0.99% in HPP cheeses (pooled data of days 60, 120 and
240). Major ethers (data not shown) were 1-methoxy-2-propanol, 2-(2-ethoxyethoxy)-ethanol and
1,1’-oxybis-2-propanol, accounting for 80.4%, 7.7% and 4.2% respectively of the overall
abundance of total ethers. No odour active ethers were found in cow or ewe raw milk cheeses by
Mallia et al. (2005), although the presence of ethers in smoked goat raw milk cheese was reported
by Guillén et al. (2004). Ethers are not considered to be key odorants in cheese and their origin is
not well known (Curioni and Bosset 2002).
Total terpenoids increased gradually in control and HPP cheeses until day 240, with few
significant differences between cheeses (Tables 4 and 6). Terpenoids accounted for 0.63% of the
overall abundance of VC in control cheeses and 0.56% in HPP cheeses (pooled data of days 60, 120
and 240). The two terpenoids found, pristine and borneol, accounted for 94.0% and 6.0% of the
overall abundance of total terpenoids, respectively (data not shown). Terpenoids, commonly present
16
in the volatile fraction of cheeses (Carbonell et al. 2002; Fernández-García et al. 2004), are
considered to come from the animal diet (Toso et al. 2002).
Total hydrocarbons hardly varied from day 60 to day 120 in control and HPP cheeses, with
slightly higher levels in HPP cheeses, and declined considerably from day 120 to day 240, time at
which 600 MPa cheeses showed significantly (p<0.05) higher levels than control cheese (Tables 4
and 6). Linear hydrocarbons are generally formed through the oxidation of FFA. Total
hydrocarbons accounted for 0.26% of the overall abundance of VC in control cheeses and 0.46% in
HPP cheeses (pooled data of days 60, 120 and 240). The most abundant hydrocarbons were
2,2,4,6,6-pentamethyl heptane, hexane and 3,3-dimethylhexane which accounted for 50.5%, 19.5%
and 13.0%, respectively, of the overall abundance of this chemical group (data not shown).
Hydrocarbons, in spite of being commonly found in the volatile fraction of raw milk cheese
varieties (Carbonell et al. 2002; Fernández-García et al. 2002; Fernández-García et al. 2004), are
not considered as main key odorants (Thierry et al. 1999).
Total aldehydes did not vary significantly with time of ripening and showed slight differences
between control and HPP cheeses (Tables 4 and 6). Total aldehydes accounted for 0.09% of the
overall abundance of VC in control cheeses and 0.18% in HPP cheeses (pooled data of days 60, 120
and 240). The principal aldehydes were acetaldehyde, 3-methylbutanal and benzaldehyde which
accounted for 38.5%, 32.6% and 19.1%, respectively, of the overall abundance of total aldehydes
(data not shown). Acetaldehyde is mostly formed through the metabolism of lactose and may also
derive from glycine, while 3-methylbutanal is derived from leucine and benzaldehyde from the
catabolism of aromatic amino acids (Yvon and Rijnen 2001), although it also may be formed
through the oxidation of toluene. Acetaldehyde shows green odour notes while 3-methylbutanal has
pleasant fruity odour notes al low concentrations which turn into green malty at high levels and is a
potent odorant in Camembert, Cheddar, Emmental and Gruyère cheeses (Curioni and Bosset 2002).
17
Total sulphur compounds suffered a slight decline from day 60 to day 120 followed by a drastic
decline from day 120 to day 240, with no significant differences between cheese until day 240, time
at which HPP cheeses had higher (p<0.05) levels than control cheese (Tables 4 and 6). Total
aldehydes accounted for 0.09% of the overall abundance of VC in control cheeses and 0.18% in
HPP cheeses (pooled data of days 60, 120 and 240). The two sulphur compounds found,
dimethylsulphide and dimethylsulfoxide, accounted for 80.2% and 19.8%, respectively, of the
overall abundance of this chemical group (data not shown). Most sulphur compounds derive from
methanethiol, a compound of putrid and faecal-like aroma which is formed from methionine by the
action of cystathionine or methionine lyases (Weimer et al. 1999). Lactococcus, Lactobacillus,
Brevibacterium, Micrococcus, Corynebacterium, Pseudomonas and probably other genera of Gram-
negative bacteria produce sulphur compounds in cheese. At low concentrations, sulphur compounds
contribute with their cowy, feedy, garlic, onion, cooked cabbage, cauliflower, mashed potato odour
notes of to the aroma of Limburger, Camembert, Cheddar, Blue and ewes milk cheeses (Bonnarme
et al. 2000; Fernández-García et al. 2004). However, high concentrations of sulphur compounds
may result unpleasant because of their very low perception thresholds (Weimer et al. 1999).
The three miscellaneous compounds found in the present work, chloroform, propanamine and
methoxy-phenyl-oxime, were minor volatiles which accounted for only 0.058%, 0.039%, and
0.008%, respectively, of the overall abundance of VC in control and HPP cheeses (data not shown).
Chloroform, a compound with sweet hay odour notes found in different cheese varieties (Curioni
and Bosset 2002), declined from day 60 onwards and was generally present at higher levels in HPP
cheeses. Propanamine, a compound with ammonia and fish-like odour notes found in Camembert
cheese (Molimard and Spinnler 1996), was not detected until day 60 and was only present in control
and 400 MPa cheeses, at levels which increased until day 120 and decreased afterwards. Methoxy-
phenyl-oxime, a compound without characteristic odour notes which has been detected in Gouda
cheese (Jung et al. 2013), followed a similar trend to that of chloroform.
18
Principal Component Analysis
Principal component analysis (PCA) was firstly carried out on the concentrations of individual
carboxylic acids. The objective was to discriminate control and HPP cheeses on days 14 to 240 after
manufacture, and to ascertain their evolution during the time elapsed after manufacture, on the basis
of their concentrations of carboxylic acids. Function 1, formed by C4:0, C6:0, C8:0, C10:0, C12:0,
C14:0; C16:0, C18:0, C18:1; C18:2 and C18:3 acids, explained 73.8% of the variance, while
function 2, formed by C3:0 and benzoic acids, explained 11.6% and function 3, formed by C2:0
acid, explained 7.1%. The distribution of control and HPP cheeses on the plane defined by functions
1 and 2 of the PCA of carboxylic acids is shown on Fig. 1. Function 1 evolved from negative values
(at 14, 21 and 60 days) to positive values (at 120, 180 and 240 days). Function 2 followed a very
particular trend, evolving from negative to positive values on the left hemiplane (from day 14 to day
60) and from positive to negative values on the right hemiplane (from day 120 to day 240). PCA of
carboxylic acids did not achieve a clear discrimination between sampling times or treatments.
PCA was also performed on the levels of individual volatile compounds. The objective was to
discriminate control and HPP cheeses on days 60, 120 and 240 after manufacture, and to ascertain
their evolution during the time elapsed after manufacture, on the basis of the individual compounds
present in the volatile fraction. Function 1, formed by 4 hydrocarbons, 2 acids, 2 sulfur compounds,
6 ketones, 2 aromatic compounds, 1 aldehyde, 1 ester, 4 alcohols, 1 terpenoid, 1 ether, chloroform
and methoxy-phenyl-oxime, and with a negative coefficient by 1 acid, 3 esters and 5 alcohols,
explained 37.7% of the variance, while function 2, formed by 2 esters, 1 ketone, 7 alcohols, 1
aldehyde, 1 acid, 1 ether, 1 terpenoid, and propanamine, explained 21.3%, and function 3, formed
by 4-hexenoic acid, explained 6.5%. The distribution of control and HPP cheeses on the plane
19
defined by functions 1 and 2 of the PCA of the 70 individual volatile compounds (not shown) was
similar to that obtained with the PCA of the 13 groups of volatile compounds (Fig. 2).
A third PCA was carried out on the levels of the 13 groups of volatile compounds. The objective
was to discriminate control and HPP cheeses on days 60, 120 and 240 after manufacture, and to
ascertain their evolution during the time elapsed after manufacture, on the basis of the levels of the
main groups of compounds present in the volatile fraction. Function 1 (PC1), mainly formed by the
groups of volatile acids, aldehydes, esters, ketones, hydrocarbons, benzenic compounds and sulphur
compounds, explained 41.9% of the variance, while function 2 (PC2), formed by the groups of
alcohols, terpenoids and miscellaneous compounds, explained 23.5% and function 3, mainly formed
by the group of ethers, explained 12.8%. The distribution of control and HPP cheeses on the plane
defined by functions 1 and 2 of the PCA of the groups of volatile compounds is shown on Fig. 2.
Until day 120 of storage, control cheeses remained located on the right side of the plane, with
positive PC1 values characterized by high amounts of acids, aldehydes, esters, ketones,
hydrocarbons, benzenic and sulphur compounds, associated with cheesy, fruity, green, sweet,
creamy and pleasant flavour notes. Simultaneously, they evolved from negative PC2 values on days
14 and 21 to positive PC2 values from day 60 onwards, with high levels of alcohols, terpenoids and
miscellaneous compounds, associated with herbal, pungent, sweet and alcoholic flavour notes. On
day 240, control cheese was located at the upper-left quadrant with negative PC1 and positive PC2
values, characterized by herbal, pungent, sweet and alcoholic notes but deficient in cheesy, creamy,
fruity and pleasant notes. HPP induced lower PC2 values, in particular when applied at the higher
pressure level. The 400 MPa cheeses showed on day 60 a volatile profile similar to that of 21-day
control cheese, with cheesy, creamy and pleasant flavour notes. From day 120 onwards, they were
located at the upper-left quadrant, together with the 240-day control cheese. The 600 MPa cheeses
maintained the cheesy, creamy and pleasant flavour notes profile until day 120. From day 60 to day
240, they evolved from positive to negative PC1 values, the latter associated with mild flavour
20
notes. They remained until day 240 in the lower hemiplane, with negative PC2 values, without
development of pungent and alcoholic notes.
Odour Characteristics
Odour intensity of control and HPP cheeses at each of the sampling times did not differ
significantly, but there was an increase in odour intensity with the time elapsed after manufacture
with scores ranging from 4.56 to 5.02 on day 60 and from 5.78 to 6.34 on day 240 (data not shown).
Odour quality showed no significant differences between HPP and control cheeses at each of the
sampling times and did not vary significantly with the time elapsed after manufacture, with scores
ranging from 5.47 to 5.98 on day 60 and from 5.62 to 6.01 on day 240 (data not shown). Similarly,
the scores of putrid, rancid and acid odour attributes did not vary significantly with HPP treatments
or with the time elapsed after manufacture. Contrarily to these results, beneficial effects of HPP at
400 or 600 MPa on the odour characteristics of cheese made from raw ewe milk were observed
during the refrigerated storage of cheeses until day 240 (Calzada et al. 2014a). In the present work,
no accumulation of undesirable volatile compounds occurred in control cheese, in contrast to what
happened in the control cheese made from raw ewe milk, a fact which would explain why there
were no beneficial effects on odour characteristics due to HPP treatment.
Colour Parameters
Colour parameters of control and HPP cheeses are shown on Table 7. Lightness (L*) generally
reached the highest values in 600 MPa cheeses, in particular in 600W3 cheese. It tended to decrease
with time in control and HPP cheeses. Redness (a*) showed higher levels in control cheese than in
HPP cheeses, in particular in 600 MPa cheeses, which exhibited the lowest values at all sampling
21
times. It declined sharply from day 180 to day 240 in control and HPP cheeses. Yellowness (b*)
showed few significant differences between cheeses at each of the sampling times. It increased in
control and HPP cheeses along the 240-day period.
Previous works on the effect of HPP on cheese colour generally dealt with cheeses ripened for
shorter periods. Ávila et al. (2008) reported few differences attributable to HPP at 400 MPa in the
colour parameters of Hispánico cheese ripened for 50 days; during this period, L* hardly varied,
while a* tended to decrease and b* to increase, with higher values in the HPP cheese. Voigt et al.
(2012) obtained higher L* and b* values, and similar a* values, for Cheddar cheese made from milk
treated at 600 MPa than for control cheese. Minor differences between the colour parameters of
control and HPP Brie cheeses were recorded from day 30 to day 90, while on day 120 significantly
higher L* values, and lower a* and b* values, were recorded for HPP cheeses (Calzada et al.
2014c). Picon et al. (2013) recorded the lowest L* and a* values in cheeses made from curds treated
at 500 MPa while the highest b* values were found for control cheese and cheese made from curd
treated at 200 MPa. According to these authors, cheese colour parameters are influenced by the
microstructural changes caused by HPP, the lower amount of available water in HPP-treated cheese
and the concentration of cheese components associated with moisture loss during ripening.
Conclusions
High-pressure-processing of unpasteurized milk cheese lowered esterase activity, in particular when
a pressure level of 600 MPa was applied. The concentration of total FFA on day 60 was not affected
by HPP, but lower concentrations of total FFA were recorded on day 240 for HPP cheeses than for
control cheese, with more marked differences for short-chain FFA. The levels of most of the
individual volatile compounds found in cheese were significantly influenced by HPP. The level of
total alcohols in 240-day cheeses was lowered by HPP at 600 MPa, which increased the levels of
22
total aldehydes, ketones and hydrocarbons. Total sulphur compounds were lowered by HPP at 400
and 600 MPa while total acids, esters, ethers, benzenic compounds and terpenoids were not affected
by HPP. The odour characteristics of cheeses (odour quality and intensity) were hardly influenced
by HPP. Colour parameters were affected by HPP, with higher L* values and lower a* values
recorded in 600 MPa cheeses.
Acknowledgements This research was funded by AGL 2009-07801 project (Ministry of Science
and Innovation, Spain). The authors thank the PDO dairy for providing the cheeses and Hiperbaric
for HPP treatments. J. Calzada was the recipient of a FPI grant (Ministry of Science and Innovation,
Spain).
References
Arora, G., Cormier, F., & Lee, B. (1995). Analysis of odor-active volátiles in Cheddar cheese
headspace by multidimensional GC/MS/sniffing. Journal of Agricultural and Food Chemistry,
43, 748-752.
Ávila, M., Garde, S., Fernández-García, E., Medina, M., & Nuñez, M. (2006). Effect of high-
pressure treatment and a bacteriocin-producing lactic culture on the odor and aroma of Hispánico
cheese: correlation of volatile compounds and sensory analysis. Journal of Agricultural and
Food Chemistry, 54, 382-389.
Ávila, M., Calzada, J., Garde, S., & Nuñez, M. (2007). Effect of a bacteriocin-producing
Lactococcus lactis strain and high-pressure treatment on the esterase activity and free fatty acids
in Hispánico cheese. International Dairy Journal, 17, 1415-1423.
Ávila, M., Garde, S., & Nuñez, M. (2008). Effect of a bacteriocin-producing lactic culture and high
pressure treatment on the color of Hispánico cheese. Milchwissenschaft, 63, 406-409.
23
Bonnarme, P., Psoni, L., & Spinnler, H. E. (2000). Diversity of L-methionine catabolism pathways
in cheese-ripening bacteria. Applied and Environmental Microbiology, 66, 5514-5517.
Buffa, M., Guamis, B., Pavia, M., Trujillo, A. J. (2001). Lipolysis in cheese made from raw,
pasteurized or high-pressure-treated goats’ milk. International Dairy Journal, 11, 175-179.
Calzada, J., del Olmo, A., Picon, A., Gaya, P., & Nuñez, M. (2013). High-pressure-processing
decelerates lipolysis and formation of volatile compounds in ovine milk blue-veined cheese.
Journal of Dairy Science, 96, 7500-7510.
Calzada, J., del Olmo, A., Picon, A., Gaya, P., & Nuñez, M. (2014a). High-pressure-processing for
the control of lipolysis, volatile compounds and off-odours in raw milk cheese. Food and
Bioprocess Technology, 7, 2207-2217.
Calzada, J., del Olmo, A., Picon, A., Gaya, P., & Nuñez, M. (2014b). Effect of high-pressure
processing on the microbiology, proteolysis, biogenic amines and flavour of cheese made from
unpasteurized milk. Food and Bioprocess Technology DOI 10.1007/s11947-014-1406-7
Calzada, J., del Olmo, A., Picon, A., & Nuñez, M. (2014c). Effect of high-pressure-processing on
the lipolysis and volatile compounds of Brie cheese during ripening and refrigerated storage.
International Dairy Journal, 39, 232-239.
Carbonell, M., Nuñez, M., & Fernández-García, E. (2002). Evolution of the volatile components of
ewe raw milk La Serena cheese during ripening. Correlation with flavor characteristics. Lait, 82,
683-698.
Collins, Y. F., McSweeney, P. L. H., & Wilkinson, M. G. (2003). Lipolysis and free fatty acid
catabolism in cheese: a review of current knowledge. International Dairy Journal, 13, 841-866.
Curioni, P. M. G., & Bosset, J. O. (2002). Key odorants in various cheese types as determined by
gas chromatography-olfactometry. International Dairy Journal, 12, 959-984.
24
Delgado, F. J., González-Crespo, J., Cava, R., & Ramírez, R. (2012). High-pressure treatment
applied throughout ripening of a goat cheese caused minimal changes on free fatty acids content
and oxidation in mature cheese. Dairy Science and Technology, 92, 237-248.
Fernández-García, E., Carbonell, M., & Nuñez, M. (2002). Volatile fraction and sensory
characteristics of Manchego cheese. 1. Comparison of raw and pasteurized milk cheese. Journal
of Dairy Research 69, 579-593.
Fernández-García, E., Carbonell, M., Gaya, P., & Nuñez, M. (2004). Evolution of the volatile
components of ewes raw milk Zamorano cheese. Seasonal variation. International Dairy
Journal, 14, 701-711.
Fernández-García, E., Carbonell, M., Calzada, J., & Nuñez, M. (2006). Seasonal variation of the
free fatty acids contents of Spanish ovine milk cheeses protected by a designation of origin: A
comparative study. International Dairy Journal, 16, 252-261.
Fox, P. F., & Wallace, J. M. (1997). Formation of flavor compounds in cheese. Advances in Applied
Microbiology, 45, 17-85.
Garde, S., Arqués, J. L., Gaya, P., Medina, M., & Nuñez, M. (2007). Effect of high pressure
treatments on the proteolysis and texture of ewes’ raw milk La Serena cheese. International
Dairy Journal, 17, 1424-1433.
Gaya, P., Sánchez, C., Nuñez, M., Fernández-García, E. (2005). Proteolysis during ripening of
Manchego cheese made from raw or pasteurized ewes’ milk. Seasonal variation. Journal of
Dairy Research, 72, 287-295.
Guillén, M. D., Ibargoitia, M. L., Sopelana, P., Palencia, G., & Fresno, M. (2004). Components
detected by means of solid-phase microextraction and gas chromatography/mass spectrometry in
the headspace of artisan fresh goat cheese smoked by traditional methods. Journal of Dairy
Science, 87, 284-299.
25
Guillén, M. D., & Sopelana, P. (2004). Occurrence of polycyclic aromatic hydrocarbons in smoked
cheese. Journal of Dairy Science, 87, 556-564.
Huppertz, T., Fox, P. F. , & Kelly, A. L. (2004). Susceptibility of plasmin and chymosin in Cheddar
cheese to inactivation by high pressure. Journal of Dairy Research, 71, 496-499.
Iammarino, M., Di Taranto, A., Palermo, C., & Muscarella, M. (2011). Survey of benzoic acid in
cheeses: contribution to the estimation of an admissible maximum limit. Food Additives and
Contaminants, Part B, 4, 231-237.
Juan, B., Ferragut, V., Buffa, M., Guamis, B., & Trujillo, A. (2007). Effects of high pressure on
proteolytic enzymes in cheese: relationship with the proteolysis of ewe milk cheese. Journal of
Dairy Science, 90, 2113-2125.
Jung, H. J., Ganesan, P., Lee, S. J., & Kwak, H. S. (2013). Comparative study of flavor in
cholesterol-removed Gouda cheese and Gouda cheese during ripening. Journal of Dairy Science,
96, 1972-1983.
Liu, S.-Q., Holland, R., & Crow, V. L. (2004). Esters and their biosynthesis in fermented dairy
products: a review. International Dairy Journal, 14, 923-945.
Mallia, S., Fernández-García, E., & Bosset, J. O. (2005). Comparison of purge and trap and solid
phase microextraction techniques for studying the volatile aroma compounds of three European
PDO hard cheeses. International Dairy Journal, 15, 741-758.
Malone, A. S., Wick, C., Shellhammer, T. H., & Courtney, P. D. (2003). High pressure effects on
proteolytic and glycolytic enzymes involved in cheese manufacturing. Journal of Dairy Science,
86, 1139-1146.
McSweeney, P. L. H., & Sousa, M. J. (2000). Biochemical pathways for the production of flavour
compounds in cheese during ripening: A review. Lait, 80, 293-324.
Molimard, P., & Spinnler, H. E. (1996). Compounds involved in the flavour of surface mold-
ripened cheeses: origins and properties. Journal of Dairy Science, 79, 169-184.
26
Norton, T., & Sun, D.-W. (2008). Recent advances in the use of high pressure as an effective
processing technique in the food industry. Food and Bioprocess Technology, 1, 2-34.
O'Reilly, C. E., O’Connor, P. M., Kelly, A. L., Beresford, T. P., & Murphy, P. M. (2000). Use of
hydrostatic pressure for inactivation of microbial contaminants in cheese. Applied and
Environmental Microbiology, 66, 4890-4896.
Picon, A., Alonso, R., van Wely, K. H. M., & Nuñez, M. (2013). Microstructural, textural and
colour characteristics during ripening of Hispánico cheese made using high-pressure-treated
ovine milk curd. Food and Bioprocess Technology, 6, 3056-3067.
Shao, Y. W., & Ramaswamy, H. S. (2011). Clostridium sporogenes-ATCC 7955 spore destruction
kinetics in milk under high pressure and elevated temperature treatment conditions. Food and
Bioprocess Technology, 4, 458-468.
Sieber, R., Bütikofer, U., & Bosset, J. O. (1995). Benzoic acid as a natural compound in cultured
dairy products and cheese. International Dairy Journal, 5, 227-246.
Sulejmani, E., Rafajlovska, V., & Guneser, O. (2014). Characterization of volatiles in Beaten
cheeses “Bieno sirenje” by SPME/GC-MS: influence of geographical origin. Journal of the
Serbian Chemical Society, 79, 927-939.
Tejada, L., Gómez, R., Vioque, M., Sánchez, E., Mata, C., & Fernández-Salguero, J. (2000). Effect
of freezing and frozen storage on the sensorial characteristics of Los Pedroches, a Spanish ewe
milk cheese. Journal of Sensory Studies, 15, 251-262.
Thierry, A., Maillard, M.-B., & Le Quéré, J.-L. (1999). Dynamic headspace analysis of Emmental
aqueous phase as a method to quantify changes in volatile flavor compounds during ripening.
International Dairy Journal, 9, 453-463.
Toso, B., Procida, G., & Stefanoni, B. (2002). Determination of volatile compounds in cows’ milk
using headspace GC-MS. Journal of Dairy Research, 69, 569-577.
27
Van Hekken, D. L., Tunick, M. H., & Park, Y. W. (2005). Effect of frozen storage on the
proteolytic and rheological properties of soft caprine milk cheese. Journal of Dairy Science, 88,
1966-1972.
Voigt, D. D., Chevalier, F., Donaghy, J. A., Patterson, M. F., Qian, M. C., & Kelly, A. L. (2012).
Effect of high-pressure treatment of milk for cheese manufacture on proteolysis, lipolysis,
texture and functionality of Cheddar cheese during ripening. Innovative Food Science and
Emerging Technologies, 13, 23-30.
Weimer, B., Seefeldt, K., & Dias, B. (1999). Sulfur metabolism in bacteria associated with cheese.
Antonie van Leeuwenhoek, 76, 247-261.
Yvon, M., & Rijnen, L. (2001). Cheese flavour formation by amino acid catabolism. International
Dairy Journal, 11, 185-201.
28
Table 1 Concentrations of acetic, propionic and benzoic acids during ripening of control and HPP
cheeses made from unpasteurized milk
Acid Days Control cheese HPP cheeses1
400W2 600W2 400W3 600W3
Acetic2 21 1.121 ± 0.051a 1.071 ± 0.014a 0.991 ± 0.021a 1.051 ± 0.033a 1.064 ± 0.017a
60 1.436 ± 0.024c 1.064 ± 0.027a 1.083 ± 0.047a 1.268 ± 0.031b 1.122 ± 0.037ab
120 1.785 ± 0.038c 1.316 ± 0.063b 1.004 ± 0.023a 1.367 ± 0.029b 1.003 ± 0.032a
180 1.868 ± 0.105c 1.154 ± 0.080ab 0.893 ± 0.039a 1.416 ± 0.045b 0.995 ± 0.040a
240 2.102 ± 0.034c 1.272 ± 0.028b 0.912 ± 0.021a 1.348 ± 0.032b 0.917 ± 0.028a
Propionic2 21 0.349 ± 0.029a 0.398 ± 0.008a 0.311 ± 0.029a 0.340 ± 0.014a 0.350 ± 0.007a
60 0.358 ± 0.020a 0.349 ± 0.007a 0.378 ± 0.055a 0.398 ± 0.018a 0.359 ± 0.018a
120 0.378 ± 0.016a 0.343 ± 0.011a 0.355 ± 0.015a 0.381 ± 0.016a 0.364 ± 0.002a
180 0.377 ± 0.005a 0.306 ± 0.010a 0.348 ± 0.020a 0.354 ± 0.051a 0.379 ± 0.030a
240 0.304 ± 0.010a 0.287 ± 0.027a 0.317 ± 0.010a 0.299 ± 0.014a 0.366 ± 0.043a
Benzoic2 21 0.244 ± 0.041a 0.283 ± 0.018a 0.213 ± 0.033a 0.237 ± 0.027a 0.224 ± 0.010a
60 0.302 ± 0.004a 0.290 ± 0.015a 0.335 ± 0.041a 0.316 ± 0.036a 0.294 ± 0.009a
120 0.345 ± 0.026a 0.341 ± 0.017a 0.328 ± 0.012a 0.358 ± 0.015a 0.343 ± 0.014a
180 0.398 ± 0.010a 0.300 ± 0.008a 0.337 ± 0.015a 0.362 ± 0.060a 0.356 ± 0.032a
240 0.314 ± 0.018a 0.285 ± 0.028a 0.303 ± 0.002a 0.279 ± 0.016a 0.340 ± 0.043a
1
Codes for HPP cheeses are 400W2, HPP at 400 MPa on day 14; 600W2, HPP at 600 MPa on day
14; 400W3, HPP at 400 MPa on day 21; 600W3, HPP at 600 MPa on day 21.
2
Concentrations are expressed in mg/g cheese DM, as mean ± SEM of duplicate determinations on
two cheese-making trials. Means in the same row at the same sampling date bearing the same
superscript do not differ significantly.
29
Table 2 Esterase activity1 during ripening of control and HPP cheeses made from unpasteurized
milk
Days Control cheese HPP cheeses2
400W2 600W2 400W3 600W3
1 4.63 ± 0.18
14 6.32 ± 0.19b 4.42 ± 0.28a 3.22 ± 0.50a
21 7.71 ± 0.06d 4.63 ± 0.14c 2.73 ± 0.12b 4.21 ± 0.06c 2.25 ± 0.10a
60 7.90 ± 0.09c 3.32 ± 0.09b 1.36 ± 0.09a 3.39 ± 0.11b 1.40 ± 0.16a
120 10.75 ± 0.19c 4.26 ± 0.30b 1.80 ± 0.04a 3.91 ± 0.06b 1.79 ± 0.02a
180 11.21 ± 0.23c 3.50 ± 0.24b 0.74 ± 0.13a 2.82 ± 0.24b 0.66 ± 0.05a
240 10.72 ± 0.15c 3.18 ± 0.08b 0.97 ± 0.10a 3.51 ± 0.09b 1.20 ± 0.03a
1
Esterase activity is expressed in pmol of α-naphthol per min and g, as mean ± SEM of duplicate
determinations on two cheese-making trials. Means in the same row at the same sampling date
bearing the same superscript do not differ significantly.
2
30
Table 3 Concentrations of short-chain (SC), medium-chain (MC) and long-chain (LC) free fatty
acids (FFA) during ripening of control and HPP cheeses made from unpasteurized milk
Acids Days Control cheese HPP cheeses1
400W2 600W2 400W3 600W3
SC FFA2 21 0.038 ± 0.004a 0.032 ± 0.002a 0.030 ± 0.002a 0.035 ± 0.003a 0.034 ± 0.002a
60 0.070 ± 0.003b 0.050 ± 0.002a 0.045 ± 0.004a 0.058 ± 0.005ab 0.050 ± 0.004a
120 0.113 ± 0.005b 0.098 ± 0.008b 0.057 ± 0.003a 0.107 ± 0.002b 0.059 ± 0.002a
180 0.143 ± 0.009b 0.117 ± 0.014b 0.069 ± 0.007a 0.120 ± 0.009b 0.075 ± 0.004a
240 0.190 ± 0.009c 0.147 ± 0.010b 0.087 ± 0.003a 0.148 ± 0.006b 0.086 ± 0.003a
MC FFA2 21 0.190 ± 0.007a 0.198 ± 0.005a 0.188 ± 0.006a 0.181 ± 0.010a 0.185 ± 0.006a
60 0.228 ± 0.005a 0.234 ± 0.010a 0.257 ± 0.010a 0.239 ± 0.011a 0.240 ± 0.006a
120 0.296 ± 0.010a 0.303 ± 0.008a 0.282 ± 0.002a 0.295 ± 0.007a 0.280 ± 0.002a
180 0.356 ± 0.010a 0.333 ± 0.013a 0.325 ± 0.012a 0.335 ± 0.012a 0.316 ± 0.002a
240 0.400 ± 0.005c 0.384 ± 0.006bc 0.350 ± 0.005a 0.370 ± 0.010ab 0.361 ± 0.006ab
LC FFA2 21 0.657 ± 0.035a 0.699 ± 0.012a 0.671 ± 0.020a 0.634 ± 0.026a 0.632 ± 0.018a
60 0.745 ± 0.010a 0.746 ± 0.023a 0.871 ± 0.025b 0.810 ± 0.030ab 0.800 ± 0.014ab
120 0.982 ± 0.021a 1.006 ± 0.029a 0.969 ± 0.006a 0.996 ± 0.034a 0.938 ± 0.015a
180 1.174 ± 0.033a 1.120 ± 0.044a 1.089 ± 0.046a 1.112 ± 0.037a 1.078 ± 0.010a
240 1.307 ± 0.020c 1.244 ± 0.017bc 1.141 ± 0.025a 1.210 ± 0.026ab 1.165 ± 0.008ab
1
2
Concentrations are expressed in mg/g cheese DM, as mean ± SEM of duplicate determinations on
two cheese-making trials. Means in the same row at the same sampling date bearing the same
superscript do not differ significantly.
31
Table 4 Levels of the main groups of volatile compounds in control cheese on days 14 and 21 of
ripening.
Chemical group1 Time of ripening

