Food Chemistry 120 (2010) 850–857

Contents lists available at ScienceDirect

Food Chemistry
journal homepage: www.elsevier.com/locate/foodchem

Oligosaccharides of pitaya (dragon fruit) flesh and their prebiotic properties

S. Wichienchot a,*, M. Jatupornpipat b, R.A. Rastall c
a
Nutraceutical and Functional Food Research and Development Center, Faculty of Agro-Industry, Prince of Songkla University, Hat Yai, Songkhla 90112, Thailand
b
Department of Applied Biology, Faculty of Science, King Mongkut’s Institute of Technology Ladkrabang, Bangkok 10520, Thailand
c
School of Food Biosciences, The University of Reading, Whiteknights, Reading, UK

a r t i c l e i n f o a b s t r a c t

Article history: The major carbohydrates of white and red-flesh pitayas (dragon fruit) were glucose, fructose and some
Received 26 May 2009 oligosaccharides (total concentrations of 86.2 and 89.6 g/kg, respectively). The molecular weight distribu-
Received in revised form 2 November 2009 tion of the extract was affected by the extraction solvent. The maximum oligosaccharides content
Accepted 9 November 2009
(27.40%), which included fractions with molecular weights of 273–275, 448–500 and 787–911 Da, were
obtained using 80% ethanol extraction at room temperature (28 ± 2 °C). The low molecular weight frac-
tion, including glucose and fructose, was successfully removed by yeast cultivation. The molecular
Keywords:
weights of mixed oligosaccharides (716, 700, 490 and 474 Da) were confirmed by mass spectrometry.
Pitaya
Dragon fruit
The mixed oligosaccharides showed that they were resistant to hydrolysis by artificial human gastric
Oligosaccharides juice and human a-amylase, giving maximum hydrolysis of 4.04% and 34.88%, respectively. The mixed
Prebiotic oligosaccharides were also found capable of stimulating the growth of lactobacilli and bifidobacteria.
Probiotic Ó 2009 Elsevier Ltd. All rights reserved.

1. Introduction vide nutrients that may prevent nutrition-related diseases and im-
prove physical and mental well-being of the consumers.
Pitaya (Hylocereus undatus (Haw.)) or dragon fruit is native to Prebiotics are non-digestible oligosaccharides that beneficially
Mexico, Central and South America (Benzing, 1990; Haber, 1983) affect the host by stimulating the growth and/or activity of one
and has been grown in Vietnam for at least 100 years, following or a limited number of bacteria in the colon, thus improving host
its introduction by the French (Mizrahi, Nerd, & Nobel, 1997). A health (Gibson & Roberfroid, 1995). Ten samples from thirteen
member of the Cactaceae, it has trailing cladode stems modified fruits and vegetables and their parts from southern Thailand were
to act as leaves and bears spectacular ovoid fruit year-round which reported as potential sources of natural prebiotics with the highest
has a bright red colour when mature, and contains white, crimson, oligosaccharide content being 9.81% (w/w) (Thammarutwasik
or pale yellow-flesh, depending on the cultivar, interspersed with et al., 2009). Dragon fruit was not included in that study. Chicory
small black seeds. In Thailand, one of the widely grown varieties and artichoke root have been used as natural sources of inulin
is H. undatus or red pitaya with white-flesh. Other varieties that and oligofructose in the commercial production of prebiotics, hav-
have been commercialised are Hylocereus polyrhizus (red pitaya ing oligosaccharide yields in the range of 18–20% (Gibson & Rastall,
with red-flesh) and Hylocereus megalanthus (yellow pitaya) (Bar- 2006). As a potential source of high-yielding oligosaccharide for
beau, 1990). Currently, there is much interest in developing this commercial prebiotic production, dragon fruit was selected for
crop for fresh fruit export beyond the ‘local’ Asian markets of Sin- investigation, with respect to optimal extraction method, sugar
gapore, Hong Kong, Taiwan, Philippines, Malaysia and Thailand content and prebiotic properties.
(Hoa, Clark, Waddell, & Woolf, 2006; Mizrahi et al., 1997).
Dragon fruit has been reported as a source of beta-carotene,
lycopene and vitamin E, with average concentrations of 1.4, 3.4 2. Materials and methods
and 0.26 lg/100 g edible portion, respectively (Charoensiri, Kon-
gkachuicha, Suknicom, & Sungpuag, 2009). The seed of dragon fruit 2.1. Samples, chemicals, media and microorganisms
contains 50% essential fatty acids, i.e., 48% linoleic acid (C18:2) and
1.5% linolenic acid (C18:3) (Ariffin et al., 2008). Thus, dragon fruit White-flesh dragon fruits ( H. undatus) were imported from
has potential for use as a source of functional ingredients to pro- Hoang Hau Dragon Fruit Farm, Co. Ltd., Vietnam. The fruits were
medium size packed in a padded paper box. Red-flesh dragon fruits
(H. polyrhizus) were obtained from a local market in Chumphon,
* Corresponding author. Tel.: +66 74286389; fax: +66 74212889. Thailand. Unless otherwise stated, all chemical reagents and en-
E-mail addresses: santad.w@psu.ac.th, wsantad@yahoo.com (S. Wichienchot). zyme were obtained from Sigma–Aldrich Co. Ltd. Cheesecloth bags

