Bioresource Technology 90 (2003) 249–254

Protein size distribution and inhibitory effect of wheat

hydrolysates on Neutraseâ
R. Kammoun *, S. Bejar, R. Ellouz
Laboratoire de Microbiologie Industrielle, Centre de Biotechnologie de Sfax, BP ‘‘K’’, 3038 Sfax, Tunisia
Received 15 April 2002; received in revised form 6 May 2003; accepted 8 May 2003

Abstract
Neutraseâ , used for hydrolysis of wheat proteins, was inhibited by end-product in a competitive uncompetitive way. The in-
hibition ratio depends on the progress of protein hydrolysis (degree of hydrolysis) and it remains constant beyond a degree of
hydrolysis of 7.5%. The inhibitor was separated, on Sephadex G-25 column, in three fractions (>2.4, 2.4–0.5 and <0.5 kDa) gen-
erating an inhibition ratio of 29%, 46% and 67% respectively. The peptides size distribution (<1, 1–2, 2–3 and >3 kDa) of fractions
was determined using size exclusion-high performance liquid chromatography. The analysis of obtained data, using a simple
mathematical regression, showed a correlation factor of 0.98 between the inhibition ratio and the peptides less than 1 kDa and 0.99
when considering the peptides lower than 1 kDa and higher than 3 kDa.
Ó 2003 Elsevier Ltd. All rights reserved.

Keywords: Inhibition; Neutraseâ ; Peptides; Protein hydrolysis; Peptide size

1. Introduction nursing infants or sick adults and as stimulants for

persons liable to develop allergy. According to various
The increasing demand for wheat to produce flour, authors (Gonzalez-Tello et al., 1994b; Loosen et al.,
glucose and fructose syrups, involves the simultaneous 1990; Samuelsson and Poulsen, 1987) peptide fractions
creation of large quantities of non-starch by-products allowing a better absorption should contain a limited
rich in protein. Nowadays, the major market for these range of molecular weights and a high di- and tri-pep-
protein rich by-products is animal feed. However, in- tide content. In addition, to avoid allergenic effects, the
creasing the value of these materials would be a pro- molecular weight of peptides should be below 2000 Da.
mising way to satisfy the increasing needs of new Another important factor to be taken into consider-
functional molecules essential in many applications ation is the bitterness of peptides. As a matter of fact,
(Piot et al., 1988) especially in infant feeding (Mannheim it was reported that peptides larger than 1000 Da,
and Cheryan, 1990). derived from whey proteins, are bitter. Therefore, the
Usually protein hydrolysis studies concern the im- size of the much sought-after peptides for the above-
provement of functional properties of the proteins, in mentioned application would be preferably less than
particular, their solubility (Adler-Nissen, 1982). Some 1000 Da.
studies, dealing with the production of low molecular Enzymes used for protein hydrolysis can be inhibited
peptides and in particular the fraction rich in di and tri- during reaction. Hence, several studies concerning,
peptides, have been reported. These peptides have the mainly, inhibition of Alcalaseâ , trypsin, papain and
advantage of being absorbed in the intestine without any Neutraseâ , have been reported. Perea and Uglade (1996)
digestion in the stomach (Gonzalez-Tello et al., 1994b) have indicated, in their study of the continuous hydro-
and have low allergenic effects. This explains their lysis of the whey in a membrane reactor, that Alcalaseâ
preferential use in many formulas such as diets for (Subtilisine A: a serine protease) is inhibited by small
peptides. However for the same substrate, the kinetic
*
Corresponding author. Tel.: +216-74-27-4110; fax: +216-74-27-
study of the enzymatic hydrolysis reported by Gonzalez-
5970. Tello et al. (1994a) shows the presence of an irreversible
E-mail address: radhouan.kammoun@cbs.rnrt.tn (R. Kammoun). serine-protease inhibitor. From bovine milk (from

