Physicochemical characterization of fructooligosaccharides and evaluation of their suitability as a potential sweetener for diabetics

Article abstract of DOI:10.1016/j.carres.2007.10.012

Fructooligosaccharides (FOSs) were prepared from sucrose using fungal fructosyl transferase (FTase) obtained from Aspergillus oryzae MTCC 5154. The resulting mixture consisted of glucose (28-30%), sucrose (18-20%) and fructooligosaccharides (50-54%) as indicated by HPLC analysis. Identification of oligomers present in the mixture of fructooligosaccharides was carried out using NMR spectroscopy and LC-MS. No compounds other than mono-, di-, tri-, tetra- and pentasaccharides were identified in the FOS mixture prepared using FTase. NMR and LC-MS spectra proved the absence of any toxic microbial metabolites of Aspergillus species in FOS thereby emphasizing its safe use as a food ingredient. Animal studies conducted on streptozotocin-induced diabetic rats suggested that the use of FOS as an alternative non-nutrient sweetener is without any adverse effects on various diabetes-related metabolic parameters. Despite the high free-sugar content associated with it, FOS did not further aggravate the hyperglycemia and glucosuria in diabetic animals, even at 10% levels. On the other hand, by virtue of its soluble fibre effect, it has even alleviated diabetic-related metabolic complications to a certain degree.

Full text of DOI:10.1016/j.carres.2007.10.012

Available online at www.sciencedirect.com
Carbohydrate Research 343 (2008) 56–66
Physicochemical characterization of fructooligosaccharides
and evaluation of their suitability as a potential
sweetener for diabetics
M. J. Mabel,a P. T. Sangeetha,a Kalpana Platel,b K. Srinivasanb and S. G. Prapullaa,*
aDepartment of Fermentation Technology and Bioengineering, Central Food Technological Research Institute, Mysore 570 020, India
bDepartment of Biochemistry and Nutrition, Central Food Technological Research Institute, Mysore 570 020, India
Received 17 April 2007; received in revised form 7 October 2007; accepted 15 October 2007
Available online 23 October 2007
Abstract—Fructooligosaccharides (FOSs) were prepared from sucrose using fungal fructosyl transferase (FTase) obtained from
Aspergillus oryzae MTCC 5154. The resulting mixture consisted of glucose (28–30%), sucrose (18–20%) and fructooligosaccharides
(50–54%) as indicated by HPLC analysis. Identification of oligomers present in the mixture of fructooligosaccharides was carried
out using NMR spectroscopy and LC–MS. No compounds other than mono-, di-, tri-, tetra- and pentasaccharides were identified
in the FOS mixture prepared using FTase. NMR and LC–MS spectra proved the absence of any toxic microbial metabolites of
Aspergillus species in FOS thereby emphasizing its safe use as a food ingredient. Animal studies conducted on streptozotocin-
induced diabetic rats suggested that the use of FOS as an alternative non-nutrient sweetener is without any adverse effects on various
diabetes-related metabolic parameters. Despite the high free-sugar content associated with it, FOS did not further aggravate the
hyperglycemia and glucosuria in diabetic animals, even at 10% levels. On the other hand, by virtue of its soluble fibre effect, it
has even alleviated diabetic-related metabolic complications to a certain degree.
ꢀ 2007 Elsevier Ltd. All rights reserved.
Keywords: Aspergillus oryzae; Fructooligosaccharides; Physicochemical characterization; Non-nutrient sweetener
1. Introduction
source of FTase used for its production. They can be
classified as inulin-type FOSs, which have a b-(2!1)
linkage, such as 1-kestose, nystose and fructofuranosyl
nystose, and neo-FOSs, which have a b-(2!6) linkage,
such as neokestose. The authors have reported the pro-
duction of FOSs using FTase from Aspergillus oryzae
MTCC 5154 and the influence of various parameters
involved in the process5,6 and Aspergillus pullulans
CFR 77.7,8
The physicochemical characteristics of any food
ingredient have a significant impact on its applicability.
The first objective of the present investigation was to
characterize FOSs produced enzymatically using FTase
from A. oryzae MTCC 5154. The next objective was
to evaluate their efficiency of application as a sweetener
for diabetics by conducting animal studies.
