The Journal of Nutrition Vol. 128 No. 8 August 1998, pp. 1385-1391

Dietary Guar Gum and Pectin Stimulate Intestinal Microbial Polyamine Synthesis in Rats¹

Jutta Noack², Brigitta Kleessen, Jürgen Proll, Gerhard Dongowski, and Michael Blaut

German Institute of Human Nutrition Potsdam-Rehbrücke, Department of Gastrointestinal Microbiology, 14558 Bergholz-Rehbrücke, Germany

	ABSTRACT

Abstract Introduction Methods Results Discussion References

The effects of two highly fermentable dietary fibers (guar gum and pectin) on the type and concentrations of cecal polyamines as affected by the intestinal microflora were studied in groups of germ-free (n = 10/group) and conventional rats (n = 6/group). Both germ-free and conventional rats were randomly assigned to one of three treatments as follows: 1) fiber-free control diet, 2) control diet + 10% guar gum and 3) control diet + 10% pectin. In germ-free rats, guar gum and pectin had no effect on cecal polyamine concentrations. Putrescine was confirmed to be the major endogenous polyamine within the gut lumen. In cecal contents of conventional rats, both guar gum and pectin led to the appearance of cadaverine and to elevated putrescine concentrations in comparison with the fiber-free control diet (1.35 ± 0.15 and 2.27 ± 0.32, respectively, vs. 0.20 ± 0.03 µmol/g dry weight, P < 0.05). The cecal cadaverine concentration was higher in pectin- than in guar-fed rats (8.20 ± 0.89 vs. 1.92 ± 0.27 µmol/g dry weight, P < 0.05). Counts of total bacteria, bacteroides, fusobacteria and enterobacteria were higher (P < 0.05) in rats fed guar gum and pectin. Bifidobacteria were found exclusively in guar-fed rats. In vitro studies on selected species representing the numerically dominant population groups of the human gut flora (bacteroides, fusobacteria, anaerobic cocci and bifidobacteria) were examined for their ability to synthesize intracellular polyamines. These experiments demonstrated the ability of bacteroides, fusobacteria and anaerobic cocci to synthesize high amounts of putrescine and spermidine. Calculations based on these results suggest that the intestinal microflora are a major source of polyamines in the contents of the large intestine.

	INTRODUCTION

Abstract Introduction Methods Results Discussion References

The polyamines putrescine, spermidine and spermine are synthesized by both eukaryotic and prokaryotic cells. The physiologic importance of polyamines arises from their implied participation in a variety of cellular processes such as the regulation of the synthesis of DNA, RNA and protein as well as cell growth and differentiation (Bardocz et al. 1995, McCormack and Johnson 1991, Seiler 1990). Studies with humans and animals have demonstrated the role of polyamines in the maturation of the intestinal mucosa (Capano et al. 1994, Wild et al. 1993) and their adaptive growth after injury or surgery (Kummerlen et al. 1994, Luk and Baylin 1983). The polyamines in the intestinal lumen originate from endogenous and exogenous sources. Food polyamines, one of the most important exogenous sources, are rapidly absorbed in the small intestine (Bardocz et al. 1993) and are not available to meet the high metabolic demand for polyamines of the mucosal tissue in the large bowel. The production of polyamines by the intestinal microflora may play a major role in providing polyamines for these purposes. The importance of gut bacteria as a possible polyamine source is supported by human experimental studies (Hessels et al. 1989, Satink et al. 1989). Furthermore, the intestinal contents of germ-free and conventional rats fed a polyamine-deficient diet differed basically in their polyamine composition (Noack et al. 1996).

Indigestible polysaccharides pass into the large bowel where they serve as energy substrates for the resident microbial flora. Microbial fermentation of various types of dietary fiber within the large intestine results in the production of short-chain fatty acids and a lower pH, which in turn may modify the composition and metabolic activity of the intestinal microbial flora (Campbell et al. 1997, Edwards and Eastwood 1995). Furthermore, short-chain fatty acids, particularly butyric acid, play an important role in the regulation of colonic epithelial cell proliferation (Lupton and Kurtz 1993). The effects of bacterial carbohydrate fermentation on the production of short-chain fatty acids and intestinal mucosal cell proliferation have been studied in detail, whereas there is no information available on the influence of carbohydrates on bacterial polyamine synthesis in the large intestine.