Day 14 Day 21
Ketones 10593 ± 1180 8792 ± 1213
Acids 4298 ± 600 5127 ± 624
Alcohols 2518 ± 403 4209 ± 525
Benzenic compounds 203.6 ± 24.1 223.9 ± 37.2
Esters 175.2 ± 25.2 238.8 ± 27.1
Ethers 69.7 ± 9.6 105.5 ± 13.6
Terpenoids 50.52 ± 5.53 59.91 ± 8.13
Hydrocarbons 24.23 ± 5.01 50.72 ± 8.22
Aldehydes 17.60 ± 2.08 26.30 ± 3.53
Sulphur compounds 17.92 ± 1.82 22.66 ± 3.79
1
Levels of VC groups (sums of individual VC) are expressed in relative abundance (compound peak
area x 102 / internal standard peak area), as mean ± SEM of triplicate determinations on two cheese
making trials.
32
Table 5 Levels of total ketones, acids, alcohols, benzenic compounds and esters during ripening of
control and HPP cheeses made from unpasteurized milk
Chemical group Days Control cheese HPP cheeses1
400W2 600W2 400W3 600W3
Ketones 60 5211 ± 829 a 21057 ± 2433 c 14344 ± 1893 bc 18262 ± 1555 c 10874 ± 1758 ab
120 4049 ± 543 a 2858 ± 530 a 10428 ± 1327 b 3669 ± 605 a 10622 ± 937 b
240 1381 ± 175 a 759 ± 62 a 7432 ± 740 b 980 ± 116 a 6990 ± 992 b
Acids 60 5392 ± 840 a 7334 ± 1201 a 5494 ± 733 a 7595 ± 1122 a 5625 ± 738 a
120 5197 ± 524 a 5255 ± 811 a 5070 ± 482 a 5515 ± 724 a 5247 ± 618 a
240 3018 ± 369 a 4520 ± 539 a 4588 ± 675 a 3475 ± 272 a 4718 ± 516 a
Alcohols 60 11663 ± 1109 b 3432 ± 463 a 2522 ± 360 a 4052 ± 436 a 3402 ± 474 a
120 16181 ± 1448 d 7395 ± 627 b 1983 ± 176 a 12055 ± 1442 c 2399 ± 124 a
240 9582 ± 1026 b 8955 ± 1219 b 1508 ± 198 a 8547 ± 751 b 1998 ± 267 a
Benzenic compounds 60 388.8 ± 45.0 a 506.0 ± 59.3 a 417.6 ± 57.3 a 469.2 ± 51.5 a 402.0 ± 40.9 a
120 334.8 ± 31.8 a 307.4 ± 33.2 a 350.6 ± 57.5 a 389.0 ± 43.9 a 341.7 ± 38.3 a
240 206.2 ± 31.3 a 286.4 ± 52.5 a 285.7 ± 54.9 a 211.1 ± 19.5 a 280.8 ± 29.0 a
Esters 60 276.2 ± 41.4 a 349.3 ± 55.5 a 295.4 ± 47.1 a 314.4 ± 38.1 a 280.3 ± 34.7 a
120 267.2 ± 24.2 a 228.7 ± 27.5 a 188.9 ± 15.9 a 232.0 ± 25.9 a 214.1 ± 16.4 a
240 204.7 ± 25.0 a 167.2 ± 22.4 a 154.7 ± 26.9 a 133.7 ± 13.2 a 160.6 ± 21.8 a
1
Codes for HPP cheeses are 400W2, HPP at 400 MPa on day 14; 600W2, HPP at 600 MPa on day 14;
400W3, HPP at 400 MPa on day 21; 600W3, HPP at 600 MPa on day 21.
2
Levels are expressed in relative abundance (compound peak area x 102 / internal standard peak area),
as mean ± SEM of triplicate determinations on two cheese making trials. Means in the same row at the
same sampling date bearing the same superscript do not differ significantly.
33
Table 6 Levels of total ethers, terpenoids, hydrocarbons, aldehydes and sulphur compounds during
ripening and refrigerated storage of control and HPP cheeses made from unpasteurized milk
Chemical group Days Control cheese HPP cheeses1
400W2 600W2 400W3 600W3
Ethers 60 148.3 ± 12.9 a 266.4 ± 66.9 a 208.0 ± 39.1 a 279.3 ± 44.2 a 131.0 ± 16.2 a
120 103.3 ± 14.2 a 245.1 ± 37.1 a 224.1 ± 27.2 a 176.1 ± 41.6 a 159.9 ± 13.7 a
240 91.3 ± 11.1 a 214.7 ± 65.4 a 133.3 ± 18.5 a 155.1 ± 36.2 a 205.5 ± 75.4 a
Terpenoids 60 113.4 ± 8.4 a 115.8 ± 14.0 a 83.32 ± 6.62 a 119.4 ± 15.0 a 84.31 ± 7.15 a
120 143.8 ± 9.6 b 126.0 ± 13.3 ab 90.46 ± 8.03 a 126.7 ± 12.9 ab 93.08 ± 5.91 a
240 151.2 ± 14.7 a 152.0 ± 13.8 a 120.3 ± 10.6 a 135.3 ± 9.9 a 118.0 ± 15.7 a
Hydrocarbons 60 74.65 ± 5.37 a 97.59 ± 9.97 ab 98.68 ± 17.43ab 146.25 ±19.26b 113.61 ±6.37ab
120 72.51 ± 10.37 a 138.13 ±18.61b 105.66 ±14.71ab 104.29 ±10.49ab 131.98 ±14.62b
240 22.69 ± 2.20 a 30.10 ± 3.67 a 53.59 ± 2.77 b 39.22 ± 5.13 ab 54.26 ± 6.27 b
Aldehydes 60 20.46 ± 2.66 a 31.79 ± 3.53 ab 42.16 ± 5.57 bc 32.70 ± 2.84 ab 51.16 ± 4.82 c
120 19.44 ± 2.25 a 30.54 ± 5.73 ab 38.94 ± 1.96 b 27.93 ± 3.97 ab 39.43 ± 4.18 b
240 19.60 ± 2.65 a 29.33 ± 3.75 ab 41.39 ± 5.15 bc 17.35 ± 1.52 a 51.97 ± 6.91 c
Sulphur compounds 60 12.08 ± 2.22 a 22.74 ± 4.31 a 18.36 ± 2.29 a 17.78 ± 3.11 a 18.59 ± 3.61 a
120 8.41 ± 0.96 a 12.06 ± 2.08 a 12.12 ± 0.80 a 9.97 ± 1.29 a 13.55 ± 2.64 a
240 3.69 ± 0.59 a 8.16 ± 1.28 b 8.79 ± 1.21 b 5.56 ± 0.76 ab 8.90 ± 1.29 b
1
2
Levels are expressed in relative abundance (compound peak area x 102 / internal standard peak area),
as mean ± SEM of triplicate determinations on two cheese making trials. Means in the same row at the
same sampling date bearing the same superscript do not differ significantly.
34
Table 7 CIE-LAB colour parameters during ripening of control and HPP cheeses made from
unpasteurized milk
Parameter Days Control cheese HPP cheeses1
400W3 600W3 400W5 600W5
L* (lightness)2 21 85.13 ± 0.35a 85.24 ± 0.22ab 86.57 ± 0.39 b 85.29± 0.47ab 86.10 ± 0.32 ab
60 83.31 ± 0.21a 84.44 ± 0.40bc 84.92 ± 0.26 c 84.13± 0.21abc 83.76 ± 0.27 ab
120 82.84 ± 0.37abc 82.16 ± 0.35ab 83.66 ± 0.28 c 82.07± 0.31a 83.33 ± 0.25 bc
180 80.60 ± 0.49a 80.57 ± 0.63a 82.87 ± 0.52 b 80.72± 0.40a 82.80 ± 0.46 b
240 76.58 ± 0.47 a 77.99 ± 0.68 a 78.94 ± 0.70 a 77.24 ± 0.67 a 77.99 ± 0.56 a
a* (redness)2 21 1.99 ± 0.04 c 1.67 ± 0.02 b 1.25 ± 0.33 a 1.75 ± 0.04 b 1.35 ± 0.03 a
60 2.13 ± 0.03c 1.85 ± 0.05 b 1.46 ± 0.03 a 1.88 ± 0.03 b 1.42 ± 0.04 a
120 2.00 ± 0.07 c 1.72 ± 0.03 b 1.09 ± 0.05 a 1.59 ± 0.02 b 1.06 ± 0.04 a
180 2.12 ± 0.03c 2.06 ± 0.07 c 1.16 ± 0.04 a 1.75 ± 0.04 b 1.12 ± 0.03 a
240 1.63 ± 0.06 c 1.36 ± 0.09bc 0.72 ± 0.10 a 1.17 ± 0.07 b 0.56 ± 0.10 a
b* (yellowness)2 21 19.01 ± 0.44 a 18.17 ± 0.36 a 18.77 ± 0.49 a 18.22 ± 0.47 a 18.75 ± 0.40 a
60 19.31 ± 0.24ab 18.68 ± 0.16 a 19.06 ± 0.23 ab 19.27 ± 0.29 ab 19.68 ± 0.24 b
120 22.12 ± 0.27 a 22.14 ± 0.37 a 21.58 ± 0.26 a 22.16 ± 0.27 a 22.11 ± 0.37 a
180 22.42 ± 0.25 a 22.53 ± 0.42 a 22.32 ± 0.36 a 22.15 ± 0.23 a 21.46 ± 0.29 a
240 24.13 ± 0.17 a 24.14 ± 0.16 a 23.65 ± 0.19 a 23.97 ± 0.16 a 23.70 ± 0.27 a
1
2
Results are expressed as mean ± SEM of six determinations on the cheese interior in two cheese
making trials. Means in the same row at the same sampling date bearing the same superscript do not
differ significantly.
35
Figure legends
Fig. 1. Distribution of HPP and control cheeses made from unpasteurized milk on the plane defined
by functions 1 and 2 of the principal component analysis performed on the concentrations of
individual carboxylic acids. Each symbol represents the averaged value of the two cheese-making
trials. Treatments are as follows: control, C; 400W2, black circle; 600W2, black triangle; 400W3,
open circle; 600W3, open triangle. Days after manufacture are indicated on each of the symbols.
Fig. 2. Distribution of HPP and control cheeses made from unpasteurized milk on the plane defined
by functions 1 and 2 of the principal component analysis performed on the total levels of the groups
of volatile compounds. Each symbol represents the averaged value of the two cheese-making trials.
Treatments are as follows: control, C; 400W2, black circle; 600W2, black triangle; 400W3, open
circle; 600W3, open triangle. Days after manufacture are indicated on each of the symbols.
36
Figure 1
1.5
60d 60d
1.0 21d 120d 120d 180d
Function 2 (11.6%)
60d
120d
180d
C C
0.5
14d
C 60d C 180d
60d 120d 120d
0.0
14d C 240d
180d
21d 21d
-0.5 21d 180d
21d 240d
240d
C
-1.0 14d
240d 240d
-1.5
-2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 2.5
Function 1 (73.8%)
37
Figure 2
2.5
120d
2.0 C
1.5
Function 2 (23.5%)
120d 60d
240d
1.0
C C
0.5 240d 120d
240d
0.0 60d 60d
60d
-0.5 240d 240d 120d 120d 21d
C 60d
-1.0 14d
-1.5
C
-1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0
Function 1 (41.9%)
38
Capítulo 11. Discusión general.

243

Fotografía: cabina de cata para la evaluación organoléptica de los quesos (superior) y
aspecto de las muestras preparadas para cata de queso azul Roncari-blue (centro
izquierda), Torta del Casar (centro derecha), queso Brie (inferior izquierda) y queso
Arzúa-Ulloa (inferior derecha).
Discusión general
La maduración es una de las fases más importantes en la elaboración del queso, ya

que durante este periodo se desarrolla el sabor, el aroma, la textura y la apariencia de
las distintas variedades de queso (McSweeney, 2004). El correcto equilibrio de
compuestos responsables del aroma y sabor determina el momento óptimo para el
consumo de cada variedad (Mulder, 1952). Este equilibrio se alcanza durante la
maduración mediante diversos procesos bioquímicos que afectan a los componentes
químicos del queso como se describe en la Introducción. Una vez alcanzado el
equilibrio óptimo, estos procesos continúan, pudiendo provocar un desequilibrio de
compuestos responsables del aroma y sabor y consecuentemente una pérdida de
calidad del producto, limitando la vida útil de este. El empleo de temperaturas de
refrigeración ralentiza aunque no llega a impedir este fenómeno conocido como
sobremaduración. Apenas existen estudios enfocados a evitar este fenómeno, aunque sí
se ha conseguido alargar la vida útil del queso fresco mediante el empleo de altas
presiones (Daryaei et al., 2006, Evert-Arriagada et al., 2014). Además del problema de la
sobremaduración, la abundancia de aminoácidos libres junto a la presencia de
microorganismos con actividad descarboxilasa, puede desembocar en la formación y
acumulación de aminas biógenas, con los consecuentes problemas para la salud del
consumidor (Linares et al., 2011).
En el presente trabajo de investigación se ha estudiado el efecto de las altas

presiones hidrostáticas sobre distintas variedades de queso a lo largo de un prolongado
periodo de tiempo post-tratamiento, con el fin de evitar la sobremaduración y
aumentar la vida útil de las variedades estudiadas. Se examinaron diferentes parámetros
al objeto de evaluar la proteolisis, lipolisis y catabolismo de aminoácidos y ácidos
grasos libres (AGL), además de parámetros microbiológicos, texturométricos,
colorimétricos y organolépticos así como el pH y extracto seco (ES). Se seleccionaron
cuatro variedades de quesos, que se describen a continuación. El queso azul (Roncari-
blue) fabricado con leche pasteurizada de oveja y madurado con mohos (Penicillium
roqueforti) en su interior, cuyos resultados se recogen en los capítulos 2 y 3. La Torta
del Casar, fabricado con leche cruda de oveja y cuajo vegetal, cuyos resultados se
recogen en los capítulos 4, 5 y 6. El queso Brie, fabricado con leche pasteurizada de
vaca y madurado con mohos (Penicillium camemberti) en su superficie, cuyos
245

Discusión general
resultados se recogen en los capítulos 7 y 8. Y el queso Arzúa-Ulloa, fabricado con leche

cruda de vaca, cuyos resultados se recogen en los capítulos 9 y 10.
Los quesos estudiados fueron tratados con dos niveles de presión (400 y 600 MPa)
aplicados en distintos momentos de la maduración, seleccionados en función de la
variedad. Los quesos tratados por alta presión, así como sus respectivos controles no
tratados, fueron evaluados en distintos momentos de la maduración y del
almacenamiento en refrigeración, elegidos también en función de la variedad de queso.
 Queso azul (Roncari-blue)
El queso azul se caracteriza por la presencia de Penicillium roqueforti en su interior.

En el queso azul investigado (Roncari-blue), elaborado con leche pasteurizada de oveja,
se estudió la evolución de las variables microbiológicas, químicas y sensoriales durante
la maduración y el almacenamiento en refrigeración hasta el día 360. Mediante los
análisis microbiológicos, se comprobó una disminución de los niveles de bacterias
lácticas y bacterias mesófilas totales durante la maduración y el almacenamiento en
refrigeración, desde valores de 9,54 y 8,76 log ufc/g, respectivamente, a día 1, hasta
8,02 y 7,99 log ufc/g a día 180, y 6,57 y 5,75 log ufc/g a día 360. Los niveles de P.
roqueforti sufrieron también un descenso gradual, desde 7,73 log ufc/g a día 21 hasta
7,10 log ufc/g a día 180 y 5,66 log ufc/g a día 360. El pH aumentó desde 4,76 a día 1
hasta 6,24 a día 60, al igual que se ha observado en queso Gorgonzola (Gobbetti et al.,
1997), y posteriormente descendió hasta 5,34 a día 360. El extracto seco aumentó desde
46,63 % a día 1 hasta 53,95 % a día 90 y posteriormente se mantuvo en estos mismos
valores hasta el día 360.
Las αS-, β-, κ- y p-κ-caseínas sufrieron una fuerte reducción del 93, 87, 78 y 46 %
respectivamente, entre los días 1 y 21. Esta elevada hidrólisis de caseínas es debida a la
presencia de plasmina residual, que mantiene parte de su actividad tras los tratamientos
térmicos de pasteurización (Ismail & Nielsen, 2010), junto a la presencia de cuajo,
enzimas de las bacterias lácticas y principalmente las enzimas de P. roqueforti (Cantor et
al., 2004). La αS- y la κ-caseína no volvieron a detectarse a partir del día 90, mientras que
la β- y la p-κ-caseína continuaron disminuyendo a lo largo del periodo de
almacenamiento en refrigeración hasta el día 360. Los péptidos hidrófilos aumentaron
ligeramente durante todo el almacenamiento en refrigeración hasta alcanzar un valor
246

Discusión general
de 29,81 UA/mg ES a día 360, mientras que los péptidos hidrófobos descendieron de
forma gradual hasta el día 360, alcanzando un valor de 4,54 UA/mg ES. El ratio de
péptidos hidrófobos/hidrófilos descendió gradualmente durante todo el
almacenamiento en refrigeración. La actividad aminopeptidasa alcanzó valores máximos
de 24,83 nmoles de p-nitroanilina/min·g a día 42 con el sustrato Leu-p-NA y de 24,58
nmoles de p-nitroanilina/min·g a día 63 con el sustrato Lys-p-NA. Posteriormente la
actividad aminopeptidasa fue descendiendo hasta valores de 3,26 y 3,21 nmoles de p-
nitroanilina/min·g a día 360 con los sustratos Leu-p-NA y Lys-p-NA, respectivamente. La
alta actividad de las exopeptidasas de P. roqueforti hicieron que la concentración de
aminoácidos libres aumentase desde 1,49 mg/g ES en el día 1 hasta 61,20 mg/g ES a
día 90, siendo estas concentraciones más altas que las encontradas en queso
Gorgonzola (Gobbetti et al., 1997) pero inferiores a las encontradas en otros quesos
azules como Stilton (Madkor et al., 1987b) o Picón Bejes-Tresviso (Prieto et al., 2000).
Posteriormente los niveles de aminoácidos libres continuaron aumentando hasta el día
360, llegando a alcanzar una concentración de 116,76 mg/g ES. De igual manera, la
proteolisis global aumentó durante la maduración y el almacenamiento en
refrigeración.
Las aminas biógenas, procedentes de la descarboxilación de los aminoácidos,

aumentaron desde 0,007 mg/g ES de aminas biógenas totales a día 90 hasta 0,218
mg/g ES a día 360, siendo estos valores más bajos que la media encontrada en otros
quesos azules elaborados con diferentes tipos de leche (Flórez et al., 2006, Fernández et
al., 2007). A día 90 únicamente se detectó β-feniletilamina, que aumentó hasta una
concentración de 0,06 mg/g ES a día 360, mientras que la triptamina, tiramina y
putrescina únicamente se encontraron a partir del día 180, alcanzando valores a día 360
de 0,07, 0,05 y 0,03 mg/g ES, respectivamente. La espermidina sólo se detectó en los
días 270 y 360, alcanzando una concentración de 0,01 mg/g ES a día 360.
La actividad esterasa alcanzó un máximo de 38,53 pmoles de α-naftol/min·g a día

42, disminuyendo posteriormente y volviendo a aumentar hasta alcanzar 33,74 pmoles
de α-naftol/min·g a día 180. Los niveles de ácidos grasos libres totales aumentaron
hasta 83,82 mg/g ES a día 360, con valores durante la maduración similares a los
encontrados en Gorgonzola y Stilton (Madkor et al., 1987a, Gobbetti et al., 1997). Los
247

Discusión general
ácidos grasos libres de cadena corta (C4:0 - C8:0) aumentaron hasta 8,88 mg/g ES a día
360, mientras que los ácidos grasos libres de cadena media (C10:0 - C14:0) y de cadena
larga (C16:0 - C18:3) aumentaron hasta 18,31 y 56,63 mg/g ES respectivamente a día 360,
siendo estos valores muy superiores a los encontrados en variedades de queso no
maduradas por mohos (Ávila et al., 2007, Delgado et al., 2011), debido a la elevada
actividad que presentan las lipasas de P. roqueforti (Cantor et al., 2004), aunque
menores a los encontrados en otros quesos azules (Prieto et al., 2000). El ácido oleico
fue el mayoritario durante la maduración y el almacenamiento en refrigeración,
alcanzando valores de 27,03 mg/g ES a día 360. El ácido etanoico, procedente
principalmente del metabolismo de lactosa, lactato y citrato (McSweeney & Sousa,
2000), aumentó hasta el día 63 alcanzando una concentración de 1,72 mg/g ES y
posteriormente descendió hasta 1,39 mg/g ES a día 360.
El grupo mayoritario de compuestos volátiles fue el de los ácidos volátiles, seguido

de alcoholes y cetonas. Los grupos de compuestos más característicos de los quesos
azules son el de las cetonas, que aportan aromas florales y frutales, seguido del grupo
de los alcoholes, que aportan aromas a fresco, frutales y hierba (Curioni & Bosset, 2002,
Flórez et al., 2006, Voigt et al., 2010). La menor concentración de cetonas encontradas
en este estudio puede deberse a la toxicidad que presentan las altas concentraciones de
ácidos grasos, especialmente los de cadena larga, sobre el micelio del moho, lo que
puede reducir la formación de cetonas (Kinsella & Hwang, 1976). Los niveles de ácidos
volátiles, ésteres, terpenos, compuestos nitrogenados y azufrados aumentaron entre los
días 180 y 360. Los niveles de alcoholes y cetonas también aumentaron, aunque más
ligeramente. Por otro lado, los aldehídos y los compuestos bencénicos descendieron y
los hidrocarburos se mantuvieron en los mismos niveles.
En el análisis sensorial se determinó que únicamente la intensidad de sabor y el

parámetro de sabor umami aumentaron durante el almacenamiento en refrigeración. La
calidad de sabor, así como el resto de los parámetros sensoriales, se mantuvieron en
valores similares hasta el día 360.
El almacenamiento en refrigeración indujo leves cambios químicos en el queso azul,

que no se vieron reflejados en una pérdida de calidad de sabor del queso y únicamente
provocaron un aumento de la intensidad de sabor. A pesar de no observarse efectos
248

Discusión general
negativos de sobremaduración, sí se produjo una acumulación de aminas biógenas que,

a pesar de encontrarse en niveles bajos, podrían plantear problemas de salud en
individuos con un sistema de detoxificación deficiente o en individuos que se
encuentren bajo tratamiento con fármacos inhibidores de este sistema.
En queso azul se aplicaron tratamientos de altas presiones de 400 y 600 MPa a las 3
semanas (400-3S y 600-3S), a las 6 semanas (400-6S y 600-6S) y a las 9 semanas (400-
9S y 600-9S) de maduración. La aplicación de estos tratamientos indujo una serie de
cambios respecto al control en los diferentes tiempos (Tablas 5 y 6). Comparando los
quesos tratados a lo largo del almacenamiento en refrigeración con el queso control en
su momento óptimo (180 días), se pueden apreciar los cambios debidos al efecto de las
altas presiones (Tabla 7).
Las reducciones de los niveles de microorganismos provocadas por las altas

presiones, en especial con los tratamientos de 600 MPa, que redujeron los niveles de P.
roqueforti por debajo del límite de detección, no lograron atenuar la fuerte hidrólisis de
αS-caseína, que ya a día 90 no se detectó en el queso control. En los quesos tratados
con 400 MPa a las 9 semanas y los tratados con 600 MPa a las 3 y 6 semanas se
detectaron pequeñas cantidades de αS-caseína a día 180. Los quesos tratados con 600
MPa a las 3 semanas presentaron niveles de β-caseína similares a los del control de 180
días, pero únicamente hasta el día 270. Los tratamientos de altas presiones atenuaron
levemente la hidrólisis de p-κ-caseína, pero en ningún caso consiguieron mantener
niveles similares a los del queso control de 180 días. Por lo que respecta a los péptidos,
únicamente los quesos tratados con 600 MPa a las 3 y 6 semanas consiguieron
mantener hasta el día 360 niveles de péptidos hidrófilos similares o inferiores a los del
control de 180 días, mientras que a día 360 los niveles de péptidos hidrófobos de todos
los quesos tratados fueron superiores a los del control de 180 días, haciendo que el
ratio de péptidos hidrófobos/hidrófilos también fuese mayor. La reducción de la
actividad aminopeptidasa provocada por las altas presiones se vio reflejada en los
niveles de aminoácidos libres de los quesos tratados a las 3 semanas y de los tratados
con 600 MPa a las 6 y 9 semanas, que mantuvieron niveles inferiores o similares a los
del queso control de 180 días hasta el final del almacenamiento en refrigeración, al
igual que sucedió con la proteolisis global.
249

Discusión general
Tabla 5. Tendencias de las características microbiológicas y químicas del queso azul tratado por alta presión
con respecto al queso control.
Parámetro 400-3S 600-3S 400-6S 600-6S 400-9S 600-9S
ES NS NS NS NS NS NS
pH ↓ 90 a 360 d ↓ 90 a 360 d NS NS ↓ 180 d ↓ 180 d
Aerobios totales ↓ 90 a 270 d ↓↓ hasta 360 d ↓ 180 a 360 d ↓↓ hasta 360 d ↓ 180 a 360 d ↓↓ hasta 360 d
Bacterias lácticas ↓ 90 a 270 d ↓↓ hasta 360 d ↓ 180 a 360 d ↓↓ hasta 360 d ↓ 180 a 360 d ↓↓ hasta 360 d
P. roqueforti ↓ hasta 360 d ↓↓ hasta 360 d ↓ hasta 360 d ↓↓ hasta 360 d ↓ hasta 360 d ↓↓ hasta 360 d
α-caseína ↑ 90 d ↑ 90 d ↑ 90 d ↑ 90 d NS NS
β-caseína ↑ 90 d ↑ 180 y 270 d NS ↑ 360 d NS NS
κ-caseína NS NS NS NS NS NS
p-κ-caseína NS ↑ 270 d NS NS NS NS
Péptidos hidrófilos ↑ 90 d NS ↑ 90 d NS ↑ 360 d ↑ 90 d
↑ 90, 180 y ↑ 90, 180 y
Péptidos hidrófobos ↑ 90 a 360 d ↑ 360 d ↑ 270 y 360 d ↑ 270 y 360 d
360 d 360 d
↑ 90, 180 y ↑ 90, 180 y
Ratio hidrófobos/filos ↑ 90 a 360 d ↑ 360 d ↑ 270 d ↑ 270 y 360 d
360 d 360 d
Act. aminopeptidasa (Lys) ↓ hasta 360 d ↓ hasta 360 d ↓ hasta 360 d ↓ hasta 360 d ↓ hasta 360 d ↓ hasta 360 d
Act. aminopeptidasa (Leu) ↓ hasta 360 d ↓ hasta 360 d ↓ hasta 360 d ↓ hasta 360 d ↓ hasta 360 d ↓ hasta 360 d
Aminoácidos libres
↓ 90 a 360 d ↓ 90 a 360 d NS ↓ 270 y 360 d ↑ 180 d ↓ 360 d
totales
Proteolisis global (OPA) ↓ 180 a 360 d ↓ 90 a 360 d NS ↓ 180 y 360 d NS ↓ 180 y 360 d
Tiramina ↓ 360 d ↓ 360 d ↓ 360 d ↓ 360 d ↓ 360 d ↓ 360 d
Putrescina NS NS NS ↑ 270 y 360 d NS ↑ 270 y 360 d
Triptamina ↑ 360 d NS NS ↑ 360 d NS ↑ 360 d
↓ 180 d,
Feniletilamina ↓ 180 d, ↑ 360 d ↑ 360 d ↓ 180 d ↓ 180 d ↓ 180 d, ↑ 270 d
↑ 270 y 360 d
Aminas totales NS ↓ 180 d ↓ 180 a 360 d ↓ 180 d NS ↑ 360 d
↓ 21, 42, 180 a ↓ 42, 270 y ↓ 63, 180 a
Actividad esterasa ↓ hasta 360 d ↓ 42 d ↓ 360 d
360 d 360 d 360 d
↓ 42, 63, 180
Etanoico + propanoico ↓ 42 a 360 d ↓ hasta 360 d ↓ 42 y 63 d ↓ 180 d ↓ 90 a 360 d
a 360 d
↓ 42, 63, 180 a ↓ 21 a 63, 180
AGL cadena corta ↑ 63 y 90 d ↑ 90 d ↑ 63 d ↑ 90 d
360 d a 360 d
↓ 42, 180 y ↓ 21, 42, 180
AGL cadena media NS NS NS NS
270 d a 360 d
↓ 42, 180 y ↓ 21, 42, 180
AGL cadena larga NS NS NS NS
270 d a 360 d
Ácidos volátiles ↓ 180 y 360 d ↓ 180 y 360 d NS NS NS NS
Alcoholes volátiles ↓ 180 y 360 d ↓↓ 180 y 360 d ↓ 180 d ↓↓ 180 y 360 d ↓ 180 y 360 d ↓↓ 180 y 360 d
Aldehídos volátiles NS ↓ 180 d NS ↓ 180 d NS ↓ 180 d
Cetonas volátiles NS ↓ 180 y 360 d ↓ 180 d NS ↓ 180 d ↓ 360 d
Ésteres volátiles ↓ 180 y 360 d ↓↓ 180 y 360 d ↓ 360 d ↓ 180 y 360 d ↓ 180 y 360 d ↓ 180 y 360 d
Hidrocarburos volátiles NS ↓ 180 d NS NS ↓ 180 d ↓ 180 y 360 d
Bencénicos volátiles ↓↓ 180 y 360 d ↓↓ 180 y 360 d NS NS ↓ 180 y 360 d ↓ 360 d
Azufrados volátiles NS NS NS NS NS NS
Nitrogenados volátiles ↓ 360 d ↓ 360 d NS NS NS NS
Terpenos volátiles NS ↓ 180 y 360 d NS NS NS NS
↑ , ↓ ; aumento o disminución, respectivamente, significativos respecto al queso control en el tiempo indicado a
continuación. NS; ausencia de efecto significativo (P > 0,05, test de Tukey) para todos los tiempos. OPA; o-
ftalaldehído (técnica espectrofotométrica).
250

Discusión general
La acumulación de aminas biógenas totales únicamente se vio atenuada por los

tratamientos de 400 MPa aplicados a las 6 semanas de maduración, mientras que los
tratamientos de 600 MPa aplicados a las 9 semanas aumentaron los niveles de aminas
biógenas totales. En ninguno de los quesos tratados con altas presiones se lograron
mantener las aminas biógenas totales a día 360 en niveles similares a los del control de
180 días. Las altas presiones lograron reducir la acumulación de tiramina a día 360,
aunque sus niveles fueron superiores a los del queso control de 180 días. Los quesos
tratados con 400 MPa a las 3 semanas y los tratados con 600 MPa a las 6 y 9 semanas
alcanzaron a día 360 niveles de triptamina superiores a los del queso control, mientras
que los quesos tratados con 400 MPa a las 3 semanas y los tratados con 600 MPa a las
3 y 9 semanas alcanzaron niveles superiores de β-feniletilamina a día 360 y los quesos
tratados con 600 MPa a las 6 y 9 semanas alcanzaron a día 360 niveles superiores de
putrescina. La liberación de enzimas al medio, debida a la lisis de microorganismos con
actividad descarboxilasa, pudo causar el aumento de los niveles de aminas biógenas en
algunos de los quesos tratados.
Tabla 6. Tendencias de las características sensoriales del queso azul tratado por alta presión con respecto al
queso control.
Parámetro 400-3S 600-3S 400-6S 600-6S 400-9S 600-9S
Calidad sabor ↓ 90 d ↓ 90 a 360 d NS NS NS NS
Intensidad sabor NS ↓ 180 a 360 d NS NS NS NS
Sabor umami NS NS NS NS NS NS
Sabor ácido NS NS NS NS NS NS
Sabor amargo NS NS NS NS NS NS
Sabor dulce NS NS NS NS NS NS
Sabor salado NS NS NS NS NS NS
continuación. NS; ausencia de efecto significativo (P > 0,05, test de Tukey) para todos los tiempos.
La reducción de la actividad esterasa causada por los tratamientos de 600 MPa y el

tratamiento de 400 MPa aplicado a las 3 semanas atenuó la formación de ácidos grasos
de cadena corta, pero únicamente los quesos tratados a las 3 semanas mantuvieron a
día 360 niveles de ácidos grasos de cadena corta similares o inferiores a los del queso
control de 180 días. Los niveles de ácidos grasos libres de cadena media de los quesos
tratados con 400 MPa a las 3 semanas y los tratados con 600 MPa se mantuvieron hasta
251