Fructooligosaccharides (FOSs) are non-digestible car-
bohydrates that possess interesting functional and
physiological attributes like low sweetness, non-cario-
genicity, low caloric value, prebiotic, hypolipidemic
and hypocholestrolemic properties.1 They also facilitate
mineral absorption2 and inhibit the growth of patho-
genic bacteria in the colon.3 Although FOSs are present
in trace amounts in natural foods like onions, aspara-
gus, wheat, banana, tomato and honey, commercial pro-
duction is accomplished using microbial transferase like
fructosyl transferase (FTase). Many fungal and bacterial
strains are reported to produce FTase.4 The structure
and linkage of FOSs differ depending on the microbial
Diabetes mellitus is an endocrine disorder affecting
millions of people worldwide and is becoming very
*
Corresponding author. Tel.: +91 0821 2515792; fax: +91 0821
2517233; e-mail: sgpsh@yahoo.co.in
0008-6215/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2007.10.012

M. J. Mabel et al. / Carbohydrate Research 343 (2008) 56–66
57
Table 1. Composition of the basal synthetic diet
metabolic parameters to evaluate the suitability of its
use as a sweetener.
Component
g/10 kg
Vitaminized starch
Salt mixture
Cane sugar
100
400
1000
2100
990
2. Experimental
2.1. Chemicals
Casein
Groundnut oil
Vitaminized oil
Corn starch
10
5400
All the chemicals used were of analytical grade. FOS
standards, 1-kestose, 1-nystose and 1-fructofuranosyl
nystose were from Wako Pure Chemical Industries,
Ltd (Osaka, Japan, gift from Dr. Sosaku Ichikawa).
Streptozotocin, glucose oxidase, horseradish peroxidase,
o-dianisidine, albumin, Coomassie Brilliant Blue G-250,
thiosemicarbazide, cadmium sulfate and diacetyl mon-
oxime were from Sigma Chemical Co., USA. All other
chemicals used were of analytical grade.
common with changing lifestyles. Dietary intervention is
one of the main therapies proposed in the case of type 2
diabetes patients, and hence non-digestible dietary fibres
and polysaccharides are gaining importance for the
treatment of diabetic subjects.9,10 It is in this context
that fructooligosaccharides (FOSs) may be considered
as a non-nutritive sweetener.
Thus, the second objective of this investigation was to
evaluate the efficacy of application of FOS as a sweet-
ener for diabetics. This was carried out in an animal
study involving streptozotocin-induced diabetic rats. In
this study, we have examined the effect of dietary FOSs
(at two dose levels, 5% and 10%) on diabetes-related
2.2. Fructooligosaccharide (FOS) production using FTase
from A. oryzae MTCC 5154
FTase was produced by submerged fermentation using
A. oryzae MTCC 5154. The medium and parameters
Table 2. 13C chemical shifts (ppm) for FOS produced using FTase from A. oryzae MTCC 5154
Carbon atom
Chemical shifts (ppm)
Sucrose
Glucose
Fructose
GF2
92.9 (26)
GF3
92.6 (44)
a-D Glucopyranoside
C-1
96.3 (b) (1)
92.9 (a) (2)
72.8 (3)
74.3 (4)
69.6 (5)
73.0 (a) (6)
76.3 (b) (7)
60.8 (8)
64.35 (9)
92.5 (14)
C-2
C-3
C-4
C-5
—
71.4 (15)
72.8 (16)
70.0 (17)
73.2 (18)
70.1 (27)
72.8 (28)
71.9 (29)
71.4 (30)
70.0 (45)
73.0 (46)
71.8 (47)
71.5 (48)
76.2 (10)
75.3 (11)
81.0 (12)
C-6
63.7 (13)
60.6 (19)
60.6 (31)
60.6 (49)
b-D Fructofuranoside
C-10
C-20
C-30
C-40
C-50
C-60
62.6 (20)
104.0 (21)
78.0 (22)
74.4 (23)
81.1 (24)
62.7 (25)
61.8 (32)
103.6 (33)
77.2 (34)
74.9 (35)
81.7 (36)
61.2 (37)
61.4 (50)
103.6 (51)
77.3 (52)
74.8 (53)
81.6 (54)
61.2 (55)
b-D Fructofuranoside
C-100
C-200
C-300
C-400
C-500
C-600
61.1 (38)
103.4 (39)
77.1 (40)
74.3 (41)
81.5 (42)
62.5 (43)
60.9 (56)
103.6 (57)
77.1 (58)
74.5 (59)
81.5 (60)
61.8 (61)
b-D Fructofuranoside
C-1000
C-2000
C-3000
C-4000
C-5000
C-6000
60.85 (62)
103.4 (63)
76.8 (64)
74.3 (65)
81.4 (66)
62.6 (66)
Some of the C-2–C-5 signals of fructose units are interchangeable.

