22

HEALTH BENEFITS OF NON-DIGESTIBLE

OLIGOSACCHARIDES

M. B. Roberfroid

Unite de Biochimie Toxicologique et Cancerologique

Department of Pharmaceutical Sciences
Universite Catholique de Louvain
BCTC-UCL 7369, Avenue Mounier 73
B-1200 Brussels, Belgium

ABSTRACT

Non-digestible oligosaccharides are complex carbohydrates of the non-a-glucan type

which, because of the configuration of their osidic bonds, resist hydrolysis by salivary and
intestinal digestive enzymes. In the colon they are fermented by anaerobic bacteria. Among
the non-digestible oligosaccharides, the chicory fiuctooligosaccharides occupy a key position
and, in most european countries, they are recognised as natural food ingredients. The other
major products are the short chain fructooligosaccharides and galactooligosaccharides
obtained by enzymatic synthesis using sucrose and lactose as substrates respectively, the soy-
bean oligosaccharides, the xylooligosaccharides produced by partial hydrolysis ofxylans and
polydextrose or pyrodextrins prepared by a chemical treament of carbohydrates. The most
well known effect of most non-digestible oligosaccharides, and in particular of the fructo-
oJigosaccharides, is the selective stimulation of the growth of Bijidobacteria thus modifying
significantly the composition of the colonic microbiota. Such a modification, which has
clearly been demonstrated in human volunteers, is meant to be benificial in part because it is
accompanied by a significant reduction in the number of bacteria reported to have pathogenic
potential. Within the framework of research and development of "functional foods", such an
effect justifies a "functional claim" for fructooligosaccharides namely "bifidogenesis". They
are also typical "prebiotics". Besides their bifidogenic effect, the chicory fructooligosaccha-
rides have additional nutritional properties on digestive physiological parameters like colonic
pH and stool bulking which justify their classification as dietary fibers. Moreover, in experi-
mental models, it has also been reported that they improve the bioavailability of essentiel
minerals and that they reduce serum triglyceridemia by lowering hepatic lipogenesis. Such
effects demonstrate interactions between the chicory fructooligosaccharides and key func-
tions in the body but their significance for humans still need to be proven before being used to
justify additional claims.

DietGlY Fiber in Health and Disease, edited by Kritchevsky and Bonfield

Plenum Press, New York, 1997 211
212 M. B. Roberfroid

1. INTRODUCTION

1.1. Functional Foods: Concept and Strategy!

A "functional food" is a "food which contains (in adequate concentration) one or
a combination of components which affects functions in the body so as to have positive
cellular or physiological effects which may, in due course, justify functional or health
claims". Research and development of "functional food" is part of the science of nutrition
and the benefit associated with its intake has to be evaluated in the context of a balanced
diet. The strategy for research and development of a "functional food" or a "functional
food ingredient" requires:
a. The identification of its interaction with a function in the body
b. The understanding of the mechanism of this interaction
c. The demonstration of the functional effect in relevant biological systems
d. The use of the scientific data to formulate sound hypotheses
e. Testing of these hypotheses in human nutrition studies, the functional food being
ingested as part of a 'balanced diet.

1.2. The Non-Digestible Oligosaccharides: Definition

Among the components likely to be used in functional foods, the non-digestible oligo-
saccharides show interesting properties and some are already recognised and included in
foods as ingredients 2 • By definition a non-digestible oligosaccharide (NDO) is:

"an oligomeric carbohydrate the osidic bond of which is in a spatial configuration which
makes it resistant to the hydrolytic activity of the intestinal digestive enzymes, but still senti-
tive to hydrolysis by the enzymes of the colonic bacteria which then ferment its monomers to
produce short chain carboxylic acids and gases but also cellular energy for metabolic activi-
ties, growth and proliferation".

The NDOs belong to the category of non a-glucans and, if they are hydrolysed-
fermented only by a specific group of colonic bacteria, they are classified as "prebiotics"
a concept we have recently introduced3 • According to this concept, a "prebiotic" is:

"a non-digestible food ingredient which beneficially affects the host by selectively stimulating
the growth and/or activating the metabolism of one or a limited number of health-promoting
bacteria in the intestinal tract thus improving the host's intestinal physiology".