0308-8146/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.foodchem.2009.11.026
 S. Wichienchot et al. / Food Chemistry 120 (2010) 850–857 851

were prepared (23  14 cm). Anaerobic gas pack and Reinforced 1999). The molecular weight distribution of mixed oligosaccha-
Clostridium Medium were supplied by Oxoid, Ltd., Basingstoke, rides after removal of low molecular weight fraction by yeast cul-
UK. MRS broth was purchased from Merck, Co., Ltd., Darmstadt, tivation was confirmed by mass spectrometry (Micromass, UK)
Germany and Rogosa SL Agar was purchased from Difco, Co., Ltd., with ESI detector in positive mode (modified from Hernandez,
USA. Bifidobacterium bifidum NCIMB 702715 was obtained from Muiz-Matute, Olano, Moreno, & Sanz, 2009).
the National Collections of Industrial Food and Marine Bacteria,
Aberdeen, UK. Lactobacillus delbrueckii BCC 13296 and Saccharomy- 2.4. Purification method
ces cerevisiae BCC 12652 were obtained from BIOTEC Culture Col-
lection, Bangkok, Thailand. The extract obtained by 80% ethanol extraction at ambient tem-
perature was distilled by rotary evaporator at 175 mbar, 60 °C and
2.2. Determination of sugar composition rotating speed of 45 rpm. The concentrated extract was diluted
with distilled water in the ratio of 1:1 and redistilled again. This
Reducing sugar concentration was determined using the step was repeated twice to removal residual ethanol in the extract.
dinitrosalicylic acid (DNS) assay and expressed as maltose equiva- Ethanol-free extract was adjusted to 20° Brix by addition of dis-
lents (Robertson et al., 2001). The assay was calibrated with malt- tilled water before 5% (v/v) inoculum (18 h culture of S. cerevisiae
ose standards from 0 to 1 mg/ml. Sample or glucose standard BCC 12652) was inoculated. Cultivation was carried out at room
solution (1 ml) was added to DNS solution (1 ml) and water temperature (28 ± 2 °C) for 48 h to remove mostly glucose and
(1 ml). The mixture was boiled for 15 min and after cooling on some fructose. Culture broth was then diluted with sterile distilled
ice for 2 min, water (9 ml) was added before mixing and measuring water to 10° Brix and cultivation was carried out under the same
the absorbance at 540 nm. Content of reducing sugar in the sample conditions for 6 h to remove mostly the remaining fructose in
was compared to the standard curve of glucose. The total sugar the crude extract (modified from Hernandez et al., 2009).
concentration, expressed as glucose equivalents, was determined
using the phenol–sulphuric acid assay (Dubois, Gilles, Hamilton, 2.5. Effect of extraction methods
Rebers, & Smith, 1956). The assay was calibrated with D-glucose
standards from 0 to 100 lg/ml. Sample or glucose standard solu- White-flesh dragon fruits ( H. undatus) and red-flesh dragon
tion (1 ml) was added to 5% (w/v) phenol solution (1 ml). Concen- fruits ( H. polyrhizus) were used as raw material for this study.
trated sulphuric acid (5 ml) was then added and left for 10 min The dragon fruit was peeled and the flesh was extracted with 3 dif-
prior to vigorous mixing. The samples and standards were left for ferent solvents (distilled water, 20% ethanol and 80% ethanol).
20 min at room temperature before reading the absorbance at Extraction by water at room temperature was done by mixing
490 nm. The concentration of carbohydrate in the sample was cal- water and the flesh in the ratio of 2:1, stirring continuously and
culated by comparison with the standard curve of glucose. Mono- leaving for 1 h at room temperature. The mixture was then filtered
saccharide and disaccharide were analysed by HPLC. Sample was through a cheesecloth bag while being pressed with a paddle blen-
prepared in solution (1% w/v) and filtered through 0.45-lm der (Seward-Stomacher 400) to remove the seeds. Viscous juice ob-