0960-8524/$ - see front matter Ó 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0960-8524(03)00130-5
250 R. Kammoun et al. / Bioresource Technology 90 (2003) 249–254

which the whey derives) Weber and Nielsen (1991) have product per minute at 25 °C using denatured hemoglo-
isolated a native serine-type protease inhibitor having a bin as the substrate and Folin reagent. This enzyme was
molecular weight of 56–65 kDa. In the case of soybean used at its optimal pH and temperature (6.5 and 40 °C)
proteins, the presence of more than five trypsin inhibi- (Kammoun, 2000).
tors was reported (Hsu et al., 1977). Product inhibition
of Alcalaseâ was also reported in the kinetic study of 2.3. Hydrolysis
enzymatic hydrolysis of corn gluten proteins (Hardwick
and Glatz, 1989). Ozbek and Lovitt (1999) have noticed The substrate was hydrated in distilled water, heated
an inhibition of PROMOD 144 (a papain preparation) at 100 ° C for 10 min and subsequently cooled at room
by the hydrolysis product of wheat protein. temperature. The pH was adjusted to 6.5 with NaOH
As far as we know, no study has been reported about (0.1 N) solution for subsequent hydrolysis. The experi-
the nature of the inhibitor of Neutraseâ found in the ments were conducted in a 50 ml stirred batch reactor
reaction mixture despite the extensive use of this enzyme (700 rpm) fitted with both temperature and pH controls.
in protein hydrolysis. The reaction was started by adding the enzyme. Protein
In our laboratory, we developed a process for pro- enzymatic hydrolysis was achieved by means of a pH-
ducing glucose syrup from gruel (by-product of wheat stat technique (Adler-Nissen, 1982). The pH of the re-
milling) via an amylolytic hydrolysis that leads to a by- action medium was maintained constant by the 718
product rich in proteins (Bejar et al., 1992). In previous STAT Titrino unit (Metrohm-Switzerland). Solutions of
work, we examined various enzymes in by-product 0.5, 0.1 and 0.01 N of NaOH were used to adjust and
protein hydrolysis (Kammoun, 2000). This study en- regulate the reaction medium. At the end of the hy-
abled us to identify the effective enzymatic systems for drolysis, the amount of free amino-groups in the hy-
hydrolyzing protein matter and for building a kinetic drolyzate was determined by the ninhydrin reagent
model for the control of hydrolysis (Kammoun et al., (Masson et al., 1986). Then, the hydrolysate was cen-
2001). More specifically, we demonstrated that the trifuged at 4500g for 15 min, and the supernatant was
Neutraseâ : EC 3.4.24.28, a commercial bacterial prote- separated by filtration through a 0.45 lm pore-size filter.
ase, is suitable to produce small peptides, although it is Finally, it was freeze-dried and the protein content was
inhibited by the hydrolysis products (Kammoun et al., determined by KjeldahlÕs method.
1999). In the present study, the aims were to define the
effects and the nature of Neutraseâ inhibitory products 2.4. Determination of hydrolysis degree
and constitutes a contribution to the understanding of
the inhibition process that limits the wheat protein hy- The degree of hydrolysis (DH), defined as being the
drolysis. ratio between the numbers of cleaved and total peptide
bonds, was routinely measured by the pH-stat method
(DH is proportional to the volume of base added to
2. Methods maintain a pH constant) (Adler-Nissen, 1982). When we
hydrolyzed proteins at a pH near the pKa of a-amino-
2.1. Reagents and substrate groups, the DH determined by the above-mentioned
method was not accurate. By measuring the DH using a
Reagents used in this study were analytic grade and relationship based on the amount of free amino-groups
purchased from Sigma-Aldrich (Germany). The by- (ninhydrin reagent) and the volume of base consump-
product of amylolytic hydrolysis of gruel (a by-product tion (pH-stat method) reported to protein mass in per-
of wheat milling) was obtained from the ‘‘Centre de cent (B
N =mp
100) for Neutraseâ 0.5 l at a pH 6.5
Biotechnologie de Sfax’’ (CBS). The average content of and 40 °C (Fig. 1), we established the following corre-
this product is 58% (based on dry matter) obtained by lation:
using KjeldahlÕs method as described by AOAC (1975). B
N
Its water content, as calculated by dehydration at 60 °C DHð%Þ ¼ 0:179