Discusión general
el día 360 en niveles similares o inferiores a los del queso control de 180 días, mientras
que los niveles de ácidos grasos libres de cadena larga se mantuvieron en niveles
ligeramente superiores a los del queso control de 180 días en los quesos tratados con
400 MPa a las 3 semanas y los tratados con 600 MPa a las 6 y 9 semanas, y en niveles
inferiores en los quesos tratados con 600 MPa a las 3 semanas. Esta reducción de los
niveles de ácidos grasos libres se ha observado en otras variedades de queso al ser
tratados por alta presión (Saldo et al., 2003, Rynne et al., 2008). Los ácidos etanoico y
propanoico se mantuvieron en niveles más bajos que los del queso control de 180 días
en todos los quesos tratados, especialmente en los quesos tratados a las 3 semanas y
los tratados con 600 MPa a las 6 y 9 semanas.
Los distintos grupos de compuestos volátiles se vieron afectados de diferente

manera por los tratamientos de altas presiones. Así, los quesos tratados con 400 MPa a
las 3 y 6 semanas tuvieron niveles de cetonas más altos que los del queso control de
180 días, mientras que los tratados con 600 MPa se mantuvieron en niveles más bajos, y
únicamente en los quesos tratados con 400 MPa a las 9 semanas se mantuvieron en
niveles similares a los del control de 180 días. Todos los quesos tratados, especialmente
los tratados con 600 MPa, mantuvieron niveles de alcoholes inferiores a los del queso
control de 180 días. Únicamente los tratamientos aplicados a las 3 semanas
consiguieron atenuar el aumento de ácidos volátiles manteniendo niveles similares a los
del control de 180 días. Todos los tratamientos atenuaron la formación de ésteres,
manteniéndose los quesos tratados con 400 MPa a las 6 y 9 semanas en niveles
ligeramente inferiores a los del queso control de 180 días y el resto de quesos tratados
en niveles muy inferiores. Los tratamientos de 600 MPa y el tratamiento de 400 MPa
aplicado a las 3 semanas mantuvieron hasta el día 360 niveles de compuestos
nitrogenados y azufrados similares o inferiores a los del queso control de 180 días. El
aumento de los niveles de terpenos se consiguió atenuar con los tratamientos de altas
presiones, manteniéndose los niveles de los quesos tratados a las 6 y 9 semanas
similares a los del control de 180 días. El descenso de los aldehídos que se dio en el
queso control se vio atenuado por las altas presiones, aunque todos los quesos tratados
se mantuvieron en niveles más bajos que los del control de 180 días. Los quesos
252

Discusión general
tratados mantuvieron los niveles de hidrocarburos y compuestos bencénicos por debajo

de los del queso control de 180 días.
Tabla 7. Valores de las características microbiológicas, químicas y sensoriales en queso azul control de 180
días, quesos tratados por alta presión de 180-360 días y queso control de 360 días.
400-3S 600-3S 400-6S 600-6S 400-9S 600-9S
Control Control
Parámetro 6 a 12 m 6 a 12 m 6 a 12 m 6 a 12 m 6 a 12 m 6 a 12 m
180 d 360 d
(rango) (rango) (rango) (rango) (rango) (rango)
pH 5,5 5,1 - 5,2 4,9 - 5,1 5,2 - 5,4 5,2 - 5,4 5,3 - 5,4 5,3 - 5,4 5,3
a
Aerobios totales 8,0 5,4 - 6,1 3,9 - 4,3 4,6 - 6,7 3,5 - 3,8 4,1 - 5,9 3,8 - 4,1 5,6
Bacterias lácticasa 8,0 6,4 - 6,7 3,2 - 4,0 5,6 - 7,5 3,2 - 3,5 5,5 - 7,1 3,3 - 3,9 6,6
a
Lactobacilos 4,9 3,4 - 3,7 1,0 - 1,8 2,6 - 4,5 1,2 - 1,5 2,5 - 4,1 1,3 - 1,9 3,6
a
P. roqueforti 7,1 2,8 - 4,4 nd 3,4 - 5,9 nd 2,6 - 5,8 nd 5,7
α-caseínab 0,0 0,0 - 0,5 0,0 - 0,2 0,0 - 0,07 0,0 - 0,3 0,0 - 0,06 0,0 - 0,0 0,0
β-caseínab 0,80 0,13 - 1,20 0,17 - 1,98 0,16 - 0,62 0,32 - 0,66 0,07 - 0,33 0,09 - 0,39 0,05
b
p-κ-caseína 3,5 1,3 - 2,9 2,0 - 4,5 0,6 - 2,6 1,2 - 3,9 0,6 - 1,7 1,3 - 2,6 1,2
Péptidos hidrófilosc 28,1 27,8 - 31,8 25,9 - 29,4 26,7 - 33,1 27,6 - 29,2 27,9 - 36,1 27,1 - 31,9 29,8
Péptidos hidrófobosc 4,8 4,1 - 7,6 6,5 - 9,1 4,8 - 6,8 5,4 - 7,1 5,9 - 6,9 5,2 - 7,9 4,5
Ratio hidrófobos/filos 0,17 0,15 - 0,24 0,25 - 0,31 0,16 - 0,23 0,19 - 0,24 0,19 - 0,22 0,19 - 0,25 0,15
Act. aminopeptidasa (Lys) d 6,3 1,0 - 3,7 1,0 - 3,2 1,3 - 3,3 0,9 - 2,4 1,7 - 2,9 0,8 - 2,2 3,2
d
Act. aminopeptidasa (Leu) 6,7 1,1 - 4,4 1,4 - 4,8 1,4 - 3,3 0,9 - 2,6 1,8 - 2,9 0,9 - 2,7 3,3
b
Aminoácidos libres totales 88,6 60,1 - 81,5 36,5 - 54,3 91,0 - 116,7 81,0 - 95,8 92,9 - 123,5 79,0 - 94,8 116,8
Proteolisis global (OPA)e 8,3 6,5 - 8,1 4,5 - 5,5 8,6 - 11,4 7,3 - 8,8 8,9 - 10,4 7,0 - 9,0 10,4
b
Tiramina 0,007 0,008 - 0,032 0,007 - 0,033 0,005 - 0,021 0,006 - 0,032 0,006 - 0,031 0,005 - 0,027 0,052
b
Putrescina 0,017 0,015 - 0,034 0,016 - 0,040 0,013 - 0,033 0,020 - 0,047 0,013 - 0,038 0,019 - 0,050 0,033
Triptaminab 0,062 0,067 - 0,103 0,046 - 0,077 0,054 - 0,075 0,043 - 0,104 0,069 - 0,094 0,058 - 0,120 0,071
b
Feniletilamina 0,013 0,008 - 0,074 0,012 - 0,078 0,005 - 0,066 0,008 - 0,063 0,004 - 0,080 0,007 - 0,092 0,061
b
Aminas biógenas totales 0,10 0,10 - 0,24 0,08 - 0,23 0,08 - 0,18 0,08 - 0,24 0,09 - 0,23 0,09 - 0,29 0,22
Actividad esterasaf 33,6 8,4 - 12,5 4,3 - 7,7 26,2 - 34,5 6,0 - 28,4 11,6 - 32,0 8,5 - 19,8 28,2
b
Etanoico + propanoico 1,7 0,5 - 0,6 0,3 - 0,4 1,1 - 1,4 0,7 - 0,9 0,9 - 1,4 0,6 - 0,9 1,4
b
AGL cadena corta 5,1 1,9 - 5,2 1,6 - 2,2 6,4 - 11,8 4,4 - 7,2 6,0 - 10,0 4,6 - 7,6 8,9
AGL cadena mediab 12,2 2,7 - 11,8 3,8 - 4,5 14,3 - 25,7 9,3 - 13,6 12,3 - 18,6 8,0 - 14,4 18,3
AGL cadena largab 36,1 10,6 - 38,3 12,2 - 15,1 41,1 - 66,9 31,2 - 44,2 39,8 - 54,7 32,9 - 45,0 56,6
g
Ácidos volátiles 2465 1517 - 2955 1441 -2090 2631 - 4455 2506 -3611 2804 - 4390 2844 - 3971 4211
Alcoholes volátilesg 1452 744 - 806 209 - 239 481 - 512 1070 - 1281 775 - 881 242 - 332 1496
Aldehídos volátilesg 15 11 - 15 5-9 7 - 15 7 - 12 9 - 10 5 - 11 11
g
Cetonas volátiles 1285 809 - 1717 397 - 670 355 - 1837 847 - 1005 649 - 1280 534 - 726 1364
Ésteres volátilesg 868 200 - 420 47 - 61 698 - 760 284 - 311 506 - 841 173 - 190 1493
Bencénicos volátilesg 17 8-9 5-6 12 - 14 13 - 14 10 - 11 11 - 14 16
g
Hidrocarburos volátiles 10 8-8 6-6 6-7 7-7 6-6 5-6 10
Terpenos volátilesg 1,5 1,0 - 1,1 0,6 - 0,9 1,4 - 1,7 1,6 - 1,7 1,4 - 1,5 1,4 - 1,5 1,9
Nitrogenados volátilesg 3,4 2,3 - 2,9 2,4 - 3,1 3,9 - 4,1 3,3 - 3,7 3,8 - 4,4 3,3 - 3,5 5,4
Calidad de sabor 7,2 6,2 - 6,5 5,2 - 5,5 7,1 - 7,3 6,5 - 6,6 6,6 - 7,2 6,6 - 7,5 7,0
Intensidad de sabor 7,3 7,1 - 7,3 6,3 - 6,5 7,0 - 7,9 7,1 - 7,4 7,3 - 7,4 6,8 - 7,5 7,6
nd: no detectado, OPA: o-ftalaldehído, a: (log ufc/g), b: (mg/g ES), c: (UA/mg ES), d:(nmol de p-nitroanilina/min g), e: (unidades de
absorbancia), f: (pmol α-naftol/min g), g: (abundancia relativa).
253

Discusión general
Las altas presiones atenuaron levemente el aumento de la intensidad de sabor,

excepto en los quesos tratados con 400 MPa a las 6 semanas que presentaron valores
más altos a día 360. Los quesos tratados con 600 MPa a las 3 semanas fueron los que
obtuvieron menor puntuación de intensidad de sabor. La calidad de sabor de los
quesos tratados se mantuvo en valores similares o ligeramente inferiores a los del
queso control de 180 días, excepto en los quesos tratados con 600 MPa a las 3 semanas
que recibieron la puntuación de calidad de sabor más baja.
Las altas presiones provocaron el hundimiento de la parte central del queso, que
llegó a ser de un 36 % respecto al borde en los quesos tratados con 400 MPa a las 3
semanas.
Considerando estos resultados junto con los del análisis de componentes

principales (recogido en el capítulo 3), que discrimina única y escasamente entre
tiempos, se puede concluir que los tratamientos de altas presiones afectaron muy
levemente al queso azul. El queso control no experimentó apenas variaciones entre su
momento óptimo de consumo (6 meses) y el final del periodo de almacenamiento en
refrigeración (12 meses). Los cambios bioquímicos durante el almacenamiento en
refrigeración no afectaron a la calidad global del queso, aunque sí se produjo un cierto
aumento de aminas biógenas. Las altas presiones no consiguieron evitar la formación
de aminas biógenas y sin embargo se produjo cierta pérdida de calidad de sabor en los
quesos tratados con 600 MPa a las 3 semanas. Se puede así concluir que la aplicación
de altas presiones no representa una mejora adicional en la conservación del queso
azul, al menos en lo que respecta a la variedad estudiada y en las condiciones
empleadas en el presente trabajo.
 Torta del Casar
La Torta del Casar se caracteriza por ser un queso elaborado con leche cruda de
oveja coagulada con cuajo vegetal obtenido de las flores del cardo (Cynara
cardunculus). En la Torta del Casar se estudió la evolución de las variables
microbiológicas, químicas, de textura y sensoriales durante la maduración y el
almacenamiento en refrigeración hasta los 240 días. Mediante los análisis
microbiológicos, se observó un aumento en los niveles de bacterias mesófilas totales y
bacterias lácticas entre los días 1 y 21, desde 7,74 y 7,56 log ufc/g respectivamente
254

Discusión general
hasta 9,51 y 9,46 log ufc/g. Posteriormente los niveles fueron reduciéndose lentamente
hasta 8,61 y 8,42 log ufc/g a día 240. De forma similar, los niveles de enterococos
aumentaron entre el día 1 y 21 desde 6,50 log ufc/g hasta 7,32 log ufc/g, reduciéndose
hasta 6,80 log ufc/g a día 240, y los niveles de lactobacilos aumentaron entre el día 1 y
el día 35 de 5,14 log ufc/g hasta 8,13 log ufc/g, manteniéndose en los mismos niveles
hasta el día 240. Los niveles de Micrococcaceae, coliformes y bacterias gram negativas
descendieron desde 6,39, 5,52 y 6,01 log ufc/g respectivamente a día 1 hasta 5,75, 2,87
y 5,37 log ufc/g a día 240. Los niveles de estafilococos coagulasa positivos descendieron
desde 5,23 log ufc/g a día 1 hasta 2,30 log ufc/g a día 180, manteniéndose por debajo
del límite de detección a día 240. Los valores de pH descendieron de 6,49 a día 1 hasta
5,19 a día 21, y aumentaron posteriormente hasta valores de 6,13 al final del periodo de
almacenamiento en refrigeración, mientras que el extracto seco aumentó desde 45,56 %
a día 1 hasta 54,67 % a día 240.
El empleo de cuajo vegetal en la elaboración de este queso hizo que la

concentración de la κ-CN disminuyese un 96 % entre los días 1 y 21, no volviéndose a
detectar a partir del día 35. La αS- y la β-CN sufrieron reducciones del 79 y 62 %
respectivamente entre los días 1 y 35. Descensos similares se han registrado en la
misma variedad de queso (Delgado et al., 2010b) y en queso Serra elaborado con cuajo
vegetal (Macedo & Malcata, 1997). La p-κ-caseína sufrió reducciones menores durante
la maduración y el posterior almacenamiento en refrigeración. Los péptidos hidrófilos
aumentaron con el tiempo, alcanzando un valor de 41,99 UA/mg ES a día 240, mientras
que los hidrófobos se mantuvieron en niveles en torno a 25 UA/mg ES, haciendo que el
ratio de péptidos hidrófobos/hidrófilos se redujese a lo largo del tiempo hasta un valor
de 0,64 a día 240. La actividad aminopeptidasa, con Leu–p-nitroanilida y Lys-p-
nitroanilida como sustratos, alcanzó valores máximos de 80,67 y 38,10 nmoles de p-
nitroanilina/min·g respectivamente a día 60, siendo valores muy superiores a los
obtenidos en queso de la Serena durante la maduración (Garde et al., 2007), y
posteriormente se fue reduciendo hasta el final del periodo de refrigeración alcanzando
valores de 6,42 y 4,67 nmoles de p-nitroanilina/min·g con Leu–p-nitroanilida y Lys-p-
nitroanilida respectivamente a día 240. La concentración de aminoácidos libres aumentó
desde 4,82 mg/g ES a día 60, valor similar a los obtenidos en Torta del Casar y queso de
255

Discusión general
la Serena durante la maduración (Garde et al., 2007, Delgado et al., 2010b) hasta 23,30
mg/g ES a los 240 días. La proteolisis global (estimada por el método OPA) aumentó
igualmente a lo largo del tiempo.
La actividad tirosina descarboxilasa se triplicó entre los días 60 y 180, provocando

un aumento de la concentración de tiramina desde 0,37 mg/g ES a día 60 hasta 1,09
mg/g ES a día 240. El resto de aminas biógenas detectadas aumentaron igualmente
durante el almacenamiento en refrigeración. La putrescina, triptamina, β-feniletilamina,
cadaverina e histamina aumentaron desde 0,29, 0,07, 0,07, 0,16 y 0,14 mg/g ES,
respectivamente, a día 60 hasta 0,95, 0,83, 0,29, 0,69 y 0,30 mg/g ES a día 240. Las
aminas biógenas totales aumentaron desde 1,09 mg/g ES a día 60 hasta 3,69 mg/g ES a
día 240, siendo niveles muy superiores a los encontrados en diferentes variedades de
quesos de leche cruda (Ordóñez et al., 1997, Fernández-García et al., 1999, Fernández-
García et al., 2000).
La actividad esterasa aumentó hasta 9,64 pmoles de α-naftol/min·g a día 60 y

posteriormente se redujo hasta 5,28 pmoles de α-naftol/min·g a día 120,
manteniéndose en estos niveles hasta el día 240. Los niveles de ácidos grasos libres
aumentaron a lo largo de la maduración y del almacenamiento en refrigeración. En esta
variedad de queso la hidrólisis de triglicéridos es debida, principalmente, a la acción de
la lipoproteína lipasa presente en la leche, la cual tiene preferencia por hidrolizar los
ácidos grasos situados en la posición sn-3 del triglicérido, donde se sitúan
preferentemente ácidos grasos de menos de 10 átomos de carbono (Collins et al., 2003,
Deeth, 2006), los cuales contribuyen a su sabor característico. Los ácidos grasos libres
de cadena corta fueron los más abundantes durante la maduración y el
almacenamiento en refrigeración, al igual que se observó en otro estudio sobre esta
variedad de queso (Delgado et al., 2009) en el que aumentaron hasta 8,69 mg/g ES a
día 240. Los ácidos grasos libres de cadena media y larga llegaron a alcanzar 0,88 y 3,02
mg/g ES, respectivamente, al final del almacenamiento en refrigeración. El ácido
etanoico, procedente del metabolismo de lactosa, lactato y citrato (McSweeney &
Sousa, 2000), fue el más abundante hasta el día 180, aportando su típico aroma a
vinagre, mientras que a día 240 el más abundante fue el ácido butírico, aportando
aroma típico de queso rancio (Curioni & Bosset, 2002). Los ácidos de cadena ramificada,
256

Discusión general
2-metilpropanoico y 3-metilbutanoico, procedentes del catabolismo de la valina y

leucina respectivamente (Yvon & Rijnen, 2001), aumentaron a lo largo de la maduración
y el periodo de almacenamiento en refrigeración aportando aromas a sudor, rancio y
pútrido.
El grupo de compuestos volátiles más abundante al final de la maduración fue el de

los ácidos volátiles, mientras que el más abundante durante el almacenamiento en
refrigeración fue el grupo de los alcoholes, siendo el 2-butanol, formado por reducción
de la 2-butanona (Molimard & Spinnler, 1996), el alcohol mayoritario. Todos los grupos
aumentaron durante el almacenamiento en refrigeración, excepto el de ácidos volátiles,
que disminuyó. Cabe destacar el drástico aumento del grupo de compuestos azufrados,
provocado principalmente por el aumento de dimetildisulfuro, que no se había descrito
en esta variedad de queso (Delgado et al., 2010a), pero sí en queso de la Serena,
aunque en menor concentración (Carbonell et al., 2002). Este compuesto, que aporta al
queso aromas de azufre y pútrido, tiene un bajo umbral de percepción (Curioni &
Bosset, 2002). La gran acumulación de dimetildisulfuro durante el almacenamiento en
refrigeración fue una de las causas de la pérdida de calidad de sabor del queso control
durante este periodo.
Durante el almacenamiento en refrigeración la firmeza y la elasticidad descendieron

a lo largo del tiempo, al igual que en otro estudio sobre la Torta del Casar (Delgado et
al., 2010b), debido a la fuerte hidrólisis de la matriz de caseínas, especialmente αS1-
caseína, y al aumento de pH, según la teoría de Creamer & Olson (1982). La calidad de
sabor y de olor disminuyeron significativamente, mientras que la intensidad de sabor y
de olor, el sabor amargo y el olor a pútrido y rancio aumentaron durante el
almacenamiento en refrigeración.
El almacenamiento en condiciones de refrigeración no fue suficiente para frenar los

procesos bioquímicos responsables de la formación de los compuestos implicados en el
aroma y sabor de la Torta del Casar. Esto provocó el desequilibrio de algunos de estos
compuestos, produciéndose una pérdida de la calidad del sabor y del olor a partir de
los 120 días. Junto a esta pérdida de calidad, también se encontró que los niveles de
aminas biógenas presentes en el queso control a partir de los 120 días fueron excesivos,
pudiendo provocar efectos nocivos para la salud del consumidor.
257

Discusión general
Tabla 8. Tendencias de las características microbiológicas y químicas de la Torta del Casar tratada por alta
presión con respecto al queso control.
Parámetro 400-3S 600-3S 400-5S 600-5S
ES NS ↑ 120 a 240 d ↑ 120 a 240 d ↑ 120 a 240 d
pH ↓ 120 a 240 d ↓ 120 a 240 d ↓ 120 a 240 d ↓ 120 a 240 d
Aerobios totales ↓ hasta 240 d ↓↓ hasta 240 d ↓ hasta 240 d ↓↓ hasta 240 d
Bacterias lácticas ↓ hasta 240 d ↓↓ hasta 240 d ↓ hasta 240 d ↓↓ hasta 240 d
Lactobacilos ↓ hasta 240 d ↓↓ hasta 240 d ↓ hasta 240 d ↓↓ hasta 240 d
Micrococos ↓ 120 a 240 d ↓ 120 a 240 d ↓ hasta 240 d ↓↓ hasta 240 d
Gram negativos ↓ hasta 240 d ↓↓ hasta 240 d ↓↓ hasta 240 d ↓↓ hasta 240 d
Coliformes ↓ hasta 240 d ↓↓ hasta 240 d ↓↓ hasta 240 d ↓↓ hasta 240 d
Estafilococos coagulasa + ↓↓ hasta 240 d ↓↓ hasta 240 d ↓↓ hasta 240 d ↓↓ hasta 240 d
Enterococos ↓ hasta 240 d ↓↓ hasta 240 d ↓ hasta 240 d ↓↓ hasta 240 d
Actividad de cardosinas NS NS ↑ 240 d NS
Actividad peptidolítica NS ↓ 60 d NS ↓ 60 d
α-caseína NS ↑ 35 y 60 d NS ↑ 60 d
β-caseína NS ↑ 120 a 240 d NS ↑ 180 y 240 d
κ-caseína NS NS NS NS
p-κ-caseína NS ↑ 180 d NS ↑ 180 y 240 d
Péptidos hidrófilos ↑ 60 d, ↓ 240 d ↓ 240 d ↓ 60 y 240 d ↓ 240 d
Péptidos hidrófobos ↑ 180 d NS ↓ 60 y 120 d ↓ 120 d
Ratio hidrófobos/filos ↓ 60 d ↓ 60 d NS NS
Act. aminopeptidasa (Lys) ↓ 60 d ↓ 35, 60 y 120 d ↓ 35 y 60 d ↓ 35 y 60 d
Act. aminopeptidasa (Leu) NS ↓ 35, 60 y 120 d ↓ 35 y 60 d ↓ 35 y 60 d
Aminoácidos libres totales ↑ 35 y 60 d ↓ 240 d ↑ 180 d NS
Proteolisis global (OPA) NS ↓ 35 a 240 d NS ↓ 60 a 240 d
Actividad TDC ↓ 60 y 180 d ↓ 60 y 180 d ↓ 60 y 180 d ↓ 60 y 180 d
Tiramina NS ↓ 60 a 240 d NS ↓ 60 a 240 d
Putrescina ↓ 35 a 120 d ↓↓ 35 a 240 d ↓ 60 a 240 d ↓↓ 60 a 240 d
Triptamina ↓ 35 y 60 d ↓ 35, 60, 120 y 240 d ↓ 35, 60 y 240 d ↓↓ 35 a 240 d
Feniletilamina ↓ 35 a 120 d ↓ 35 a 240 d ↓ 35 a 240 d ↓ 35 a 240 d
Cadaverina ↓ 35 a 240 d ↓ 35 a 240 d ↓ 60 a 240 d ↓ 60 a 240 d
Histamina ↓ 35 y 60 d ↓↓ 35 a 240 d ↓ 60 d ↓↓ 60 a 240 d
Aminas totales ↓ 60 a 240 d ↓↓ 60 a 240 d ↓ 60 a 240 d ↓↓ 60 a 240 d
Actividad esterasa ↓ 35, 180 y 240 d ↓↓ 35 a 240 d ↓ 35 y 240 d ↓ 35 a 240 d
Etanoico + propanoico ↓↓ 35 a 240 d ↓↓ 35 a 240 d ↓ 120 y 240 d ↓ 120 y 240 d
AGL cadena ramificada ↓↓ 35 a 240 d ↓↓ 35 a 240 d ↓ 120 y 240 d ↓ 120 y 240 d
AGL cadena corta ↓ 35 a 240 d ↓ 35 a 240 d ↓ 120 a 240 d ↓ 120 a 240 d
AGL cadena media ↓ 120 a 240 d ↓ 120 a 240 d ↓ 120 a 240 d ↓ 120 a 240 d
AGL cadena larga ↓ 120 a 240 d ↓ 120 a 240 d ↓ 120 a 240 d ↓ 120 a 240 d
Ácidos volátiles ↓ 60 a 240 d ↓ 60 a 240 d NS NS
Alcoholes volátiles ↓ 60 y 120 d ↓ 60 a 240 d ↓ 120 d ↓ 120 y 240 d
Aldehídos volátiles ↓ 120 y 240 d ↓ 120 y 240 d ↓ 120 y 240 d ↓ 120 y 240 d
Cetonas volátiles ↓↓ 60 d NS ↓ 60 y 120 d ↓ 120 d
Ésteres volátiles ↓↓ 60 a 240 d ↓↓ 60 a 240 d ↓ 60 a 240 d ↓ 60 a 240 d
Éteres volátiles ↓ 120 d ↓ 120 d ↓ 120 d NS
Hidrocarburos volátiles NS NS NS NS
Bencénicos volátiles NS NS NS NS
Azufrados volátiles ↓↓ 120 y 240 d ↓↓ 120 y 240 d ↓↓ 120 y 240 d ↓↓ 120 y 240 d
continuación. NS; ausencia de efecto significativo (P > 0,05, test de Tukey) para todos los tiempos. OPA; o-ftalaldehído
(técnica espectrofotométrica), TDC; tirosina descarboxilasa.
258

Discusión general
Con el objetivo de reducir la pérdida de calidad y la acumulación de aminas

biógenas en la Torta del Casar se aplicaron tratamientos de 400 y 600 MPa a las 3
semanas (400-3S y 600-3S) y a las 5 semanas (400-5S y 600-5S) de maduración. Estos
tratamientos ocasionaron una serie de cambios significativos respecto al control en los
diferentes tiempos (Tablas 8 y 9). Por otro lado, comparando los quesos tratados a lo
largo del almacenamiento en refrigeración con el queso control en su momento óptimo
(60 días), se puede apreciar que en la mayoría de los casos se logró evitar o atenuar los
efectos negativos asociados a la sobremaduración (Tablas 10 y 11).
Tabla 9. Tendencias de las características de textura y sensoriales de la Torta del Casar tratada por alta
Parámetro 400-3S 600-3S 400-5S 600-5S
Fracturabilidad NS NS NS NS
Firmeza ↓↓ 35 y 180 d ↑ 120 d NS NS
Elasticidad ↓↓ 35 a 180 d ↑ 120 d NS NS
Calidad sabor ↑ 120 a 240 d ↑ 120 a 240 d ↑ 120 a 240 d ↑ 120 a 240 d
Intensidad sabor NS ↓ 60 a 240 d NS ↓ 180 y 240 d
Sabor umami NS ↓ 120 d NS NS
Sabor ácido NS NS NS NS
Sabor amargo NS ↓ 120 a 240 d ↓ 120 a 240 d ↓ 120 a 240 d
Sabor dulce NS NS NS ↑ 180 d
Sabor salado NS NS NS NS
Calidad de olor ↑ 240 d ↑ 240 d ↑ 120 a 240 d ↑ 240 d
Intensidad de olor ↓ 240 d ↓ 240 d ↓ 240 d ↓ 240 d
Olor pútrido ↓ 240 d ↓ 120 a 240 d ↓ 120 a 240 d ↓ 120 a 240 d
Olor rancio ↓ 120 a 240 d ↓ 120 a 240 d ↓ 120 a 240 d ↓ 120 a 240 d
La aplicación de tratamientos de altas presiones provocó una reducción de los

niveles de microorganismos, que en el caso de los tratamientos de 600 MPa fue
acompañado de la inactivación de las enzimas proteolíticas, provocando una reducción
de la hidrólisis de caseínas. Los niveles de αS-, β- y p-κ-caseínas en los quesos tratados
con 600 MPa se mantuvieron durante todo el periodo de almacenamiento en
refrigeración en niveles similares o superiores a los del queso control de 60 días. La
hidrólisis de caseínas, tanto en el queso control como en los tratados por alta presión,
es debida a la acción de la plasmina, que mantiene parte de su actividad incluso
después de ser tratada a 800 MPa (Malone et al., 2003), y a las cardosinas, que no
259

Discusión general
perdieron actividad por las altas presiones sino que, por el contrario, aumentó a día 240
con el tratamiento de 400 MPa aplicado a las 5 semanas. Sin embargo, en queso de la
Serena tratado a 300 y 400 MPa, sí que se redujo la hidrólisis de caseínas (Garde et al.,
2007).
Los péptidos hidrófobos se mantuvieron en los quesos tratados a las 3 semanas en

niveles similares a los del queso control a los 60 días durante el almacenamiento en
refrigeración, mientras que en los quesos tratados a las 5 semanas se mantuvieron en
niveles inferiores. Por el contrario, los niveles de péptidos hidrófilos fueron superiores a
los del queso control de 60 días durante el almacenamiento de todos los quesos
tratados, de igual forma que sucedió en queso de la Serena con los tratamientos
aplicados a día 2 (Garde et al., 2007). La reducción de la actividad aminopeptidasa
provocada por los tratamientos de altas presiones no consiguió que durante el
almacenamiento en refrigeración los niveles de aminoácidos libres de los quesos
tratados se mantuviesen por debajo de los niveles del queso control de 60 días, al igual
que ocurrió en queso de la Serena con los tratamientos aplicados a día 2 (Garde et al.,
2007). La inactivación de algunas de las enzimas responsables del catabolismo de los
aminoácidos fue posiblemente la causa de su acumulación en los quesos tratados. El
efecto de las altas presiones sobre los aminoácidos se vio igualmente reflejado en la
proteolisis global.
La reducción de la actividad de la tirosina descarboxilasa debida a las altas

presiones hizo que los niveles de tiramina durante el almacenamiento de los quesos
tratados con 600 MPa fuesen similares a los del queso control de 60 días. Todos los
tratamientos consiguieron mantener niveles de cadaverina hasta los 240 días similares a
los del queso control de 60 días. La triptamina y la β-feniletilamina se mantuvieron en
niveles similares o ligeramente superiores a los del queso control a día 60 durante el
almacenamiento de los quesos tratados con 600 MPa y de los tratados con 400 MPa a
las 5 semanas. Excepto el tratamiento de 400 MPa aplicado a las 3 semanas, las altas
presiones redujeron los niveles de putrescina, aunque únicamente los tratamientos de
600 MPa mantuvieron niveles inferiores a los del queso control de 60 días. Los
tratamientos de 600 MPa redujeron además los niveles de histamina, manteniendo
niveles inferiores a los del queso control de 60 días.
260

Discusión general
Tabla 10. Valores de las características microbiológicas y químicas en Torta del Casar control de 60 días,
quesos tratados por alta presión de 60-240 días y queso control de 120-240 días (sobremadurado).
400-3S 600-3S 400-5S 600-5S Control