membrane; 20 ll sample were injected in a Zorbax Carbohydrate tained was further evaporated by a rotary evaporator under
column (4.6  150 mm, 5 lm resin). Mobile phase was acetoni- reduced pressure at 40 °C. Sample was dried with a freeze dryer
trile:water (75:25) and flow rate was 0.5 ml/min. Refractive Index before sugar composition, molecular weight distribution and pre-
(RI) detector was used and column oven temperature was 35 °C. biotic properties were analysed. Hot water extraction was done
Glucose and fructose were used as monosaccharide standards by mixing hot water and the flesh in the ratio of 2:1, stirring con-
and sucrose was used as a disaccharide standard. Qualitative anal- tinuously and leaving for 1 h in a controlled temperature water
ysis of sugar in the sample was determined by comparison to the bath (85 ± 2 °C). The mixture was then filtered through a cheese-
retention time of standard sugar. The concentration of sugar in cloth bag while being pressed by a stomacher to remove the seeds.
the sample was calculated by comparison with peak area of the The sample was then dried and further analysed as mentioned
standard curve of the respective sugar. above. Extraction by 20% and 80% (v/v) ethanol were done using
the same manner and the mixture was left for 1 h at room temper-
2.3. Determination of molecular weight distribution of ature. The sample was then dried and again analysed as above.
oligosaccharides
2.6. Effect of artificial human gastric juice hydrolysis
Molecular weight of sample was analysed by gel permeation
chromatography (GPC) with refractive index detection. The analyt- Mixed oligosaccharides obtained by 80% ethanol extraction and
ical system consisted of Ultrahydrogel linear column (MWCO the juice without extraction from white-flesh were tested for acid
1000–20,000,000) and Ultrahydrogel 120 column (MWCO 100– resistance, compared to an inulin standard as a commercial prebi-
5000) connected in series with decreasing pore size. The flow rate otic reference. Sample was dissolved in RO water to give a 1% (w/v)
was 0.6 ml/min, the eluent was 0.1 M sodium nitrate and the oper- solution. Artificial human gastric juice was mimicked by using
ating temperature was 30 °C. Samples and standards (pullulan hydrochloric acid buffer containing (in g/l): NaCl, 8; KCl, 0.2; Na2H-
with different molecular weight, maltoheptaose, and glucose) were PO42H2O, 8.25; NaHPO4, 14.35; CaCl22H2O, 0.1; MgCl26H2O, 0.18.
filtered through a 0.45-lm syringe filter and the injection volume The pH of the buffer was adjusted to 1, 2, 3, 4 and 5 using 5 M HCl
was 20 ll. The system was calibrated with 8 standards consisting (Korakli, Ganzle, & Vogel, 2002). HCl buffer (5 ml) at each pH was
of pullulan with average molecular weights of 788, 404, 112, added to the sample solution (5 ml) and the reaction mixture
47.3, 22.8, 5.9 kDa, maltopentaose (0.828 kDa) and glucose was incubated in a water bath at a controlled temperature of
(0.180 kDa). External standards of maltopentaose (0.828 kDa) at 37 ± 1 °C for 6 h. Sample (1 ml) was taken periodically at 0, 0.5, 1,
concentrations of 1–10 mg/ml were used to establish the area re- 2, 4 and 6 h. Reducing sugar content in the sample was determined
sponse for quantification of the products. Oligosaccharides yield by DNS method (Robertson et al., 2001) and total sugar was deter-
was calculated by the summary of the percentage area of the oligo- mined by phenol–sulphuric acid method (Dubois et al., 1956). Per-
saccharides fraction. The molecular weight (MW) of oligosaccha- centage hydrolysis of sample was calculated based on reducing
rides was determined by comparing sample retention time to sugar liberated and total sugar content of the sample (Korakli
those of the standard curve (Mountzouris, Gilmour, Jay, & Rastall, et al., 2002):
852 S. Wichienchot et al. / Food Chemistry 120 (2010) 850–857