100 ð1Þ
mp
for 24 h, is 13%.
where B (liter) represents the volume of base needed to
2.2. Enzyme keep the pH constant; mp (kg) is the protein mass, N
(mol/l) is the normality of the base and 0.179 (kg/mol) is
The enzyme used was Neutraseâ 0.5 l (a liquid food the coefficient of the correlation.
grade preparation from Novozymes A/S Denmark): an
endo-proteolytic enzyme of Bacillus subtilis. This en- 2.5. Gel filtration chromatography
zyme is a metallo-protease preparation. Its density is
1.25 g/ml. It contains 0.5 Anson Unit/g (AU/g). 1 Anson The molecular weight distribution of samples was
unit releases 0.001 A750 nm as TCA soluble hydrolysis investigated by gel filtration on Sephadex G-25 (Sigma,
 R. Kammoun et al. / Bioresource Technology 90 (2003) 249–254 251

tion consisted of plotting the inverse of the initial speed

(1=V0 ) versus the inverse of the initial substrate con-
centration (1=S0 ) in the absence or in the presence of the
inhibitor (using the same quantity of the inhibitor). For
the establishment of these curves, we followed the ki-
netic of base volume added to maintain a pH constant.
This kinetic curve allows the determination of the tan-
gent at the origin. At a given normality of the base used
to control the pH, the slope value of the tangent is
proportional to the initial speed of hydrolysis. For these
experiments, we have used an end-hydrolysis product
issued from a hydrolysis having 9.3% as hydrolysis de-
gree (the hydrolysis time is about 10 h).
Fig. 1. Relationship between the DH (%) using Neutraseâ 0.5 l (40 °C,
pH 6.5), determined by ninhydrin method and the base consumption
per the protein quantity expressed as B
N =mp
100 and measured 2.8. Determination of inhibition ratio
by pH-stat.
The inhibition ratio of different products has been
evaluated by measuring the hydrolysis degree after 10
France) using a 75
2.0 cm column and insulin (5808 min by means of the following equation:
Da), vitamin B12 (1350 Da), Thiamine (377 Da) and
Gly-Tyr (238 Da) as standards. The injection quantity DH10  DHI10
Inhibition ratio10 min ð%Þ ¼ ð2Þ
was 100 mg of dried product and the elution buffer was DH10
0.02 M sodium phosphate (pH 7.2) (Amiot et al., 1981).
Inhibition ratio10 min : percent of inhibition after 10 min
This elution was carried out at a rate of 7 ml/h using a
of hydrolysis, DH10 , DHI10 : degree of hydrolysis mea-
UV detector set at 280 nm. The recovery volume was 2
sured after 10 min in absence and presence of the in-
ml/tube. The chromatogram obtained for each protein
hibitor respectively.
hydrolysate was divided in three parts of known mole-
cular weight intervals. The corresponding fractions of
these parts were pooled and the area under each frac- 2.9. Statistical analysis
tion, proportional to the amounts of products released
by the protein hydrolysis, was determined. The pooled All experiments were replicated three times. Inhibi-
fractions were concentrated by evaporation under vac- tion ratio10 min was regressed on percentage of fraction
uum followed by freeze-drying. using a linear regression model. This was achieved by
performing a hypothesis test of Person correlation co-
1=2 1=2
2.6. SE-HPLC efficient (r) based on the fact that rðn  2Þ =ð1  r2 Þ
has a Student distribution with n  2 degrees of freedom
The range of molecular weights in samples was de- (n: number of points). This yields a critical value for r at
termined by size exclusion-high performance liquid the 5% significance level of rC ¼ 0:95. All analysis were
chromatography (SE-HPLC) using the KW 802.5 Sho- performed using EXCEL Software (Office 2000 edition)
dex column (30 cm
8 mm internal diameter) purchased for WINDOWS.
from Waters (Saint-Quentin, France). The solvent was a
phosphate buffer (0.05 M) containing 0.02 M of Na2 SO4
(pH 7.5). Elution was carried out at flux of 0.5 ml/min 3. Results and discussion
using a UV detector at 280 nm. The calibration of the
Shodex KW 802.5 column was achieved by injection of 3.1. Effect of hydrolysis degree on inhibition
10 ll volume containing 360 lg of the standards: Thy-
roglobulin 670 kDa, Gamma globulin 158 kDa, Oval- Several end-products having different hydrolysis de-
bumin 44 kDa, Myoglobin 17 kDa, Insulin 5 808 Da, grees were prepared. Each end-product was used to test
Vitamin B12 1 350 Da, Lys-Try-Lys 437.5 Da, Penicillin its inhibitory effect on Neutraseâ 0.5 l. The inhibition (%
G 356 Da, Gly-Try 238 Da, L-carnosine 226 Da and of inhibition after 10 min and noticed as inhibition
Cresol 108 Da. ratio10 min ) versus degree of hydrolysis showed two kinetic
zones: a DH-dependent zone, where a large increase in
2.7. Identification of the enzyme inhibition type inhibition ratio10 min was observed at DH lower than
7.5%, and a DH-independent zone, where DH level had
Lineweaver–Burk representation was used to identify little or no effect on inhibition ratio10 min , was observed
the type of inhibition (Palmer, 1984). This representa- with a DH greater than 7.5% (Fig. 2).
252 R. Kammoun et al. / Bioresource Technology 90 (2003) 249–254

