To investigate the influence of different pre- and probiotics on faecal β-glucuronidase and β-glucosidase activity, as one of the claimed beneficial effects of pre- and probiotics is the hypothesis that these substrates are able to reduce the production of toxic and carcinogenic metabolites by suppressing specific enzyme activities in the colon.
Department of Gastrointestinal Research, University Hospital Gasthuisberg, KU Leuven, Belgium.
Design and subjects:
The effect was evaluated in a randomized, crossover study in 53 healthy volunteers who were randomly assigned to one of five treatment groups.
At the start and after a 4-week treatment period, the healthy volunteers collected faeces during 72 h. Lactulose and oligofructose-enriched inulin (OF-IN) were chosen as prebiotics, whereas Lactobacillus casei Shirota, Bifidobacterium breve and Saccharomyces boulardii were selected as probiotics. Two synbiotic combinations were evaluated as well. The enzyme activity was assessed spectrophotometricly.
Lactulose and OF-IN significantly decreased β-glucuronidase activity, whereas a tendency to a decreased β-glucuronidase activity was observed after L. casei Shirota and B. breve intake. To the contrary, B. breve increased β-glucosidase levels. Supplementation with the synbiotic did not appear to be more beneficial than either compound alone. No influence of S. boulardii was noted.
Administration of lactulose, OF-IN, L. casei Shirota or B. breve resulted in a decrease of the β-glucuronidase activity, which is considered beneficial for the host.
During the past decades, evidence is accumulating that both genetic and environmental factors may contribute to the progression and aetiology of colonic cancer in human beings (Moore and Moore, 1995; Heavey et al., 2004). Several risk factors have been identified such as dietary fat and high consumption of red meat, and have been shown to be associated with an increased risk of colon cancer (Goldin and Gorbach, 1976; Bingham, 1999). By contrast, a high intake of fruits, vegetables and whole grain cereals has been associated with a reduced risk (Bingham, 1999). The interaction between these factors, in particular diet, is an area of growing interest.
Any compound that has escaped digestion in the small intestine, any substance entering the colon by diffusion from the blood stream or via the biliary tract, or any substance secreted directly into the lumen of the colon, is a potential substrate for bacterial fermentation (Goldin, 1990). Hydrolysis of glycosidic bonds is one of the best-known examples of bacterial metabolism. Glycosides in the gut originate either from the diet or are excreted by the liver via the bile. The diet contains numerous plant glycosides, comprised predominantly of flavonoids (Takada et al., 1982; Goldin, 1990). Glycosides excreted from the liver include compounds that are detoxified by glucuronide formation and are subsequently secreted into the bowel via the bile (Rowland et al., 1985). The intestinal microbiota can hydrolyse the glycosidic bond resulting in the release of active aglycons, some of which are potentially toxic or carcinogenic (Takada et al., 1982). The principal glycosidases produced by the intestinal flora are β-glucuronidase and β-glucosidase, the carcinogenic potential of which has been described in a number of studies (Goldin, 1990).
Modulation of bacterial enzyme activity has been described as one of the mechanisms through which pre- and probiotics exert their beneficial effects. Bifidobacteria and lactobacilli are the most widely studied probiotic genera and have been shown to exert cancer protective effects in vitro and in vivo (Lidbeck et al., 1992; Pool-Zobel et al., 1993). These organisms have low activities of enzymes involved in the conversion of procarcinogens into potentially carcinogenic compounds (Nakamura et al., 2002). Several studies have described a significant reduction in the bacterial enzyme activity after probiotic administration (Goldin et al., 1980; Goldin and Gorbach, 1984; Ling et al., 1994; Bouhnik et al., 1996; Guerin-Danan et al., 1998; Spanhaak et al., 1998; Haberer et al., 2003), whereas other studies did not find significant effects (Marteau et al., 1990; Tannock et al., 2000; Goossens et al., 2003). On the other hand, prebiotics have been shown to increase intestinal bifidobacteria concentrations and to suppress faecal activities of carcinogen-metabolizing enzymes in humans and rats (Bauer et al., 1979; Mallett et al., 1987; Ballongue et al., 1997; Van Dokkum et al., 1999; Hughes and Rowland, 2001).
In the present study, the influence of long-term administration of two different prebiotic substrates (i.e. lactulose and oligofructose-enriched inulin (OF-IN)), three different probiotic organisms (i.e. Lactobacillus casei Shirota, Bifidobacterium breve and Saccharomyces boulardii) and two synbiotic combinations (i.e. lactulose in combination with S. boulardii and L. casei Shirota in combination with OF-IN) on β-glucuronidase and β-glucosidase activity was evaluated.