Hydrolysisð%Þ centration of 108 cell/ml. The culture was incubated at 37 °C for

reducing sugar released  100 48 h in anaerobic conditions. Sample was taken at 0, 24 and 48 h
¼ ð1Þ for bacterial enumeration by standard direct count using a haem-
total sugar content  initial reducing sugar content
ocytometer (Swanson, Petran, & Hanlin, 2001). B. bifidum 702715
where reducing sugar released is the difference between its final was sub-cultured in Reinforced Clostridial Medium at 37 °C for
and initial content. 24 h prior to use as inoculum. One millilitre of sample solution
was prepared (1% w/v) and aseptically added to Reinforced Clos-
2.7. Effect of human a-amylase hydrolysis tridial Medium (3 ml) in a sterilised screwed-cap vial with rubber
septum. Cultivation was carried out under anaerobic conditions, by
The enzyme activity was determined by the Sigma quality con- flushing with nitrogen gas over the culture medium then closing
trol test procedure for a-amylase (EC. 3.2.1.1). Enzyme was pre- with a screwed-cap. Inoculum (1 ml) was injected by a sterilised
pared in solution containing 2 units/ml using sodium phosphate syringe into the sealed vial via a septum. Cultivation was carried
buffer (20 mM) in sodium chloride (6.7 mM), pH 4, 5, 6, 7 or 8.
Sample was prepared as a 1% (w/v) solution in sodium phosphate Table 1
buffer. Five millilitres of enzyme solution were added to 5 ml of Physical and chemical properties of two varieties of pitaya.
sample solution. The reaction mixture was incubated in a con- Characteristic White-flesh dragon fruit Red-flesh dragon fruit
trolled temperature water bath (37 ± 1 °C) for 6 h. Sample (1 ml)
Fruit dimension (cm)a
was taken at 0, 0.5, 1, 2, 4 and 6 h to determine reducing sugar Length 134 ± 5.0a 127 ± 5.5b
and total sugar. Percentage hydrolysis was calculated based on Diameter 94.0 ± 9.0a 66.0 ± 4.0b
reducing sugar liberated and the total sugar content of the sample Fruit weight (g)a
as above. Flesh weight 305 ± 75.0a 215 ± 35.0b
Skin weight 100 ± 30.0a 75.0 ± 25.0b
a
2.8. Effect of probiotic growth Sweetness (Brix) 12.5 ± 0.55a 14.8 ± 0.75b
b
Sugar content (g/kg)
Mixed oligosaccharides obtained from white-flesh and a refer- Glucose 353 ± 0.74a 401 ± 1.27b
Fructose 238 ± 0.84a 158 ± 0.32b
ence prebiotic (inulin) were used as carbon source for the growth
Oligosaccharides 86.2 ± 0.93a 89.6 ± 0.76a
of 2 probiotic strains. Inoculum of L. delbrueckii BCC 13296 was pre-
pared by cultivation on MRS broth at 37 °C for 24 h. One millilitre Different letters in the same row mean a significant difference (p 6 0.05).
a
Average of 15 fruits.
of sample solution was prepared (1% w/v) and added to MRS broth b
Average of triplicate analyses by HPLC.
(3 ml), then the inoculum (1 ml) was added, to obtain a final con-

Fig. 1. Sugar composition of white-flesh – (a) and red-flesh (b) dragon fruit extracts, analysed by HPLC.
 S. Wichienchot et al. / Food Chemistry 120 (2010) 850–857 853

out under anaerobic conditions at 37 °C for 72 h. Sample was taken (353 ± 0.7 g/kg) was significantly lower than of red fleshed fruit
at 0, 24, 48 and 72 h for bacterial enumeration by a standard direct (401 ± 1.27 g/kg) (p 6 0.05). In contrast, fructose of white-flesh
count using haemocytometer (Swanson et al., 2001). (238 ± 0.84 g/kg) was significantly higher than red-flesh (158 ±
0.32 g/kg) (p 6 0.05), as shown in Table 1. Oligosaccharides content
in red-flesh (89.6 ± 0.76 g/kg) was slightly higher than white-flesh
3. Results and discussion (86.2 ± 0.93 g/kg) but not significantly. Molecular weight distribu-
tions of crude extract obtained from either white or red-flesh ana-
3.1. Chemical composition and molecular weight distribution of lysed by gel permeation chromatography showed exactly the same
oligosaccharides values (273–275, 448–500 and 787–911 Da by 80% ethanol extrac-
tion and 273–275 and 787–911 Da by distilled water extraction,
Sugars of white- and red-flesh dragon fruit analysed by HPLC Fig. 2). The low molecular weight fraction (273–275 Da), glucose
consisted mostly of glucose, fructose and some oligosaccharides, and fructose were successfully removed by two-step yeast
as shown in Fig. 1. Glucose concentration of white-flesh (S. cerevisiae) cultivation and molecular weight distribution of the

Fig. 2. Molecular weight distribution of white-flesh dragon fruit extracts obtained by water (a) and 80% ethanol (b) extraction at ambient temperature, analysed by GPC.
854 S. Wichienchot et al. / Food Chemistry 120 (2010) 850–857