Table 1
Inhibition ratio of the Neutraseâ 0.5 l caused by the high DH end-
product and their derivative inhibitory fractions
Size (kDa) Inhibition ratio10 min (%)a
Entire inhibitor 45.5 ± 1.1
Fraction I >2.4 29.2 ± 0.7
Fraction II 2.4–0.5 46.2 ± 1.2
Fraction III <0.5 67.7 ± 1.7
The highest inhibition ratio is in bold.
a
Means values and standard deviations for three experiments.

However, the slopes of the three curves were very dif-

ferent indicating a substantial inhibition of Neutraseâ
Fig. 2. Hydrolysis degree effect (DH) on Neutraseâ 0.5 l inhibition.
The inhibition was calculated after 10 min in presence of the same 0.5 l by the high DH end-product. In fact, the slopes of
quantity of inhibitor. Reaction conditions: 40 °C; pH 6.5; enzyme 0.07 the three curves were respectively, 2995, 8919 and
g/g wheat protein, inhibitor 0.33 g/g wheat protein; protein concen- 12 460. The three curves, when continued until the axis,
tration in the reactor 19.6 g/l and time range between 2 and 14 h. intersected at a point near the 1=V0 -axis indicating that
the inhibition was of a competitive–uncompetitive type.
Consequently, the inhibition ratio depended on the
degree of hydrolysis that reached a maximum value and 3.3. Effect of peptide fractions on the inhibition ratio
remained constant beyond 7.5% independent of the
hydrolysis degree. This behavior was probably linked to The observed differences in the inhibition ratio could
the size of peptides contained in each product. be related to the size and the amount of the peptideÕs
fractions responsible for the most substantial inhibition
of Neutraseâ 0.5 l. A high DH end-product was first
3.2. Identification of inhibition type
separated into three fractions, noted as I, II and III
(<0.5, 0.5–2.4 and >2.4 kDa), by gel filtration on
For this study, a high DH end-product (DH > 7.5%)
Sephadex G-25 gel. Each fraction was then tested on
was used as inhibitor. The representation of the inverse
Neutraseâ 0.5 l inhibition during the hydrolysis. Data
of the initial speed of hydrolysis (1=V0 ) versus the inverse
showed that fractions containing small peptides are
of the initial substrate concentration (1=S0 ) in the pres-
more inhibitory than those containing medium or higher
ence and the absence of the inhibitor, according to Li-
ones. Therefore, the more the degree of hydrolysis in-
neweaver–Burk, has been applied to identify the type of
creases, generating a higher amount of low molecular
Neutraseâ inhibition (Fig. 3). We have tested two dif-
weight peptides, the more important the inhibition ratio
ferent inhibitor concentrations (4 and 5.8 g/l, which
is (Table 1).
correspond to 0.2 and 0.29 g of inhibitor, respectively).
Results showed that the Lineweaver–Burk plots, in
3.4. SE-HPLC analysis
the presence and in the absence of inhibitor, were linear.
For a better characterization of molecular weight of
12000 10000
the different peptide fractions, size exclusion-high per-
8000
10000 formance liquid chromatography (SE-HPLC) was used.
1/Vo (min.ml/mmol)