58
M. J. Mabel et al. / Carbohydrate Research 343 (2008) 56–66
were as reported earlier.6 FOS production was carried
out by incubating 2.5 L of FTase with 7.5 L of the
substrate (60ꢁBx sugar) at 55 ꢁC for 18 h. The reaction
was terminated by raising the temperature to 85 ꢁC for
15 min, and the samples were analyzed by HPLC
(LC-6A, Shimadzu, Kyoto, Japan) with a refractive
index detector using a polar bonded-phase column
(Exsil NH2, 4.6 mm · 25 cm, 5 lm) at room tempera-
ture. The mobile phase used was 80:20 acetonitrile–
water at a flow rate of 1.0 mL/min. The retention times
of the individual FOSs were compared with that of stan-
dards for identification and quantification. The lyophi-
lized sample was used (Heto Drywinner DW3, Heto-
Halten, Denmark) for determining the chemical
structure.
amino column (250 cm · 4.6 mm, 5 lm) at 25 ꢁC with
80:20 acetonitrile–water as the mobile phase. Electro-
spray-ionization (ESI) of the sample was carried out in
the positive-ionization mode at 51.0 psi nebulizer pres-
sure, 10 L minꢀ1 dry gas flow rate at 360 ꢁC dry gas tem-
perature. The detector used was an ion-trap analyzer.
2.5. Animal studies
Male Wistar rats (12 weeks old) weighing 150–160 g
were used for the induction of diabetes by streptozoto-
cin. Streptozotocin solution was prepared in fresh citrate
buffer (0.1 M, pH 4.5) and injected intraperitoneally at a
concentration of 60 mg/kg body weight, in a volume of
1 mL per rat. The animals were given 5% glucose solu-
tion for 24 h following streptozotocin injection to pre-
vent initial drug-induced hypoglycemic mortality.
Hyperglycemic rats (having at least two times the
normal blood glucose levels) were sorted out after deter-
mining fasting blood glucose values, one week after
streptozotocin administration. Three groups of hyper-
glycemic rats (10 per group) and in parallel, three groups
of normal rats, were housed in individual stainless steel
metabolic cages with free access to food and water.
The basal semisynthetic diet was of the composition
shown in Table 1. The test material (FOS) was included
at 5% and 10% levels, at the expense of corn starch.
Consequently, there were six rat groups, Normal
control, Normal FOS 5%, Normal FOS 10%, diabetic
2.3. Purification of FOS
Lyophilized FOS (1 g) was dissolved in 2 mL of triple-
distilled water, centrifuged (4000 rpm for 10 min) and
loaded to a BioGel P-2 (BioRad Laboratories, Califor-
nia, USA) gel-filtration column (100 · 1 cm). Elution
was carried out using triple-distilled water at a flow rate
of 5 mL hꢀ1, and 1-mL fractions were collected. The
fractions were analyzed by HPLC.
2.4. Analytical procedures
2.4.1. NMR analysis. 1H and 13C NMR spectra were
recorded at 500 MHz on a Bruker DRX 500-MHz spec-
trometer (500.13 MHz proton and 125 MHz carbon
frequencies). Proton and carbon 90ꢁ pulse widths were
12.25 and 10.5 ls, respectively. About 50 mg of the sam-
ple dissolved in D2O was used for recording the spectra.
Chemical shifts were expressed in ppm relative to the
internal standard trimethylsilane (TMS).
Two-dimensional heteronuclear multiple quantum
coherence transfer spectra (2D HMQCT) were recorded
in the magnitude mode with the sinusoidal shaped z gra-
dients of strengths 25.7, 15.42 and 20.56 G/cm in the
ratio of 5:3:4 applied for a duration of 1 ms each with
a gradient recovery delay of 100 ls to defocus unwanted
coherences. The t1 was incremented in 256 steps. The
size of the computer memory used to accumulate the
2D data was 4 K. The spectra were processed using
unshifted and p/4 shifted sine bell window function in
F1 and F2 dimensions, respectively.
2.4.2. LC–MS analysis of purified oligomers of
FOS. LC–MS was performed on a Bruker Esquire
3000 mass spectrometer (Bruker Instruments, Germany)
connected to an Agilent 1100 HPLC system (Hewlett–
Packard Company, Pennsylvania) fitted with an Exsil
ꢁBx = degrees Brix, a measurement of the mass ratio of sugar to
water in a solution.