mixed oligosaccharides was confirmed by mass spectrometry. The while 80% ethanol extraction elevated the yield to 27.4% (Table
mixed oligosaccharides consisted of four components of 716, 700, 2). Thus extraction methods had a highly significant effect
490 and 474 Da with relative percentages of 100, 68, 45 and 21, (p 6 0.01) on the oligosaccharide yield obtained. It was concluded
respectively (Fig. 3). Thus, the degree of polymerisation of mixed that the optimal extraction method for oligosaccharides from
oligosaccharides was 3–4, which was in the same range as some white-flesh was 80% (w/v) ethanol (ethanol:flesh was 2:1, v/v) at
fructooligosaccharides, i.e., 1-kestose, 6-kestose and neokestose ambient temperature. To obtain higher purity of pitaya oligosac-
(1 glucose unit and 2 fructose units), or nystose, bifurcose and neo- charides, a further purification step is required, i.e., chromatogra-
bifurcose (1 glucose unit and 3 fructose units), or stachyose (3 glu- phy or membrane filtration, which has been used for commercial
cose units and 1 fructose unit). Some plants producing prebiotic prebiotic production (Kono, 1993), whereas in this work purifica-
oligosaccharides, such as onion, garlic, dahlia, tulip, oat (Ritsema tion was done with a biological approach using yeast. The results
& Smeekensy, 2003), soy bean and mung bean, contained oligosac- showed that S. cerevisiae was very efficient at removing mostly glu-
charides in the same range of molecular weight as this study; cose and some of fructose in the crude extract within 48 h whereas
chicory and artichoke contained higher molecular weight oligosac- removal of residual fructose required 6 h after the extract was
charides (Eggleston & Cote, 2003; Xiaoli et al., 2008). The molecular diluted twice (data not shown). However, this method was less
weight of the pitaya oligosaccharides was in the same range as efficient for the removal of lactose and galactose from galactooligo-
some commercial prebiotics, such as fructooligosaccharide syn- saccharide mixtures (Hernandez et al., 2009). This is because the
thesised from sucrose (DP = 2–4), oligofructose (DP = 4) derived native strain of S. cerevisiae could metabolize common sugars such
from enzymatic hydrolysis of inulin, and soybean oligosaccharide as glucose, fructose, sucrose and maltose, but not lactose, melibi-
obtained from soybean extraction (DP = 4–5) (Vernazza, Rabiu, & ose and xylose as a carbon source.
Gibson, 2006).
3.3. Effect of artificial human gastric juice hydrolysis
3.2. Effect of extraction methods
Mixed oligosaccharides obtained with 80% ethanol extraction
Though white- and red-flesh dragon fruit were not significantly from white-flesh were hydrolysed with artificial human gastric
different in terms of quality and quantity of their oligosaccharides, juice and showed to be resistant to the juice. Percentage of hydro-
white-flesh variety was less expensive and more abundant in Thai- lysis increased with decreasing pH of artificial gastric juice (Fig. 4).
land, and it was thus chosen for subsequent studies. For the study The degree of hydrolysis at pH of 1, 2, 3, 4 and 5 was 4.07%, 2.43%,
on extraction methods, it was found that solvent had marked ef- 1.66%, 0.85% and 0.02%, respectively. The maximum hydrolysis
fects on the molecular weight of the oligosaccharides obtained. Oli- (4.07%) occurred after 2 h of incubation at pH 1. Pitaya oligosaccha-
gosaccharides obtained by water extraction, either at ambient rides showed high acid resistance compared to the reference prebi-
temperature (28 ± 2 °C) or 85 ± 2 °C had molecular weights of otic (inulin), which gave maximum hydrolysis of 55.00%, 23.22%,
787–911 Da while with 80% ethanol extraction, a fraction of 35.75% and 5.96% at pH 1, 2, 3, 4 and 5, respectively. Food is usually
448–500 Da was also found (Fig. 2). The yield of oligosaccharides retained in the human stomach where gastric juice (pH 2–4) is
obtained by water extraction at ambient temperature was 15.1% released within 2 h; thus, when pitaya oligosaccharides are

Fig. 3. Molecular weight distribution of white-flesh dragon fruit extract after removal of low molecular weight sugar by yeast cultivation, analysed by mass spectrometry.
 S. Wichienchot et al. / Food Chemistry 120 (2010) 850–857 855

Table 2
Sugar contents of white-flesh dragon fruit obtained by different extraction methods.

Sugar contenta Extraction by water Extraction by ethanol

Ambient Hot 20% (w/v) 80% (w/v)
Mono- and disaccharide (%) 84.9 ± 1.8a 83.6 ± 2.1a 79.4 ± 2.8b 72.6 ± 1.6c
Oligosaccharides (%) 15.1 ± 0.5a 16.4 ± 0.6a 20.6 ± 1.3b 27.4 ± 1.5c

Different letters in the same row mean a significant difference (p 6 0.05).

a
% area of fractions analysed by GPC in triplicate.

a 4.5
a 40
4.0 35
3.5
30

Hydrolysis (%)
Hydrolysis (%)

3.0
25
2.5
20
2.0
15
1.5
1.0 10

0.5 5

0.0 0
0 1 2 3 4 5 6 0 1 2 3 4 5 6
Time (h)
Time (h)
b 450
400
b 50
45
350 40
Hydrolysis (%)

Hydrolysis (%)