6000

4000 The SE-HPLC patterns obtained for the high DH end-

8000 2000
product of hydrolysis and the derived fractions I, II and
0
6000 -0.7 -0.5 -0.3 -0.1
-2000
0.1 0.3
III are presented in Fig. 4(a)–(d) respectively. These
figures reveal that the peptide size distribution of the
4000
high DH end-product was different from those of the
2000 derivative fractions. To give a more precise interpreta-
tion, the peptide size distribution of the high DH end-
0 product and its different fractions was determined while
0 0.1 0.2 0.3 0.4
1/So (ml/g) using four range fractions: 0–1 kDa (fraction 1), 1–2
kDa (fraction 2), 2–3 kDa (fraction 3) and >3 kDa
Fig. 3. Lineweaver–Burk plot of Neutraseâ 0.5 l inhibition type. Data (fraction 4) (Table 2).
are presented as inverse of the initial hydrolysis speed (1=V0 ) in absence
In this study, it is apparent that fraction I is mainly
(O) and presence of 0.2 g (M), 0.29 g (
) of inhibitor versus the inverse
of the substrate concentration (1=S0 ). Reaction conditions: 40 °C; pH composed of fraction 3 (2–3 kDa) whereas the entire
6.5; enzyme 0.88 AU/l; protein concentration in the reactor ranges high DH end-product is mostly composed of fraction 1
between 5 and 65 g/l; time 10 min and volume mixture 0.03 l. (0–1 kDa). The peptides size distribution of fraction II is
 R. Kammoun et al. / Bioresource Technology 90 (2003) 249–254 253

Fig. 4. Size exclusion-high performance liquid chromatography elution profiles of the entire (a) and the derivative fraction of Neutraseâ 0.5 l in-
hibitor eluted from G-25 gel filtration column: fraction I (b), fraction II (c) and fraction III (d).

Table 2
Peptide percentage, inhibition ratio and molecular weight distribution of entire end-product and different inhibitor fractions
Fraction Weight ratio (%) Inhibition Chromatographic area of fractions in %a
ratio10 min (%)
Fraction 1 Fraction 2 Fraction 3 Fraction 4
(0–1 kDa) (1–2 kDa) (2–3 kDa) (>3 kDa)
Y X
Entire 100 45.5 ± 1.25 53.2 ± 8.2 25.4 ± 1.4 11.8 ± 0.6 9.4 ± 0.2
I 33 29.2 ± 0.7 11.9 ± 1.8 16.2 ± 0.9 42 ± 2.3 29.9 ± 0.7
II 30 46.2 ± 1.25 57.7 ± 8.9 39 ± 2.1 ND 2.6 ± 0.1
III 37 67.7 ± 1.7 88.7 ± 13.7 3.5 ± 0.2 2.7 ± 0.2 5.1 ± 0.1
The most important peptides percentages are in bold.
a
Means values and standard errors for three chromatograms.