Figure 1. HPLC profile of the FOS mixture.

M. J. Mabel et al. / Carbohydrate Research 343 (2008) 56–66
59
control, diabetic FOS 5%, and diabetic FOS 10%. The
rats were maintained on these diets for a period of six
weeks. The diet was given ad libitum. After induction
of diabetes, rats were fed with two types of diet, contain-
ing 5% and 10% FOS, respectively. FOS is added replac-
ing corn starch in each case. During the dietary regimen,
fasting blood glucose levels were monitored at fort-
nightly intervals by collecting blood from the retro-
orbital plexus. Body weights were recorded at weekly
intervals. The amounts of glucose, protein, urea and
creatinine excreted in the urine were monitored at
weekly intervals in the 24-h urine samples collected. At
the end of the feeding period, the animals were sacri-
ficed, and blood samples were collected and used for
the determination of plasma glucose, urea, albumin
and creatinine.
2.6. Metabolic parameters
Plasma and urinary glucose was determined by the
glucose oxidase method employing glucose oxidase, per-
oxidase and o-dianisidine.11 Urinary protein was mea-
sured by dye-binding method described by Bradford,12
while plasma albumin was also quantitated by a
dye-binding method described by Gustafsson.13 Urinary
urea was assayed by the method of Wyeberga et al.,14
Figure 2. 1H NMR spectra and 13C NMR spectra of FOS (in D2O) produced using FTase from A. oryzae MTCC 5154.

60
M. J. Mabel et al. / Carbohydrate Research 343 (2008) 56–66
while creatinine was measured by the method of Folin
and Wu.15 Plasma urea was estimated by the method
of Levine et al.16
and 92.8 ppm for a). Unreacted sucrose was found to
be present in substantial amount (12% by HPLC). In
GF2 and GF3, the anomeric glucose linkage was
observed at 92.8 ppm, indicating that the linkage is a.
Correspondingly, the C-20, C-200 and C-2000 signals ob-
served for fructose were all observed around 103 ppm,
indicating a b-linkage. The C-1 and C-6 signals of fruc-
tose were in the range of 60.9–61.8 ppm. All the fructose
signals in both GF2 and GF3 were detected by NMR
spectroscopy. From the 2D HMQCT spectrum, a few
proton signals could also be detected (Table 3). Thus,
NMR analysis conclusively proved the presence of
GF2 and GF3 in FOS prepared using FTase produced
by A. oryzae MTCC 5154.
2.7. Statistical analysis
Results were expressed as mean SEM, and compari-
sons between groups were made by means of an
unpaired Student’s t-test.17 Differences were considered
significant when p < 0.05.
3. Results and discussion
3.1. Physicochemical characterization of FOS
The oligomers present in the FOS mixture were puri-
fied using column chromatography. The oligo-, di- and
monosaccharides eluted sequentially based on their
decreasing molecular weight. The different fractions
containing pure kestose, nystose and fructofuranosyl
nystose were analyzed by LC–MS. There have been
many reports on the analysis of FOS by HPLC, GC–
MS, NMR, etc.21–24 In this work, attempts have been
made to explore the application of LC–MS for the direct
analysis of FOS. The MS results were obtained in the
positive-ionization mode, using a sodium salt contained
FOS production was carried out using FTase obtained
from A. oryzae MTCC 5154 as described previously.18
FOS resulting from the experiment is a mixture of
mono-, di-, tri-, tetra- and pentasaccharides as observed
by HPLC analysis (Fig. 1).18
Studies on the proton and carbon chemical align-
ments of kestose (GF2) have already been reported.19,20
The chemical shift values of the 13C NMR spectra of
FOS are shown in Table 2. FOS used in the study was
the lyophilized reaction product resulting from 18 h of
enzymatic synthesis from sucrose using FTase. The sam-
ple was analyzed for structure by recording 2D
HMQCT spectra, and the structures were assigned
based on the literature information.21–24 As shown in
Figure 2a and b, the spectra clearly showed the presence
of free glucose, a small amount of free fructose, unre-
acted sucrose, GF2 (kestose) and GF3 (nystose). The
presence of GF2 and GF3 was clearly detected from both
1H proton and 13C spectral correlation observed in 2D
HMQCT (Fig. 3). In GF2 and GF3, the linkage between
glucose and fructose is 1!2 and a for the 1-linked gly-
cosidic linkage and b for the (1!2)-linked glycosidic
linkage of fructose.