300 35
250 30
200 25

150 20

100 15
10
50
5
0
0 1 2 3 4 5 6 0
Time (h) 0 1 2 3 4 5 6
Time (h)
Fig. 4. Resistance of white-flesh dragon fruit oligosaccharides (a) and inulin (b) to
various artificial gastric juices; incubation for 6 h at 37 °C. , pH 1; j, pH 2; N, pH 3; Fig. 5. Resistance of white-flesh dragon fruit oligosaccharides (a) and inulin (b) to
, pH 4; s, pH 5. human salivary a-amylase (1 U/ml) at various pH values, incubated at 37 °C for 6 h.
, pH 4; j, pH 5; N, pH 6; , pH 7; s, pH 8.

consumed, 96% of them is estimated to reach the intestine. This re-

sult is comparable to that of kojiooligosaccharide, which had 100% Table 3
resistance to artificial gastric acid (Nakada, Nishimoto, Chaen, & Microbial counts of Lactobacillus delbrueckii BCC 13296 using extracts obtained from
Fukuda, 2003). Gluco-oligosaccharide produced by Gluconobacter white-flesh dragon fruit and inulin as carbon sources.

oxydans NCIMB 4943 showed 98.4% resistant in the acidic condi- Sample Microbial counts (cell/ml)
tions of the human stomach (Wichienchot, Prasertsan, Hongpatta- 0h 24 h 48 h
rakere, Gibson, & Rastall, 2006). Dextran has been reported to be
Crude extract 9.72  107a 6.31  108a 2.55  109a
highly resistant to acidic conditions in the human digestive tract Oligosaccharides extract 9.02  107a 6.42  108a 6.17  109b
(Debnam, Denholm, & Grimble, 1998). Inulin 9.14  107a 7.47  108b 8.24  108c

Average of triplicate analyses.

3.4. Effect of human a-amylase hydrolysis Different letters in the same column mean a significant difference (p 6 0.05).

The hydrolysis of oligosaccharides from white-flesh dragon fruit

with human salivary a-amylase at pH 4–8 showed that higher pH oligosaccharides had significantly higher enzymatic resistance
gave a significantly higher degree of hydrolysis (p 6 0.05); how- compared to inulin (p 6 0.01). It was estimated that about 50% of
ever, there was no significant difference at pH 4 and 5. The maxi- the pitaya oligosaccharides consumed would reach the colon since
mum hydrolysis at pH 4, 5, 6, 7 and 8 was 2.90%, 3.08%, 3.28%, some of them were hydrolysed by a-amylase (16%), by acid in
6.88% and 11.18%, respectively after 30 min incubation (Fig. 5). stomach (2.5%) and by brush-border enzymes in small intestine
The maximum hydrolysis of a reference prebiotic (inluin) in- (30%). Typically, the carbohydrates were mainly digested in the
creased at higher pH, giving values of 5.72%, 8.03%, 7.39%, 12.32% small intestine (30%) where some brush-border enzymes, i.e.,
and 16.22% at pH 4, 5, 6, 7 and 8, respectively. Thus, pitaya isomaltase, glucoamylase, maltase, sucrase and lactase, hydrolyse
856 S. Wichienchot et al. / Food Chemistry 120 (2010) 850–857

Table 4
Microbial counts of Bifidobacterium bifidum NCIMB 702715 using extracts obtained from white-flesh dragon fruit and inulin as carbon source.

Sample Microbial counts (cell/ml)

0h 24 h 48 h 72 h
Crude extract 1.42  108a 1.88  109a 2.79  109a 2.04  109a
Oligosaccharides extract 1.70  108a 1.66  109a 2.22  109b 2.51  109b
Inulin 1.23  108a 1.34  109b 2.42  109b 2.76  109b

Average of triplicate analyses.

Different letters in the same column mean a significant difference (p 6 0.05).

a-1,4- and a-1,6-linked glucosaccharides present in the small acid conditions in human stomach, partial resistance to human sal-
intestine and yield monosaccharides as end-products (Johnson & ivary a-amylase and the capability to stimulate the growth of lac-
Schmit, 2005). It has been reported that 88% of inulin and oligo- tobacilli and bifidobacteria. Thus, pitaya is a potential source of
fructose reach the colon, using both the ileostomy patient model prebiotics, which may be used as an ingredient in functional food
and the intubation model (Cummings & Macfarlane, 2002). It has and nutraceutical products.
also been reported that, using an in vitro technique, at least 60%
of gluco-oligosaccharide produced by G. oxydans NCIMB 4943
Conflict of interest statement
reaches the colon. Therefore, the oligosaccharides in this experi-
ment appeared to have partial enzymatic resistance.
The work described in the manuscript is original. It has never
been presented, submitted or published in any publication before
3.5. Effect of pitaya oligosaccharides on probiotic growth and does not encroach on any patents.