In addition to their prebiotic effect, some of the NDOs may also have interesting
systemic effects acting, in particular, as modulators of the metabolism of lipids and carbo-
hydrates and/or endocrine secretions. Up to now, effects of the NDOs can justify "func-
tional claims", but not "health claims" which will require further and extensive research.
As opposed to an "health claim" which directly concerns disease prevention, a "functional
claim" is limited to a change in a particular function of the body not implying that this
change has any proven relevance to the prevention of a disease l .
The most important class of dietary NDOs is that composed of short (up to 8 mono-
mers with a mean degree of polymerisation of about 5) and medium (up to 50-60 units
with a mean degree of polymerisation of about 10 or 20) chain-length homopolymers
(table 1). The chicory NDOs (inulin and its hydrolysate oligofructose) are fructooligo-
Health Benefits of Non-Digestible Oligosaccharides 213

saccharides classified as natural food ingredients4 • Other NDOs are synthetic products
resulting from the enzymatic and/or chemical modifications of natural disaccharides like
saccharose or lactose and polysaccharides like starch or xylans.

1.3. Aim of the Paper

The aim of the present paper is to review the nutritional and physiological properties of
the chicory fructooligosaccharides which are, so far, the most studied NDOs. Even though
more partly, similar evidences exist also for other NDOs, and they will be referenced. These
properties will be discussed by reference to the strategy for research and development of a
functional food described above (see 1.1) with the objective to identify which functional
interactions have been identified, which mechanisms have been explored, which hypotheses
are made, which human data are available and, eventually, which claims are justified.

2. BEHAVIOR AND EFFECTS OF CHICORY

FRUCTOOLIGOSACCHARIDES IN THE
GASTROINTESTINAL TRACT

2.1. Non-Digestibility
The evidence that chicory fructooligosaccharides resist hydrolysis by digestive
enzymes and are not absorbed in the upper part of the intestinal tract comes from in vivo
human studies performed in ileostomy subjects which demonstrate that 90-95% of the
ingested oligosaccharides are recovered at the end of the small intestine, the small loss of
material being due to microbial contamination of the ileon5 •6. A study in rats fed 14C_
labelled Neosugar® provided similar evidence for the resistance to intestinal hydrolysis
and absorption 7•

2.2. Fermentation by Bifidobacteria

That chicory fructooligosaccharides are quantitatively fermented by the colonic
microbiota during their passage through the colon is supported by many experiments
which showed that these carbohydrates could not be detected in the faeces of animals or of
human volunteers who consumed them (sometimes more than 50g/dayy4-16. We have con-
firmed these results in rats fed a diet containing 10% and even 20%, and in human volun-
teers consuming 109 or 30g of chicory fructooligosaccharides (unpublished data).
Using an in vitro model of mixed faecal culture, Wang and Gibson ll have demon-
strated that, as compared to other carbohydrates like polydextrose, pectins, fructose or glu-
cose, the chicory fructooligosaccharides are selectively hydrolysed and fermented by the
Bifidobacteria, one of the health promoting population in the colonic microbiota. Because
of such a selectivity, the Bifidobacteria are stimulated to overgrow and they become pre-
dominant. Using pure cultures, the same authors have demonstrated that the specificity of
Bifidobacteria is due to their rather unique capacity to secrete a f3-fructosidase which
hydrolyses the f3 1-2 osidic bonds of the chicory fructooligosaccharides 9 • lo • The same
specificity of Bifidobacteria to selectively hydrolyse and ferment fructooligosaccharides
has been reported for the synthetic product NeosugarR. That the same selective stimulation
of colonic Bifidobacteria to overgrow also occurs in vivo has clearly been demonstrated
 N
...
""