Table 3. 1H chemical shifts (ppm) for FOS produced using FTase from
A. oryzae MTCC 5154
Chemical shifts (ppm)
Glucose (free) Fructose (free) GF2
GF3
a-D Glucopyranoside
H-1
5.24 (a) (1)
4.65 (b) (2)
3.78 (3)
3.72 (6)
4.23 (7)
3.74 (8)
5.43 (9)
5.41 (18)
H-2
H-3
H-4
H-5
H-6
3.49 (10) 3.42 (19)
3.75 (11) 3.72 (20)
3.52 (4)
3.69 (5)
3.56 (12) 3.57 (21)
3.83 (13)
b-D Fructofuranoside
H-10
H-20
H-30
H-40
H-50
H-60
Three kestoses are known so far. They are 1-kestose
[b-D-fructofuranosyl-(2!1)-b-D-fructofuranosyl-(2!1)-
a-D-glucopyranoside], 6-kestose [b-D-fructofuranosyl-
(2!6)-b-D-fructofuranosyl-(2!1)-a-D-glucopyranoside]
and neokestose [b-D-fructofuranosyl-(2!6)-a-D-fructo-
furanosyl–(1!2)-glucopyranoside]. The data clearly
showed that 1-kestose [b-D-fructofuranosyl-(2!1)-b-D-
3.48 (14) 3.43 (22)
3.88 (15) 3.81 (23)
3.86 (24)
b-D Fructofuranoside
H-100
H-200
H-300
H-400
H-500
H-600
3.76 (25)
4.2 (16)
4.05 (17)
4.21 (26)
fructofuranosyl-(2!1)-a-D-glucopyranoside
(inulin
type)] is the major product detected for GF2. The six-
linkage was not detected in any fructose unit of GF3.
Compared to GF3, the concentration of GF2 was found
to be higher correlating with the HPLC data.18 The
NMR spectra showed less intense peaks, which merged
with the noise, probably indicating small amounts of
GF4 (fructofuranosyl nystose).
b-D Fructofuranoside
H-1000
H-2000
H-3000
H-4000
H-5000
H-6000
The free glucose was present both in a and b forms as
observed by the anomeric C-1 signals (96.3 ppm for b
3.9 (27)

M. J. Mabel et al. / Carbohydrate Research 343 (2008) 56–66
61
in the mobile phase to facilitate ionization. Hence, all
the masses are presented as sodium adducts. Figure 4
shows the HPLC profile of purified 1-kestose with its
mass spectrum in the inset. The signals at m/z 527.1 cor-
respond to 1-kestose, and those at m/z 365.1 correspond
to sucrose, whereas the minor signal at m/z 203.0 corre-
sponds to glucose. The HPLC profile of purified nystose
and its mass spectrum represented in Figure 5 shows a
major signal of m/z 689.1. Figure 6 shows the mass spec-
trum of fructofuranosyl nystose corresponding to m/z
851.1. The actual masses of kestose, nystose and fructo-
furanosyl nystose were calculated as 504, 666 and 828,
respectively.
served in the diabetic control group during the first fort-
night of the experiment, the mortality was only 10%
during this period in the FOS-fed group. Thus the sur-
vival rate was higher due to FOS feeding, especially at
the 10% levels, while considering the mortality in subse-
quent weeks. There was a steady decline in the body
weight in the diabetic control group during the six-week
experimental period. Such a trend was not observed in
either of the FOS-treated groups (Fig. 7).
FOS treatment did not adversely influence fasting
blood sugar values in diabetic animals, in spite of the
presence of free blood sugar (50%) associated with it,
as judged by fasting blood glucose (FBG) values at fort-
nightly intervals (Fig. 8). This observation was also cor-
roborated by the urinary glucose excretion pattern
throughout the experimental duration (Fig. 9).
3.2. Evaluation of the suitability of FOS as a sweetener in
an animal study
Increased protein catabolism is a characteristic feature
of diabetes mellitus. Thus hypo-albuminemia (decreased
titres of albumin in serum) should be expected during
diabetes mellitus. Lower serum albumin levels (15%)
While all the non-diabetic animals survived the duration
of the experiment, considerable mortality was observed
among the diabetic rats. Whereas 80% mortality was ob-
Figure 3. 2D HMQCT spectrum of FOS (in D2O) produced using FTase from A. oryzae MTCC 5154.

62
M. J. Mabel et al. / Carbohydrate Research 343 (2008) 56–66
Figure 6. HPLC elution profile of purified fructofuranosyl nystose.
