The present experiment showed that four-week scFOS ingestion, with a dose of 8 g/d, is well tolerated and leads to significant increase in faecal bifidobacteria and cholesterol excretion in healthy elderly. The sc-FOS bifidogenic effect has been extensively demonstrated in adults [6, 26–28], but rarely in elderly [19, 21, 22]. Among the very few available studies about functional foods in elderly, one recent double-blind trial testing a symbiotic (B. lactis BL-01, B. bifidum BB-02 and an inulin-based prebiotic) also found promising results . Significant increase in total bifidobacteria counts was indeed observed in the symbiotic group compared with the placebo group.
We also found increasing Clostridium spp. after sc-FOS ingestion discontinuation. Clostridium spp. is a major component of normal anaerobic microflora and it can not be considered as a deleterious or beneficial genus. Since some toxinogenic subspecies of Clostridium difficile are related to an increased risk of pseudo membranous colitis and/or infection in older people, it would have been interesting to measure the sc-FOS effects on these subspecies. However, we did not perform those analyses, for they were out of our study scope. Further studies may investigate this point, using adequate measurement methods for species concentrations and toxinogenic properties (cellular cultures, biomolecular analysis). Culture-based enumeration of microbiota does not usually allow for bacterial species measurement, but mainly bacterial genus.
In our study, several parameters were assessed with the objective to better understand scFOS physiological effects in healthy elderly, such as transit time, stool characteristics, and colonic environment. We did not find scFOS ingestion changed faecal weight and oro-fecal transit time in elderly. Gibson et al. have shown that prebiotics can increase stool output: they studied 8 volunteers under controlled diet, and showed that with 15 g/d fructo-oligosaccharides, stool output significantly increased from 136 to 154 g/d . Other two human studies did not demonstrate increasing stool output [2, 30]. but the diet was not controlled in none of these studies, which may have hidden any slight effect. In the study of Alles et al., 12 healthy subjects were given 4.8–19.2 g/d oligomate (52% galacto-oligosaccharides), which did not result in any change in bowel habit. However, the subjects started with unusually high faecal weights under controlled diet, 272 ± 26 g/d On the other hand, studies using probiotics demonstrated bifidobacteria could reduce human colonic transit time, but not all bifidobacteria strains have the same effects . This specific strain-dependent effect could explain the reason why our prebiotic, which stimulates global endogenous bifidobacteria, had no effect.
In our study, the microbial transformation of cholesterol into coprostanol was not influenced by scFOS ingestion. Another study observed that sterol and fatty acid biohydrogenation by intestinal microflora is altered by oligosaccharide fermentation . Coprostanol production results from intestinal anaerobic bacteria action . Concerning bile acid metabolism, no differences were observed during the three periods. Furthermore, the use of poorly digestible carbohydrates in rats, hamsters and pigs demonstrated that prevention of microbial conversion of bile acids depended on the carbohydrate dose in the diet [23, 34] This suggests that the low carbohydrate dose, 8 g/d scFOS, used in this experiment is unable to modify microbial conversion of bile acids.
Endogenous or exogenous bile acids, as well as dietary cholesterol are carcinogenic factors involved in colon cancer in laboratory animals [35, 36] Various epidemiological studies suggest those steroids could also be involved in colon cancer in men [12, 37]. According to these studies, low scFOS dose ingestion by humans, which prevented microbial conversion of cholesterol into cytotoxic molecule, (coprostanol, potentially carcinogenetic), could be interesting for humans. In our study, the intake of 8 g/d scFOS led to increasing faecal cholesterol. The mechanism of such increase could be related to decreasing cholesterol bacterial transformation, although we failed to find any significant sc-FOS effect on cholesterol bacterial metabolism. Moreover, the low scFOS dose used in our study was also probably not sufficient to significantly reduce microbial conversion of bile acids. However, in our previous study evaluating a higher scFOS dose (12.5 g/d), we also failed to show any significant effect in bile acids and neutral sterol . These negative results could be explained by a questionable capacity of various bifidobacteria to take up cholesterol into their cellular membrane .