Materials and methods
Fifty-three healthy volunteers (25 women and 28 men, age range 19–26 years) participated in the study. None of the subjects had a history of gastrointestinal or metabolic disease or previous surgery (apart from appendectomy). The subjects were free of antibiotics or any other medical treatment influencing gut transit or intestinal flora for at least 3 months before the start of the study. The Ethics Committee of the University of Leuven approved the study protocol and all subjects gave their written informed consent before participation.
The healthy volunteers were randomly assigned to one of five treatment groups of a placebo-controlled trial to study the influence of long-term pre- and/or probiotic intake on bacterial enzyme activity. The study consisted of three (groups 1–4) or four (group 5) experimental periods of 4 weeks. The experimental periods of groups 1, 2 and 3 were separated by a 3-day interval during which no substrates were taken, whereas in groups 4 and 5, the experimental periods were separated by a 2-week washout period. During each treatment period, each subject received two different substrates twice a day as indicated in Table 1. In groups 1 and 2, the influence of different doses of the same pre- and probiotic substrates was evaluated, whereas in group 3 the influence of the synbiotic combination of the same pre- and probiotic was evaluated as well. In groups 4 and 5, the effect of B. breve and L. casei Shirota as probiotics, respectively, and OF-IN as prebiotic was investigated. Throughout the study, the volunteers consumed their usual diet, taking care that the diet remained as stable as possible over the three/four periods. In addition, they were advised to avoid intake of fermented milk products and food components containing high quantities of fermentable carbohydrates. Three days before the start of the study and during 3 days after each treatment/washout period, all stools were collected during 72 h for determination of total faecal output and faecal dry weight.
Lactulose (Duphalac, Solvay Pharma & Cie, Brussels, Belgium) and OF-IN (Raftilose-Synergy1, Orafti, Tienen, Belgium) were chosen as prebiotic substrates (Jenkins et al., 1999; Roberfroid, 1999; Schumann, 2002; Bouhnik et al., 2004). As probiotic substrates, L. casei Shirota cells, as a single strain fermented milk product (6.5 × 109/65 ml), B. breve Yakult cells (both provided by Yakult, Yakult Honsha Co. Ltd, Tokyo, Japan) and S. boulardii cells (Perenterol, Biodiphar, Dübendorf, Switserland) were chosen (Spanhaak et al., 1998; Marteau et al., 2001; Fioramonti et al., 2003). The placebo for the L. casei Shirota was an identical milk product without the L. casei Shirota strain (provided by Yakult, Yakult Honsha Co. Ltd, Tokyo, Japan). In all other cases, maltodextrin (AVEBE BA Food, Foxhol, The Netherlands) was used as placebo.
General laboratory chemicals including phenolphthalein β-glucuronide and p-nitrophenyl β-pyranoside were purchased from Sigma-Aldrich (St Louis, MO, USA).
Bacterial enzyme activities
The β-glucuronidase and β-glucosidase activities were determined aerobically according to the method of Goldin et al. (1980). Fresh faecal samples (100 mg/ml) were suspended in cold 0.1 M potassium phosphate buffer (pH 7.0). The faecal suspension was first homogenized, then centrifuged at 1200 g for 10 min and the supernatant was passed through a 1.2-μm filter (Pall Newquay, Cornwall, UK). The filtrate was sonicated for 10 min and subsequently centrifuged at 1200 g for 15 min. The supernatant was used for the assessment of β-glucuronidase and β-glucosidase activities.
Faecal supernatant (0.1 ml) was added to 0.9 ml reaction mixture (0.02 M potassium phosphate buffer, 0.1 mM ethylenediaminetetraacetic acid and 0.05 mM phenolphthalein β-glucuronide). The enzyme reaction was run for 60 min at 37°C and was stopped by addition of 5 ml of 0.2 M glycine buffer (pH 10.4) containing 0.2 M NaCl. For each sample, a control (faecal supernatant with reaction mixture without phenolphthalein β-glucuronide substrate) was determined which was incubated for the same time period. The amount of phenolphthalein released from the enzymatic reaction was quantified by measuring the ultraviolet (UV) absorption at 550 nm with an UV-VIS spectrophotometer (Kontron Uvikon 810, Beun De Ronde Serlabo, Drogenbos, Belgium). Concentrations were calculated, after correction for controls, from a standard curve of phenolphthalein and were expressed as milligrams of phenolphthalein per gram faeces per hour.
Faecal supernatant (0.2 ml) was added to 0.8 ml reaction mixture (0.1 M phosphate-buffered saline, 1 mM p-nitrophenyl β-pyranoside). The reaction was run for 60 min at 37°C and was stopped by addition of 5 ml of 0.1 M NaOH. For each sample, a control (faecal supernatant with reaction mixture without p-nitrophenyl β-pyranoside substrate) was determined which was incubated for the same time period. UV absorption was measured at 420 nm. Concentrations were calculated, after correction for controls, from a standard curve of p-nitrophenol and were expressed as milligrams of p-nitrophenol per gram faeces per hour.