Inset: mass spectrum of fructofuranosyl nystose.
Figure 4. HPLC elution profile of purified 1-kestose. Inset: mass
spectrum of 1-kestose.
this albumin destruction, as indicated by higher than
normal serum albumin levels in diabetic rats.
Blood creatinine levels were higher in diabetic rats
compared to normal animals, which is an indication of
increased muscle wasting during this metabolic disease
(Table 4). FOS treatment at either of the dietary levels
significantly lowered blood creatinine concentration in
diabetic rats. This indicates that FOS has countered
the muscle wasting normally associated with diabetes
mellitus, and this is in concurrence with the maintenance
of body weight observed in FOS-fed rats. Proteinuria,
which is an indicator of progressive kidney damage
during diabetes, was considerably lowered by dietary
FOS towards the end of the experimental period,
especially at 10% levels (Fig. 10).
There have been a few studies on the effect of FOS on
diabetes. The study by Yamashita et al.25 demonstrated
that the daily intake of fructooligosaccharides for four-
teen days by type-2 diabetic patients caused an 8% de-
crease in fasting blood glucose concentrations, a 6%
decrease in total cholesterol and a 10% decrease in
LDL cholesterol. The authors claim that these observa-
tions of decrease in fasting blood glucose and serum to-
tal cholesterol levels were statistically significant. Several
studies have been conducted in rats since, all showing
that fructooligosaccharides lower serum triglycerides
and total cholesterol as well as LDL and VLDL choles-
terol.26–29 Although fructooligosaccharides showed var-
ious effects on metabolism, the exact mechanism is not
explained. Fructooligosaccharides may possibly bind
to nutrients such as carbohydrates and lipids and to
Figure 5. HPLC elution profile of purified nystose. Inset: mass
spectrum of nystose.
were seen in diabetic control animals compared to their
normal counterparts (Table 4). FOS treatment protected

M. J. Mabel et al. / Carbohydrate Research 343 (2008) 56–66
63
the absorptive mucosal surface, and the absorption of
the nutrients from the intestinal mucosa may be pre-
vented. Since fructooligosaccharides alter the composi-
tion of bacterial flora in the intestine,30 the digestion
of carbohydrates and the metabolism of bile acids may
be changed. Hence, absorption of carbohydrates and
lipids will be reduced, resulting in lowering of blood glu-
cose along with serum lipids.25
A study by Luo et al.31 showed that chronic consump-
tion of 20 g of fructooligosaccharides by healthy
humans increased basal hepatic glucose production but
had no effect on insulin-stimulated glucose metabolism
180
160
140
120
100
80
60
40
20
0
0
1
2
3
4
5
6
Week
Diabetic - Control
Diabetic - FOS 5%
Diabetic - FOS 10%
Figure 7. Body weight gain in diabetic rats maintained on FOS diet. Values are mean SEM of six animals in each group.
160
140
120
100
80
60
40
20
0
0
2
4
6
Week
Diabetic - Control
Diabetic - FOS 5%
Diabetic - FOS 10%
Figure 8. Blood glucose in diabetic rats maintained on FOS. Fasting blood glucose values in normal animals were 40–45 mg/dL. Values are
mean SEM of six animals in each group.

64
M. J. Mabel et al. / Carbohydrate Research 343 (2008) 56–66
or serum lipids. Pedersen et al.32 studied the effect of
inulin on blood lipids and did not find any changes in
total cholesterol, HDL cholesterol, LDL cholesterol or
triglyceride. Thus the effects of fructooligosaccharides
on blood glucose and serum lipids are not clear. Alles
et al.33 have evidenced that consumption of fructooligo-
saccharides for three weeks by patients with type-2
diabetes does not favourably affect blood glucose and
serum lipid concentrations. Even in our present study,
we did not observe any major influence of FOS on blood
glucose. The FOS preparation used in the present study
was around 50% in terms of FOS, and the remainder
consisted of free sugars. Despite the high free-sugar con-
tent associated with it, FOS did not further aggravate
the hyperglycemia and glucosuria in diabetic animals
even at 10% levels. This may be attributed to the fact
that, FOS as soluble fibre can adsorb nutrients such as
glucose, thus retarding their absorption. As a result,
the glucose flux in the blood following a carbohydrate
meal is retarded, and the amount of insulin needed by
the body is lowered. By virtue of its soluble fibre effect,
FOS has even alleviated diabetic-related metabolic com-
plications to a certain degree. Thus the present study
suggests that the use of FOS as an alternative sweetener
is without any adverse effects.