Control
Parámetro 60 a 240 d 60 a 240 d 60 a 240 d 60 a 240 d 120 a 240 d
60 días
(rango) (rango) (rango) (rango) (rango)
ES (%) 51,2 51,6 - 54,9 51,7 - 56,3 53,0 - 57,0 52,4 - 57,3 52,2 - 54,7
pH 5,4 5,4 - 5,7 5,4 - 5,6 5,3 - 5,7 5,4 - 5,6 5,8 - 6,1
Aerobios totalesa 9,4 8,0 - 8,5 5,4 - 6,3 7,1 - 7,9 4,3 - 5,0 8,6 - 9,2
Bacterias lácticasa 9,2 8,0 - 8,1 5,0 - 5,7 6,8 - 7,7 3,7 - 4,0 8,4 - 8,9
a
Lactobacilos 8,3 7,7 - 8,0 3,3 - 4,4 6,6 - 6,9 2,3 - 3,1 8,0 - 8,4
Enterococosa 7,2 5,8 - 6,5 2,4 - 3,4 5,2 - 6,4 nd - 2,5 6,8 - 7,1
a
Micrococos 6,6 5,2 - 6,3 3,7 - 6,3 3,1 - 4,7 2,4 - 3,6 5,8 - 6,0
a
Gram negativos 6,9 nd - 3,2 nd nd nd 5,4 - 6,5
Coliformesa 4,4 nd - 1,0 nd nd nd 2,4 - 3,4
a
Estafilococos coagulasa + 2,6 nd nd nd nd nd - 2,2
α-caseína b 5,8 3,2 - 6,4 5,0 - 8,6 3,1 - 5,9 5,1 - 8,8 3,8 - 5,5
β-caseína b 81,9 39,3 - 80,9 84,8 - 103,2 40,2 - 87,5 81,2 - 102,4 41,8 - 58,1
b
p-κ-caseína 16,1 10,3 - 15,4 13,8 - 15,1 10,3 - 15,5 14,7 - 16,9 11,3 - 13,5
Péptidos hidrófilosc 13,2 17,2 - 33,2 15,4 - 27,7 12,8 - 32,8 14,4 - 31,0 24,1 - 42,0
c
Péptidos hidrófobos 24,7 23,9 - 26,4 24,2 - 24,7 20,0 - 22,7 19,9 - 24,4 22,9 - 26,7
Ratio hidrófobos/filos 1,9 0,8 - 1,6 0,9 - 1,6 0,6 - 1,7 0,8 - 1,6 0,6 - 1,1
Act. aminopeptidasa (Lys)d 38,1 5,0 - 25,7 2,8 - 7,2 4,3 - 9,3 3,8 - 14,1 4,7 - 11,3
Act. aminopeptidasa (Leu)d 80,7 5,5 - 67,5 4,5 - 19,1 4,4 - 41,3 5,8 - 35,2 6,4 - 31,0
b
Aminoácidos libres totales 4,8 7,4 - 25,4 5,0 - 16,5 6,5 - 26,0 5,5 - 17,5 13,6 - 23,3
Proteolisis global (OPA)e 0,8 0,9 - 3,4 0,7 - 2,0 0,9 - 3,1 0,8 - 2,0 1,7 - 3,5
f
TDC 0,7 0,03 - 0,2 0,05 - 0,1 0,1 - 0,7 0,00 - 0,06 3,0
b
Tiramina 0,4 0,4 - 1,0 0,2 - 0,4 0,3 - 0,8 0,3 - 0,5 0,6 - 1,1
Putrescina b 0,3 0,2 - 1,2 0,1 - 0,1 0,2 - 0,5 0,2 - 0,2 0,6 - 1,0
Triptamina b 0,07 0,01 - 0,35 0,00 - 0,13 0,03 - 0,13 0,03 - 0,09 0,20 - 0,38
b
Feniletilamina 0,07 0,02 - 0,25 0,00 - 0,12 0,04 - 0,12 0,03 - 0,12 0,18 - 0,29
Cadaverina b 0,16 0,05 - 0,09 0,04 - 0,13 0,09 - 0,13 0,09 - 0,15 0,49 - 0,69
Histamina b 0,14 0,10 - 0,41 0,00 - 0,02 0,09 - 0,38 0,04 - 0,08 0,20 - 0,30
b
Aminas totales 1,1 0,7 - 3,3 0,4 - 0,9 0,8 - 2,0 0,7 - 1,0 2,3 - 3,7
Actividad esterasag 9,6 2,1 - 8,5 0,8 - 1,7 2,9 - 7,6 1,1 - 5,8 5,3 - 5,6
b
Etanoico + propanoico 6,7 4,1 - 5,5 3,7 - 3,9 6,3 - 7,2 5,7 - 7,1 7,7 - 8,8
b
AGL cadena ramificada 0,8 0,3 - 0,4 0,3 - 0,3 0,6 - 0,8 0,6 - 0,8 0,8 - 1,2
AGL cadena corta b 1,9 0,6 - 1,1 0,7 - 2,3 1,0 - 2,0 1,0 - 2,6 3,4 - 8,7
AGL cadena media b 0,4 0,4 - 0,6 0,4 - 0,6 0,3 - 0,6 0,4 - 0,5 0,7 - 0,9
b
AGL cadena larga 1,4 1,4 - 2,1 1,2 - 1,9 1,3 - 1,9 1,4 - 1,7 2,6 - 3,0
Ácidos volátilesh 12275 5239 - 6680 5246 - 7201 7571 - 9878 8537 - 10962 9483 - 10444
Alcoholes volátilesh 9065 5191 - 14701 4866 - 7450 9071 - 14065 7879 - 10067 13490 - 15914
h
Aldehídos volátiles 10,7 9,3 - 12,6 10,8 - 13,3 9,7 - 13,3 9,3 - 12,3 15,8 - 17,1
Cetonas volátilesh 1429 152 - 1836 1411 - 2475 387 - 2117 1095 - 1625 1778 - 2147
Ésteres volátilesh 3651 1411 - 1533 862 - 1488 2753 - 3522 1502 - 2403 3984 - 5536
h
Éteres volátiles 14,4 11,3 - 18,1 11,9 - 16,2 16,1 - 24,3 17,4 - 24,6 20,1 - 23,7
Azufrados volátilesh 14,1 7,3 - 66,6 7,5 - 78,1 3,1 - 10,2 8,0 - 14,6 1158 - 6583
nd: no detectado, OPA: o-ftalaldehído, TDC: actividad tirosina descarboxilasa, : (log ufc/g), : (mg/g ES), : (UA/mg ES), d:(nmol de
a b c
p-nitroanilina/min g), e: (unidades de absorbancia), f: (mUA/g ES) g: (pmol α-naftol/min g), h: (abundancia relativa).
261

Discusión general
La reducción de la actividad esterasa provocada por los tratamientos de altas

presiones consiguió que los niveles de ácidos grasos libres de cadena corta,
mayoritariamente ácido butanoico, fueran similares o inferiores a los del queso control
de 60 días. Las altas presiones consiguieron atenuar la formación de ácidos grasos de
cadena media y larga, aunque sus niveles no fueron inferiores a los del queso control de
60 días. La reducción de los niveles de ácidos grasos libres por las altas presiones se ha
observado previamente en otras variedades de queso (Saldo et al., 2003, Ávila et al.,
2007, Voigt et al., 2010, Delgado et al., 2012). En los quesos tratados, los niveles de
ácido etanoico, propanoico y ácidos de cadena ramificada fueron inferiores o similares a
los del queso control de 60 días. Debido a la reducción de los niveles de ácidos grasos
de cadena corta, ácidos ramificados y ácido etanoico y propanoico, se consiguió reducir
la aparición de los sabores no deseados que estos compuestos aportan al queso
cuando se encuentran en grandes cantidades (Curioni & Bosset, 2002).
Tabla 11. Valores de las características de textura y sensoriales en Torta del Casar control de 60 días, quesos
tratados por alta presión de 60-240 días y queso control de 120-240 días (sobremadurado).
400-3S 600-3S 400-5S 600-5S Control
Control
60 días
Firmeza (J) 0,014 0,002 - 0,006 0,017 - 0,048 0,009 - 0,020 0,008 - 0,028 0,009 - 0,021
2
Elasticidad (N/mm ) 0,021 0,001 - 0,009 0,016 - 0,044 0,005 - 0,021 0,010 - 0,029 0,011 - 0,025
Calidad de sabor 6,4 5,0 - 7,0 5,1 - 6,0 5,7 - 6,7 5,8 - 6,7 0,9 - 3,7
Intensidad de sabor 6,4 6,2 - 7,2 5,1 - 6,2 6,2 - 7,2 5,8 - 6,3 7,0 - 7,3
Sabor umami 3,3 3,0 - 4,6 2,5 - 3,8 3,3 - 4,1 3,1 - 3,9 3,9 - 4,4
Sabor amargo 2,5 2,6 - 3,9 1,5 - 3,0 2,0 - 3,9 2,0 - 2,6 3,9 - 4,6
Sabor dulce 0,4 0,3 - 0,6 0,3 - 0,8 0,2 - 0,5 0,4 - 0,5 0,1 - 0,4
Calidad de olor 6,4 4,8 - 6,2 4,9 - 6,0 5,3 - 6,5 5,6 - 6,2 1,9 - 4,6
Intensidad de olor 5,9 5,7 - 6,4 5,6 - 6,1 5,9 - 6,7 5,7 - 6,5 6,7 - 7,9
Olor pútrido 1,0 1,5 - 2,9 0,9 - 2,4 1,0 - 2,0 1,2 - 2,3 2,6 - 6,8
Olor rancio 0,5 0,8 – 1,6 0,5 - 1,8 0,4 - 1,5 0,6 - 1,4 1,9 - 3,3
Los tratamientos por altas presiones afectaron de forma diferente a los distintos
grupos de compuestos volátiles. Las altas presiones provocaron que los ácidos volátiles
se mantuvieran durante el almacenamiento en refrigeración en niveles inferiores a los
del queso control de 60 días. Los tratamientos evitaron el drástico aumento de
compuestos azufrados registrado en el queso control, manteniendo los niveles de los
quesos tratados en valores similares a los del queso control de 60 días. Las altas
262

Discusión general
presiones atenuaron la formación de ésteres, manteniendo en niveles inferiores a los

del queso control de 60 días todos los quesos tratados hasta el final del
almacenamiento en refrigeración. También consiguieron mantener los niveles de
aldehídos en valores similares o ligeramente superiores a los del queso control de 60
días. Los tratamientos aplicados a las 3 semanas consiguieron controlar el incremento
de éteres, manteniendo niveles similares a los del queso control de 60 días, en especial
con el tratamiento de 600 MPa, mientras que los alcoholes se mantuvieron en niveles
inferiores a los del queso control de 60 días en los quesos tratados con 600 MPa. Los
quesos tratados con 600 MPa a las 5 semanas mantuvieron niveles de cetonas similares
o ligeramente superiores a los del queso control de 60 días hasta el día 240. Sin
embargo, los tratamientos de altas presiones no consiguieron mantener en valores
similares a los del control de 60 días los niveles de hidrocarburos y compuestos
bencénicos.
En lo que respecta a las características de textura, las altas presiones no lograron

estabilizar ni mantener en valores similares a los del queso control de 60 días los
parámetros de firmeza y elasticidad. Las altas presiones consiguieron evitar el aumento
de la intensidad de olor, y de olor pútrido y rancio, así como el desarrollo de sabor
amargo. Únicamente los tratamientos de 600 MPa consiguieron reducir el aumento de
la intensidad de sabor, gracias a sus efectos sobre los parámetros microbiológicos y
químicos de la Torta del Casar. De modo que se logró controlar la considerable pérdida
de calidad de sabor y olor, manteniendo a los quesos tratados con valoraciones de sus
características sensoriales similares a las del queso control en su momento óptimo.
Por otra parte, el análisis de componentes principales (PCA) de ácidos grasos y de

grupos de compuestos volátiles (recogido en el capítulo 6) indicó que la Torta del Casar
tratada a las 5 semanas de maduración mantuvo un perfil de ácidos grasos y de
compuestos volátiles muy similar al del queso control de 60 días. Por consiguiente, se
puede concluir que la Torta del Casar tratada por altas presiones a las 5 semanas,
especialmente con 600 MPa, mantiene en gran medida las características de esta
variedad de queso en su momento óptimo de consumo (60 días) hasta los 8 meses de
almacenamiento en refrigeración, evitando además la formación y acumulación de
aminas biógenas.
263

Discusión general
 Queso Brie
El queso Brie se caracteriza por la presencia de P. camemberti en su superficie.

Tiene un periodo de maduración de unos 21 días y sufre una rápida evolución del sabor
así como un aumento de intensidad de sabor durante su vida útil. En esta variedad de
queso se estudió la evolución de las variables microbiológicas, químicas, de textura,
color y sensoriales durante la maduración y el almacenamiento en refrigeración hasta el
día 120. Mediante los análisis microbiológicos, se observó una disminución gradual de
los niveles de bacterias lácticas y bacterias mesófilas totales durante la maduración y el
almacenamiento en refrigeración, desde 8,99 y 8,75 log ufc/g respectivamente a día 1
hasta 8,43 y 8,46 log ufc/g a día 21 y 7,88 y 7,90 log ufc/g a día 120. Los niveles de P.
camemberti se mantuvieron prácticamente estables a lo largo de todo el periodo de
estudio, con niveles máximos de 6,95 log ufc/g a los 60 días, mientras que las levaduras
no se detectaron hasta el día 90, encontrándose entonces en niveles de 6,52 log ufc/g.
El pH del queso sufrió un rápido aumento desde 5,55 a día 21 hasta 7,55 a día 60 y
posteriormente aumentó hasta 7,87 a día 120. El extracto seco del queso se redujo de
45,48 a día 21 a 42,27 a día 60 y se mantuvo en estos valores hasta el día 120.
La presencia de plasmina, que mantiene parte de su actividad tras los tratamientos

térmicos de pasteurización (Ismail & Nielsen, 2010), junto a la presencia de cuajo
residual y de P. camemberti en la superficie, provocaron descensos de αS-, β- y κ-
caseínas del 41, 37 y 36 %, respectivamente, entre los días 1 y 14 de maduración.
Posteriormente sus niveles se fueron reduciendo gradualmente hasta el día 30,
momento en el que el pH alcanzó un valor de 6,3 favorable para la actividad de la
plasmina residual y de la metaloproteinasa de P. camemberti (Lenoir & Auberger, 1977),
lo que ocasionó una mayor hidrólisis de αS-, β- y κ-caseínas. Por otro lado, la p-κ-
caseína se mantuvo en los mismos niveles entre los días 14 y 90, descendiendo
posteriormente. Los niveles de péptidos aumentaron durante el almacenamiento en
refrigeración del queso Brie. Los péptidos hidrófilos aumentaron gradualmente hasta
alcanzar 47,35 UA/mg ES a día 120, mientras que los péptidos hidrófobos sufrieron un
drástico aumento a partir del día 30, alcanzando una concentración de 72,41 UA/mg ES
a día 90, coincidiendo con el aumento de pH que provocó una mayor hidrólisis de
caseínas. El ratio de péptidos hidrófobos/hidrófilos aumentó también drásticamente
264

Discusión general
entre los días 30 y 90. La relación encontrada entre este ratio y el sabor amargo en
queso (Gomez et al., 1997) se vio reflejada en la valoración sensorial del amargor, que
alcanzó a día 90 la mayor puntuación, de 6,41 sobre 10. La actividad aminopeptidasa
aumentó hasta el día 60, alcanzando valores de 7,36 y 8,26 nmoles de p-
nitroanilina/min·g con Leu-p-NA y Lys-p-NA como sustratos, respectivamente. La alta
actividad de las exopeptidasas de los mohos del género Penicillium hizo que los niveles
de aminoácidos liberados durante la proteolisis aumentasen en función del tiempo,
desde valores de 1,94 mg/g ES a día 1 hasta 3,88 y 78,24 mg/g ES al final de la
maduración (día 21) y del periodo de almacenamiento en refrigeración (día 120)
respectivamente, siendo estos niveles más bajos a los encontrados en quesos
madurados con mohos en su interior (Madkor et al., 1987b, Prieto et al., 2000). La
proteolisis global se mantuvo en niveles similares hasta el día 30, y a partir del día 60
aumentó hasta el final del almacenamiento en refrigeración.
La actividad esterasa aumentó durante la maduración y el almacenamiento en

refrigeración desde 0,83 pmoles de α-naftol/min·g a día 1 hasta 4,86 a día 21 y 18,14 a
día 120. Esta variedad de queso se caracteriza por una elevada lipolisis, debida a la
presencia de P. camemberti, aunque es menor que en los quesos azules al encontrarse
el moho únicamente en la superficie (Spinnler & Gripon, 2004). Los niveles de los ácidos
grasos libres aumentaron durante la maduración y el almacenamiento en refrigeración,
llegando a alcanzar concentraciones a los 120 días de 3,48, 5,03 y 21,58 mg/g ES de
ácidos grasos de cadena corta, media y larga, respectivamente, siendo estos valores
superiores a los encontrados en otras variedades de queso (Ávila et al., 2007, Juan et al.,
2007, Voigt et al., 2012) e inferiores a los encontrados en queso azul (Prieto et al., 2000).
Los ácidos grasos libres de cadena larga fueron el grupo mayoritario, con el oleico
como el más abundante, el cual alcanzó niveles de 0,45 mg/g ES al final de la
maduración, y de 13,10 mg/g ES a día 120. El ácido etanoico aumentó durante el
almacenamiento en refrigeración, hasta 0,40 mg/g ES a día 120. También se produjo un
aumento de los ácidos ramificados, que llegaron a alcanzar 1,03 mg/g ES al final del
El grupo más abundante de compuestos volátiles a día 30 fue el de los ácidos

volátiles, mientras que a día 60 y 120 fue el de las cetonas. El compuesto mayoritario
265

Discusión general
dentro las cetonas fue la acetoína a día 30, la acetona a día 60 y la 2-pentanona a día
120. Las cetonas, son compuestos característicos de esta variedad de queso y proceden
principalmente de la β-oxidación parcial de los ácidos grasos libres, aportando al queso
aromas frutales y florales (Molimard & Spinnler, 1996, Curioni & Bosset, 2002, Spinnler
& Gripon, 2004). Los distintos grupos de compuestos volátiles evolucionaron de distinta
manera durante el almacenamiento en refrigeración. Los niveles de ácidos e
hidrocarburos descendieron, los ésteres y éteres se mantuvieron en niveles estables y
los alcoholes, aldehídos, compuestos bencénicos, nitrogenados y azufrados aumentaron
a lo largo del almacenamiento en refrigeración.
La migración de iones calcio (Ca2+) del interior a la superficie, característica de esta

variedad de quesos, provoca la desestabilización de la matriz de caseínas (Spinnler &
Gripon, 2004), lo que unido a la disminución de los niveles de caseínas y del extracto
seco, hizo que los tres parámetros de textura estudiados (fracturabilidad, firmeza y
elasticidad) disminuyesen durante la maduración y alcanzasen valores muy bajos
durante el posterior almacenamiento en refrigeración. El estudio del color mostró un
ligero aumento del parámetro b* (tendencia al amarillo) tanto en la superficie como en
el interior, mientras que el parámetro a* (tendencia al rojo) aumentó de forma drástica
entre los días 90 y 120, y la L* (luminosidad) disminuyó durante el almacenamiento en
refrigeración. La calidad de sabor y de olor disminuyeron significativamente durante el
almacenamiento en refrigeración. La pérdida de calidad de sabor fue acompañada de
un fuerte aumento de la intensidad de sabor y olor, y del sabor amargo y umami.
El almacenamiento en condiciones de refrigeración no impidió la sobremaduración

del queso Brie, provocando un desequilibrio de los compuestos responsables del sabor
y el aroma que se vio reflejado en la pérdida de calidad de olor y sabor ya a día 60. Los
parámetros de sabor amargo y umami aumentaron en función del tiempo reduciendo la
calidad de sabor del queso, al tiempo que aumentó la intensidad de sabor y olor.
Con el objetivo de reducir la pérdida de calidad de sabor y olor del queso Brie se
aplicaron tratamientos de 400 y 600 MPa a las 2 semanas (400-2S y 600-2S) y a las 3
semanas (400-3S y 600-3S) de maduración. Estos tratamientos dieron lugar a una serie
de cambios significativos respecto al control (Tablas 12 y 13). Por otro lado, al comparar
los quesos tratados a lo largo del almacenamiento en refrigeración con el queso control
266

Discusión general
en su momento óptimo (30 días), se observó que en la mayoría de los casos fue posible
evitar o atenuar los efectos negativos de la sobremaduración (Tablas 14 y 15).
Tabla 12. Tendencias de las características microbiológicas y químicas del queso Brie tratado por alta presión
con respecto al queso control.
Parámetro 400-2S 600-2S 400-3S 600-3S
ES ↑ 60 a 120 d ↑ 60 a 120 d ↑ 60 a 120 d ↑ 90 y 120 d
pH ↓↓ 30 a 120 d ↓↓ 30 a 120 d ↓↓ 30 a 120 d ↓↓ 30 a 120 d
Aerobios totales ↓ hasta 120 d ↓↓ hasta 120 d ↓ hasta 120 d ↓↓ hasta 120 d
Bacterias lácticas ↓ hasta 120 d ↓↓ hasta 120 d ↓ hasta 120 d ↓↓ hasta 120 d
P. camemberti ↓ hasta 120 d ↓↓ hasta 120 d ↓ hasta 120 d ↓↓ hasta 120 d
Levaduras ↑ 60 d, ↓ 90 y 120 d ↓↓ 90 y 120 d ↓ 90 d ↓↓ 90 y 120 d
α-caseína NS ↑↑ 90 y 120 d NS ↑ 90 d
β-caseína ↑ 120 d ↑↑ 60 a 120 d ↑ 60 d ↑↑ 60 a 120 d
κ-caseína ↑ 120 d NS NS NS
p-κ-caseína NS NS NS NS
Péptidos hidrófilos ↑ 14 a 60 d, ↓ 120 d ↑ 14 a 60 d ↑ 60 d, ↓ 120 d ↑ 60 d
Péptidos hidrófobos ↓↓ 60 a 120 d ↓↓ 60 a 120 d ↓↓ 60 a 120 d ↓↓ 60 a 120 d
↓ 14 a 30 d, ↓ 14 a 30 d, ↓ 21 y 30 d, ↓ 21 y 30 d,
Ratio hidrófobos/filos
↓↓ 60 a 120 d ↓↓ 60 a 120 d ↓↓ 60 a 120 d ↓↓ 60 a 120 d
Act. aminopeptidasa (Lys) ↓ hasta 120 d ↓ hasta 120 d ↓ 120 d ↓ 60 a 120 d
Act. aminopeptidasa (Leu) ↓ 21 a 120 d ↓ 21 a 120 d ↓ 120 d ↓ 30 a 120 d
Aminoácidos libres totales ↑↑ 21 a 120 d ↑ 21 a 90 d ↑↑ 21 a 120 d ↑ 30 a 120 d
Proteolisis global (OPA) ↑ 21 a 90 d ↑ 21 a 60 d ↑↑ 30 a 120 d ↑ 30 a 90 d
Actividad esterasa ↑ 60 a 120 d ↑ 90 a 120 d ↑↑ 21 a 120 d NS
↑ hasta 30 d, ↓ 90 y
Etanoico + propanoico ↑↑ hasta 120 d ↑↑ hasta 120 d ↑ 30 d, ↓ 90 y 120 d
120 d
AGL cadena ramificada ↓↓ 60 a 120 d ↓↓ 60 a 120 d ↓↓ 60 a 120 d ↓↓ 90 y 120 d
AGL cadena corta ↑ 21 d, ↓ 30 a 120 d ↑ 21 d, ↓ 30 a 120 d ↑ 21 d, ↓ 60 a 120 d ↑ 21 d, ↓ 30 a 120 d
AGL cadena media ↓ 30, 90 y 120 d ↓ 30 a 120 d ↑ 21 d, ↓ 90 y 120 d ↓ 30, 90 y 120 d
AGL cadena larga ↓↓ 30 a 120 d ↓↓ 30 a 120 d ↑ 21 a 60 d, ↓ 120 d ↑ 21 d, ↓ 90 y 120 d
Ácidos volátiles ↑ 30 a 120 d ↑ 30 a 120 d ↑ 60 y 120 d ↑ 60 y 120 d
Alcoholes volátiles ↑↑ 30 a 120 d ↑ 30 a 120 d ↑↑ 30 a 120 d ↑ 30 a 120 d
Aldehídos volátiles ↑ 30 y 60 d ↑ 30 y 60 d ↑ 60 d ↑↑ 60 y 120 d
Cetonas volátiles ↑ 30 d, ↓↓ 120 d ↑ 60 d, ↓↓ 120 d ↓ 30 d, ↓↓ 120 d ↑ 60 d, ↓↓ 120 d
Ésteres volátiles ↑ 30 a 120 d ↑↑ 30 a 120 d ↑ 60 a 120 d ↑ 30 a 120 d
Éteres volátiles ↑↑ 30 a 120 d ↑↑ 30 a 120 d ↑↑ 30 a 120 d ↑↑ 30 a 120 d
Hidrocarburos volátiles NS ↑ 60 d ↑ 60 d NS
Bencénicos volátiles ↓ 120 d ↓ 120 d ↓ 120 d ↓ 120 d
Azufrados volátiles ↓↓ 120 d ↓↓ 120 d ↓↓ 120 d ↓↓ 120 d
Nitrogenados volátiles ↓↓ 120 d ↓↓ 120 d ↓↓ 120 d ↓↓ 120 d
(técnica espectrofotométrica).
267

Discusión general
La reducción de los niveles de microorganismos, especialmente de P. camemberti,

por las altas presiones junto a la inactivación de enzimas debidas a los tratamientos de
600 MPa, consiguieron atenuar la fuerte hidrólisis de αS-, β- y κ-caseínas característica
del queso control, aunque no lograron mantener niveles similares a los del queso
control de 30 días durante todo el periodo de almacenamiento en refrigeración. Las
altas presiones provocaron un aumento de los péptidos hidrófilos, que alcanzaron
niveles superiores a los del queso control de 30 días durante el almacenamiento en
refrigeración, mientras que mantuvieron los péptidos hidrófobos a niveles similares a
los del queso control de 30 días, de forma que el ratio de péptidos
hidrófobos/hidrófilos de los quesos tratados fue menor que el del queso control de 30
días, evitando el aumento de sabor amargo que se dio en el control. La reducción de la
actividad aminopeptidasa provocada por las altas presiones no se vio reflejada en una
disminución de los niveles de aminoácidos libres, de manera similar a lo ocurrido en
queso Hispánico (Ávila et al., 2006). La inactivación de las enzimas responsables del
catabolismo de aminoácidos posiblemente hizo que éstos se acumulasen en mayor
cantidad en los quesos tratados, al igual que sucedió con la proteolisis global.
La aplicación de altas presiones provocó un aumento de la actividad esterasa,

especialmente en los quesos tratados con 400 MPa. Este aumento no se vio reflejado en
la liberación de ácidos grasos de cadena corta, ya que los quesos tratados mantuvieron
niveles similares o ligeramente superiores a los del queso control de 30 días durante
todo el periodo de almacenamiento en refrigeración. Los quesos tratados a las 2
semanas mostraron niveles de ácidos grasos libres de cadena media y larga inferiores a
los del queso control de 30 días, mientras que los niveles de los ácidos grasos de
cadena larga de los quesos tratados a las 3 semanas fueron superiores a día 60 a los del
queso control de 30 días, aunque posteriormente disminuyeron hasta niveles inferiores
a los del control de 30 días. Un patrón similar de ácidos grasos libres se dio en queso de
leche de oveja tratado por alta presión (Juan et al., 2007). Los quesos tratados con 600
MPa mantuvieron niveles de ácido etanoico similares a los del queso control de 30 días
durante todo el almacenamiento en refrigeración, mientras que los tratados con 400
MPa mostraron niveles superiores. Con las altas presiones se logró evitar el drástico
268

Discusión general
incremento de ácidos grasos de cadena ramificada que se dio en el control a partir del
día 60.
Tabla 13. Tendencias de las características de textura, color y sensoriales del queso Brie tratado por alta
Parámetro 400-2S 600-2S 400-3S 600-3S
Fracturabilidad ↑↑ 30 a 120 d ↑↑ 30 a 120 d ↓ 21 d ↓ 21 d
Firmeza ↓ 14 d, ↑↑ 30 a 120 d ↓ 14 d, ↑↑ 30 a 120 d ↓ 21 d, ↑ 60 a 120 d ↓ 21 d, ↑ 60 a 120 d
Elasticidad ↓ 14 d, ↑↑ 30 a 120 d ↓ 14 d, ↑↑ 30 a 120 d ↓ 21 d, ↑ 60 a 120 d ↓ 21 d, ↑ 60 a 120 d
L* interior ↑ 90 y 120 d ↑ 90 y 120 d ↑ 90 y 120 d ↑ 90 y 120 d
a* interior ↓ 90 y 120 d ↓↓ 60 a 120 d ↓ 120 d ↓↓ 60 a 120 d
b* interior ↓ 120 d ↓ 90 y 120 d ↓ 120 d ↓ 90 y 120 d
Calidad sabor ↑↑ 60 a 120 d ↑↑ 60 a 120 d ↑↑ 60 a 120 d ↑↑ 60 a 120 d
↑ 30 y 60 d, ↓ 90 y
Intensidad sabor ↑ 30 d, ↓ 90 y 120 d ↓ 90 y 120 d ↓ 90 y 120 d
120 d
Sabor umami NS NS NS NS
Sabor amargo ↓ 60 a 120 d ↓ 60 a 120 d ↓ 60 a 120 d ↓ 60 a 120 d
Sabor dulce NS NS NS NS
Calidad de olor ↑↑ 60 a 120 d ↑↑ 60 a 120 d ↑↑ 60 a 120 d ↑↑ 60 a 120 d
Intensidad de olor ↓ 120 d ↓ 120 d ↓ 120 d ↓ 120 d
Los tratamientos por altas presiones afectaron de diferente forma a los distintos
grupos de compuestos volátiles. Las altas presiones evitaron el fuerte incremento de
compuestos volátiles nitrogenados, azufrados y bencénicos que se dio en el queso
control, manteniendo estos compuestos en niveles similares a los del queso control de
30 días hasta el final del almacenamiento en refrigeración. Se controló así la aparición
de sabores no deseados, tales como los aportados por el dimetildisulfuro cuando se
encuentra en concentraciones elevadas. Igualmente sucedió con los niveles de cetonas
volátiles en los quesos tratados, que se mantuvieron por debajo de los del queso
control de 30 días, excepto en los quesos tratados con 600 MPa a las 3 semanas, que a
día 60 tuvieron niveles superiores, aunque a día 120 descendieron por debajo de los
niveles del queso control de 30 días. Por otro lado los niveles de ácidos volátiles,
alcoholes, aldehídos, ésteres y éteres de los quesos tratados fueron superiores durante
todo el almacenamiento en refrigeración a los del queso control de 30 días.
269