Results are expressed as mean values and standard deviations. The statistical analysis was performed with SPSS software (SPSS 12.0 for Windows; SPSS Inc., Chicago, IL, USA). Given the low number of subjects in the treatment groups, non-parametric statistical analysis was used regardless of the distribution of results (Wilcoxon test). Each time the enzyme activity levels before and after intake of a substrate were compared. When the results of different groups were combined, the statistical evaluation of the data was performed by applying a Student's t-test. The level for statistical significance was set at P<0.05.
Fifty participants completed the study. One male and one female subject withdrew from the study because of the necessity of antibiotic intake (groups 1 and 5) and another female volunteer withdrew from the experiment because of personal reasons (group 2). Data from these subjects were excluded from analysis. No statistically significant differences were observed between the demographic characteristics of the five groups, as shown in Table 2.
Administration of lactulose resulted in a decreased β-glucuronidase activity in groups 1, 2 and 3 compared with baseline, although this reduction was only statistically significant in group 1 (P=0.007). However, combination of the results of groups 1 and 3 (n=21; both groups consumed 2 × 10 g lactulose per day during the first ingestion period) resulted in a significant decrease of the enzyme activity after lactulose intake (from 0.73±0.27 to 0.50±0.29 (P=0.001)). No dose-dependent effect of lactulose was observed. In these groups, intake of the probiotic S. boulardii cells did not result in a significant difference in the β-glucuronidase activity. Administration of the synbiotic combination of S. boulardii cells with lactulose resulted in a decreased β-glucuronidase activity, although this effect was not significant.
The probiotic intake of B. breve cells in group 4 resulted in a slight decrease of the β-glucuronidase activity compared with baseline. In group 5, dietary intervention with L. casei Shirota cells resulted in a reduction of the faecal β-glucuronidase activity compared with baseline, although this effect was not significant. Intake of OF-IN in groups 4 and 5 resulted in a statistically significant reduction of the faecal β-glucuronidase activity compared with washout 1, respectively P=0.018 and P=0.011. Administration of L. casei Shirota cells in combination with OF-IN decreased the faecal β-glucuronidase activity, but not significantly. Placebo intake did not result in a significant effect on faecal β-glucuronidase activity in the different groups.
Intake of the prebiotic substrates lactulose and OF-IN did not result in a significant effect on faecal β-glucosidase activity. In addition, no influence of S. boulardii and L. casei Shirota, either as such or in combination with a prebiotic, on β-glucosidase activity was observed, whereas administration of B. breve cells significantly increased β-glucosidase activity compared with baseline (P=0.015).
The assessment of bacterial enzyme activities is often used to demonstrate diet-related changes in the colon and may provide complementary information on the effect of dietary intervention on the modulation of the gut microbiota (Goldin and Gorbach, 1976; Ling et al., 1994). The detection of diet-related changes in the activity of these enzymes may be of toxicological importance to the host. The decreased β-glucosidase and β-glucuronidase activities may confer benefit by limiting the microbial production of aglycons involved in the pathogenesis of colorectal cancer (Goldin, 1990).
Most studies described in literature in which bacterial enzyme activities were determined, have investigated the influence of the actual presence of a pre- and/or probiotic substrate on bacterial enzyme activity. To the contrary, in our study, we investigated whether long-term dietary intervention with different pre- and/or probiotic substrates would be able to alter the bacterial activity in the large intestine. As the tests were performed in the absence of either the pre- and/or probiotic, the observed effects cannot be attributed to the actual fermentation of the pre- and/or probiotic substrate. During the second or third experimental period, placebo was administered to all groups, except for the volunteers of group 3. At the end of this period, no statistical significant differences were observed for the bacterial enzyme activities compared with baseline or washout levels. In the second period of group 3 as well as in the fourth experimental period of group 5, the influence of a synbiotic combination was tested to investigate whether previous effects could be maintained.
Furthermore, in this study, the effects of pre- and/or probiotics were evaluated under normal conditions, as pre- and probiotics are often recommended as food supplements for healthy people under medically stable circumstances. Therefore, the healthy volunteers did not consume standardized diets, but were asked to continue their normal dietary habits. As a consequence, no data are available on the components reaching the colon during the period of faecal collection. However, as the volunteers were not able to discriminate effective treatment from placebo, it was considered unlikely that the observed significant effects were due to casual changes in the diet, and as a consequence, in the components reaching the colon.