In conclusion, identification of oligomers present in a
mixture of fructooligosaccharides has been carried out
using NMR spectroscopy and LC–MS. No compounds
other than mono-, di-, tri-, tetra- and pentasaccharides
were identified in the FOS mixture prepared using
FTase produced by A. oryzae MTCC 5154. NMR and
LC–MS spectra proved the absence of any toxic micro-
bial metabolites in FOS, thereby emphasizing its safe use
as a food ingredient. In this study, it was also shown
that the intake of fructooligosaccharides for a period
of six weeks did not adversely influence the sugar values
in diabetic animals. FOS was also found to protect ani-
mals from the destruction of albumin. It also countered
Table 4. Blood albumin, urea and creatinine values in diabetic rats maintained on FOS diets for 6 weeksa
Animal group
Albumin (g/dL)
Urea (mg/dL)
Creatinine (mg/dL)
Diabetes control
Diabetes FOS 5%
Diabetes FOS 10%
Normal control
Normal FOS 5%
Normal FOS 10%
6.67 0.37b
7.85 0.38c
8.66 0.50c
7.84 0.27
7.66 0.13
8.67 0.32
116.3 5.20a
112.1 6.85
110.6 6.07
96.3 3.24
92.0 1.69
102.4 1.67
9.35 0.63b
5.92 0.59c
5.84 0.71c
5.75 0.65
5.03 0.49
5.42 0.29
a Values are mean SEM of five rats in each group.
b Significantly different from normal control.
c Significantly different from diabetes control.
7
6
5
4
3
2
1
0
0
1
2
3
4
5
6
Week
Diabetic - Control
Diabetic - FOS 5%
Diabetic - FOS 10%
Figure 9. Urinary glucose excretion in diabetic rats maintained on FOS. Glucose excretions in normal animals were 0.3–0.5 mg/day. Values are
mean SEM of six animals in each group.

M. J. Mabel et al. / Carbohydrate Research 343 (2008) 56–66
65
18
16
14
12
10
8
6
4
2
0
0
1
2
3
4
5
6
Week
Diabetic - Control
Diabetic - FOS 5%
Diabetic - FOS 10%
Figure 10. Urinary protein excretion in diabetic rats maintained on FOS. Protein excretions in normal animals were 2.5–3.5 mg/day. Values are
mean SEM of six animals in each group.
2. Ohta, A.; Ohtsuki, M.; Baba, S.; Adachi, T.; Sakata, T.;
Sakaguchi, E. J. Nutr. 1995, 125, 2417–2424.
3. Gmeiner, M.; Kneifel, W.; Kulbe, K. D.; Wouters, R.; De
Boever, P.; Nollet, L.; Verstraete, W. Appl. Microbiol.
Biotechnol. 2000, 53, 219–223.
4. Prapulla, S. G.; Subhaprada, V.; Karanth, N. G. Adv.
Appl. Microbiol. 2000, 47, 299–337.
5. Indian Patent 439/DEL/2001, 30-03-2001.
6. Sangeetha, P. T.; Ramesh, M. N.; Prapulla, S. G. Sci.
Aliment. 2002, 22, 277–287.
7. Indian Patent. 411/DEL/2001, 29-03-2001.
8. Sangeetha, P. T.; Ramesh, M. N.; Prapulla, S. G. Asian J.
Microbiol. Biotech. Environ. Sci. 2003, 5, 313–318.
9. Anderson, J. W. In Dietary Fiber in Health and Disease;
Vahouny, G. V., Kritchevsky, D., Eds.; Plenum Press:
New York, 1982; p 151.
the muscle wasting normally associated with diabetes
mellitus, and which is in concurrence with the mainte-
nance of body weight observed in FOS-fed rats.
Thus the present study on rats shows that FOS maybe
used as a sweetener by diabetics or as an adjunct to
dietary therapy with various beneficial effects on glucose
and protein metabolism. However, in-depth studies are
needed to assess the effect of daily intake of fructooligo-
saccharides by diabetics over a longer period of time.