Discusión general
Tabla 14. Valores de las características microbiológicas y químicas en queso Brie control de 30 días, quesos
tratados por alta presión de 30-120 días y queso control de 60-120 días (sobremadurado).
400-2S 600-2S 400-3S 600-3S Control
Control
30 días
ES (%) 45,3 46,8 - 50,5 47,8 - 49,5 46,0 - 49,4 45,7 - 48,4 42,1 - 42,4
pH 6,3 5,3 - 5,5 5,4 - 5,5 5,5 - 5,7 5,7 - 5,8 7,6 - 7,9
Aerobios totalesa 8,3 6,3 - 6,6 3,1 - 4,8 6,8 - 7,7 4,6 - 6,5 7,9 - 8,3
Bacterias lácticasa 8,3 6,4 - 6,7 2,6 - 4,5 6,9 - 7,7 5,0 - 5,4 7,9 - 8,3
a
Lactobacilos 8,0 6,0 - 6,3 2,3 - 4,0 6,3 - 7,2 4,4 - 5,0 7,3 - 8,1
P. camembertia 6,2 nd - 4,5 nd nd - 3,0 nd 6,0 - 7,0
Levadurasa nd nd - 2,6 nd nd - 2,3 nd nd - 6,9
b
α-caseína 87,2 12,1 - 87,0 25,4 - 98,5 5,2 - 85,2 17,2 - 84,7 1,7 - 68,3
β-caseínab 101,6 36,4 - 104,1 54,5 - 108,2 19,5 - 95,4 33,7 - 99,7 5,1 - 62,4
κ-caseínab 28,3 9,6 - 27,9 13,3 - 28,4 2,9 - 25,8 3,4 - 25,7 1,9 - 21,4
c
Péptidos hidrófilos 15,7 27,8 - 47,4 22,6 - 48,7 17,6 - 44,5 17,8 - 42,5 24,0 - 47,4
Péptidos hidrófobosc 6,3 5,8 - 7,8 3,7 - 7,6 5,0 - 6,1 4,6 - 5,9 30,3 - 72,4
Ratio hidrófobos/filos 0,4 0,16 - 0,22 0,11 - 0,21 0,12 - 0,26 0,14 - 0,26 0,8 - 1,7
d
Act. aminopeptidasa (Lys) 5,1 1,8 - 4,1 1,3 - 3,8 1,7 - 6,4 1,5 - 4,2 5,7 - 7,9
Act. aminopeptidasa (Leu)d 4,2 1,6 - 3,6 1,1 - 3,5 1,5 - 5,4 1,7 - 3,9 5,1 - 7,9
Aminoácidos libres totalesb 11,0 27,6 - 101,8 17,0 - 78,0 18,3 - 123,4 18,7 - 94,6 28,6 - 78,2
e
Proteolisis global (OPA) 0,6 1,8 - 7,5 1,3 - 5,8 1,4 - 9,6 1,0 - 7,2 1,7 - 5,7
Actividad esterasaf 8,8 5,0 - 33,3 3,6 - 27,6 12,6 - 44,1 4,7 - 22,1 9,0 - 18,1
Etanoico + propanoicob 0,1 0,5 - 0,7 0,2 - 0,3 0,3 - 0,8 0,1 - 0,2 0,2 - 0,5
b
AGL cadena ramificada 0,03 0,01 - 0,07 0,01 - 0,18 0,01 - 0,06 0,01 - 0,22 0,16 - 1,00
AGL cadena cortab 1,1 0,4 - 1,5 0,4 - 1,1 1,2 - 1,6 0,5 - 1,5 2,1 - 3,5
b
AGL cadena media 1,5 0,3 - 1,7 0,2 - 1,1 0,6 - 2,4 0,4 - 2,5 1,8 - 5,0
b
AGL cadena larga 9,8 1,1 - 6,5 1,4 - 5,0 8,9 - 22,1 5,3 - 13,7 11,2 - 21,6
Ácidos volátilesg 796 1688 - 3294 1272 - 2773 957- 3723 833 - 3206 70 - 85
g
Alcoholes volátiles 182 656 - 1120 715 - 809 745 - 1252 483 - 553 310 - 358
g
Aldehídos volátiles 5,8 17,0 - 43,7 11,1 - 35,0 7,5 - 34,6 6,7 - 73,7 8,7 - 39,1
Cetonas volátilesg 725 249 - 1458 561 - 688 168 - 344 533 - 1252 322 - 2674
Ésteres volátilesg 14,4 112,4 - 172,7 234,3 - 302,7 67,1 - 128,4 80,2 - 176,1 12,5 - 14,3
g
Éteres volátiles 31,7 138,9 - 188,5 132,4 - 183,3 91,6 - 161,6 89,6 - 172,3 24,8 - 33,6
Bencénicos volátilesg 18,7 16,0 - 16,7 13,8 - 16,9 15,8 - 18,0 16,0 - 21,2 16,6 - 31,6
g
Hidrocarburos volátiles 24,6 24,8 - 27,6 25,0 - 28,8 27,1 - 28,2 24,7 - 26,3 19,8 - 21,6
g
Azufrados volátiles 8,2 7,3 - 9,1 8,2 - 9,1 7,8 - 8,7 7,3 - 8,1 9,0 - 232,9
Nitrogenados volátilesg 3,2 3,5 - 5,7 2,8 - 3,0 3,0 - 5,8 2,4 - 7,4 3,9 - 422,1
a b c d e
nd: no detectado, OPA: o-ftalaldehído, : (log ufc/g), : (mg/g ES), : (UA/mg ES), :(nmol de p-nitroanilina/min g), : (unidades de
absorbancia), f: (pmol α-naftol/min g), g: (abundancia relativa).
Por lo que respecta a la textura, la mayor concentración de caseínas, junto a los

niveles más altos de extracto seco en los quesos tratados y su menor pH (Sousa &
McSweeney, 2001) hicieron que en todos los quesos tratados se evitase la pérdida de
elasticidad y firmeza que se dio en el queso control, y además en los tratados a las 2
270

Discusión general
semanas se logró eviar la pérdida de fracturabilidad. Respecto al color, todos los

tratamientos evitaron la pérdida de luminosidad del interior que se dio en el queso
control, así como el incremento de los parámetros a* y b* del interior del queso. La
pérdida de la capa de moho superficial provocada por las altas presiones hizo que los
parámetros de color de la superficie fuesen significativamente diferentes a los del queso
control. Las altas presiones atenuaron el incremento de intensidad de sabor y olor, así
como del sabor amargo, aunque no se mantuvieron en niveles similares a los del
control de 30 días durante todo el periodo de almacenamiento en refrigeración. Las
altas presiones evitaron la drástica disminución de la calidad de sabor y olor,
consiguiendo mantener durante el almacenamiento en refrigeración valores similares o
ligeramente inferiores a los del control de 30 días.
Tabla 15. Valores de las características de textura, color y sensoriales en queso Brie control de 30 días,
quesos tratados por alta presión de 30-120 días y queso control de 60-120 días (sobremadurado).
400-2S 600-2S 400-3S 600-3S Control
Control
30 días
Fracturabilidad (N) 3,4 2,4 - 6,3 1,0 - 8,4 0,0 - 0,6 0,0 - 0,3 0,0 - 0,0
Firmeza (J) 0,027 0,034 - 0,059 0,031 - 0,057 0,010 - 0,034 0,011 - 0,033 0,001 - 0,002
2
Elasticidad (N/mm ) 0,052 0,074 - 0,103 0,043 - 0,095 0,015 - 0,044 0,020 - 0,054 0,001 - 0,002
L* interior 86,2 84,4 - 85,4 83,1 - 84,6 82,1 - 85,5 82,0 - 85,1 75,9 - 83,9
a* interior 1,5 1,2 - 1,3 0,4 - 1,4 1,5 - 1,7 0,6 - 1,4 1,2 - 2,7
b* interior 19,3 19,2 - 21,5 19,6 - 20,2 18,8 - 21,9 19,0 - 21,1 20,2 - 22,6
Calidad de sabor 6,0 4,9 - 6,0 4,6 - 5,9 5,1 - 6,0 3,8 - 6,0 0,2 - 3,4
Intensidad de sabor 5,0 6,0 - 6,8 5,9 - 6,9 5,5 - 7,0 5,0 - 7,2 6,3 - 8,6
Sabor amargo 1,4 1,3 - 4,3 1,3 - 4,2 1,4 - 3,8 1,1 - 4,4 4,3 - 6,4
Calidad de olor 6,6 4,3 - 6,7 4,5 - 6,5 4,9 - 6,9 4,6 - 6,5 1,4 - 5,9
Intensidad de olor 5,1 5,4 - 6,8 5,4 - 6,4 5,2 - 6,6 5,3 - 6,7 6,4 - 7,9
El análisis de componentes principales de compuestos volátiles individuales

(recogido en el capítulo 8) indicó que los quesos tratados a las 2 semanas mantuvieron
un perfil de compuestos volátiles similar a los del control de 30 días durante todo el
almacenamiento en refrigeración. En base a todos estos resultados se puede concluir
que la presurización aplicada a las 2 semanas permite obtener quesos que mantienen
en gran medida las características del queso Brie en su momento óptimo de consumo
(30 días) hasta los 4 meses de almacenamiento en refrigeración. Sin embargo, la
271

Discusión general
pérdida de la capa superficial de moho causada por las altas presiones representaría un
serio impedimento a la hora de comercializar esta variedad de queso.
 Queso Arzúa-Ulloa
El queso Arzúa-Ulloa es habitualmente un queso tierno de leche cruda o

pasteurizada de vaca, que también se puede comercializar como queso curado. En
queso de esta variedad, elaborado con leche cruda, se estudió la evolución de las
variables microbiológicas, químicas, de textura, color y sensoriales durante la
maduración y el almacenamiento en refrigeración hasta el día 240. Mediante los análisis
microbiológicos, se observó una reducción de los niveles de bacterias mesófilas totales
desde 9,2 log ufc/g a día 1 hasta 8,0 log ufc/g a día 21, manteniéndose posteriormente
en niveles estables hasta el final del periodo de refrigeración. Los niveles de bacterias
lácticas, igualmente descendieron desde 8,8 log ufc/g a día 1 hasta 8,0 log ufc/g a día
14 y posteriormente se mantuvieron en los mismos niveles hasta el día 240. Los niveles
de lactobacilos aumentaron desde 4,1 log ufc/g a día 1 hasta 8,0 log ufc/g a día 60,
manteniéndose en los mismos niveles hasta el día 240. Los niveles de enterococos
descendieron durante la maduración y el almacenamiento en refrigeración desde 6,31
log ufc/ a día 1 hasta 4,71 log ufc/g a día 240. Los niveles de Micrococcaceae,
coliformes y bacterias gram negativas descendieron desde 6,92, 7,50 y 7,52 log ufc/g
respectivamente a día 1, hasta 4,69, 3,81 y 3,90 log ufc/g a día 60, y a 4,08, 1,46 y 2,04
log ufc/g a día 240. Los niveles de estafilococos coagulasa positivos descendieron entre
los días 1 y 14 desde 4,83 hasta 2,66 log ufc/g, no volviéndose a detectar durante el
posterior almacenamiento en refrigeración. Los valores de pH aumentaron
gradualmente desde 5,26 a día 1, hasta 5,36 a día 60 y 5,68 a día 240, mientras que el
extracto seco aumentó desde 47,78 % a día 1, hasta 53,13 % a día 60 y 65,43 % a día
240.
Los niveles de caseínas disminuyeron progresivamente a lo largo de la maduración

y del almacenamiento en refrigeración por efecto de la plasmina, el cuajo residual y las
enzimas del cultivo iniciador, junto a la actividad proteolítica de microorganismos
presentes en la leche como Enterococcus y Micrococcus (Centeno et al., 1995). Entre los
días 1 y 240 las αS-, β-, κ- y p-κ-caseínas sufrieron una reducción del 46, 64, 65 y 35 %,
respectivamente. Los péptidos hidrófilos aumentaron durante la maduración y el
272

Discusión general
almacenamiento, llegando a alcanzar valores de 8,72 UA/mg ES a día 240, mientras que
los péptidos hidrófobos aumentaron hasta el día 60 con valores de 7,69 UA/mg ES,
reduciéndose posteriormente hasta 4,37 UA/mg ES al final del periodo de
almacenamiento en refrigeración. El ratio de péptidos hidrófobos/hidrófilos descendió
de forma progresiva, desde 1,13 a día 14 hasta 0,50 a día 240. La actividad
aminopeptidasa alcanzó valores de 5,87 nmoles de p-nitroanilina/min·g a día 14 con el
sustrato Leu-p-NA, manteniéndose hasta el día 240 a niveles similares, mientras que
con el sustrato Lys-p-NA el máximo se alcanzó a día 21 con un valor de 11,39 nmoles
de p-nitroanilina/min·g, descendiendo posteriormente hasta 6,85 nmoles de p-
nitroanilina/min·g a día 240. La concentración de aminoácidos libres aumentó desde
2,39 mg/g ES a día 14, valores similares a los obtenidos para esta variedad de queso
elaborado con leche pasteurizada (Rodríguez-Alonso et al., 2011), hasta 31,68 mg/g ES
a día 240. De igual manera, la proteolisis global aumentó durante la maduración y el
La actividad tirosina descarboxilasa aumentó desde 0,04 mUA/g ES a día 1 hasta

2,62 mUA/g ES a día 240, provocando un aumento de tiramina desde 0,004 mg/g ES a
día 21 hasta 0,372 mg/g ES a día 240. Igualmente, la putrescina aumentó a lo largo del
tiempo desde 0,008 mg/g ES a día 21 hasta 0,207 mg/g ES a día 240, mientras que la
cadaverina que a día 1 se encontró en una concentración de 0,014 mg/g ES aumentó
hasta 0,113 mg/g ES a día 21, y a partir del día 60 se mantuvo en niveles similares. La
histamina únicamente se encontró a los 180 y 240 días, en concentraciones de 0,09 y
0,026 mg/g ES respectivamente. La concentración total de aminas biógenas alcanzó a
día 240 una concentración de 0,686 mg/g ES.
La actividad esterasa aumentó de 4,63 pmoles de α-naftol/min·g a día 1 hasta 10,75

pmoles de α-naftol/min·g a día 120, estabilizándose en estos niveles hasta el día 240.
Los niveles de ácidos grasos libres aumentaron durante la maduración y el
almacenamiento en refrigeración. Los ácidos grasos libres de cadena corta aumentaron
hasta 0,19 mg/g ES a día 240 y los de cadena media y larga a hasta 0,40 y 1,31 mg/g ES,
respectivamente, a día 240. Por otro lado, el ácido etanoico, aumentó hasta 2,10 mg/g
ES a día 240. El ácido benzoico, que puede formarse a partir del ácido hipúrico, de la
273

Discusión general
fenilalanina o de la auto-oxidación del benzaldehído (Sieber et al., 1995), aumentó hasta

0,398 mg/g ES a día 180.
El grupo de compuestos volátiles más abundante en todos los tiempos analizados

fue el de los alcoholes, al igual que en otros quesos gallegos (Rodríguez-Alonso et al.,
2009), seguido de los ácidos volátiles y de las cetonas. Dentro del grupo de los
alcoholes el compuesto más abundante fue el 2-butanol, procedente del metabolismo
del diacetilo, que aporta aromas florales (Mallia et al., 2005). Los niveles de ácidos
volátiles, cetonas, ésteres, éteres, hidrocarburos, compuestos bencénicos y azufrados
descendieron durante el almacenamiento en refrigeración, los aldehídos se
mantuvieron en niveles estables y los terpenos aumentaron. El nivel de alcoholes
aumentó entre los días 60 y 120, descendiendo posteriormente.
Durante el almacenamiento en refrigeración los parámetros de textura, elasticidad y

firmeza aumentaron en función del tiempo hasta el día 180 y descendiendo a día 240,
mientras que la fracturabilidad únicamente se pudo determinar a día 240. La
luminosidad en el interior del queso descendió a lo largo del tiempo. La tendencia al
rojo del interior aumentó gradualmente hasta el día 180, reduciéndose a día 240,
mientras que la tendencia al amarillo del interior del queso fue aumentando a lo largo
del tiempo hasta el día 240. La intensidad de sabor fue aumentando con el tiempo,
mientras que la calidad se mantuvo hasta el día 180, descendiendo a día 240. Los
parámetros de sabor amargo y umami aumentaron con el tiempo de almacenamiento.
El almacenamiento en condiciones de refrigeración consiguió mantener la calidad

de sabor del queso Arzúa-Ulloa hasta los 6 meses, aunque se siguieron produciendo
cambios en los parámetros químicos, de textura y color durante todo este periodo. A
los 8 meses se observó una pérdida de la calidad de sabor, acompañada de un aumento
de la intensidad de sabor, así como del sabor amargo. Junto a la pérdida de calidad del
sabor también se encontraron niveles de aminas biógenas al final del periodo de
almacenamiento en refrigeración, que podrían resultar nocivos para la salud del
consumidor.
274

Discusión general
Tabla 16. Tendencias de las características microbiológicas y químicas del queso Arzúa-Ulloa tratado por
alta presión con respecto al queso control.
Parámetro 400-2S 600-2S 400-3S 600-3S
ES NS NS NS NS
pH ↓ 180 a 240 d ↑ 14 d, ↓↓ 180 a 240 d ↓ 240 d ↓↓ 180 a 240 d
Aerobios totales ↓ hasta 21 d ↓↓ hasta 240 d ↓ 21 d ↓↓ hasta 240 d
Bacterias lácticas ↓ hasta 21 d ↓↓ hasta 240 d ↓ 21 d ↓↓ hasta 240 d
Lactobacilos ↓ hasta 21 d ↓↓ hasta 240 d ↓ 21 d ↓↓ hasta 240 d
Micrococos ↓ hasta 240 d ↓ hasta 240 d ↓ hasta 240 d ↓↓ hasta 240 d
Gram negativos ↓↓ hasta 240 d ↓↓ hasta 240 d ↓↓ hasta 240 d ↓↓ hasta 240 d
Coliformes ↓↓ hasta 240 d ↓↓ hasta 240 d ↓↓ hasta 240 d ↓↓ hasta 240 d
Estafilococos coagulasa + ↓↓ 14 d ↓↓ 14 d ↓↓ 14 d ↓↓ 14 d
Enterococos ↓ 14, 120 a 240 d ↓↓ hasta 240 d ↓ 120 a 240 d ↓↓ hasta 240 d
Levaduras ↓↓ hasta 240 d ↓↓ hasta 240 d ↓↓ hasta 240 d ↓↓ hasta 240 d
α-caseína ↓ 120 a 240 d NS ↓ 120 a 240 d NS
β-caseína ↓ hasta 240 d ↓ 60 a 240 d ↓ 60 a 240 d ↓ 60 a 240 d
κ-caseína ↓ 14, 60 a 240 d ↓ 60 a 240 d ↓ 60 a 240 d ↓ 60 a 240 d
p-κ-caseína ↓ 21 a 240 d ↓ 21 a 240 d ↓ 60 a 240 d ↓ 60 a 240 d
Péptidos hidrófilos ↑ 21 a 240 d ↑ 14, 60 a 240 d ↑ hasta 240 d ↑ hasta 240 d
Péptidos hidrófobos ↑ 60 a 240 d ↑ 60 a 240 d ↑ 60 a 240 d ↑ 60 a 240 d
Ratio hidrófobos/filos ↑ 240 d ↓ 21 d, ↑ 180 y 240 d ↓ 21 d, ↑ 240 d ↓ 21 d, ↑ 120 a 240 d
Act. aminopeptidasa (Lys) ↓ hasta 240 d ↓ hasta 240 d ↓ hasta 240 d ↓ hasta 240 d
Act. aminopeptidasa (Leu) ↓ hasta 240 d ↓ hasta 240 d ↓ hasta 240 d ↓ hasta 240 d
Aminoácidos libres totales ↓ 180 y 240 d ↓↓ 60 a 240 d ↓ 180 y 240 d ↓↓ 60 a 240 d
Proteolisis global (OPA) ↓ 180 y 240 d ↓↓ 21 a 240 d ↓ 180 y 240 d ↓↓ 60 a 240 d
Actividad TDC ↓ hasta 240 d ↓↓ hasta 240 d ↓ hasta 120 d ↓↓ hasta 240 d
Tiramina ↓ 21, 60 180 y 240 d ↓↓ 21 a 240 d ↓ 60 y 240 d ↓↓ 21 a 240 d
Putrescina ↓ 21, 60 y 240 d ↓↓ 21 a 240 d ↓ 21, 60 y 240 d ↓↓ 21 a 240 d
Cadaverina ↓ 21 a 240 d ↓ 21 a 180 d ↓ 21, 60 180 y 240 d ↓ 21, 120, 180 y 240 d
Histamina ↑ 120 d, ↓ 180 y 240 d ↓ 180 y 240 d ↓ 180 y 240 d ↓ 180 y 240 d
Aminas totales ↓ 21, 60 180 y 240 d ↓↓ 21 a 240 d ↓ 21, 60 y 240 d ↓↓ 21 a 240 d
Actividad esterasa ↓ hasta 240 d ↓↓ hasta 240 d ↓ hasta 240 d ↓↓ hasta 240 d
Etanoico + propanoico ↓ 14, 60 a 240 d ↓↓ hasta 240 d ↓ hasta 240 d ↓↓ hasta 240 d
AGL cadena corta ↓ 60 y 240 d ↓↓ 60 a 240 d ↓ 60 y 240 d ↓↓ 60 a 240 d
AGL cadena media NS ↓ 240 d ↓ 240 d ↓ 240 d
AGL cadena larga NS ↑ 60 d, ↓ 240 d ↓ 240 d ↓ 240 d
Ácidos volátiles NS NS NS NS
Alcoholes volátiles ↓ 60 y 120 d ↓↓ hasta 240 d ↓ 60 y 120 d ↓↓ hasta 240 d
Aldehídos volátiles ↑ hasta 240 d ↑↑ hasta 240 d ↑ 60 y 120 d ↑↑ hasta 240 d
Cetonas volátiles ↑↑ 60 d ↑ hasta 240 d ↑↑ 60 d ↑ hasta 240 d
Ésteres volátiles NS NS NS NS
Éteres volátiles ↑ 120 d NS NS ↑↑ 240 d
Hidrocarburos volátiles ↑ 120 d ↑ 240 d ↑ 60 d ↑ 120 y 240 d
Bencénicos volátiles NS NS NS NS
Azufrados volátiles ↑ 240 d ↑ 240 d NS ↑ 240 d
Terpenos volátiles NS ↓ 120 d NS ↓ 120 d
Otros volátiles ↓ 120 d ↓ hasta 240 d ↓ 60 d ↓ 120 y 240 d
(técnica espectrofotométrica), TDC; tirosina descarboxilasa.
275

Discusión general
Con el objetivo de evitar la pérdida de calidad de sabor que se dio a los 240 días y
la acumulación de aminas biógenas, se aplicaron tratamientos de 400 y 600 MPa a las 2
semanas (400-2S y 600-2S) y a las 3 semanas (400-3S y 600-3S) de maduración. Estos
tratamientos indujeron una serie de cambios respecto al control en los diferentes
tiempos considerados (Tablas 16 y 17). Por otro lado, comparando los quesos tratados a
lo largo del almacenamiento en refrigeración con el queso control en su momento
óptimo (21 a 60 días), se puede apreciar que en muchos casos se logró atenuar los
efectos negativos asociados a la sobremaduración (Tablas 18 y 19).
Tabla 17. Tendencias de las características de textura, color y sensoriales del queso Arzúa-Ulloa tratado por
alta presión con respecto al queso control.
Parámetro 400-2S 600-2S 400-3S 600-3S
Fracturabilidad ↑↑ 180 y 240 d ↑↑ 180 y 240 d ↑↑ 180 y 240 d ↑↑ 180 y 240 d
Firmeza ↓ 14 y 21 d, ↑ 240 d ↑ 120 y 240 d ↓ 21 d ↑ 120 d
↓ 14 y 21 d, ↑ 180 y
Elasticidad ↑ 240 d ↓ 21 d, ↑ 240 d ↑ 120 d
240 d
L* interior ↑ 60 d ↑ 21 a 240 d ↑ 60 d ↑ 21 a 240 d
a* interior ↓ hasta 120 d ↓↓ hasta 240 d ↓ hasta 240 d ↓↓ hasta 240 d
b* interior NS ↑ 14 d NS NS
Calidad sabor ↓ 120 d NS NS NS
Intensidad sabor NS NS NS NS
Sabor umami NS NS NS NS
Sabor amargo ↑ 60 a 240 d ↑ 240 d ↑ 120 d NS
Sabor dulce NS NS NS NS
Los tratamientos de altas presiones disminuyeron los niveles de microorganismos,

en especial los tratamientos de 600 MPa. Sin embargo, no se consiguió atenuar la
hidrólisis de caseínas, sino que los tratamientos la incrementaron de manera similar a lo
observado en queso Cheddar (O'Reilly et al., 2003, Rynne et al., 2008). Las altas
presiones pueden modificar la estructura de las caseínas haciéndolas más susceptibles a
la hidrólisis por la plasmina, que mantiene su actividad tras la aplicación de
tratamientos de hasta 800 MPa (Malone et al., 2003, Huppertz et al., 2004), lo que pudo
dar lugar a una mayor hidrólisis de las caseínas en los quesos tratados. Los niveles de
péptidos hidrófilos en quesos tratados fueron superiores a los del queso control de 60
276

Discusión general
días. Los niveles de péptidos hidrófobos también fueron por lo general superiores a los
del control de 60 días, sin embargo los quesos tratados con 400 MPa presentaron a los
180 y 240 días niveles similares o ligeramente inferiores a los del queso control de 60
días. El ratio de péptidos hidrófobos/hidrófilos se mantuvo en niveles inferiores a los del
queso control de 60 días hasta el final del almacenamiento en refrigeración. La fuerte
reducción de la actividad aminopeptidasa que provocaron las altas presiones, consiguió
atenuar la formación de aminoácidos libres, que en el caso de los quesos tratados con
600 MPa se encontraron en niveles similares a los del queso control de 60 días hasta el
final del periodo de refrigeración, al igual que sucedió con la proteolisis global.
La reducción de la actividad tirosina descarboxilasa por efecto de las altas

presiones, en especial los tratamientos de 600 MPa, atenuó la formación de tiramina,
que se mantuvo en niveles no detectables en los quesos tratados con 600 MPa hasta el
final del almacenamiento en refrigeración. Igualmente, la formación de putrescina se
atenuó con los tratamientos de 400 MPa y se eliminó completamente en los quesos
tratados con 600 MPa. Por el contrario, la formación de cadaverina no se vio tan
afectada por las altas presiones, aunque sí se observó un ligero descenso en los quesos
tratados.
La reducción de la actividad esterasa por las altas presiones, en especial con los
tratamientos de 600 MPa, consiguió disminuir la formación de ácidos grasos libres de
cadena corta, que se mantuvieron en los quesos tratados con 600 MPa en niveles
similares o ligeramente superiores a los del queso control de 60 días durante todo el
almacenamiento en refrigeración. Por el contrario, los niveles de ácidos grasos libres de
cadena media y larga fueron superiores a los del queso control de 60 días, ya que no se
observó ningún efecto de las altas presiones sobres estos ácidos grasos, de igual
manera que sucedió en queso Cheddar (Rynne et al., 2008). Las altas presiones
consiguieron mantener los niveles de ácido etanoico y propanoico por debajo de los
del queso control de 60 días, especialmente los tratamientos de 600 MPa. Los niveles de
ácido benzoico no se vieron afectados por los tratamientos de altas presiones.
277

Discusión general
Tabla 18. Valores de las características microbiológicas y químicas en queso Arzúa-Ulloa control de 21 y 60
días, quesos tratados por alta presión de 60-240 días y queso control de 120-240 días.
Control 400-2S 600-2S 400-3S 600-3S Control
Parámetro 21 y 60 días 60 a 240 d 60 a 240 d 60 a 240 d 60 a 240 d 120 a 240 d
pH 5,4 y 5,4 5,3 - 5,3 5,3 - 5,7 5,4 - 5,6 5,3 - 5,4 5,4 - 5,7
a
Aerobios totales 8,0 y 7,9 7,7 - 7,9 3,2 - 4,4 7,6 - 7,9 3,4 - 4,7 7,9 - 8,1
Bacterias lácticasa 8,1 y 8,1 7,8 - 7,9 2,9 - 4,5 7,6 - 7,8 3,1 - 4,7 7,6 - 8,0
Lactobacilosa 7,0 y 8,0 7,3 - 8,0 2,0 - 4,3 7,6 - 8,1 3,1 - 4,7 7,9 - 8,0
a
Enterococos 5,5 y 4,9 2,0 - 3,9 nd 2,4 - 4,9 nd 4,7 - 5,0
Micrococosa 5,5 y 5,0 nd - 2,6 nd - 2,2 nd - 3,4 nd 4,1 - 4,7
Gram negativosa 6,2 y 3,9 nd nd nd nd 2,0 - 3,3
a
Coliformes 5,8 y 3,8 nd nd nd nd nd - 2,9
Estafilococos coagulasa +a nd nd nd nd nd nd
a
Levaduras 3,5 y 3,2 nd nd nd nd 2,5 - 3,0
b
α-caseína 73,3 y 67,6 32,9 - 55,2 52,4 - 57,2 30,4 - 48,4 48,4 - 62,5 47,7 - 55,2
β-caseínab 73,6 y 62,8 7,3 - 25,9 14,0 - 25,3 8,9 - 25,5 14,5 - 29,9 32,4 - 52,3
b
κ-caseína 16,7 y 13,7 5,4 - 8,8 4,4 - 9,3 5,4 - 7,6 5,2 - 9,9 8,1 - 9,8
b
p-κ-caseína 12,8 y 12,0 1,8 - 3,6 3,9 - 4,6 1,9 - 3,5 4,8 - 5,1 10,4 - 10,8
Péptidos hidrófilosc 7,0 y 7,6 11,7 - 12,5 10,8 - 12,0 10,6 - 12,5 10,8 - 11,7 7,2 - 8,7
c
Péptidos hidrófobos 7,5 y 7,7 7,5 - 10,3 9,5 - 10,0 7,5 - 9,8 9,5 - 10,6 4,4 - 5,9
Ratio hidrófobos/filos 1,1 y 1,0 0,6 - 0,9 0,8 - 0,9 0,6 - 0,9 0,8 - 0,9 0,5 - 0,8
Act. aminopeptidasa (Lys)d 11,4 y 9,9 0,8 - 1,7 0,2 - 1,1 0,7 - 1,7 0,2 - 1,3 6,9 - 9,9
d
Act. aminopeptidasa (Leu) 5,5 y 5,4 0,5 - 0,8 0,3 - 0,6 0,5 - 0,8 0,3 - 0,7 4,6 - 5,0
b
Aminoácidos libres totales 3,8 y 8,8 8,0 - 22,0 6,2 - 11,7 9,8 - 22,8 6,4 - 11,4 16,7 - 31,7
Proteolisis global (OPA)e 0,4 y 1,2 1,1 - 3,0 0,7 - 1,6 1,1 - 3,2 0,9 - 1,8 2,2 - 4,2
f
TDC 0,2 y 0,9 0,04 - 1,1 0,0 - 0,0 0,08 - 1,9 0,0 - 0,0 1,4 - 2,6
b
Tiramina 0,004 y 0,109 0,012 - 0,176 0,000 - 0,000 0,068 - 0,244 0,000 - 0,005 0,178 - 0,372
Putrescinab 0,008 y 0,097 0,007 - 0,161 0,000 - 0,001 0,029 - 0,183 0,000 - 0,002 0,147 - 0,207
b
Cadaverina 0,113 y 0,088 0,032 - 0,049 0,044 - 0,081 0,032 - 0,069 0,042 - 0,090 0,057 - 0,082
b
Histamina 0,000 y 0,000 0,000 - 0,005 0,000 - 0,000 0,000 - 0,004 0,000 - 0,000 0,000 - 0,026
Aminas totalesb 0,13 y 0,29 0,07 - 0,35 0,05 - 0,08 0,16 - 0,32 0,04 - 0,10 0,34 - 0,69
g
Actividad esterasa 7,7 y 7,9 3,2 - 4,3 0,7 - 1,8 2,8 - 3,9 0,7 - 1,8 10,7 - 11,2
b
Etanoico + propanoico 1,5 y 1,8 1,4 - 1,7 1,2 - 1,5 1,7 - 1,8 1,3 - 1,5 2,2 - 2,4
AGL cadena cortab 0,04 y 0,07 0,05 - 0,15 0,04 - 0,09 0,06 - 0,15 0,05 - 0,09 0,11 - 0,19
b
AGL cadena media 0,19 y 0,22 0,23 - 0,38 0,26 - 0,35 0,24 - 0,37 0,24 - 0,36 0,30 - 0,40
b
AGL cadena larga 0,65 y 0,75 0,75 - 1,24 0,87 - 1,14 0,81 - 1,21 0,80 - 1,17 0,98 - 1,31
Alcoholes volátilesh 4209 y 11663 3432 - 8955 1508 - 2522 4052 - 12055 1998 - 3402 9582 - 16181
Aldehídos volátilesh 26,3 y 20,5 29,3 - 31-8 38,9 - 42,2 17,4 - 32,7 39,4 - 52,0 19,4 - 19,6
h
Cetonas volátiles 8792 y 5211 759 - 21057 7432 - 14344 980 - 18262 6990 - 10874 1381 - 4049
Éteres volátilesh 106 y 148 215 - 395 133 - 224 155 - 279 131 - 546 91 - 103
h
Azufrados volátiles 23 y 12 8 - 23 9 - 18 6 - 18 9 - 19 4-8
h
Hidrocarburos volátiles 51 y 75 30 - 138 54 - 106 39 - 146 54 - 132 23 - 73
nd: no detectado, OPA: o-ftalaldehído, TDC: actividad tirosina descarboxilasa, a: (log ufc/g), b: (mg/g ES), c: (UA/mg ES), d:(nmol de
p-nitroanilina/min g), e: (unidades de absorbancia), f: (mUA/g ES) g: (pmol α-naftol/min g), h: (abundancia relativa).
278