composed mainly of fraction 1 and 2 (0–2 kDa) while Table 3

that of fraction III is largely made up of fraction 1. Correlations between Inhibition ratio10 min (%) and the percentage of
peptides matter contained in one or more fractions of the Neutraseâ
0.5 l inhibitors (the entire product and the derivatives fractions)
Chromatogram Sign of slope Correlation
3.5. Statistical treatments of results
fraction coefficient (r)
1 + 0.98a
The effect of inhibitor is directly related to peptide
2 ) 0.41
distribution and therefore linked indirectly to peptide 3 ) 0.92
fractions. The correlations between the Inhibition 4 ) 0.76
ratio10 min and the peptide percentages of each fraction
1+2 + 0.78
(Y versus X in Table 2), according to the least square 1+3 + 0.93
method (Carnahan et al., 1990), were calculated (Table 1+4 + 0.99a
3). A significant correlation (P < 0:05), with a standard 3+2 ) 0.99a
deviation r ¼ 0:98, was obtained between Inhibition 3+4 ) 0.78
ratio10 min and the percentage of peptides contained in Combinations were made between fractions that give a correlation
fraction 1 of all fractions. It is noted that fraction 1 coefficient higher than 0.9.
a
Correlations are significant at P < 0:05.
contained peptides of up to eight amino acid residues. A
positive correlation was also obtained with the peptide
percentages of fraction 3. The correlation was lower again significant (P < 0:05) and greater than 0.9 with
with fraction 4 (r ¼ 0:76) and became insignificant fractions 1 + 4 or 3 + 2. The higher the percentage of
(P > 0:05) with fraction 2ðr ¼ 0:41Þ. In the Table 3, we peptides of fractions 1 and 4 is, the more significant the
also combined some fractions that gave correlation co- Inhibition ratio10 min is (slope+). Inversely, the lower the
efficients higher than 0.9 when compared with all other percentage of peptides of fractions 3 and 2 is, the more
fractions. In that case, the standard deviations were significant the Inhibition ratio10 min is (slope)).
254 R. Kammoun et al. / Bioresource Technology 90 (2003) 249–254