Table 1. Non-digestible oligosaccharides

Name (abbreviation) Chemical structure Osidic bond Origin Trade name
Fructooligosaccharides (FOS) Inulin Glucosyl (fructosyl)n fructose(n=2-760) ~H2 Plant Raftiline®
Fibrulin®
Oligofructose Glucosyl (fructosyl)n b 1-72 Plant and enzymatic hydrolysis of inulin Raftilose®
Fructose (fructosyl)m fructose(n=I-76, 11=2-77)
Neosugar Glucosyl (fructosyl)n fructose(n=I-73) b 1-72 Enzymatic synthesis from sucrose Neosugar®
Actilight®
Galactooligosaccharides (GOS or TOS) Glucosyl (galactosyl)n galactose(n=I-73) b 1-76 Enzymatic synthesis from lactose Oligomate®
Transgalacto oligosaccharides (Galactosyl)n galactose (n=2) ~ 1-76 Enzymatic synthesis from lactose Cup-Oligo®
Isomalto oligosaccharides (IMO) (Glucosyl)n glucose(n= 2-77) a 1-76 Enzymatic rearrangement of maltose Isomalto®
IX1-74
Polydextrose Randomly branched + citric acid(n=2-7100?) I glucose pyrolysis - citric acid Polydextrose®
Pyrodextrins Complex mixture Pyrolysis of corn or potato starch I
Soyoligosaccharides(SOS) Raffinose + Stachyose(n=3-4) Enzymatic synthesis + pyrolysis Soya-Oligo®

~
!=
:=
Q
r:t'
..:;.
Q
s:
Health Benefits of Non-Digestible Oligosaccharides 215

by Gibson et al. 11 who reported that, after ingesting chicory fructooligosaccharides

(3x5g/day) for two weeks, the composition of the faecal flora of human volunteers had
been modified dramatically, the ratio BijidobacterialBacteroides changing from 114 to
4/1. A more recent study has shown the same effect after the ingestion of 8g/day for
5 weeks (N.Guggenbuhl et al. Institut Paul Lambin Brussels, Belgium personal communi-
cation). Similar studies performed with the synthetic NeosugarlZ.1 3 have led essentially to
the same conclusions. Fructooligosaccharides in general and chicory fructooligosaccha-
rides in particular are thus bifidogenic dietary factors.

2.3. Bulking Effect

Both in experimental animal (rat)17 and in human volunteers I I, adding chicory fruc-
tooligosaccharides to the diet is accompanied by an increase in stool weight. In average,
this corresponds to an excess of wet faecal mass of 1-1.5g per g of fructooligosaccharides
eaten. Comparable values have been reported for wheat and oat bran or pectin added to
rat's diet l8 . In human studies, fructooligosaccharides are less efficient than wheat bran l9-22
but as efficient as pectin or guar gumZ3 - 25 (table 2).

2.4. Chicory Fructooligosaccharides as Prebiotics

and Functional Food Ingredients
Based on the data available, it can thus be hypothesized that chicory as well as other
similar fructooligosaccharides have a significant impact on both the biomass and the compo-
sition of the gut micro flora by selectively stimulating the growth of the Bifidobacteria which
are potentially health promoting and, consequently, by reducing the number of potentially
harmful anaerobes. These food ingredients are thus likely to be beneficial and for these
reasons, they have classified as prebiotics. But the exact role of the gut microflora in health
as well as the definiton of an healthy colonic micro biota still remain open questions 26 •
By reference to the strategy for research and development of a "functional food", it
has been demonstrated that:
a. chicory and similar fructooligosaccharides interact specifically with a key com-
ponent of the colonic physiological functions i.e. the biomass and the composi-
tion of the microbiota,

Table 2. Bulking effect of chicory fructooligosaccharides as compared to dietary fiber

Species Carbohydrates Bulking index' Reference

Rats Oligofructose (10%) +1 17
Inulin (10%)
Pectin (I 0%) +1 18
Oat bran (14%) +0.7 18
Wheat bran (13%) +1.3-2.0 18
Humans Oligofructose (I5g) +1.2 II
Inulin (15g) +2.1
Pectin +1-2 23-25
Guar gum +1-2 23-25
Wheat bran (I5g) +2.5-5 19-22
'Bulking index = g of increase in fresh faecal weight Ig of carbohydrates eaten
() Indicates either % added to rat diet or g eaten per day.
216 M. B. Roberfroid

b. both in experimentam systems and in human volunteers, such an interaction

leads to a dramatic change in the composition of the colonic microflora in which
the Bifidobacteria become the predominant population.
The obvious conclusion of these data is that chicory and similar fructooligosaccha-
rides are functional food ingredients for which, at the present time functional but not
health claims can be justified.
Moreover, the non-digestibility of these carbohydrates as well as their stool bulking
effect justify their classification as dietary fibre or complex carbohydrates.