Discusión general
Los tratamientos de altas presiones afectaron de forma diferente a los distintos

grupos de compuestos volátiles. Los tratamientos de 600 MPa atenuaron la disminución
de cetonas, manteniendo niveles superiores a los del queso control de 60 días y
provocaron que los niveles de alcoholes se mantuviesen estables durante el
almacenamiento en refrigeración, con niveles inferiores a los del queso control de 60
días. Las altas presiones disminuyeron los niveles de terpenos, manteniendo los quesos
tratados con 600 MPa niveles similares a los del queso control de 60 días. Las altas
presiones atenuaron la disminución de ácidos volátiles, éteres, hidrocarburos y
compuestos azufrados y bencénicos. Únicamente se consiguieron mantener niveles de
éteres similares a los del queso control de 60 días, a día 120, en los quesos tratados a
las 3 semanas y a día 240, en los tratados con 600 MPa a las 2 semanas y con 400 MPa
a las 3 semanas. Los niveles de ácidos volátiles y compuestos azufrados en los quesos
tratados se mantuvieron en niveles similares a los del queso control de 60 días
únicamente hasta el día 120. Los tratamientos de altas presiones hicieron que los
aldehídos se mantuvieran en niveles superiores a los del queso control de 60 días,
excepto en los quesos tratados con 400 MPa a las 3 semanas, que a día 240
presentaron niveles similares. Los ésteres de los quesos tratados se mantuvieron en
niveles inferiores a los del queso control de 60 días hasta el día 240.
Tabla 19. Valores de las características de textura, color y sensoriales en queso Arzúa-Ulloa control de 21 y
60 días, quesos tratados por alta presión de 60-240 días y queso control de 120-240 días.
Control 400-2S 600-2S 400-3S 600-3S Control
Parámetro 21 y 60 días 60 a 240 d 60 a 240 d 60 a 240 d 60 a 240 d 120 a 240 d
Fracturabilidad (N) 0,0 y 0,0 0,0 - 21,2 0,0 - 24,0 0,0 - 19,4 0,0 - 19,0 0,0 - 2,4
Firmeza (J) 0,03 y 0,04 0,05 - 0,18 0,05 - 0,20 0,04 - 0,17 0,05 - 0,18 0,07 - 0,18
Elasticidad (N/mm2) 0,03 y 0,05 0,06 - 0,38 0,06 - 0,41 0,05 - 0,37 0,05 - 0,33 0,10 - 0,26
L* en interior 85,1 y 83,3 78,0 - 84,4 78,9 - 84,9 77,2 - 84,1 78,0 - 83,8 76,6 - 82,8
a* en interior 2,0 y 2,1 1,4 - 2,1 0,7 - 1,5 1,2 - 1,9 0,6 - 1,4 1,6 - 2,1
b* en interior 19,0 y 19,3 18,7 - 24,1 19,1 - 23,7 19,3 - 24,0 19,7 - 23,7 22,1 - 24,1
Calidad de sabor 7,0 y 7,2 5,5 - 6,3 5,4 - 6,5 5,6 - 6,5 5,5 - 6,7 5,5 - 7,3
Sabor amargo 1,1 y 1,4 3,3 - 4,8 2,7 - 4,7 2,9 - 4,3 2,7 - 4,3 1,8 - 2,6
Las altas presiones hicieron que los parámetros de textura (fracturabilidad, firmeza
y elasticidad) aumentasen en los quesos tratados en mayor medida que en el queso
control, manteniendo en todo momento valores superiores a los del queso control de
279

Discusión general
60 días. El parámetro b* en el interior del queso no se vio afectado por las altas
presiones. La pérdida de luminosidad en el interior del queso control se vio atenuada
por los tratamientos de altas presiones y únicamente los quesos tratados con 600 MPa
mantuvieron niveles similares a los del queso control de 60 días hasta el día 180. El
parámetro a* fue inferior en los quesos tratados, manteniéndose en valores inferiores a
los del queso control de 60 días durante todo el periodo de almacenamiento en
refrigeración. Las altas presiones no lograron controlar el aumento de la intensidad de
sabor ni la disminución de la calidad de sabor que se dio a los 240 días, y además
favorecieron el aumento de sabor amargo.
El análisis de componentes principales (recogido en el capítulo 10) indicó que los

quesos tratados con 600 MPa mantuvieron hasta los 120 días un perfil de compuestos
volátiles muy similar al del queso control de 21 días. De los resultados obtenidos se
puede concluir que los tratamientos de 600 MPa ralentizan y atenúan ciertos cambios
que se producen durante el almacenamiento en refrigeración, consiguiendo que el
queso Arzúa-Ulloa tratado con 600 MPa mantenga en gran medida las características
de esta variedad de queso en su momento óptimo (21 a 60 días) hasta los 6 meses de
almacenamiento en refrigeración, y evitando la acumulación de aminas biógenas.
280

Discusión general
 Bibliografía
and taste of Hispánico cheese. Journal of Dairy Science 89, 2882-2893.
Ávila, M., Calzada, J., Garde, S. & Nuñez, M. (2007). Effect of a bacteriocin-producing
Lactococcus lactis strain and high-pressure treatment on the esterase activity and
free fatty acids in Hispánico cheese. International Dairy Journal 17, 1415-1423.
Cantor, M. D., van den Tempel, T., Hansen, T. K. & Ardö, Y. (2004). Blue cheese. In
Cheese: Chemistry, Physics and Microbiology, pp. 175-198: Academic Press.
Carbonell, M., Nuñez, M. & Fernández-García, E. (2002). Evolution of the volatile
components of ewe raw milk La Serena cheese during ripening. Correlation with
flavour characteristics. Lait 82, 683-698.
Centeno, J. A., Cepeda, A. & Rodríguez-Otero, J. L. (1995). Identification and
preliminary characterization of strains of enterococci and micrococci isolated from
Arzúa raw cows'-milk cheese. Nahrung-Food 39, 55-62.
13, 841-866.
Creamer, L. K. & Olson, N. F. (1982). Rheological evaluation of maturing cheddar
cheese. Journal of Food Science 47, 631-636.
Curioni, P. M. G. & Bosset, J. O. (2002). Key odorants in various cheese types as
determined by gas chromatography-olfactometry. International Dairy Journal 12,
959-984.
Daryaei, H., Coventry, M. J., Versteeg, C. & Sherkat, F. (2006). Effects of high-pressure
treatment on shelf life and quality of fresh lactic curd cheese. Australian Journal of
Dairy Technology 61, 186-188.
Deeth, H. C. (2006). Lipoprotein lipase and lipolysis in milk. International Dairy Journal
16, 555-562.
Delgado, F. J., González-Crespo, J., Ladero, L., Cava, R. & Ramírez, R. (2009). Free
fatty acids and oxidative changes of a Spanish soft cheese (PDO 'Torta del Casar')
during ripening. International Journal of Food Science and Technology 44, 1721-1728.
Delgado, F. J., González-Crespo, J., Cava, R., García-Parra, J. & Ramírez, R. (2010a).
Characterisation by SPME-GC-MS of the volatile profile of a Spanish soft cheese
P.D.O. Torta del Casar during ripening. Food Chemistry 118, 182-189.
Delgado, F. J., Rodríguez-Pinilla, J., González-Crespo, J., Ramírez, R. & Roa, I.
(2010b). Proteolysis and texture changes of a Spanish soft cheese ('Torta del Casar')
manufactured with raw ewe milk and vegetable rennet during ripening.
International Journal of Food Science and Technology 45, 512-519.
281

Discusión general
Delgado, F. J., González-Crespo, J., Cava, R. & Ramírez, R. (2011). Free fatty acids and
oxidative changes of a raw goat milk cheese through maturation. Journal of Food
Science 76, C669-C673.
Delgado, F. J., González-Crespo, J., Cava, R. & Ramírez, R. (2012). High-pressure
treatment applied throughout ripening of a goat cheese caused minimal changes
on free fatty acids content and oxidation in mature cheese. Dairy Science &
Technology 92, 237-248.
Fernández, M., Linares, D. M., del Río, B., Ladero, V. & Alvarez, M. A. (2007). HPLC
quantification of biogenic amines in cheeses: correlation with PCR-detection of
tyramine-producing microorganisms. Journal of Dairy Research 74, 276-282.
Fernández-García, E., Tomillo, J. & Nuñez, M. (1999). Effect of added proteinases and
level of starter culture on the formation of biogenic amines in raw milk Manchego
cheese. International Journal of Food Microbiology 52, 189-196.
Fernández-García, E., Tomillo, J. & Nuñez, M. (2000). Formation of biogenic amines in
raw milk Hispánico cheese manufactured with proteinases and different levels of
starter culture. Journal of Food Protection 63, 1551-1555.
Flórez, A., Ruas-Madiedo, P., Alonso, L. & Mayo, B. (2006). Microbial, chemical and
sensorial variables of the Spanish traditional blue-veined Cabrales cheese, as
affected by inoculation with commercial Penicillium roqueforti spores. European Food
Research and Technology 222, 250-257.
Garde, S., Arqués, J. L., Gaya, P., Medina, M. & Nuñez, M. (2007). Effect of high-
pressure treatments on proteolysis and texture of ewes' raw milk La Serena cheese.
Gobbetti, M., Burzigotti, R., Smacchi, E., Corsetti, A. & De Angelis, M. (1997).
Microbiology and biochemistry of Gorgonzola cheese during ripening. International
Dairy Journal 7, 519-529.
Gomez, M. J., Garde, S., Gaya, P., Medina, M. & Nuñez, M. (1997). Relationship
between level of hydrophobic peptides and bitterness in cheese made from
pasteurized and raw milk. Journal of Dairy Research 64, 289-297.
499.
Ismail, B. & Nielsen, S. S. (2010). Invited review: Plasmin protease in milk: Current
knowledge and relevance to dairy industry. Journal of Dairy Science 93, 4999-5009.
Juan, B., Ferragut, V., Buffa, M., Guamis, B. & Trujillo, A.-J. (2007). Effects of high-
pressure treatment on free fatty acids release during ripening of ewes' milk cheese.
Journal of Dairy Research 74, 438-445.
282

Discusión general
Kinsella, J. E. & Hwang, D. H. (1976). Enzymes of Penicillium roqueforti involved in the

biosynthesis of cheese flavor. CRC Critical Reviews in Food Science and Nutrition 8,
191-228.
Lenoir, J. & Auberger, B. (1977). Les caractères du système protéolytique de Penicillium
caseicolum. II. Caractérisation dune protéase neutre. Le Lait 57, 471-491.
Linares, D. M., Martín, M. C., Ladero, V., Álvarez, M. A. & Fernández, M. (2011).
Biogenic amines in dairy products. Critical Reviews in Food Science and Nutrition 51,
691-703.
Macedo, A. C. & Malcata, F. X. (1997). Hydrolysis of alpha(s) and beta-caseins during
ripening of Serra cheese. Food Chemistry 58, 43-48.
Madkor, S., Fox, P. F., Shalabi, S. I. & Metwalli, N. H. (1987a). Studies on the ripening
of Stilton cheese: Lipolysis. Food Chemistry 25, 93-109.
Madkor, S., Fox, P. F., Shalabi, S. I. & Metwalli, N. H. (1987b). Studies on the ripening
of Stilton cheese: Proteolysis. Food Chemistry 25, 13-29.
Mallia, S., Fernández-García, E. & Bosset, J. O. (2005). Comparison of purge and trap
and solid phase microextraction techniques for studying the volatile aroma
compounds of three European PDO hard cheeses. International Dairy Journal 15,
741-758.
McSweeney, P. L. H. & Sousa, M. J. (2000). Biochemical pathways for the production
of flavour compounds in cheeses during ripening: A review. Lait 80, 293-324.
McSweeney, P. L. H. (2004). Biochemistry of cheese ripening: Introduction and
overview. Cheese: Chemistry, Physics and Microbiology, 347-360.
Molimard, P. & Spinnler, H. E. (1996). Compounds involved in the flavor of surface
mold-ripened cheeses: Origins and properties. Journal of Dairy Science 79, 169-184.
Mulder, H. (1952). Taste and flavour forming substancesin cheese. Netherlands Milk and
O'Reilly, C. E., Kelly, A. L., Oliveira, J. C., Murphy, P. M., Auty, M. A. E. & Beresford,
T. P. (2003). Effect of varying high-pressure treatment conditions on acceleration of
ripening of cheddar cheese. Innovative Food Science & Emerging Technologies 4, 277-
284.
Ordóñez, A. I., Ibáñez, F. C., Torre, P. & Barcina, Y. (1997). Formation of biogenic
amines in Idiazabal ewe's-milk cheese: effect of ripening, pasteurization, and
starter. Journal of Food Protection 60, 1371-1375.
Prieto, B., Franco, I., Fresno, J. M., Bernardo, A. & Carballo, J. (2000). Picon Bejes-
Tresviso blue cheese: an overall biochemical survey throughout the ripening
process. International Dairy Journal 10, 159-167.
283

Discusión general
Rodríguez-Alonso, P., Centeno, J. A. & Garabal, J. I. (2009). Comparison of the volatile

profiles of Arzúa-Ulloa and Tetilla cheeses manufactured from raw and
pasteurized milk. LWT - Food Science and Technology 42, 1722-1728.
Rodríguez-Alonso, P., Centeno, J. A. & Garabal, J. I. (2011). Biochemical study of
industrially produced Arzúa-Ulloa semi-soft cows' milk cheese: Effects of storage
under vacuum and modified atmospheres with high-nitrogen contents.
Rynne, N. M., Beresford, T. P., Guinee, T. P., Sheehan, E., Delahunty, C. M. & Kelly,
A. L. (2008). Effect of high-pressure treatment of 1 day-old full-fat Cheddar cheese
on subsequent quality and ripening. Innovative Food Science & Emerging Technologies
9, 429-440.
Saldo, J., Fernández, A., Sendra, E., Butz, P., Tauscher, B. & Guamis, B. (2003). High
pressure treatment decelerates the lipolysis in a caprine cheese. Food Research
Sieber, R., Bütikofer, U. & Bosset, J. O. (1995). Benzoic acid as a natural compound in
cultured dairy products and cheese. International Dairy Journal 5, 227-246.
Sousa, M. J. & McSweeney, P. L. H. (2001). Studies on the ripening of Cooleeney, an
Irish farmhouse Camembert-type cheese. Irish Journal of Agricultural and Food
Research 40, 83-95.
Spinnler, H. E. & Gripon, J. C. (2004). Surface mould-ripened cheeses. In Cheese:
Voigt, D. D., Chevalier, F., Qian, M. C. & Kelly, A. L. (2010). Effect of high-pressure
treatment on microbiology, proteolysis, lipolysis and levels of flavour compounds
in mature blue-veined cheese. Innovative Food Science & Emerging Technologies 11,
68-77.
Voigt, D. D., Chevalier, F., Donaghy, J. A., Patterson, M. F., Qian, M. C. & Kelly, A. L.
(2012). Effect of high-pressure treatment of milk for cheese manufacture on
proteolysis, lipolysis, texture and functionality of Cheddar cheese during ripening.
Innovative Food Science & Emerging Technologies 13, 23-30.
Yvon, M. & Rijnen, L. (2001). Cheese flavour formation by amino acid catabolism.
284

Capítulo 12. Conclusiones.

285

Fotografía: variedades de queso estudiadas, queso azul (superior izquierda), Torta del
Casar (superior derecha), queso Brie (inferior izquierda) y queso Arzúa-Ulloa (Inferior
derecha).
Conclusiones
1. En el queso azul, únicamente los tratamientos de 600 MPa y el de 400 MPa

aplicado a las 3 semanas consiguieron reducir la proteolisis secundaria, manteniendo
los niveles de aminoácidos libres y proteolisis global en valores similares a los del queso
control de 180 días. Todos los tratamientos redujeron los niveles de tiramina, pero
únicamente el tratamiento de 400 MPa aplicado a las 6 semanas consiguió disminuir los
niveles de aminas biógenas totales.
2. En el queso azul, la reducción de los niveles de ácidos grasos y compuestos

volátiles en los quesos tratados con 600 MPa a las 3 semanas provocó una disminución
de la calidad de sabor. Los restantes tratamientos disminuyeron los niveles de
numerosos compuestos, a pesar de lo cual se mantuvieron niveles de calidad de sabor
similares a los del queso control, que apenas sufrió variaciones durante el
almacenamiento.
3. En el queso azul, las altas presiones causaron un aumento del índice de

hundimiento en los quesos. En general, los tratamientos de altas presiones empleados
no representaron ninguna mejora adicional de cara a la conservación de este tipo de
queso.
4. En la Torta del Casar, los tratamientos de 600 MPa consiguieron frenar tanto la
proteolisis primaria como la secundaria, manteniendo los niveles de caseínas, aunque
no los niveles de aminoácidos y proteolisis global, en valores similares a los del queso
control de 60 días.
5. En la Torta del Casar, los tratamientos por altas presiones disminuyeron los
niveles de tirosina descarboxilasa y los de 600 MPa los niveles de todas las aminas
biógenas encontradas. Sin embargo la mayoría de los tratamientos por altas presiones
dieron lugar a quesos de textura más firme que la del queso control.
6. En la Torta del Casar, las altas presiones evitaron el incremento de los niveles de
ácidos grasos libres y compuestos volátiles, en especial de compuestos azufrados, que
se dio en el queso control, consiguiendo que los quesos tratados a las 5 semanas
mantuviesen hasta el día 240 un perfil de ácidos grasos libres y compuestos volátiles
similar al del queso control de 60 días, previniendo la sobremaduración.
287

Conclusiones
7. En el queso Brie, las altas presiones retrasaron la proteolisis primaria,

disminuyendo con 600 MPa la hidrólisis de caseínas y con todos los tratamientos los
niveles de péptidos hidrófobos, que aumentaron fuertemente en el queso control. En
ningún caso se consiguió reducir la proteolisis secundaria.
8. En el queso Brie, las altas presiones redujeron el incremento de ácidos grasos y,

en los quesos tratados a las 2 semanas, mantuvieron un perfil de compuestos volátiles
similar al del queso control de 30 días, estabilizando las características organolépticas y
evitando así la sobremaduración del queso.
9. En el queso Brie, las altas presiones evitaron la fuerte disminución de

fracturabilidad, firmeza y elasticidad que se dio en el queso control, especialmente con
los tratamientos aplicados a las 3 semanas. Sin embargo, todos los tratamientos
provocaron la pérdida de la capa blanca superficial de moho característica de esta
variedad de queso.
10. En el queso Arzúa-Ulloa, las altas presiones aceleraron la proteolisis primaria,

pero frenaron la proteolisis secundaria así como la formación de las aminas biógenas.
Los quesos tratados con 600 MPa mantuvieron hasta los 240 días niveles de aminas
biógenas totales inferiores a los del queso control de 21 días.
11. En el queso Arzúa-Ulloa, las altas presiones redujeron la formación de ácidos

grasos. Los tratamientos de 600 MPa mantuvieron además un perfil de compuestos
volátiles similar al del control de 21 días hasta el día 120.
12. En el queso Arzúa-Ulloa, las altas presiones indujeron un aumento de

fracturabilidad, firmeza y elasticidad. Los cambios químicos ocasionados por las altas
presiones no afectaron a la calidad del queso, aunque causaron un cierto aumento de
las puntuaciones de amargor.
288

Capítulo 13. Resumen ampliado.

289

Fotografía: quesos estudiados, placas de agar para recuento de microorganismos,
cromatograma del perfil de volátiles, reacción colorimétrica para determinación de la
actividad esterasa y secuencia de ensayo texturométrico por compresión.
Resumen ampliado
 Introducción
El queso es un derivado lácteo conocido desde la antigüedad, que ha ido

evolucionando hasta dar lugar a los cientos de variedades que se elaboran hoy en día.
La producción mundial de queso en 2012 según la FAO, fue de 20,6 millones de
toneladas, convirtiéndose en uno de los principales productos agroalimentarios
(FAOSTAT, 2014). En la Unión Europea la producción de queso alcanzó 8,7 millones de
toneladas en 2013, de los cuales únicamente se produjo en España el 1,66 %
(EUROSTAT, 2014). El consumo medio per cápita en España durante el año 2013 se
estimó en 8,2 kg/persona y año, lejos aún de la media europea de 17,2 Kg/persona y
año (MAGRAMA, 2014).
La transformación de la leche en queso comprende una serie de etapas, de las

cuales la maduración es una de las más importantes, ya que es cuando se desarrollan
las características organolépticas, de textura y apariencia típicas de cada variedad. Estas
características se alcanzan gracias a una serie de reacciones que se pueden agrupar
como glicolisis y catabolismo de lactato y citrato, proteolisis y catabolismo de
aminoácidos, y lipolisis y catabolismo de ácidos grasos libres. La fermentación de la
lactosa por los microorganismos presentes en el queso, principalmente bacterias
lácticas, da lugar a la formación de ácido láctico junto a cantidades menores de ácido
acético y propiónico, diacetilo, acetoína, etanol y otros compuestos que contribuyen al
sabor y aroma. El citrato residual, retenido en la cuajada, puede ser metabolizado por
cultivos iniciadores mesófilos para dar diacetilo, acetato, acetoína, 2,3-butanodiol y CO2
(McSweeney & Fox, 2004). La hidrólisis de triglicéridos por las esterasas y lipasas nativas
de la leche o de microorganismos presentes provoca la liberación de ácidos grasos, que
contribuyen directamente al aroma y sabor, en especial los de cadena corta y media, y
son además precursores de otros compuestos responsables del aroma y sabor (Collins
et al., 2003). La hidrólisis de caseínas por las proteasas nativas de la leche, el cuajo o los
sistemas proteolíticos de los microorganismos, da lugar a péptidos de gran tamaño,
que a su vez son hidrolizados a péptidos de menor tamaño y en última instancia a
aminoácidos libres. Tanto péptidos como aminoácidos libres contribuyen al sabor del
queso, siendo los aminoácidos además precursores de otros compuestos responsables
del sabor y aroma (Sousa et al., 2001, Marilley & Casey, 2004).
291

Resumen ampliado
Una vez que el queso llega a su punto de maduración óptimo, en el que los
compuestos responsables del aroma y sabor alcanzan un equilibrio que da al producto
su sabor característico (Mulder, 1952), la maduración continúa durante el
almacenamiento, distribución y venta haciendo que el queso llegue al consumidor con
un sabor mas fuerte o diferente al deseado por el fabricante (Wick et al., 2004). Este
fenómeno conocido como sobremaduración se da principalmente en algunas
variedades de queso que experimentan una intensa proteolisis, debido al empleo de
enzimas coagulantes de elevada actividad como las cardosinas o de microorganismos
con elevada actividad proteolítica como los mohos del género Penicillium, y limita la
vida útil del producto. Apenas existen trabajos de investigación enfocados a evitar este
fenómeno, más allá del empleo de temperaturas de refrigeración, que aunque
consiguen ralentizarlo, no lo detienen completamente.
La presencia de microorganismos con actividad descarboxilasa plantea otro

problema, como es la presencia de aminas biógenas. Las aminas biógenas son
compuestos básicos nitrogenados, de bajo peso molecular, con actividad biológica. Su
síntesis y degradación forma parte del metabolismo normal de animales, plantas y
microorganismos (ten Brink et al., 1990). Las aminas pueden ser alifáticas (cadaverina,
putrescina, espermina y espermidina), aromáticas (tiramina y β-feniletilamina) o
heterocíclicas (histamina y triptamina), y pueden clasificarse según el número de grupos
amino que contienen como mono, di, o poliaminas (Silla Santos, 1996). A pesar de
formar parte del metabolismo normal y de regular diversas funciones indispensables en
el organismo, la ingesta de niveles elevados de aminas biógenas puede tener efectos
tóxicos tales como aumento de la presión sanguínea, dolores de cabeza, calambres
abdominales y urticaria. En mamíferos existe un eficiente sistema de detoxificación en el
tracto intestinal y en condiciones normales las aminas exógenas son rápidamente
detoxificadas por acción de las amino oxidasas. No obstante, en individuos con un
sistema de detoxificación deficiente o bajo tratamiento con fármacos inhibidores de las
amino oxidasas, pequeñas cantidades de aminas biógenas pueden provocar episodios
de intoxicación que pueden llegar a ser graves (Ladero et al., 2010).
Las aminas biógenas se pueden formar por descarboxilación de aminoácidos o por

transaminación de aldehídos y cetonas. El queso es un sustrato ideal para la formación
292

Resumen ampliado
de aminas, ya que tiene un alto contenido de aminoácidos libres y puede contener

microorganismos con actividad descarboxilasa, así como unas condiciones favorables
para el crecimiento de estos microorganismos y para la formación de aminas (Loizzo et
al., 2013).
Las altas presiones se emplean actualmente como método de pasteurización no

térmico, que mejora las características nutricionales de los alimentos frente a los
tratamientos térmicos de pasteurización. Además de transmitirse homogéneamente a
todo el producto de forma instantánea, otra de las ventajas que ofrecen las altas
presiones sobre los tratamientos térmicos es la posibilidad de ser aplicadas
directamente al producto elaborado, impidiendo la contaminación posterior del interior
del alimento. La demanda existente por parte del consumidor de alimentos más
naturales y mínimamente procesados convierte a las altas presiones en una tecnología
adecuada para satisfacer esta demanda. Existen en la actualidad gran número de
productos tratados por altas presiones en el mercado, como zumos y bebidas,
productos lácteos y cárnicos, platos preparados, rellenos para sándwiches, pescados y
mariscos, salsas, aguacate y guacamole. Las altas presiones se han aplicado con éxito
en leche y queso para la eliminación de microorganismos patógenos contaminantes
(Rodriguez et al., 2005, López-Pedemonte et al., 2007, Yang et al., 2012). También se
han empleado con el objetivo de acelerar la maduración en diferentes variedades de
queso, provocando la lisis de las bacterias lácticas y la liberación de enzimas al medio
(O’Reilly et al., 2000, Saldo et al., 2002, Ávila et al., 2006). Únicamente se han empleado
altas presiones para alargar la vida útil en queso fresco (Evert-Arriagada et al., 2012,
Evert-Arriagada et al., 2014). Las altas presiones además, de la capacidad de eliminar
microorganismos, también pueden emplearse para inactivar enzimas, pudiendo evitar el
fenómeno de la sobremaduración, así como la formación de aminas biógenas en queso
(Novella-Rodríguez et al., 2002, Malone et al., 2003, Huppertz et al., 2004).
 Objetivo
El objetivo del presente trabajo fue evaluar el efecto del tratamiento por altas
presiones hidrostáticas en diferentes variedades de queso a lo largo de un prolongado
periodo de almacenamiento en refrigeración, con el fin de evitar la sobremaduración
alargando la vida útil y prevenir la formación y acumulación de aminas biógenas.
293

Resumen ampliado
 Material y métodos
Los tratamientos de altas presiones se aplicaron en cuatro variedades de queso a

distintos tiempos de maduración, escogidos en función de la variedad. En queso azul de
leche pasteurizada de oveja, con Penicillium roqueforti en su interior, se aplicaron 400 y
600 MPa durante 5 minutos a las 3 (400-3S y 600-3S), 6 (400-6S y 600-6S) y 9 semanas
(400-9S y 600-9S) de maduración y se realizaron los análisis inmediatamente después
de los tratamientos y a los 3, 6, 9 y 12 meses. En Torta del Casar, elaborada con leche
cruda de oveja y cuajo vegetal de cardo (Cynara cardunculus), se aplicaron 400 y 600
MPa durante 5 minutos a las 3 (400-3S y 600-3S) y 5 semanas (400-5S y 600-5S) de
maduración y se realizaron los análisis inmediatamente después de los tratamientos y a
los 2, 4, 6 y 8 meses. En queso Brie de leche de vaca pasteurizada, con Penicillium
camemberti en su superficie, se aplicaron 400 y 600 MPa durante 5 minutos a las 2
(400-2S y 600-2S) y 3 semanas (400-3S y 600-3S) de maduración y se realizaron los
análisis inmediatamente después de los tratamientos y a los 1, 2, 3 y 4 meses. En el
queso Arzúa-Ulloa elaborado con leche cruda de vaca, se aplicaron 400 y 600 MPa
durante 5 minutos a las 2 (400-2S y 600-2S) y 3 semanas (400-3S y 600-3S) de
maduración y se realizaron los análisis inmediatamente después de los tratamientos y a
los 2, 4, 6 y 8 meses.
Para evaluar el efecto de las altas presiones sobre la microbiota del queso se
realizaron análisis microbiológicos mediante recuento en medio sólido de bacterias
mesófilas totales (PCA), bacterias lácticas (M17 y MRS agar), lactobacilos (Rogosa agar),
enterococos (KF agar), bacterias gram negativas (MacConkey agar), coliformes (VRBA),
Micrococcaceae (MSA), estafilococos coagulasa positivos (Baird-Parker + RPF II), Listeria
monocytogenes (Palcam agar), mohos y levaduras (CGA). Se analizó el pH mediante un
pH-metro acoplado a un electrodo de penetración y el extracto seco (ES) mediante
desecación a 102 ºC hasta peso constante. La proteolisis se evaluó mediante el análisis
de caseínas por electroforesis capilar en gel, péptidos y aminoácidos libres por
cromatografía líquida (HPLC), y proteolisis global mediante ensayo espectrofotométrico.
La actividad aminopeptidasa se determinó mediante ensayo colorimétrico medido por
espectrofotometría. La lipolisis se evaluó determinando los ácidos grasos libres
mediante extracción en fase sólida y análisis por cromatografía de gases (GC). La
294

Resumen ampliado
actividad esterasa se determinó mediante ensayo colorimétrico medido por

espectrofotometría. Para evaluar el efecto de las altas presiones sobre la formación de
aminas biógenas, se midió la actividad tirosina descarboxilasa y la concentración de
aminas biógenas mediante HPLC. Se analizaron los compuestos volátiles mediante
microextracción en fase sólida (SPME), seguida de cromatografía de gases-
espectrometría de masas (GC-MS). Se determinaron los parámetros de textura
(fracturabilidad, firmeza y elasticidad) por compresión mediante un texturómetro y los
parámetros colorimétricos (L*, a* y b*) mediante un colorímetro. Finalmente, se llevó a
cabo el análisis sensorial de sabor y olor, mediante evaluación de diversas
características y descriptores por un panel entrenado de catadores.
 Resultados
- Queso azul
El queso control, sin tratar, sufrió una reducción de los niveles de bacterias lácticas
y bacterias mesófilas totales desde 9,54 y 8,76 log ufc/g respectivamente a día 1 hasta
6,57 y 5,57 log ufc /g a día 360, mientras que los niveles de P. roqueforti descendieron
desde 7,73 log ufc/g a día 21 hasta 5,66 log ufc/g a día 360. Los tratamientos de altas
presiones redujeron los niveles de bacterias lácticas, bacterias mesófilas totales y P.
roqueforti, en especial los tratamientos de 600 MPa que llegaron a reducir la población
de P. roqueforti hasta niveles no detectables
La fuerte proteolisis primaria que se dio en el queso control, con reducciones de las
αS-, β-, κ- y p-κ-caseínas de 93, 87, 78 y 46 %, respectivamente, entre los días 1 y 21, no
pudo evitarse con los tratamientos de altas presiones. En cuanto a los niveles de
péptidos hidrófilos e hidrófobos, que aumentaron en el queso control, únicamente se
consiguieron reducir los niveles péptidos hidrófilos con los tratamientos de 600 MPa
aplicados a las 3 y 6 semanas. La actividad aminopeptidasa se redujo durante el
almacenamiento en refrigeración, y las altas presiones provocaron la reducción de esta
actividad. Los niveles de aminoácidos del queso control aumentaron durante la
maduración y el almacenamiento en refrigeración, llegando a alcanzar una
concentración de 116,76 mg/g ES a día 360, que únicamente los tratamientos de 600
MPa y el de 400 MPa aplicado a las 3 semanas consiguieron frenar, manteniendo hasta
295