Oligosaccharides extracted from white-flesh dragon fruit were

used as a carbon source for the cultivation of two probiotic strains. Statement of authorship
It was found that the oligosaccharides stimulated the growth of L.
delbrueckii BCC13296 by increasing its number from 9.02  107 to S. Wichienchot participated in the design of the study, carried
6.17  109 cell/ml within 48 h. Although inulin also stimulated out the experiments and drafted the manuscript.
the growth of this probiotic strain, it had a lower effect (Table 3). M. Jatupornpipat participated in sample analyses.
It has been reported that oligosaccharides extracted from jackfruit R. A. Rastall participated in the design of the study and drafting
seed also stimulated the growth of lactobacilli (Thammarutwasik of the manuscript.
et al., 2009). Pitaya oligosaccharides also stimulated the growth of
B. bifidum NCIMB 702715 by increasing their number from
Acknowledgements
1.70  108 to 2.51  109 cell/ml within 72 h (Table 4); however,
inulin showed a greater effect but no significant difference. Jackfruit
Financial support from Office of the National Research Council
seed oligosaccharides have been reported to stimulate the growth
of Thailand, is gratefully acknowledged. We would like to thank
of lactobacilli and bifidobacterium (Thammarutwasik et al., 2009).
B. Ooraikul, Professor Emeritus, University of Alberta, for correc-
A similar result was observed on gluco-oligosaccharide produced
tion of the manuscript.
by G. oxydans (Wichienchot, Prasertsan, Hongpattarakere, Rastall,
& Gibson, 2006). Some commercial prebiotics that have been
reported to stimulate the growth of bifidobacterium were inulin, References
oligofructose, fructooligosaccharide, galactooligosaccharide, genti-
Ariffin, A. A., Bakar, J., Tan, C. P., Rahman, R. A., Karim, R., & Loi, C. C. (2008). Essential
ooligosaccharide, isomaltooligosaccharide, lactosucrose, lactulose, fatty acids of pitaya (dragon fruit) seed oil. Food Chemistry, 114(2), 561–564.
raffinose, soybean oligosaccharide and xylooligosaccharide (Sako, Barbeau, G. C. (1990). La pitahaya rouge, un nouveau fruit exotique (the red
Matsumoto, & Tanaka, 1999). Pitaya oligosaccharides showed other pitahaya, a new exotic fruit). Fruits, 45(2), 141–147.
Benzing, D. H. (1990). Vascular epiphytes, general biology and related biota.
functional properties beside their prebiotic effect, such as reduced
Cambridge: Cambridge University Press.
caloric intake and insulinaemia, compared to digestible carbohy- Charoensiri, R., Kongkachuicha, R., Suknicom, S., & Sungpuag, P. (2009). Beta-
drates. Therefore, oligosaccharides from dragon fruit may be suit- carotene, lycopene, and alpha-tocopherol contents of selected Thai fruits. Food
Chemistry, 113, 202–207.
able for inclusion as food supplements in a wide variety of food
Cummings, J. H., & Macfarlane, G. T. (2002). Gastrointestinal effects of prebiotics.
products, e.g., dairy products, products designed for overweight British Journal of Nutrition, 87(Suppl. 2), S145–S1151.
individuals, diabetic prevention products and prebiotic products. Debnam, E. S., Denholm, E. E., & Grimble, G. K. (1998). Acute and chronic exposure of
rate intestinal mucosa to dextran promotes SGLTI-mediated glucose transport.
European Journal of Clinical Investigate, 28, 651–658.
4. Conclusion Dubois, M., Gilles, K. A., Hamilton, J. K., Rebers, P. A., & Smith, F. (1956). Calorimetric
method for determination of sugars and related substances. Analytical
Chemistry, 28, 350–356.
Oligosaccharide content in the white- and red-flesh pitaya was Eggleston, G., & Cote, G. L. (2003). Oligosaccharides in food and agriculture. In G.
86.2 and 89.6 g/kg, respectively. The optimal extraction conditions Eggleston & G. L. Cote (Eds.). Oligosaccharides in food and agriculture (Vol. 849,
pp. 1–14). Washington, DC: American Chemical Society.
for pitaya flesh were 80% (w/v) ethanol, with a ratio of solvent to Gibson, G. R., & Rastall, R. A. (2006). Prebiotics: Development and application.
flesh of 2:1 at ambient temperature (28 ± 2 °C). The oligosaccha- Chichester: John Wiley & Sons Press.
ride content in the crude extract was 27.40% relative to the sugar Gibson, G. R., & Roberfroid, M. B. (1995). Dietary modulation of the human colonic
microbiota: Introducing the concept of prebiotics. Journal of Nutrition, 125,
content. Low molecular weight fraction including glucose and fruc-
1401–1412.
tose in the crude extract was successful removed by yeast (S. cere- Haber, W. A. (1983). Hylocereus costaricensis (pitahaya silvestre), wild pitahaya. In D.
visiae) cultivation. Partial purified extract consisted of a mixture of H. Janzen (Ed.), Costarican natural history (pp. 252–253). Chicago: University of
oligosaccharides having molecular weights of 716, 700, 490 and Chicago Press.
Hernandez, O., Muiz-Matute, A. I., Olano, A., Moreno, F. J., & Sanz, M. L. (2009).
474 Da with a degree of polymerisation of 3–4. Pitaya oligosaccha- Comparison of fractionation techniques to obtain prebiotic
rides showed prebiotic properties, which included resistance to galactooligosaccharides. International Dairy Journal, 19, 531–536.
 S. Wichienchot et al. / Food Chemistry 120 (2010) 850–857 857