3. SYSTEMIC EFFECTS OF CHICORY

FRUCTOOLIGOSACCHARIDES

3.1. Effects on Lipid Metabolism

The main systemic effect of chicory fructooligosaccharides so far reported is their effect
on lipid metabolism. Such an effect had previously been suggested by Japanese studies which
had shown that the administration of short chain synthetic fructooligosaccharides reduced
serum lipids concentration. In various protocols differing in the composition ofthe basal diet,
the dose of chicory fructooligosaccharides and the duration of the feeding treatment we have
been able to demonstrate a significant decrease in serum lipidemia, thus identifYing a possible
interaction between these food ingredients and systemic lipid homeostasis. In a standard
protocol in which male rats are fed a diet containing 10% chicory fructooligosaccharides, a
significant hypotriglyceridemia (-30 to -40%) appears after 2-3 weeks but lowering choles-
terolemia requires a longer treament (10 weeks or more). The early hypotriglyceridemic
effect is exclusively due to a reduction in the VLDL fraction of the serum lipoproteins 27 •
In order to proceed with the development of chicory fructooligosaccharides as
hypolipidemic functional food ingredients, and by reference to the strategy described
above (see 1.1), it was necessary to formulate sound hypotheses to be tested in human
nutrition studies. For doing so more mechanistic informations were needed based on the
hypothesis that hypotriglyceridemia induced by chicory fructooligosaccharides feeding
resulted from a metabolic adaptation in the hepatic lipid metabolism.
Using an ex vivo protocol Kok et a1. 28 ,29 demonstrated that, as compared to control
liver cells, hepatocytes isolated from rats fed a diet supplemented with chicory fructooli-
gosaccharides (l 0%) incorporated less (-40%) 14C-Iabelled acetate in both their intracellu-
lar triglycerides and extracellular VLDL. Furthermore they reported that all the liver
enzyme activities related to lipogenesis including glucose-6-phosphate dehydrogenase and
malic enzyme which produce NADPH, fatty acyl synthase, ATP-citrate lyase and acetyl
CoA carboxylase which all catalyse key steps in the synthesis of fatty acids were signifi-
cantly reduced and to a similar extent (-40%) (table 3). Preliminary results seem also to
indicate that these changes in hepatic lipogenic activity, possibly due to a change in gene
expression, could, at least partly, be due to a decrease in postprandial insulinemia and pos-
sibly an increase in insulin sensitivity (Kok and Delzenne personal communication).
Being still at the level of experimental evidences, these systemic functional effects
of chicory fructooligosaccharides cannot yet be used to support functional claims but they
serve to formulate new hypotheses and to design protocols f0r further human nutritional
studies. Functional interactions have been identified in experimental models which now
need to be tested in humans.
Health Benefits of Non-Digestible Oligosaccharides 217

Table 3. Relative values of lipid parameters in rats fed a

chicory fructooligosaccharides containing (FOS) or a control
diet (CTRL) (See ref28,29 for experimental protocols)
Lipid parameters (FOS / CTRL) x 100
Parameters Serum Liver
Triacylglycerols 60 74
Insulin 77
Fatty acyl synthase 57
Malic enzyme 54
Glucose-6-phosphate dehydrogenase 56
AcetylCoA carboxylase 60

3.2. Bioavailability of Micronutrients

The possible interference of dietary fibers with the absorption of ions has regularly
been questionned. With chicory tructooligosaccharides the effect could be quite the oppo-
site. Indeed, in growing rats fed a diet supplemented with 10% of these 3o•31 and similar
fructooligosaccharides 32 , the balances of Ca, Mg and Fe were improved significantly pos-
sibly because of an increased intestinal absorption of the minerals. Neither Cu or Zn
bioavailability, nor lipid soluble vitamins (A and E) homeostasis were affected by the
chicory fructooligosaccharides.