Resumen ampliado
el día 360 niveles de aminoácidos libres y proteolisis global similares a los del queso
control de 180 días.
Las aminas biógenas en el queso control alcanzaron concentraciones de 0,06, 0,07,

0,05, 0,03 y 0,01 mg/g ES de β-feniletilamina, triptamina, tiramina, putrescina y
espermidina, respectivamente, a día 360. Los tratamientos de altas presiones
únicamente consiguieron reducir los niveles de tiramina.
Los tratamientos de 600 MPa y el de 400 MPa aplicado a las 3 semanas redujeron la
actividad esterasa. Los ácidos grasos libres en el queso control aumentaron durante la
maduración y el almacenamiento en refrigeración, alcanzando concentraciones de 8,88,
18,31 y 56,63 mg/g ES de ácidos grasos libres de cadena corta (C4:0 – C8:0), media (C10:0 –
C14:0) y larga (C16:0 – C18:3), respectivamente, a día 360. Excepto en los quesos tratados
con 400 MPa a las 6 y 9 semanas, el resto de tratamientos redujeron la formación de
todos los grupos de ácidos grasos libres. Únicamente los quesos tratados con 400 MPa
a las 3 semanas mantuvieron niveles de ácidos grasos libres de cadena corta similares o
inferiores a los del queso control de 180 días. Estos quesos, junto a los quesos tratados
con 600 MPa a las 6 y 9 semanas, mantuvieron niveles de ácidos grasos libres de
cadena media y larga similares a los del control de 180 días. Por otro lado, los niveles de
los tres grupos de ácidos grasos libres en los quesos tratados con 600 MPa a las 3
semanas se mantuvieron muy por debajo de los niveles del control de 180 días. El
ácido etanoico aumentó hasta el día 63, descendiendo posteriormente. Excepto los
tratamientos de 400 MPa aplicados a las 6 y 9 semanas, el resto redujeron los niveles de
este ácido.
El grupo mayoritario de compuestos volátiles fue el de los ácidos, que aumentó

durante el almacenamiento en refrigeración, seguido de alcoholes y cetonas, que
también aumentaron, aunque en menor medida. Igualmente, aumentaron los niveles de
ésteres, terpenos, compuestos azufrados y nitrogenados, mientras que los aldehídos y
compuestos bencénicos descendieron, y los hidrocarburos no variaron. Las altas
presiones redujeron los niveles de ésteres, alcoholes, hidrocarburos, compuestos
bencénicos y nitrogenados, especialmente en los quesos tratados con 600 MPa, en los
cuales además se redujo la formación de cetonas. En los quesos tratados a las 3
semanas se redujeron los niveles de ácidos volátiles, compuestos azufrados y terpenos.
296

Resumen ampliado
El análisis de componentes principales de los grupos de ácidos grasos libres y los

grupos de compuestos volátiles discriminó única y escasamente entre tiempos,
indicando que el efecto de las altas presiones sobre el queso azul fue leve.
Únicamente el tratamiento de 600 MPa aplicado a las 3 semanas redujo el ligero

aumento de intensidad de sabor que se dio en el queso control durante el
almacenamiento en refrigeración. El queso control no sufrió variaciones en la calidad de
sabor durante el almacenamiento en refrigeración y todos los quesos tratados
mantuvieron niveles similares a los del queso control, excepto los tratados con 600 MPa
a las 3 semanas que obtuvieron puntuaciones más bajas.
En el queso control se observó un hundimiento de la parte central del queso, que

llegó a ser de un 15 % respecto al borde. Las altas presiones provocaron un aumento de
este hundimiento, que en el caso de los quesos tratados con 400 MPa a las 3 semanas
llegó a ser de hasta un 36 %.
- Torta del Casar
En el queso control se observó un descenso de los niveles de bacterias mesófilas

totales, bacterias lácticas y enterococos desde 9,51, 9,46 y 7,32 log ufc/g
respectivamente a día 21, hasta 8,61, 8,42 y 6,80 log ufc/g a día 240. Los niveles de
lactobacilos aumentaron hasta 8,13 log ufc/g a día 35 y se mantuvieron en esos niveles
durante todo el almacenamiento en refrigeración. Las altas presiones redujeron los
niveles de bacterias mesófilas, bacterias lácticas, lactobacilos y enterococos,
especialmente con los tratamientos de 600 MPa. Los niveles de Micrococcaceae,
coliformes y bacterias gram negativas descendieron desde 6,39, 5,52 y 6,01 log ufc/g
respectivamente a día 1, hasta 5,57, 2,87 y 5,37 log ufc/g a día 240. Las altas presiones
redujeron los niveles de Micrococcaceae, especialmente con los tratamientos aplicados
a las 5 semanas, mientras que todos los tratamientos, excepto el de 400 MPa aplicado a
las 3 semanas, redujeron por debajo del límite de detección los niveles de coliformes y
bacterias gram negativas. Los niveles de estafilococos coagulasa positivos descendieron
en el queso control desde 5,23 log ufc/g a día 1 hasta 2,30 a día 180, no detectándose a
día 240, mientras que en los quesos presurizados estuvieron desde el día 21 por debajo
del límite de detección. En el queso control el pH aumentó durante la maduración y el
297

Resumen ampliado
almacenamiento en refrigeración hasta alcanzar valores de 6,13, siendo muy inferior el

aumento en los quesos tratados por altas presiones.
Las αS- y β-caseínas del queso control sufrieron reducciones del 79 y 62 %,

respectivamente, entre los días 1 y 35, mientras que la κ-caseína se redujo en un 96 %
entre los días 1 y 21, no volviéndose a detectar a partir del día 35. Los tratamientos de
600 MPa consiguieron frenar la proteolisis primaria, reduciendo la hidrólisis de αS- y β-
caseínas así como la formación de péptidos hidrófilos, que sí aumentaron en el queso
control. La actividad aminopeptidasa disminuyó más rápidamente en los quesos
tratados por altas presiones que en el queso control. La concentración de aminoácidos
libres aumentó durante la maduración y el almacenamiento en refrigeración, llegando a
alcanzar 23,30 mg/g ES a día 360. Los tratamientos de 600 MPa redujeron la proteolisis
global, además el tratamiento aplicado a las 3 semanas redujo la formación de
aminoácidos libres, aunque para ambos parámetros los niveles obtenidos fueron
superiores a los del queso control de 60 días.
La actividad tirosina descarboxilasa se triplicó en el queso control entre los días 60

y 180. Todos los tratamientos de altas presiones consiguieron reducirla. El incremento
de tiramina llegó hasta 1,09 mg/g ES a día 240 en el queso control, y se vio reducido
con los tratamientos de 600 MPa. La putrescina, triptamina, β-feniletilamina, cadaverina
e histamina aumentaron hasta alcanzar 0,95, 0,83, 0,29, 0,69 y 0,30 mg/g ES,
respectivamente, a día 240. Todos los tratamientos de altas presiones consiguieron
reducir los niveles de cadaverina. Los niveles de putrescina, triptamina y β-feniletilamina
se vieron igualmente reducidos, excepto en los quesos tratados con 400 MPa a las 3
semanas. Únicamente los tratamientos de 600 MPa consiguieron reducir los niveles de
histamina.
La actividad esterasa en el queso control disminuyó a partir del día 60. Los
tratamientos de 600 MPa lograron reducir esta actividad durante todo el
almacenamiento en refrigeración. Los niveles de ácidos grasos libres del queso control
aumentaron a lo largo de la maduración y del almacenamiento en refrigeración,
llegando a alcanzar concentraciones de ácidos grasos libres de cadena corta, media y
larga de 8,69, 0,88 y 3,02 mg/g ES, respectivamente, a día 240. Todos los tratamientos
redujeron la formación de ácidos grasos libres y únicamente los ácidos grasos libres de
298

Resumen ampliado
cadena corta mantuvieron hasta el día 240 niveles similares a los del queso control de
60 días. Los ácidos etanoico, propanoico y de cadena ramificada también aumentaron,
alcanzando 5,82, 2,99 y 1,15 mg/g ES a día 240. Todos los tratamientos redujeron los
niveles de ácido propanoico y ácidos de cadena ramificada, y únicamente los
tratamientos aplicados a las 3 semanas los de ácido etanoico, logrando mantener
niveles similares o inferiores a los del queso control de 60 días durante todo el
almacenamiento.
El grupo mayoritario de compuestos volátiles en el queso control fue el de los

ácidos hasta el día 180, mientras que a día 240 fue el grupo de alcoholes. Durante el
almacenamiento en refrigeración, únicamente disminuyó el grupo de ácidos volátiles,
mientras que el resto aumentaron, con un fuerte incremento de compuestos azufrados
que se dio en el queso control. Todos los tratamientos de altas presiones consiguieron
reducir los niveles de ésteres, aldehídos y especialmente de compuestos azufrados. Los
tratamientos de 600 MPa redujeron los niveles de alcoholes, mientras que los
tratamientos aplicados a las 3 semanas redujeron los niveles de ácidos volátiles.
Mediante el análisis de componentes principales de los grupos de ácidos grasos y los
grupos de compuestos volátiles se vio que los quesos tratados a las 5 semanas de
maduración mantuvieron hasta el día 240 un perfil similar al del queso control de 60
días.
Durante el almacenamiento en refrigeración descendieron los parámetros de

firmeza y elasticidad en el queso control. Ninguno de los tratamientos consiguió
estabilizar ni mantener valores similares a los del queso control de 60 días. En el queso
control se produjo un incremento de la intensidad de sabor y olor así como de sabor
amargo y de olor a pútrido y rancio durante el almacenamiento, provocando un fuerte
descenso de la calidad de sabor y olor. Todos los tratamientos redujeron el incremento
de la intensidad de olor pero únicamente los de 600 MPa consiguieron reducir la
intensidad de sabor. Las altas presiones frenaron también el incremento de olor pútrido
y rancio, evitando la fuerte pérdida de calidad de sabor y olor que se dio en el queso
control.
299

Resumen ampliado
- Queso Brie
El queso control sufrió una reducción de los niveles de bacterias lácticas y bacterias
mesófilas totales desde 8,99 y 8,75 log ufc/g respectivamente a día 1, hasta 7,88 y 7,90
log ufc/g a día 120, mientras que los niveles de P. camemberti se mantuvieron estables
a lo largo de todo el periodo de estudio. Los tratamientos de altas presiones redujeron
los niveles de bacterias lácticas, bacterias mesófilas totales y P. camemberti, en especial
los tratamientos de 600 MPa, que redujeron los niveles de P. camemberti por debajo
del límite de detección. El pH del queso control sufrió un drástico aumento a partir del
día 30, alcanzando valores de 7,87 a día 120. Con los tratamientos de altas presiones se
consiguió que el pH se mantuviera en valores estables durante todo el almacenamiento
en refrigeración.
La proteolisis primaria que se dio en el queso control, con reducciones de las αS-, β-
y κ-caseínas del 41, 37 y 36 %, respectivamente, entre los días 1 y 14, únicamente se vio
reducida por los tratamientos de 600 MPa, que frenaron la hidrólisis de αS- y β-caseína.
Todos los tratamientos evitaron el fuerte incremento de péptidos hidrófobos que se dio
en el queso control. Los tratamientos de altas presiones lograron reducir la actividad
aminopeptidásica durante todo el almacenamiento en refrigeración, excepto el de 400
MPa aplicado a las 3 semanas que únicamente consiguió valores más bajos que el
queso control a día 120. Los niveles de aminoácidos libres del queso control
aumentaron durante la maduración y el almacenamiento en refrigeración, llegando a
alcanzar una concentración de 78,24 mg/g ES a día 120. Las altas presiones no frenaron
la proteolisis secundaria, sino que causaron una mayor acumulación de aminoácidos
libres, así como una mayor proteolisis global.
La actividad estearasa en el queso control aumentó durante el almacenamiento. Las

altas presiones no consiguieron evitar este aumento y en algunos casos la aumentaron.
Los ácidos grasos libres aumentaron en el queso control durante la maduración y el
almacenamiento en refrigeración, llegando a alcanzar concentraciones de 3,48, 5,03 y
21,58 mg/g ES de ácidos grasos libres de cadena corta, media y larga respectivamente,
a día 120. Las altas presiones redujeron la formación de ácidos grasos libres de cadena
corta, y los tratamientos aplicados a las 2 semanas redujeron además los niveles de
ácidos grasos libres de cadena media y larga, mientras que los tratamientos aplicados a
300

Resumen ampliado
las 3 semanas no afectaron a los ácidos grasos de cadena media hasta el día 60 y
provocaron un aumento de los de cadena larga, reduciendo posteriormente los niveles
de estos dos grupos hasta niveles similares o inferiores a los del queso control de 30
días. Los niveles de ácidos etanoico y de cadena ramificada aumentaron hasta 0,40 y
1,03 mg/g ES, respectivamente, a día 120. Las altas presiones redujeron la formación de
los ácidos libres de cadena ramificada pero únicamente los tratamientos de 600 MPa
redujeron la formación de ácido etanoico.
El grupo de compuestos volátiles más abundante en el control a día 30 fue el de los

ácidos, mientras que a día 60 y 120 fueron las cetonas, que sufrieron un descenso entre
los días 30 y 60, aumentando posteriormente. Los niveles de alcoholes, aldehídos,
compuestos bencénicos, nitrogenados y azufrados aumentaron durante el
almacenamiento en refrigeración. Las altas presiones evitaron el incremento de cetonas,
compuestos bencénicos, nitrogenados y azufrados, y provocaron una mayor
acumulación de ácidos volátiles, alcoholes, éteres y ésteres. Mediante el análisis de
componentes principales de los compuestos volátiles individuales se vio que los quesos
tratados a las 2 semanas mantuvieron hasta el día 120 un perfil de compuestos volátiles
similar al del queso control de 30 días.
Los valores de fracturabilidad, firmeza y elasticidad descendieron durante la

maduración, manteniéndose en valores muy bajos durante el almacenamiento en
refrigeración. Las altas presiones evitaron el descenso de elasticidad y firmeza, y
únicamente los tratamientos aplicados a las 2 semanas frenaron el descenso de la
fracturabilidad. En el queso control se dio un ligero aumento del parámetro de color b*
(tendencia al amarillo) tanto en la superficie como en el interior, un fuerte aumento
entre los días 90 y 120 del parámetro a* (tendencia al rojo) y una disminución de la L*
(luminosidad). Las altas presiones frenaron el aumento de b* y a* del interior y evitaron
la pérdida de luminosidad del interior del queso. Las altas presiones provocaron la
pérdida de la capa superficial de moho característica de esta variedad de queso,
haciendo que los parámetros de color de la superficie fuesen muy diferentes de los del
queso control. Durante el almacenamiento en refrigeración se produjo una pérdida de
calidad de olor y sabor acompañada de un fuerte incremento de intensidad de olor y
sabor, así como de los sabores amargo y umami. Las altas presiones evitaron el
301

Resumen ampliado
aumento de intensidad de sabor y olor y de sabor amargo, evitando la pérdida de

calidad de sabor y olor, que se mantuvieron en niveles similares a los del queso control
de 30 días.
- Arzúa-Ulloa
El queso control sufrió una reducción de los niveles de bacterias mesófilas totales
desde 9,2 log ufc/g a día 1 hasta 8,0 log ufc/g a día 21, permaneciendo en los mismos
niveles hasta el día 240. Los niveles de bacterias lácticas descendieron desde 8,8 log
ufc/g a día 1 hasta 8,0 log ufc/g a día 14, manteniéndose en los mismos valores hasta el
día 240. Por otro lado los niveles de lactobacilos aumentaron, desde 4,1 log ufc/g a día
1 hasta 8,0 log ufc/g a día 60, no variando posteriormente. Los tratamientos de 600
MPa consiguieron reducir y mantener durante el almacenamiento en refrigeración los
niveles de bacterias mesófilas totales, lácticas y lactobacilos. Los niveles de enterococos,
Micrococcaceae, coliformes y bacterias gram negativas descendieron desde 6,31, 6,92,
7,50 y 7,52 log ufc/g respectivamente a día 1, hasta 4,71, 4,08, 1,46 y 2,04 log ufc/g a
día 240, mientras que los niveles de estafilococos coagulasa positivos descendieron
desde 4,83 log ufc/g a día 1 hasta 2,66 log ufc/g a día 14, no volviéndose a detectar a
partir del día 21. Todos los tratamientos de altas presiones consiguieron reducir los
niveles de enterococos, Micrococcaceae, coliformes, gram negativas y estafilococos,
manteniendo por debajo del límite de detección los niveles de coliformes, gram
negativas y estafilococos. Para los enterococos este efecto sólo se consiguió con los
tratamientos de 600 MPa. El pH del queso control aumentó levemente durante el
almacenamiento en refrigeración, alcanzando un valor de 5,68 a día 240. Los
tratamientos de 600 MPa frenaron este aumento manteniendo valores similares a los
del queso control de 60 días.
El queso control sufrió una progresiva proteolisis primaria, con reducciones de los
niveles de las αS-, β-, κ- y p-κ-caseínas del 46, 64, 65 y 35 %, respectivamente, entre los
días 1 y 240. Las altas presiones no sólo no frenaron la proteolisis primaria sino que la
aceleraron, provocando una mayor hidrólisis de β-, κ- y p-κ-caseínas. Los tratamientos
de 400 MPa aumentaron además la hidrólisis de αS-caseína. Esta mayor hidrólisis de
caseínas en los quesos tratados se vio reflejada en una mayor acumulación de péptidos
hidrófilos e hidrófobos que en el queso control. Todos los tratamientos de altas
302

Resumen ampliado
presiones consiguieron reducir la actividad aminopeptidasa. La concentración de

aminoácidos libres del queso control aumentó a lo largo de la maduración y del
almacenamiento en refrigeración, llegando a alcanzar niveles de 31,68 mg/g ES. Las
altas presiones consiguieron frenar la proteolisis secundaria, reduciendo los niveles de
aminoácidos libres y la proteolisis global, y manteniendo los quesos tratados con 600
MPa niveles similares a los del queso control de 60 días hasta el final del
La actividad tirosina descarboxilasa aumentó en el queso control a lo largo de la

maduración y el almacenamiento en refrigeración, reduciéndose con los tratamientos
de altas presiones, excepto con el tratamiento de 400 MPa aplicado a las 3 semanas en
los días 180 y 240. En el queso control la tiramina y la putrescina aumentaron durante la
maduración y el almacenamiento en refrigeración, llegando a alcanzar concentraciones
de 0,372 y 0,207 mg/g ES, respectivamente, a día 240, mientras que la cadaverina
aumentó hasta 0,113 mg/g ES a día 21, descendiendo posteriormente y manteniéndose
en valores de 0,800 mg/g ES hasta el día 240. La histamina no se detectó hasta el día
180, con una concentración de 0,009 mg/g ES que a día 240 aumentó hasta 0,026 mg/g
ES. Las altas presiones redujeron los niveles de tiramina y putrescina, manteniendo los
tratamientos de 600 MPa niveles no detectables. Las altas presiones consiguieron
además reducir los niveles de histamina, mientras que los de cadaverina no
mantuvieron una tendencia clara a lo largo del almacenamiento en refrigeración.
La actividad esterasa, que aumentó en el queso control a lo largo del tiempo, se vio
reducida por los tratamientos de altas presiones. Durante la maduración y el
almacenamiento en refrigeración, en el queso control se produjo un aumento de ácidos
grasos libres de cadena corta, media y larga y de ácido etanoico, hasta concentraciones
de 0,190, 0,400, 1,307 y 2,102 mg/g ES, respectivamente, a día 240. Las altas presiones
redujeron la formación de ácido etanoico y ácidos grasos libres de cadena corta, pero
no de los de cadena media y larga, que únicamente mostraron niveles menores a día
240 en los quesos tratados con 600 MPa y los tratados con 400 MPa a las 3 semanas. En
el queso control se observó un aumento de ácido benzoico que llegó a alcanzar una
concentración de 0,398 mg/g ES a día 180. Únicamente se vio una reducción de los
niveles de este ácido a día 180 en los quesos tratados a las 2 semanas.
303

Resumen ampliado
El grupo de compuestos volátiles mayoritario en el queso control fue el de los

alcoholes, que aumentó entre los días 60 y 120, descendiendo posteriormente. Los
niveles de ácidos, cetonas, ésteres, éteres, hidrocarburos, compuestos bencénicos y
azufrados descendieron durante el almacenamiento en refrigeración, mientras que los
aldehídos se mantuvieron en los niveles estables y los terpenos aumentaron. Las altas
presiones evitaron parcialmente el descenso de ácidos volátiles, éteres y compuestos
azufrados. Además, en los quesos tratados con 600 MPa se evitó el descenso de
cetonas e hidrocarburos. Mediante el análisis de componentes principales de los
distintos grupos de compuestos volátiles se vio que los quesos tratados con 600 MPa
mantuvieron hasta los 120 días un perfil similar al del queso control de 21 días.
En el queso control la elasticidad y la firmeza aumentaron hasta el día 180, mientras

que la fracturabilidad únicamente pudo determinarse a día 240. Las altas presiones
aumentaron la fracturabilidad y no lograron evitar el aumento de elasticidad y firmeza.
La intensidad de sabor aumentó durante el almacenamiento en refrigeración, al igual
que aumentaron el sabor amargo y umami, mientras que la calidad de sabor se
mantuvo hasta día 180, descendiendo levemente a día 240. Las altas presiones no
lograron frenar el aumento de la intensidad de sabor, ni la disminución de la calidad de
sabor que se dio a los 240 días, sino que por el contrario favorecieron el aumento de
sabor amargo.
 Conclusiones
En el queso azul, el queso control experimentó cambios químicos menores durante

el almacenamiento en refrigeración que no afectaron a la calidad ni a la intensidad de
sabor. Los tratamientos de altas presiones de 600 MPa y el de 400 MPa aplicado a las 3
semanas consiguieron reducir la proteolisis secundaria y, además, los tratamientos
aplicados a las 3 semanas consiguieron reducir la lipolisis. El efecto de las altas
presiones sobre la proteolisis y la lipolisis junto al efecto que ejercieron en la formación
de compuestos volátiles, únicamente provocó una pérdida de la calidad e intensidad de
sabor en los quesos tratados con 600 MPa a las 3 semanas, mientras que en el resto de
quesos no supuso ninguna diferencia. De modo que las altas presiones no representan
un beneficio adicional en la conservación del queso azul bajo las condiciones
empleadas en este trabajo.
304

Resumen ampliado
En la Torta del Casar, el queso control experimentó fuertes cambios durante el

almacenamiento en refrigeración que provocaron una pérdida de la calidad de sabor y
olor, además de producirse una gran acumulación de aminas biógenas. Los
tratamientos de altas presiones evitaron el aumento de pH que se dio en el queso
control y además, los tratamientos de 600 MPa redujeron la proteolisis. Las altas
presiones consiguieron reducir la lipolisis, así como la formación de aminas biógenas y
la formación de compuestos volátiles responsables de la pérdida de calidad que se dio
en el queso control. De modo que las altas presiones, en especial los tratamientos
aplicados a las 5 semanas, supusieron una mejora sustancial para la conservación de la
Torta del Casar, alargando su vida útil hasta 240 días, además de reducir la acumulación
de aminas biógenas.
En el queso Brie, el queso control experimentó marcados cambios químicos

durante el almacenamiento en refrigeración que provocaron una fuerte pérdida de
calidad, así como un notable incremento de la intensidad de sabor y olor. Los
tratamientos de altas presiones evitaron el fuerte incremento de pH que se dio en el
control, además de reducir la proteolisis primaria, evitando la acumulación de péptidos
hidrófobos. Las altas presiones también consiguieron frenar la lipolisis, así como la
formación de compuestos volátiles responsables de la pérdida de calidad del queso
control, principalmente compuestos azufrados. Por otro lado, los tratamientos de altas
presiones provocaron la pérdida de la capa superficial de moho. De modo que las altas
presiones, en especial los tratamientos aplicados a las 2 semanas, supusieron una
mejora sustancial para la conservación del queso Brie, aumentando su vida útil hasta
120 días, aunque la pérdida de la capa superficial de moho supone una limitación a la
hora de su comercialización.
En el queso Arzúa-Ulloa, el queso control experimentó cambios químicos durante

el almacenamiento en refrigeración que únicamente provocaron la pérdida de calidad
de sabor en el día 240, aunque se dio un aumento de intensidad de sabor y olor, así
como de sabor amargo y umami, a lo largo de este periodo. Los tratamientos de altas
presiones redujeron el aumento de pH que tuvo lugar en el queso control. A pesar de
acelerar la proteolisis primaria, los tratamientos consiguieron frenar la proteolisis
secundaria. Las altas presiones disminuyeron levemente la lipolisis, especialmente la
305

Resumen ampliado
formación de ácidos grasos libres de cadena corta, y redujeron la formación de aminas

biógenas, en especial los tratamientos de 600 MPa. El efecto de las altas presiones
sobre los compuestos volátiles hizo que los quesos tratados con 600 MPa mantuviesen
hasta el día 120 un perfil de volátiles similar al del queso control de 21 días. Las altas
presiones no influyeron sobre la calidad ni la intensidad de sabor del queso, a pesar de
que provocaron un aumento del sabor amargo. De modo que las altas presiones no
representaron una mejora de cara a la conservación del queso desde el punto de vista
de sus características sensoriales, aunque sí que lograron evitar la acumulación de
aminas biógenas.
306

Resumen ampliado
 Bibliografía
13, 841-866.
EUROSTAT (2014).
http://epp.eurostat.ec.europa.eu/portal/page/portal/eurostat/home/
110, 248-253.
FAOSTAT (2014). http://faostat3.fao.org/faostat-gateway/go/to/home/E
499.
López-Pedemonte, T., Roig-Sagués, A. X., De Lamo, S., Gervilla, R. & Guamis, B.
(2007). High hydrostatic pressure treatment applied to model cheeses made from
cow's milk inoculated with Staphylococcus aureus. Food Control 18, 441-447.
Academic Press.
307

Resumen ampliado
Novella-Rodríguez, S., Veciana-Nogués, M. T., Saldo, J. & Vidal-Carou, M. C. (2002).

Effects of high hydrostatic pressure treatments on biogenic amine contents in goat
cheeses during ripening. Journal of Agricultural and Food Chemistry 50, 7288-7292.
O’Reilly, C. E., O’Connor, P. M., Murphy, P. M., Kelly, A. L. & Beresford, T. P. (2000).
The effect of exposure to pressure of 50 MPa on Cheddar cheese ripening.
Rodriguez, E., Arques, J. L., Nuñez, M., Gaya, P. & Medina, M. (2005). Combined
effect of high-pressure treatments and bacteriocin-producing lactic acid bacteria on
inactivation of Escherichia coli O157:H7 in raw-milk cheese. Applied and
Environmental Microbiology 71, 3399-3404.
Saldo, J., McSweeney, P. L. H., Sendra, E., Kelly, A. L. & Guamis, B. (2002).
Proteolysis in caprine milk cheese treated by high pressure to accelerate cheese
ripening. International Dairy Journal 12, 35-44.
Sousa, M. J., Ardö, Y. & McSweeney, P. L. H. (2001). Advances in the study of
Yang, B. W., Shi, Y., Xia, X. D., Xi, M. L., Wang, X., Ji, B. Y. & Meng, J. H. (2012).
Inactivation of foodborne pathogens in raw milk using high hydrostatic pressure.
Food Control 28, 273-278.
308

Capítulo 14. Extended abstract.