Hoa, T. T., Clark, C. J., Waddell, B. C., & Woolf, A. B. (2006). Postharvest quality of on butyrate production during fermentation. British Journal of Nutrition, 81,
Dragon fruit (Hylocereus undatus) following disinfesting hot air treatments. S219–S223.
Postharvest Biology and Technology, 41, 62–69. Sako, T., Matsumoto, K., & Tanaka, R. (1999). Recent progress on research and
Johnson, C. D., & Schmit, G. D. (2005). Mayo clinic gastrointestinal imaging review. applications of non-digestible galacto-oligosaccharides. International Dairy
Rochester: Mayo Clinic Scientific Press. Journal, 9, 69–80.
Kono, T. (1993). Production, properties, and applications: Fructooligosaccharides. In Swanson, K. M. J., Petran, R. L., & Hanlin, J. H. (2001). Culture methods for
T. Nakakuki (Ed.). Oligosaccharides (Vol. 3, pp. 55–78). London: Gordon and enumeration of microorganisms. In F. P. Downed & K. Ito (Eds.). Compendium of
Breach. methods for the microbiological examination of foods (Vol. 4, pp. 53–62).
Korakli, M., Ganzle, M. G., & Vogel, R. F. (2002). Metabolism by bifidobacteria and Washington, DC: American Public Health Association Press.
lactic acid bacteria of polysaccharides from wheat and rye, and Thammarutwasik, P., Hongpattarakere, T., Chantachum, S., Kijroongrojana, K.,
exopolysaccharides produced by Lactobacillus sanfranciscensis. Journal of Itharat, A., Reanmongkol, W., et al. (2009). Prebiotics: A review.
Applied Microbiology, 92, 958–965. Songklanakarin Journal of Science and Technology, 31(4), 401–408.
Mizrahi, Y., Nerd, A., & Nobel, P. S. (1997). Cacti as crops. Horticulture Reviews, 18, Vernazza, C. L., Rabiu, B. A., & Gibson, G. R. (2006). Human colonic microbiology and
291–319. the role of dietary intervention: Introduction to prebiotics. In Prebiotics:
Mountzouris, K. C., Gilmour, S. G., Jay, A. J., & Rastall, R. A. (1999). A study of dextran Development and application (pp. 1–28). Chichester: John Wiley & Sons Press.
production from maltodextrin by cell suspensions of Gluconobacter oxydans Wichienchot, S., Prasertsan, P., Hongpattarakere, T., Gibson, G. R., & Rastall, R. A.
NCIB 4943. Journal of Applied Microbiology, 87, 546–556. (2006). In vitro fermentation of mixed linkage gluco-oligosaccharides produced
Nakada, T., Nishimoto, T., Chaen, H., & Fukuda, S. (2003). Kojioligosaccharides: by Gluconobacter oxydans NCIMB 4943 by the human colonic microflora. Current
Application of kojibiose, phosphorylase on the formation of various Issues in Intestinal Microbiology, 7, 7–12.
kojioligosaccharides. In G. Eggleston & G. L. Cote (Eds.). Oligosaccharides in Wichienchot, S., Prasertsan, P., Hongpattarakere, T., Rastall, R. A., & Gibson, G. R.
food and agriculture (Vol. 849, pp. 104–117). Washington, DC: American (2006). In vitro three-stage continuous fermentation of gluco-oligosaccharides,
Chemical Society. produced by Gluconobacter oxydans NCIMB 4943, by the human colonic
Ritsema, T., & Smeekensy, S. (2003). Fructans: Beneficial for plants and humans. microflora. Current Issues in Intestinal Microbiology, 7, 13–18.
Current Opinion in Plant Biology, 6, 223–230. Xiaoli, X., Liyi, Y., Shuang, H., Wei, L., Yi, S., Hao, M., et al. (2008). Determination of
Robertson, J. A., Ryden, P., Botham, R. L., Reading, L., Gibson, G. R., & Ring, S. G. oligosaccharide contents in 19 cultivars of chickpea (Cicer arietinum L.) seeds by
(2001). Structural properties of diet-derived polysaccharides and their influence high performance liquid chromatography. Food Chemistry, 111, 215–219.