3.3. Caloric Value

Like with the other non digestible carbohydrates, the colonic fermentation of chic-
ory fructooligosaccharides produces short chain carboxylic acids mainly acetate,
propionate and butyrate. But the colonic fermentation of inulin has been reported to pro-
duce 4 times more butyrate than other dietary fibre like wheat and oat brans, pea and
carrot fibers 33 , an effect which, if confirmed could be of great interest since it is hypothe-
sized that butyrate could playa role in preventing carcinogenesis. Moreover the fructoli-
gosaccharides serves as an energy source for the bacteria to support their metabolic
activity as well as their growth and proliferation. This complex process has to be taken
into consideration when evaluating the caloric value of such non-digestible oligosaccha-
rides in other words their contribution to energy metabolism of the host. The key ques-
tion is to evaluate quantitatively which part of the initial energy content of the
non-digested but fermented carbohydrate is available for the metabolism of the host,
which part is used by the bacteria for their metabolic activity, which part serves their
growth and proliferation and is, consequently lost in the faecal biomass. Based on
experimental data as well as on basic biochemical know ledges, a theoretical approach
has been used to compare the metabolic efficiency (in term of ATPproduction) and thus
the effective caloric value of a fructosyl unit in chicory fructooligosaccharides and in
sucrose (as a model of fully digested carbohydrate). As compared to sucrose, a fructosyl
unit in fructooligosaccharides has a relative caloric value of 25 to 35%, it means 1 to 1.5
Kcal/g or 4 to 6 KJ/g 17 . They are thus in the range of the values generally accepted for
non-digestible carbohydrates which vary from 0 to 2 Kcal/g or 0 to 8 KJ/g. This justifies
their use as bulking agent, sugar substitute and, for inulin, as fat replacer in miscellane-
ous food products34 .
218 M. B. Roberfroid

4. DISCUSSION AND CONCLUSION

Research and development of "functional foods" opens new perspectives in nutri-

tion and food sciences35 . The systematic investigation of the interactions between food
components or food ingredients and genomic, biochemical, cellular or physiological
functions completed by the understanding of their mechanisms, is a unique way to
improve both our knowledge and the role of nutrition in maintaining good health and
preventing diseases. But such a basic knowledge is insufficient to justify "claims" unless
it is completed by relevant nutrition studies aimed at demonstrating the relevance of
these interactions and their positive consequences in humans. In a first stage, this demon-
stration will, in most cases justify "functional claims with no direct reference to any
health benefit. A "true" health claim wil require further studies involving large popula-
tions and long term trials.
But it is anticipated that the better we understand the mechanism of interactions
between food components and biological functions, the better we will be in a position to
formulate sound hypotheses to be tested in properly designed human nutrition trials based
on which we will demonstrate significant functional effects. We will then accumulate con-
vincing arguments in favor ofliealth promotion and disease prevention.
Within the vast group of dietary carbohydrates, the non-digestible oligosaccharides
including the chicory fructooligosaccharides belong to the group of non a-glucans. From a
physiological point of view, the fructooligosaccharides resist digestion by the intestinal
enzymes but they are quantitatively fermented in the colon most exclusively by the Bifidobac-
teria which become stimulated to grow. This specific effect has been well established in
humans, and it justifies their classification as prebiotics and the functional claim of "bifido-
genesis". Moreover they can be and are indeed, in most european countries, classified as die-
tary fibre. Via their systemic effects, the chicory fructooligosaccharides might also modulate
hepatic lipid metabolism, but such an effect still needs to be demonstrated in humans.