From Table 3, we established that even if the low Amiot, J., Brisson, G.J., Savoie, L., Goulet, J., Jones, J.D., 1981.
molecular weight peptides exhibited the primary inhib- Nitrogenous products released by short term in vitro enzymatic
itory effect, the high molecular weight fractions still hydrolysis of food proteins. Nutr. Rep. Int. 24, 513–529.
AOAC (Association of Official Analytical Chemists) 1975. Biological
exhibit a secondary role on the inhibition process. evaluation of protein quality. In: Horwitz, W. (Ed.), Official
Methods of the Association of Official Analytical Chemists, 12th
ed. AOAC, Washington DC, p. 857.
Bejar, S., Ghorbel, R., Ben Amar, R., Kammoun, R., Ben Messaoud,
4. Conclusions E., Belghith, K., Gargouri, R., Ellouz, R., 1992. Proc
ed
e de
fabrication de sirop de glucose. Tunisian Patent No. 16577.
The study of Neutraseâ inhibition by end-product Carnahan, B., Luther, H.A., Wilkes, J.O., 1990. In: Robert, E. (Ed.),
displayed high sensitivity towards hydrolysis degree for Applied Numerical Methods. Krieger Publishing Company Mal-
values less than 7.5%. SE-HPLC analysis showed that abar, Florida, pp. 571–572.
Gonzalez-Tello, P., Camacho, F., Jurado, E., Paez, M.P., Guadex,
molar weight distribution of fractions has an important E.M., 1994a. Enzymatic hydrolysis of whey proteins: I Kinetic
effect on the inhibition process. The entire high DH models. Biotechnol. Bioeng. 44, 523–528.
hydrolysis end-product and fraction II are mostly Gonzalez-Tello, P., Camacho, F., Jurado, E., Paez, M.P., Guadex,
composed of relatively medium molecular weight pep- E.M., 1994b. Enzymatic hydrolysis of whey proteins: II. Molecu-
tides (<2 kDa) while fractions III and I are essentially lar-weight range. Biotechnol. Bioeng. 44, 529–534.
Hardwick, J.E., Glatz, C.E., 1989. Enzymatic hydrolysis of corn gluten
formed of peptides less than 1 kDa and >2 kDa, re- proteins. Ann. Bio. Chem. Eng. Symp., 10–20.
spectively. Hsu, H.W., Vavak, D.L., Satterlee, L.D., Miller, G.A., 1977. A
All peptides of size less than 3 kDa contribute to the multienzyme technique for estimating protein digestibility. J. Food
inhibition process but those of less than 1 kDa have the Sci. 42, 1269–1273.
main role. Inhibition is limited when the reaction me- Kammoun, R., 2000. Hydrolyse contr^ ol
ee de prot
eines de bl
e par
syst
emes mono et multi-enzymatiques pour lÕobtention de peptides
dium contains more peptides of medium molecular de faibles masses molaires. Doctorat Thesis, Institut National
weight (1–2 kDa). This study represents an attempt at Polytechnique de Lorrain, Nancy, France.
understanding Neutraseâ inhibition that will allow us to Kammoun, R., Bejar, S., Chevalot, I., Marc, I., 1999. Wheat protein
resolve this problem. Consequently, we suggest using a hydrolysis: controlled production of low molecular weight pep-
two-step process. In the first step, a less specific hydro- tides. Poster ECB9/1853 on Ninth European Congress on Biotech-
nology, July 11–15, Brussels ISBN 80521-1-5.
lytic enzyme (such as Alcalaseâ ) could be used in order Kammoun, R., Fournier, F., Le Bont
e, S., Bejar, S., Chevalot, I.,
to produce a fraction composed of peptides of less than Marc, I., 2001. Construction dÕun mod
ele de lÕhydrolyse des
3 kDa. In the second step, this fraction could be hy- prot
eines du gruau de bl
e dur par une prot
ease. R
ecents Progr
es en
drolyzed by Neutraseâ using a continuous membrane G
enie des Proc
ed
es, ISBN 2-910239-61-6. 87, 41–48.
reactor to remove peptides of molecular weight less than Loosen, P., Bressollier, P., Julien, R., Pejoan, C., Verneuil, B., 1990.
Method for preparing an enzymatic hydrolysate, WO 91/10369.
1 kDa. Mannheim, A., Cheryan, M., 1990. Continuous hydrolysis of milk
proteins in a membrane reactor. J. Food Sci. 55, 381–390.
Masson, P., Tom
e, D., Popineau, Y., 1986. Peptic hydrolysis of gluten,
glutenin and gliadin from wheat grain: kinetics and characteriza-
Acknowledgements tion of peptides. J. Sci. Food Agric. 37, 1223–1235.
Ozbek, B., Lovitt, R.W., 1999, Solubilisation studies on the enzymatic
This work was supported by funds from the Tunisian hydrolysis of wheat gluten. Ninth European Congress on Biotech-
Government (‘‘Contrat-Programme CBS-LMI’’). We nology, July 11–15, Brussels July, ISBN 80521-1-5.
thank Mr. Hedi Aouissaoui and Miss Najla Masmoudi Palmer, T., 1984. Understanding Enzymes. Elis Horwood Limited,
Chischester. pp. 89–91.
for their technical assistance with chromatographic Piot, J.M., Guillochon, D., Leconte, D., Thomas, D., 1988. Applica-
analysis. Thanks are also due to Mrs. Hamdi Guebsi tion of ultrafiltration to the preparation of defined hydrolysates of
and Moncef Affes for critically reading the manuscript. bovine Haemoglobin. J. Chem. Tech. Biotechnol. 42, 147–156.
Perea, A., Uglade, U., 1996. Continuous hydrolysis of whey proteins in
a membrane recycle reactor. Enzyme Microbial. Tech. 18, 29–34.
Samuelsson, E.G., Poulsen, O.M., 1987. A peptide preparation, a
References process for producing it and use of the peptide preparation.
European Patent Application A 1 0 226 221.
Adler-Nissen, J., 1982. Limited enzymic degradation of proteins: a new Weber, B.A., Nielsen, S.S., 1991. Isolation and partial characterization
approach in the industrial application of hydrolase. J. Chem. Tech. of a native serine-type protease inhibitor from bovine milk. J.
Biotechnol. 32, 138–156. Dairy Sci. 74, 764–774.