309

Fotografía: cuñas de los quesos estudiados, determinación de pH, liofilizado de
péptidos, colonias de P. roqueforti, equipo de altas presiones y extracción de ácidos
grasos libres mediante separación en fase solida.
Extended abstract
 Introduction
Cheese is a dairy product known since ancient times that has evolved to give rise to
the hundreds of varieties that are currently manufactured. According to FAO, world
cheese production in 2012 was 20.6 million tons, making cheese one of the main food
products (FAOSTAT, 2014). In the EU cheese production reached 8.7 million tonnes in
2013, out of which only 1.66 % was made in Spain (EUROSTAT, 2014). The averaged per
capita consumption in Spain in 2013 was estimated in 8.2 kg/person per year, still far
from the European average of 17.2 kg/person per year (MAGRAMA, 2014).
The transformation of milk into cheese comprises different stages, out of which
ripening is one of the most important, since organoleptic characteristics, texture and
typical appearance of each variety develop during this phase. These characteristics are
achieved by a series of reactions which can be grouped as glycolysis and catabolism of
lactate and citrate, proteolysis and amino acid catabolism, and lipolysis and free fatty
acids catabolism. Lactose fermentation by the microorganisms present in cheese,
especially lactic bacteria, results in the formation of lactic acid and minor amounts of
acetic and propionic acids, diacetyl, acetoin, ethanol and other compounds that
contribute to cheese flavour and aroma. The residual citrate, retained in the curd, can be
metabolized by mesophilic starter cultures, yielding diacetyl, acetate, acetoin, 2,3-
butanediol and CO2 (McSweeney & Fox, 2004). Triglyceride hydrolysis by esterases and
lipases from milk or microorganisms causes the release of fatty acids, which contribute
directly to aroma and flavour, especially the short and medium chain free fatty acids,
and are also precursors of other compounds responsible for aroma and flavour (Collins
et al., 2003). Hydrolysis of casein by native milk proteases, rennet or proteolytic systems
of microorganisms, leads to the generation of large peptides, which are further
hydrolyzed into smaller peptides and finally, into free amino acids. Both, peptides and
free amino acids, contribute to the flavour of cheese, being amino acids also precursors
of flavour and aroma compounds (Sousa et al., 2001, Marilley & Casey, 2004).
Once cheese reaches its optimum ripening time, at which the compounds
responsible for flavour and aroma reach an equilibrium which confers the product its
characteristic flavour (Mulder, 1952), ripening continues during storage, distribution and
sale, causing the cheese to reach the consumer with a stronger or different flavour than
311

Extended abstract
the manufacturer intended (Wick et al., 2004). This phenomenon, known as over-
ripening, mainly occurs in some cheese varieties which undergo an intense proteolysis,
due to the use of coagulating enzymes of high proteolytic activity as cardosins or
microorganisms with high proteolytic activity as moulds from the genus Penicillium, and
it limits the shelf-life of product. There is hardly any research leading to avoid this
phenomenon, just the use of refrigeration temperatures, which allows to slow it down
but not to stop it.
The presence in cheese of microorganisms with decarboxylase activity poses

another problem, the presence of biogenic amines. Biogenic amines are nitrogenous
basic compounds of low molecular weight with biological activity. Their synthesis and
degradation is part of the normal metabolism of animals, plants and microorganisms
(ten Brink et al., 1990). Biogenic amines can be aliphatic (cadaverine, putrescine,
spermine and spermidine), aromatic (tyramine and β-phenylethylamine) or heterocyclic
(histamine and tryptamine) and can be classified by the number of amino groups as
mono, di, or polyamines (Silla Santos, 1996). Although biogenic amines are involved in
normal metabolism and regulate various essential functions in the body, the intake of
high levels of biogenic amines may have toxic effects such as increased blood pressure,
headache, abdominal cramps and urticaria. In mammals, there is an efficient
detoxification system in the intestinal tract and under normal conditions exogenous
biogenic amines are rapidly detoxified by the action of amine oxidases. However, in
individuals with a deficient detoxification system or under treatment with amine
oxidases inhibitory drugs, small amounts of biogenic amines can induce episodes of
intoxication that can become serious (Ladero et al., 2010).
Biogenic amines are formed by decarboxylation of amino acids or by

transamination of aldehydes and ketones. Cheese is an ideal substrate for amines
formation, due to its high content of free amino acids and because it may contain
microorganisms with decarboxylase activity, and it presents favourable conditions for
the growth of these microorganisms and for the formation of amines (Loizzo et al.,
2013).
High hydrostatic pressure is currently used as a non-thermal pasteurization

procedure, which improves the nutritional characteristics of foods over thermal
312

Extended abstract
pasteurization treatments. Besides of the fact that the pressure is rapidly and
homogeneously transferred to the entire product, another advantage offered by high
pressure over thermal treatments is the possibility of being directly applied to the
elaborated product, preventing further contamination of the food interior. Consumers
demand more natural and minimally processed foods, and high pressure processing
represents an appropriate technology to meet this demand. Currently, there are many
products treated by high pressure in the market, such as juices and beverages, meat
and dairy products, prepared meals, sandwich fillings, fish and shellfish, sauces, avocado
and guacamole. High pressure treatments have been successfully applied to milk and
cheese for elimination of foodborne pathogens (Rodriguez et al., 2005, López-
Pedemonte et al., 2007, Yang et al., 2012). High pressure treatments have also been
used in order to accelerate ripening in different cheese varieties, causing lysis of lactic
acid bacteria and releasing enzymes into the cheese matrix (O’Reilly et al., 2000, Saldo
et al., 2002, Ávila et al., 2006). But high pressure treatments have been used to extend
shelf life only in fresh cheese (Evert-Arriagada et al., 2012, Evert-Arriagada et al., 2014).
In addition to the ability of high pressure treatments to eliminate microorganisms, they
can also be used to inactivate enzymes and might avoid the phenomenon of over-
ripening as well as the formation of biogenic amines in cheese (Novella-Rodríguez et al.,
2002, Malone et al., 2003, Huppertz et al., 2004).
 Objetive
The aim of this study was to evaluate the effect of high hydrostatic pressure
treatments on different cheese varieties throughout a prolonged period of refrigerated
storage, in order to avoid over-ripening, extending shelf-life and preventing the
formation and accumulation of biogenic amines.
 Material and methods
High pressure treatments were applied on four cheese varieties at different times of
ripening, that were selected according to each variety. In blue cheese, made from
pasteurized ewe’s milk and ripened with Penicillium roqueforti in the interior, 400 and
600 MPa were applied for 5 minutes at 3 (400-3W and 600-3W), 6 (400-6W and 600-
6W) and 9 weeks (400-9W and 600-9W) of ripening, and analyses were performed
immediately after treatment and at 3, 6, 9 and 12 months. In Torta del Casar, made from
313

Extended abstract
raw ewe’s milk and vegetable rennet from cardoon (Cynara cardunculus), 400 and 600
MPa were applied for 5 minutes at 3 (400-3W and 600-3W), and 5 weeks (400-5W and
600-5W) of ripening, and analyses were performed immediately after treatment and at
2, 4, 6 and 8 months. In Brie cheese, made from pasteurized cow’s milk and ripened with
P. camemberti on its surface, 400 and 600 MPa were applied for 5 minutes at 2 (400-2W
and 600-2W) and 3 weeks (400-3W and 600-3W) of ripening, and analyses were
performed immediately after treatment and at 1, 2, 3 and 4 months. In Arzúa-Ulloa
cheese, made from raw cow's milk, 400 and 600 MPa were applied for 5 minutes at 2
(400-2W and 600-2W) and 3 weeks (400-3W and 600-3W) of ripening, and analyses
were performed immediately after treatment and at 2, 4, 6 and 8 months.
To evaluate the effect of high pressure on cheese microbiota, microbiological

analyses were performed by counting on agar plates the following microbial groups:
total aerobic mesophilic bacteria (PCA), lactic acid bacteria (M17 and MRS agar),
lactobacilli (Rogosa agar), enterococci (KF agar), gram negative bacteria (MacConkey
agar), coliforms (VRBA), Micrococcaceae (MSA), coagulase-positive staphylococci (Baird-
Parker + RPF II), Listeria monocytogenes (Palcam agar), moulds and yeasts (CGA). The
pH was measured by using a pH-meter coupled to a penetration electrode and dry
matter (DM) was determined by drying at 102 °C to constant weight. Proteolysis was
evaluated by analyses of caseins by capillary gel electrophoresis, peptides and free
amino acids by liquid chromatography (HPLC) and overall proteolysis by a
spectrophotometric assay. Aminopeptidase activity was determined by a colorimetric
assay, measured spectrophotometrically. Lipolysis was evaluated by determining the
free fatty acids by solid phase extraction followed by gas chromatography. Esterase
activity was determined by colorimetric assay, measured spectrophotometrically. To
evaluate the effect of high pressure on the formation of biogenic amines, tyrosine
decarboxylase activity and biogenic amines concentration were determined by HPLC.
Volatile compounds were analysed by solid phase microextraction (SPME) followed by
gas chromatography-mass spectrometry (GC-MS). Texture parameters (fracturability,
firmness and elasticity) were determined by using a compression texturometer and the
colorimetric parameters (lightness: L*, redness: a* and yellowness: b*) were estimated by
314

Extended abstract
using a colorimeter. Finally, sensory analyses of flavour and odour were carried out
through the evaluation of different characteristics and descriptors by a trained panel.
 Results
- Blue cheese
Control cheese, non-treated, underwent a reduction in the counts of lactic acid

bacteria and mesophilic bacteria from 9.54 and 8.76 log cfu/g on day 1 to 6.57 and 5.57
log cfu/g on day 360, respectively, while P. roqueforti counts declined from 7.73 log
cfu/g on day 21 to 5.66 log cfu/g on day 360. High pressure treatments reduced the
counts of lactic acid bacteria, total mesophilic bacteria and P. roqueforti, especially the
600 MPa treatments which reduced P. roqueforti population below the detection level.
The strong primary proteolysis occurring in control cheese, with reductions

between days 1 and 21 of 93, 87, 78 and 46 %, for αS-, β-, κ- and p-κ-caseins,
respectively, could not be avoided by high pressure treatments. The levels of
hydrophilic and hydrophobic peptides increased in control cheese, and high pressure
treatments were only able to reduce the hydrophilic peptides with 600 MPa treatments
applied at 3 and 6 weeks. The aminopeptidase activity decreased in control cheese
during refrigerated storage, and high pressure caused the reduction of this activity. Free
amino acids levels in control cheese increased during ripening and refrigerated storage,
reaching a concentration of 116.76 mg/g DM on day 360, and only 600 MPa treatments
and 400 MPa treatment applied at 3 weeks were able to reduce the formation of free
amino acids, maintaining until day 360 levels of free amino acids and overall proteolysis
similar to those of control cheese at 180 days.
Biogenic amines in the control cheese reached on day 360 concentrations of 0.06,
0.07, 0.05, 0.03 and 0.01 mg/g DM for β-phenylethylamine, tyramine, tryptamine,
putrescine and spermidine, respectively. High pressure treatments only reduced the
levels of tyramine.
Esterase activity was reduced by 600 MPa treatments and 400 MPa treatment
applied at 3 weeks. Free fatty acids in the control cheese increased during ripening and
refrigerated storage, reaching on day 360 concentrations of 8.88, 18.31 and 56.63 mg/g
DM of short (C4:0 - C8:0), medium (C10:0 - C14:0) and long chain (C16:0 - C18:3) free fatty acids,
315

Extended abstract
respectively. Excepting the treatments at 400 MPa at 6 and 9 weeks, high pressure
treatments reduced the formation of all the groups of free fatty acids. Only cheeses
treated at 400 MPa at 3 weeks showed levels of short chain free fatty acids similar or
lower than those of control cheese at 180 days. These cheeses, together with those
treated with 600 MPa at 6 and 9 weeks, showed levels of medium and long chain fatty
acids similar to those of control cheese at 180 days. Moreover, the levels of the three
groups of free fatty acids in cheeses treated with 600 MPa at 3 weeks remained well
below those of control cheese at 180 days. Ethanoic acid increased until day 63,
decreasing thereafter in control cheese. High pressure treatments reduced the levels of
this acid, with the exception of 400 MPa applied at 6 and 9 weeks.
The major group of volatile compounds in control cheese was the volatile acids
group, which increased during refrigerated storage, followed by alcohols and ketones,
which also increased during storage but to a lesser extent. Levels of esters, terpenes,
sulphur and nitrogen compounds also increased during storage, while aldehydes and
benzenic compounds declined and hydrocarbons remained stable. High pressure
treatments reduced the levels of esters, alcohols, hydrocarbons, nitrogen and benzenic
compounds, especially the 600 MPa treatments which also reduced the formation of
ketones. In cheeses treated at 3 weeks the levels of volatile acids, sulphur compounds
and terpenes were reduced by high pressure treatments. Principal component analysis
of free fatty acids groups and volatile compounds groups only scarcely discriminated
between times of storage, indicating a very mild effect of high pressures on blue
cheese.
Only the treatment of 600 MPa applied at 3 weeks reduced the slight increase in
flavour intensity that occurred in control cheese during refrigerated storage. Control
cheese showed stable values for flavour quality during refrigerated storage, and all
treated cheeses showed similar scores to those of the control cheese, with the
exception of cheeses treated with 600 MPa at 3 weeks, which obtained lower scores.
Control cheese showed subsidence at the centre that reached 15 % with respect to
the edge. High pressure greatly increased it, reaching 36 % in the case of cheeses
treated with 400 MPa at 3 weeks.
316

Extended abstract
- Torta del Casar
There was a decrease of mesophilic bacteria, lactic acid bacteria and enterococci
counts in control cheese from 9.51, 9.46 and 7.32 log cfu/g on day 21 to 8.61, 8.42 and
6.80 log cfu/g on day 240, respectively. Lactobacilli counts increased to 8.13 log cfu/g
on day 35 and remained at these levels throughout the refrigerated storage. High
pressure treatments reduced the counts of mesophilic bacteria, lactic acid bacteria,
lactobacilli and enterococci, especially 600 MPa treatments. Micrococcaceae, coliform
and gram negative bacteria counts decreased from 6.39, 5.52 and 6.01 log cfu/g on day
1 to 5.57, 2.87 and 5.37 log cfu/g on day 240, respectively. High pressure reduced
Micrococcaceae counts, especially with treatments applied at 5 weeks, whereas all
treatments, with the exception of 400 MPa applied at 3 weeks, reduced coliforms and
gram negative bacteria counts below the detection level. Coagulase positive
staphylococci counts decreased in control cheese from 5.23 log cfu/g on day 1 to 2.30
on day 180 and below the detection level on day 240, whereas in pressurized cheeses,
they were below the detection level since day 21. In control cheese pH increased during
ripening and refrigerated storage reaching values of 6.13, and in treated cheeses pH
also increased but to a lesser extent.
Control cheese αS- and β-caseins suffered reductions of 79 and 62 %, respectively,

between days 1 and 35, while κ-casein suffered a reduction of 96 % between days 1 and
21, remaining below the detection level from day 35 onwards. The 600 MPa treatments
successfully arrested primary proteolysis reducing αS- and β-caseins hydrolysis and the
formation of hydrophilic peptides, which increased in control cheese. Aminopeptidase
activity decreased more rapidly in high pressure treated cheeses than in control cheese.
Free amino acids concentration increased in control cheese during ripening and
refrigerated storage, reaching 23.30 mg/g DM on day 360. The 600 MPa treatments
reduced the overall proteolysis and in addition, the 600 MPa applied at 3 weeks
treatment reduced the formation of free amino acids, but for both parameters their
levels were higher than in the 60-day control cheese.
Tyrosine decarboxylase activity tripled in control cheese between days 60 and 180,
and all high pressure treatments reduced it. Tyramine increased in control cheese
reaching 1.09 mg/g DM on day 240, and 600 MPa treatments lowered its levels.
317

Extended abstract
Putrescine, tryptamine, β-phenylethylamine, cadaverine and histamine increased to 0.95,

0.83, 0.29, 0.69 and 0.30 mg/g DM, respectively, on day 240. All high pressure
treatments reduced cadaverine levels. The levels of putrescine, tryptamine and β-
phenylethylamine were similarly reduced by high pressure treatments, with the
exception of 400 MPa at 3 weeks. Only 600 MPa treatments reduced the levels of
histamine.
Esterase activity in control cheese decreased since day 60. The 600 MPa treatments
reduced this activity throughout refrigerated storage. The levels of free fatty acids in
control cheese increased throughout ripening and refrigerated storage reaching
concentrations of short, medium and long chain free fatty acids of 8.69, 0.88 and 3.02
mg/g DM, respectively, on day 240. All treatments reduced the formation of free fatty
acids, and only short chain free fatty acids levels remained in treated cheeses until day
240 at similar levels to those of control cheese on day 60. Ethanoic, propanoic and
branched chain acids also increased, reaching 5.82, 2.99 and 1.15 mg/g DM,
respectively. All treatments reduced the levels of propanoic and branched-chain acids
and only the treatments applied at 3 weeks reduced the levels of ethanoic acid,
maintaining throughout refrigerated storage similar or lower levels than those of
control cheese on day 60.
The major group of volatile compounds in control cheese was the volatile acids
group until day 180, while on day 240 it was the group of alcohols. During refrigerated
storage, only the volatile acids group decreased, while the rest of groups increased, with
a marked increase of sulphur compounds in control cheese. High pressure treatments
reduced the levels of esters, aldehydes, and especially sulphur compounds. The 600
MPa treatments reduced the levels of alcohols, while treatments applied at 3 weeks
reduced the levels of volatile acids. Principal component analysis of free fatty acids
groups and volatile compounds groups indicated that cheeses treated at 5 weeks of
ripening showed until day 240 a similar profile to that of the 60-day control cheese.
During refrigerated storage, firmness and elasticity parameters decreased in control

cheese. None of the treatments was able to stabilize or maintain similar values to those
of control cheese on day 60. In control cheese flavour and odour intensity, bitter taste
and putrid and rancid odour increased during storage, causing a marked decline in
318

Extended abstract
flavour and odour quality. All treatments reduced the increase in odour intensity but
only 600 MPa treatments were able to reduce flavour intensity. High pressure also
slowed the increase in putrid and rancid odour, preventing the drstic loss of flavour and
odour quality that occurred in control cheese.
- Brie Cheese
In control cheese, lactic acid bacteria and mesophilic bacteria counts decreased
from 8.99 and 8.75 log cfu/g on day 1 to 7.88 and 7.90 log cfu/g on day 120,
respectively, while P. camemberti counts remained stable throughout storage. High
pressure treatments reduced counts of lactic acid bacteria, mesophilic bacteria and P.
camemberti, especially 600 MPa treatments which reduced P. camemberti counts below
the detection level. The pH of control cheese increased drastically from day 30 onwards,
reaching values of 7.87 at day 120. High pressure treatments maintained the same pH
values throughout refrigerated storage.
Primary proteolysis of control cheese, with reductions between days 1 and 14 of 41,
37 and 36 % for αS-, β- and κ-casein, respectively, was only successfully arrested by 600
MPa treatments, which reduced the hydrolysis of αS- and β-caseins. All treatments
prevented the marked increase of hydrophobic peptides that occurred in control
cheese. High pressure treatments reduced aminopeptidase activity throughout
refrigerated storage, with the exception of 400 MPa applied at 3 weeks, which only on
day 120 yielded lower values than control cheese. Free amino acid levels in control
cheese increased during ripening and refrigerated storage, reaching a concentration of
78.24 mg/g DM on day 120. High pressure treatments did not reduce secondary
proteolysis, on the contrary, they caused a greater accumulation of free amino acids as
well as a higher overall proteolysis.
Esterase activity increased in control cheese during refrigerated storage. High

pressure treatments did not reduce this increase and in some cases they increased it.
Free fatty acids accumulated in control cheese during ripening and refrigerated storage,
reaching concentrations of 3.48, 5.03 and 21.58 mg/g DM of short, medium and long
chain free fatty acids, respectively, on day 120. High pressure treatments reduced the
formation of short chain free fatty acids, and treatments applied at 2 weeks also
reduced the levels of medium and long chain free fatty acids, while treatments applied
319

Extended abstract
at 3 weeks did not affect medium chain free fatty acids and caused an increase of long
chain free fatty acids until day 60, reducing subsequently the levels of these two groups
to similar or lower values than those of control cheese on day 30. Ethanoic and
branched chain acid levels increased to 0.40 and 1.03 mg/g DM, respectively, on day
120. High pressure treatments reduced the formation of branched chain acids, but only
600 MPa treatments reduced the formation of ethanoic acid.
The major group of volatile compounds in control cheese on day 30 was the
volatile acids group, while on days 60 and 120 it was the ketones group, which suffered
a decline between days 30 and 60, increasing thereafter. The levels of alcohols,
aldehydes, benzenic, nitrogen and sulphur compounds increased during refrigerated
storage. High pressure treatments prevented the increase of ketones, benzenic,
nitrogen and sulphur compounds, and caused a higher accumulation of volatile acids,
alcohols, ethers and esters. Principal component analysis of individual volatile
compounds indicated that cheeses treated at 2 weeks showed until day 120 a profile of
volatile compounds similar to that of 30-day control cheese.
Fracturability, firmness and elasticity values decreased during ripening, remaining at

very low values during refrigerated storage. High pressure treatments prevented the
decrease of elasticity and firmness, and only treatments applied at 2 weeks slowed the
decline of fracturability. In control cheese there was a slight increase in the colour
parameter b* on the surface and the interior, a marked increase between days 90 and
120 of the parameter a* and a decrease of lightness. High pressure slowed the increase
of a* and b* parameters in the interior and prevented the loss of lightness in the cheese
interior. High pressure treatments caused the loss of the mould surface layer that is
characteristic of this cheese variety, making the colour parameters of the surface very
different from those of control cheese. During refrigerated storage there was a loss of
flavour and odour quality, accompanied by a strong increase of flavour and odour
intensity as well as of bitter and umami flavours. High pressures prevented the increase
of flavour and odour intensity and bitter flavour, avoiding the loss of flavour and odour
quality and maintaining treated cheeses with similar scores to those of control cheese
on day 30.
320

Extended abstract
- Arzúa-Ulloa
Mesophilic bacteria counts decreased in control cheese between days 1 and 21

from 9.2 log cfu/g to 8.0 log cfu/g, and then remained at stable levels until day 240.
Lactic acid bacteria levels decreased between days 1 and 14 from 8.8 log cfu/g to 8.0
log cfu/g, remaining at the same values until day 240. Furthermore lactobacilli counts
increased between days 1 and 60 from 4.1 log cfu/g to 8.0 log cfu/g, maintaining the
same levels until day 240. The 600 MPa treatments reduced and maintained the levels
of mesophilic bacteria, lactobacilli and lactic acid bacteria throughout refrigerated
storage. Enterococci, Micrococcaceae, coliforms and gram negative bacteria counts
decreased between days 1 and 240 from 6.31, 6.92, 7.50 and 7.52 log cfu/g to 4.71, 4.08,
1.46 and 2.04 log cfu/g, respectively, while coagulase positive staphylococci counts
decreased between days 1 and 14 from 4.83 log cfu/g to 2.66 log cfu/g, and were below
the detection level from day 21 onwards. High pressure treatments reduced enterococci
Micrococcaceae, coliforms, gram negative bacteria and staphylococci counts, and kept
below the detection level the counts of coliforms, gram negative bacteria and
staphylococci. Only 600 MPa treatments maintained enterococci counts below the
detection level. The pH of control cheese slightly increased during refrigerated storage,
reaching a value of 5.68 on day 240. The 600 MPa treatments reduced this increase,
maintaining similar values to those of control cheese on day 60.
Control cheese suffered a progressive primary proteolysis, with reductions of 46, 64,
65 and 35 % for αS-, β-, κ- and p-κ-casein, respectively, between days 1 and 240. High
pressure treatments not only did not reduce primary proteolysis but increased it,
causing a higher hydrolysis of β-, κ- and p-κ-caseins. The 400 MPa treatments also
increased the hydrolysis of αS-casein. This higher hydrolysis of caseins in treated cheeses
resulted in a greater accumulation of hydrophilic and hydrophobic peptides, than in
control cheese. High pressure treatments reduced aminopeptidase activity. The
concentrations of free amino acids increased in control cheese throughout ripening and
refrigerated storage, reaching levels of 31.68 mg/g DM. High pressure treatments
slowed secondary proteolysis, reducing free amino acids and overall proteolysis level,
and showing 600 MPa treated cheeses similar levels to those of 60-day control cheese
until the end of refrigerated storage.
321

Extended abstract
Tyrosine decarboxylase activity increased in control cheese during ripening and

refrigerated storage. High pressure treatments reduced this activity, with the exception
of 400 MPa applied at 3 weeks on days 180 and 240. In control cheese, tyramine and
putrescine increased during ripening and refrigerated storage, reaching concentrations
of 0.372 and 0.207 mg/g DM on day 240 respectively, while cadaverine increased to
0.113 mg/g DM on day 21, decreasing thereafter and maintaining values of 0.800 mg/g
up to day 240. Histamine was only detected on days 180 and 240, with concentrations
of 0.009 mg/g DM and 0.026 mg/g DM respectively. High pressure treatments reduced
the levels of tyramine and putrescine, and 600 MPa treatments kept their
concentrations below the detection level. High pressure treatments also reduced
histamine levels, whereas cadaverine levels did not show a clear trend throughout
refrigerated storage.
Esterase activity, which increased in control cheese throughout storage, was

reduced by high pressure treatments. During ripening and refrigerated storage short,
medium, long chain free fatty acids and ethanoic acid increased in control cheese,
reaching concentrations of 0.190, 0.400, 1.307 and 2.102 mg/g DM on day 240,
respectively. High pressure treatments reduced the formation of ethanoic acid and short
chain free fatty acids, but not medium and long chain free fatty acids that only showed
lower levels on day 240, in cheeses treated with 400 MPa at 3 weeks and in 600 MPa
cheeses. In control cheese, benzoic acid increased reaching a concentration of 0.398
mg/g DM on day 180. High pressure treatments applied at 2 weeks reduced the levels
of benzoic acid on day 180.
The major group of volatile compounds in control cheese was the alcohols, that
increased between days 60 and 120 and decreased thereafter. The levels of acids,
ketones, esters, ethers, hydrocarbons, benzenic and sulphur compounds decreased
during refrigerated storage, while aldehydes remained at the same levels and terpenes
increased. High pressure treatments partially prevented the decrease of volatile acids,
ethers, and sulphur compounds. Furthermore, in cheeses treated with 600 MPa the
decrease of ketones and hydrocarbons was avoided. Principal component analysis of
the different volatile compound groups indicated that cheeses treated at 600 MPa
showed until day 120 a similar profile to that of 21-day control cheese.
322

Extended abstract
In control cheese elasticity and firmness increased until day 180, while fracturability
could only be determined at day 240. High pressure treatments increased fracturability
and could not avoid the increase of elasticity and firmness. Flavour intensity increased
during refrigerated storage in control cheese as well as bitter and umami flavour, while
flavour quality maintained similar scores until day 180, suffering a slightly decrease on
day 240. High pressure treatments did not avoid the increase of flavour intensity or the
reduction in flavour quality which control cheese suffered on day 240, and on the
contrary, they favoured the increase of bitter flavour.
 Conclusions
In blue cheese, control cheese underwent some minor chemical changes during
refrigerated storage, which did not affect flavour quality or intensity. High pressure
treatments of 400 MPa applied at 3 weeks and 600 MPa reduced secondary proteolysis,
and treatments applied at 3 weeks also reduced lipolysis. The effect of high pressure
treatments on proteolysis and lipolysis together with the effect exerted on the
formation of volatile compounds only caused a loss of flavour quality and intensity in
cheeses treated with 600 MPa at 3 weeks, whereas in the rest of treated cheeses it did
not cause differences. High pressure treatments did not represent an additional benefit
for preserving blue cheese under the conditions employed in this work.
In Torta del Casar, control cheese underwent strong changes during refrigerated
storage which caused a loss of flavour and odour quality, along with a considerable
accumulation of biogenic amines. High pressure treatments prevented the increase of
pH that occurred in control cheese, and treatments of 600 MPa also reduced
proteolysis. High pressure treatments were able to reduce lipolysis, biogenic amines
levels and the formation of volatile compounds, which were responsible for the quality
loss in control cheese. High pressure treatments, especially those applied at 5 weeks,
represented a substantial improvement for the conservation of Torta del Casar,
extending its shelf life up to 240 days and reducing the accumulation of biogenic
amines.
In Brie cheese, control cheese underwent chemical changes during refrigerated

storage which caused a strong loss of quality, as well as a strong increase of flavour and
odour intensity. High pressure treatments prevented the marked increase of pH that
323

Extended abstract
occurred in control cheese, and reduced the primary proteolysis impeding the
accumulation of hydrophobic peptides. High pressure treatments were also able to
reduce lipolysis and the formation of volatile compounds, mainly sulphur compounds
which were responsible for the quality loss of control cheese. Furthermore, high
pressure treatments caused the loss of the mould surface layer. High pressures,
especially treatments applied at 2 weeks, represented a substantial improvement for the
preservation of Brie cheese, extending its shelf life up to 120 days, although the loss of
the mould surface layer may hinder its marketing.
In Arzúa-Ulloa cheese, control cheese underwent chemical changes during

refrigerated storage, which only caused the loss of flavour quality on day 240, although
flavour and odour intensity, as well as bitter and umami flavour, increased throughout
this period. High pressure treatments reduced the increase of pH that occurred in
control cheese. Despite the acceleration of primary proteolysis by high pressure
treatments, they were able to reduce secondary proteolysis. High pressure treatments
slightly reduced lipolysis, especially the formation of short chain free fatty acids, and
reduced the formation of biogenic amines, especially treatments of 600 MPa. Because
of the effect of high pressure treatments on volatile compounds, 600 MPa cheeses
maintained until day 120 a similar volatile profile to that of 21-day control cheese. High
pressure treatments did not influence the flavour quality and intensity of the cheeses,
although they caused an increase of bitterness. High pressure treatments, from the
standpoint of the sensory characteristics, did not represent an improvement for cheese
preservation, but avoided the accumulation of biogenic amines.
324

Extended abstract
 Bibliography
13, 841-866.
EUROSTAT (2014).
http://epp.eurostat.ec.europa.eu/portal/page/portal/eurostat/home/
110, 248-253.
FAOSTAT (2014). http://faostat3.fao.org/faostat-gateway/go/to/home/E
499.
López-Pedemonte, T., Roig-Sagués, A. X., De Lamo, S., Gervilla, R. & Guamis, B.
(2007). High hydrostatic pressure treatment applied to model cheeses made from
cow's milk inoculated with Staphylococcus aureus. Food Control 18, 441-447.
Academic Press.
325

Extended abstract
Novella-Rodríguez, S., Veciana-Nogués, M. T., Saldo, J. & Vidal-Carou, M. C. (2002).

Effects of high hydrostatic pressure treatments on biogenic amine contents in goat
cheeses during ripening. Journal of Agricultural and Food Chemistry 50, 7288-7292.
O’Reilly, C. E., O’Connor, P. M., Murphy, P. M., Kelly, A. L. & Beresford, T. P. (2000).
The effect of exposure to pressure of 50 MPa on Cheddar cheese ripening.
Rodriguez, E., Arques, J. L., Nuñez, M., Gaya, P. & Medina, M. (2005). Combined
effect of high-pressure treatments and bacteriocin-producing lactic acid bacteria on
inactivation of Escherichia coli O157:H7 in raw-milk cheese. Applied and
Environmental Microbiology 71, 3399-3404.
Saldo, J., McSweeney, P. L. H., Sendra, E., Kelly, A. L. & Guamis, B. (2002).
Proteolysis in caprine milk cheese treated by high pressure to accelerate cheese
ripening. International Dairy Journal 12, 35-44.
Sousa, M. J., Ardö, Y. & McSweeney, P. L. H. (2001). Advances in the study of
Yang, B. W., Shi, Y., Xia, X. D., Xi, M. L., Wang, X., Ji, B. Y. & Meng, J. H. (2012).
Inactivation of foodborne pathogens in raw milk using high hydrostatic pressure.
Food Control 28, 273-278.
326

Universidad Complutense de Madrid

Cargado por

Copyright:

Formatos disponibles

Universidad Complutense de Madrid

Cargado por

Información del documento

Título original

Derechos de autor

Formatos disponibles

Compartir este documento

Compartir o incrustar documentos

Opciones para compartir

¿Le pareció útil este documento?

¿Este contenido es inapropiado?

Copyright:

Formatos disponibles

Universidad Complutense de Madrid

Cargado por

Copyright:

Formatos disponibles

UNIVERSIDAD COMPLUTENSE DE MADRID

Prevención de la sobremaduración y de la formación de aminas biógenas en

MEMORIA PARA OPTAR AL GRADO DE DOCTOR

Javier Calzada Gómez

Manuel Núñez Gutiérrez

©Javier Calzada Gómez, 2015

PREVENCIÓN DE LA SOBREMADURACIÓN Y DE LA FORMACIÓN

Javier Calzada Gómez

PREVENCIÓN DE LA SOBREMADURACIÓN Y DE LA FORMACIÓN

Director: Prof. Dr. Manuel Núñez Gutiérrez

EL DOCTORANDO VºBº DEL DIRECTOR VºBº DEL CODIRECTOR

Que la Tesis doctoral titulada “Prevención de la sobremaduración y de

Madrid, 28 de Octubre de 2014

Fdo. Manuel Núñez Gutiérrez Fdo. Ana del Olmo Sánchez

A todos, gracias, se os quiere.

“Solo hay dos cosas infinitas, el universo y la estupidez

Capítulo 2. Proteolysis and biogenic amine buildup in high-pressure treated ovine 87

Capítulo 4. Using high-pressure processing for reduction of proteolysis and 117

Capítulo 5. Reducing biogenic-amine-producing bacteria, decarboxylase activity, 129

Capítulo 6. High-pressure processing for the control of lipolysis, volatile 139

Capítulo 7. Effect of high-pressure-processing on the microbiology, 153

Capítulo 9. Effect of high-pressure processing on the microbiology, 185

Capítulo 10. Effect of high-pressure-processing on the lipolysis, volatile 201

Capítulo 11. Discusión general. 243

Capítulo 12. Conclusiones. 285

Capítulo 13. Resumen ampliado. 289

Capítulo 14. Extended abstract. 309

Tabla 1 Composición de la leche de vaca, oveja y cabra. 9

Tabla 2 Etapas en la elaboración del queso. 12

Tabla 3 Características de las principales aminas biógenas. 38

Tabla 4 Concentración (mg/kg) de aminas biógenas en diferentes 42

Figura 1 Fragmentos de vasijas para el desuerado (b, d) y esquema de la 5

Figura 3 Producción (en toneladas) de queso durante el año 2012. 7

Figura 10 Cuña de Torta del Casar. 32

Figura 11 Esquema de migración y gradientes en queso Brie. 34

Figura 12 Cuña de queso Arzúa-Ulloa. 36

Figura 13 Reacciones de formación de las aminas biógenas. 44

Figura 14 Ruta y reacciones de formación de diaminas y poliaminas. 45

Figura 15 Esquema de sistemas de presurización directa (A) e indirecta (B). 56

PREVENCIÓN DE LA SOBREMADURACIÓN Y DE LA FORMACIÓN DE

Javier Calzada Gómez

Departamento de Tecnología de Alimentos

Instituto Nacional de Investigación y Tecnología Agraria y Alimentaria, Madrid

Director: Prof. Dr. Manuel Nuñez Gutiérrez, Departamento de Tecnología de Alimentos,

El queso desarrolla sus características organolépticas durante la maduración, debido

Las altas presiones hidrostáticas se emplean como método de pasteurización no

En el presente trabajo de investigación se ha evaluado el efecto de la aplicación de

En queso azul de leche pasteurizada de oveja, madurado con Penicillium roqueforti

putrescina, triptamina y β-feniletilamina. Los tratamientos de altas presiones

los tratamientos provocaron una reducción de la actividad esterasa al final del

En queso Brie de leche de vaca pasteurizada, con Penicillium camemberti en su

Financiación para el desarrollo de la presente Tesis Doctoral:

La presente Tesis Doctoral se realizó en el marco del proyecto AGL 2009-07801,

La presente tesis doctoral consta de los siguientes apartados:

 Una Introducción (Capítulo 1) redactada en español, que incluye una

 Nueve capítulos temáticos (Capítulos 2 a 10) redactados en inglés y

 Una Discusión general (Capítulo 11) y una sección de Conclusiones

 Dos resúmenes ampliados, uno redactado en español (Capítulo 13) y el

El queso es un sistema complejo en el que se suceden un gran número de