REFERENCES
I. Roberfroid M. B. (1995) World of Ingredients March-April 42-44.
2. Delzenne N.M. & Roberfroid M.B. (1994) Lebensm.- Wiss. u.- Technol. 27, 1-6.
3. Gibson G.R. & Roberfroid M.B. (1994) J. Nutr. 125, 1401-1412.
4. Van Loo J., Coussement P., De Leenheer L., Hoebregs H. & Smits G. (1995) CRC Crit. Rev. Food Sci.
Technol. 35(6), pp. 525-552.
5. Bach Knudsen K.E. & Hessov I. (1995) Br. J. Nutr. 74, 101-113.
6. Ellegiird L., Anderson H. & Bsaeus I. (1996) Eur. J. Clin. Nutr. in press.
7. Tokunaga T., Oku T. & Hosoya N. ( 1989) J. Nutr. 119,553-559.
8. Wang X. & Gibson G. (1993) J. Appl. Bacteriol. 75, 373-380.
9. McKellar R.C. & Modler H.W (1989) Appl. Microbiol. Biotechnol. 31, 537-541.
10. Mc Kellar R. c., Modler H. W. & Mullin 1. (1993) Bifidobacteria Microflora 12, 75-86.
II. Gibson G.R., Beatty E.B., Wang X. & Cummings J.H. (1995) Gastroenterology 108, 975-982.
12. Williams C.H., Witherly S.A. & Buddington R.K. (1994) Microb. Ecolog. Health Dis. 7, 91-97.
13. Buddington R. K., Williams C. H., Sui-Chi Chen & Witjerly S. W (1996) Am. J. Clin. Nutr. 63, 1-8.
14. Wise E. c., Heyl F.W (1931)1. Amer. Pharm. Ass. 20, 26-29.
15. Heupke W, Blanckenburg K. (1934) Deut. Arch. Klin. Med. 176, 182-187.
16. Beringer A. and Wenger R. (1955) Dtsch. Zeitschr. Verdauungs und Stoffwechselkrankh. 15,268-272.
17. Roberrroid M., Gibson G.R. & Delzenne N. (1993) Nutr. Rev. 51,137-146.
18. Armstrong E.F., Eastwood M.A., Brydon WY. (1993) J.Nutr. 69, 913-920.
19. Hansen L., Knudsen K.B., Eggum B.O. (1992) BrJ.Nutr. 68,451-462.
20. Jenkins DJ.A., Peterson R.D., Thome MJ., Ferguson P.W (1987) Am.1.GastroenteroI.82, 1259-1263.
Health Benefits of Non-Digestible Oligosaccharides 219

21. Tomlin J. and Read N.W. (1988) Eur.J.Clin.Nutr. 42. 857-861.

22. Muller-Lissner S.A. (1988) Br.MedJ. 296, 615--617.
23. Cummings J.H., Hill J.M., Jenkins J.A .• Pearson J.R., Wiggins H.S. (1976) Am.J.Clin.Nutr. 29.1468--1473.
24. Cummings J.H. (1978) Lancet i, 5-9.
25. Durrington et a!. (1978) Lancet ii. 394-396.
26. Roberfroid M.B., Bornet F.. Bouley C. & Cummings J.H. (1995) Nutr. Rev. 53, 127-130.
27. Fiordaliso M.F., Kok N.• Desager J.P., Goethals F.. Deboyser D., Roberfroid M. & Delzenne N. (1995)
Lipids 30, 163--167.
28. Kok N., Roberfroid M. & Delzenne N. (1995) Arch. Biochem. Physiol. 103. B19.
29. Kok N. Roberfroid M, Robert A. & Delzenne N. (1996) Br. J. Nutr. in press.
30. Delzenne N., Aertsens J.. Verplaetse H., Roccaro M. & Roberfroid M. (1995) Life Sci. 57, 1579-1587.
31. Levrat M., Remesy C. & Demigne C. (1991) J. Nutr. 121, 1730--1737.
32. Ohta A., Osakabe N, Yamada K.• Saito Y. & Hidaka H. (1993) J. Jap. Soc. Nutr. Food Sci. 46.123--129.
33. Roland N., Nugon-Baudon L., Andrieux C. & Szylit O. (1995) Br. J. Nutr. 74,239-249.
34. Franck A. (1992) Food Ingredients Europe. conference proceedings. pp. 193--197.
35. Roberfroid M.B. (1996) Nutr. Rev. in press.