The results of the present study have indicated that the prebiotic substrates lactulose and OF-IN are able to decrease the β-glucuronidase activity in the colon. The lower enzyme activity was probably due to the effect of these substrates on the relative composition of the colonic microbiota. It is generally known that dietary intervention with lactulose or inulin selectively stimulates the growth and activity of bifidobacteria (Buddington et al., 1996; Roberfroid et al., 1998; Bouhnik et al., 2004; Roberfroid, 2005). It has been shown before that increased levels of bifidobacterial species can be perceived as beneficial and are of specific interest in the prevention of colon carcinogenesis (Buddington et al., 1996). Furthermore, it has been reported that lactobacilli and bifidobacteria produce low levels of β-glucuronidase, but also of azoreductase and nitroreductase, whereas strict anaerobes (Bacteroides sp., Eubacterium sp. and Clostridium sp.) produce high levels of these enzymes (Nakamura et al., 2002). Presumably, the increase in bifidobacteria is accompanied by a decrease in bacterial species with high β-glucuronidase activity.
The results of the β-glucuronidase and β-glucosidase assays have been expressed as milligram product formed per gram wet faeces and not as a function of the amount of bacteria. As a consequence, dilution of faecal samples resulting in fewer bacteria per gram wet faeces could also result in lower enzyme activities. Several non-digestible carbohydrates, and in particular lactulose, are known to act as laxatives and hence, could cause dilution of faeces (Stephen and Cummings, 1980; Schneeman, 1999). However, the fact that no changes were induced on total faecal output nor on faecal dry weight allowed us to conclude that the changes in faecal enzyme activity were not due to a dilution of the faecal samples.
On the other hand, the influence of three different probiotics on bacterial enzyme activity was evaluated. After long-term dietary administration of L. casei Shirota cells a decrease in β-glucuronidase activity was observed, whereas no effect on β-glucosidase was noted. Although the decrease in β-glucuronidase activity was not statistically significantly decreased in the present study, the combination of the present results with data obtained in a preliminary study in 10 healthy volunteers (four females and six males, age range 20–22 years), in which an identical treatment period for L. casei Shirota cells was set up (reduction of the β-glucuronidase activity from 0.98±0.31 to 0.76±0.29 (P=0.017); unpublished results), resulted in a significant reduction of β-glucuronidase activity from 1.01±0.29 to 0.79±0.27 (n=19; P=0.010). These data confirm the results of a previous study in which the effect of treatment with L. casei Shirota on bacterial enzyme activity was investigated (Spanhaak et al., 1998). Other studies in animals and humans in which different Lactobacillus species were evaluated, did not demonstrate an effect on β-glucuronidase activity (Goldin et al., 1980; Goldin and Gorbach, 1984; Marteau et al., 1990; Ling et al., 1994; Tannock et al., 2000; Goossens et al., 2003; Haberer et al., 2003), confirming the species-specificity of the observed effects.
Administration of B. breve cells resulted in a decrease of β-glucuronidase activity. Bouhnik et al. (1996) also observed a decrease in faecal β-glucuronidase activity after supplementation of the diet with a Bifidobacterium strain. This reduced activity may be due to the relative increase in bifidobacteria, as this genus expresses a low β-glucuronidase activity in comparison with other bacterial genera (Nakamura et al., 2002). Contrary to β-glucuronidase, the β-glucosidase activity significantly increased after B. breve administration. This effect was also observed in a study of Marteau et al. (1990) after supplementation with a fermented dairy product containing L. acidophilus, B. bifidum and mesophilic cultures (Streptococcus lactis and S. cremoris) for 3 weeks. The increase in colonic β-glucosidase is possibly due to the fact that bifidobacteria, but also lactobacilli, possess higher levels of β-glucosidase activities (Rowland and Tanaka, 1993).
No influence on the bacterial enzyme activity was observed after administration of S. boulardii cells. This effect was not unexpected, as it has been shown previously that S. boulardii does not cause changes in the relative bacterial populations present in the colon of healthy volunteers (De Preter et al., 2006).
The combination of both pre- and probiotics has been proposed to increase their respective potential effects on intestinal microbial composition. However, the effect of the two synbiotic combinations evaluated in the present study did not appear to be more beneficial than either compound alone.
In conclusions, in view of the carcinogenic potential of these enzymes, the induced changes that occurred in this study by lactulose, OF-IN, L. casei Shirota and B. breve on the bacterial β-glucuronidase activity could be considered as beneficial for the host and may have important implications for health.
This work was supported by IWT-Vlaanderen, Brussels, Belgium (GBOU project no. 010054), the Fund for Scientific Research-Flanders, the University Research Councils and several companies. L De Vuyst, G Huys, J Swings and B Pot are acknowledged for providing scientific comments.
About this article
European Journal of Clinical Nutrition (2009)