Acknowledgements
The authors are thankful to late Dr. M. N. Ramesh,
Scientist, Department of Food Engineering, CFTRI,
Mysore for his discussions and guidance and Dr. Sosaku
Ichikawa, Professor, Institute of Applied Biochemistry,
University of Tsukuba, Japan for his gift of FOS stan-
dards. We are thankful to Dr. S. Divakar, Scientist,
Department of Fermentation Technology and Bioengi-
neering, for interpreting the NMR results. We are also
thankful to the Staff of Measurement Science Depart-
ment, Hindustan Lever Research Centre, Bangalore
for LC–MS analysis. M.J.M. and P.T.S. are thankful
to CSIR for the research fellowship. The authors thank
Director, CFTRI, for supporting the work.
10. Luo, J.; Yperselle, M. V.; Rizkalla, S. W.; Rossi, F.;
Bornet, F. R. J.; Slama, G. J. Nutr. 2000, 130, 1572–1577.
11. Hugget, A. S. G.; Nixon, D. A. Lancet 1957, 273, 366–370.
12. Bradford, M. M. Anal. Chem. 1976, 72, 248–254.
13. Gustafsson, J. E. C. Clin. Chem. 1976, 22, 616–622.
14. Wyeberga, D. R.; Di Giorgio, J.; Pileggi, V. J. Clin. Chem.
1971, 17, 891–895.
15. Oser, B. L. In Hawk’s Physiological Chemistry, 14th ed.;
McGraw Hill Publishing Co.: New Delhi, 1965; p 1040.
16. Oser, B. L. In Hawk’s Physiological Chemistry, 14th ed.;
McGraw Hill Publishing Co.: New Delhi, 1965; p 1039.
17. Snedecor, G. W.; Cochran, W. G. In Statistical Methods;
The Iowa State University Press: Ames, IA, USA, 1967;
p 100.
18. Sangeetha, P. T.; Ramesh, M. N.; Prapulla, S. G. Process
Biochem. 2003, 39, 753–758.
19. Calub, T. M.; Waterhouse, A. L.; Chatterton, N. J.
Carbohydr. Res. 1990, 199, 11–17.
20. Liu, J.; Waterhouse, A. L.; Chatterton, N. J. Carbohydr.
Res. 1991, 217, 43–49.
References
1. Katapodis, P.; Kalogeris, E.; Kekos, D.; Macris, B. J.;
21. Hayashi, S.; Imada, K.; Kushima, Y.; Ueno, H. Curr.
Microbiol. 1989, 19, 175–177.
Christakopoulos, P. Food Biotechnol. 2003, 17, 1–14.

66
M. J. Mabel et al. / Carbohydrate Research 343 (2008) 56–66
22. Fujita, K.; Kuwahara, N.; Tanimoto, T.; Koizumi, K.;
Iizuka, M.; Minamiura, N.; Furuichi, K.; Kitahata, S.
Biosci. Biotech. Biochem. 1994, 58, 239–243.
23. Barthomeuf, C.; Grizard, D.; Teulade, J. C. Biotechnol.
Tech. 1997, 11, 845–848.
28. Fiordaliso, M.; Kok, N.; Desager, J. P.; Goethals, F. M.;
Roberfroid, M. B.; Delzenne, N. Lipids 1995, 30, 163–
167.
29. Kok, N.; Roberfroid, M.; Robert, A.; Delzenne, N. Br. J.
Nutr. 1996, 76, 881–890.
24. Hayashi, S.; Yoshiyama, T.; Fuji, N.; Shinohara, S.
Biotechnol. Lett. 2000, 22, 1465–1469.
25. Yamashita, K.; Kawai, K.; Itakura, M. Nutr. Res. 1984, 4,
961–966.
30. Mitsuoka, T.; Hara, T. In Reports of Neosugar Confer-
ence; Hosoya Meiji Seika Co.: Tokyo, 1982; pp 63–76.
31. Luo, J.; Rizkalla, S. W.; Alamowitch, C. Am. J. Clin.
Nutr. 1996, 63, 939–945.
26. Delzenne, N.; Kok, N.; Fiordaliso, M.; Deboyser, D. M.;
Goethals, F. M.; Roberfroid, M. B. Am. J. Clin. Nutr.
1993, 57, 820S.
27. Levrat, M.; Favier, M.; Moundras, C.; Remesy, C.;
Demigne, C.; Morand, C. J. Nutr. 1994, 124, 531–538.
32. Pedersen, A.; Sandstrom, B.; Van Amelsvoort, J. M. M.
Br. J. Nutr. 1997, 78, 215–222.
33. Alles, M. S.; de Roos, N. M.; Bakx, J. C.; Van de Lisdonk,
E.; Zock, P. L.; Hautvast, J. G. A. J. Am. J. Clin. Nutr.
1999, 69, 64–69.