The purpose of this study was therefore to compare diets of two soluble indigestible dietary fibers, guar gum and pectin, with a fiber-free diet as to their effects on the polyamine concentrations in cecal contents and tissues in germ-free and conventional rats fed polyamine-deficient purified diets. The composition of the cecal microbial flora in conventional rats was studied to find out whether and in what way guar gum or pectin modifies the bacterial polyamine production in the large bowel and which microbial population groups play a role in polyamine formation. In vitro studies on selected species representing the dominant bacterial population groups of the human intestinal flora were performed to check the results obtained from the feeding study and to calculate the possible contribution of bacteria to the polyamine supply in the large intestine.

	MATERIALS AND METHODS

Abstract Introduction Methods Results Discussion References

Animals and diets.  All treatments and diets were formally approved by the Animal Welfare Committee of Brandenburg. Conventional and germ-free male Wistar rats (Central Institute of Experimental Animal Sciences, Hannover, Germany; initial body mass 200 ± 30 g) were individually housed in wire-bottomed cages in a room with 12-h light:dark cycle and at a temperature of 22 ± 2°C. The cages of the germ-free rats were arranged in sterile isolators equipped with a sterile water supply. Material from each isolator was checked weekly for sterility. The conventional and germ-free rats were randomly assigned to one of three dietary treatments. One group each of the conventional (n = 6/group) and of the germ-free rats (n = 10/group) were fed a basal diet (polyamine-deficient; fiber-free). The other groups were fed a basal diet containing either pectin (Copenhagen Pectin Lille Skensved, Denmark) or guar gum (Sigma-Aldrich Chemie GmbH, Deisenhofen, Germany) as soluble dietary fiber. The composition of the basal diet is shown in Table 1. The pectin- and guar-supplemented diets were prepared by adding 100 g of pectin or guar gum to 900 g of the basal diet. All diets contained 21, 67 and 12% of total energy as protein, wheat starch and fat, respectively. The polyamine concentrations (nmol/g dry matter) of the basal diet or of both fiber-containing diets (guar and pectin diet) were 3.1 and 2.8 for putrescine, 52.3 and 47.1 for spermidine, and 23.3 and 21.0 for spermine, respectively. Portions of 25 g (dry matter) of the diets were sealed in polyethylene bags. For sterilization, the packaged diets were subjected to 25 kGy of gamma radiation. Rats were adapted to the diet for 7 d and then fed the respective diet for 6 d. The rats were given free access to the diets and water. After adaptation to the pectin or guar gum diets, rats consumed higher amounts of these diets than of the basal diet. This may be interpreted as compensation for the lower energy density of these diets. Therefore, all feeding groups had a comparable nutrient and energy intake. Food intake was determined daily. The remaining food was dried to constant weight at 105°C. The food intake was calculated as g dry weight/(rat·d). Each rat was weighed at the beginning and at the end of the study.

Sample collection and chemical analyses.  On d 7 of the experimental period, rats were individually killed by ether anesthesia and dissection of the vena jugularis at 0800 h. To determine the cecal polyamines, luminal contents were removed from the cecum and 0.6-g aliquots were used to measure the pH with a microprocessor pH meter 537 A (Wissenschaftlich-Technische Werkstätten GmbH, Weilheim, Germany). In addition, the entire cecal tissue was collected and promptly rinsed with ice-cold isotonic NaCl. All samples were frozen in liquid nitrogen and subsequently stored at 85°C. Our previous studies had indicated a complete recovery of polyamines after lyophilization. Therefore, before analysis, the intestinal contents and tissues were lyophilized and stored at 4°C in a desiccator. The extraction of the polyamines from the intestinal samples was performed by homogenizing ~0.2 g of the lyophilized sample in 1.5 mL of 50 g/L trichloroacetic acid (TCA) and centrifuging samples at 26,000 × g for 30 min at 5°C. The supernatant was collected and the sediment was washed with 1.5 mL of 50 g/L TCA and recentrifuged. Before analysis, aliquots of the combined supernatants were centrifuged through Ultrafree MC-filter units with a pore size of 0.2 µm (Millipore, Eschborn, Germany). The diets were subjected to the same extraction procedure used for the intestinal contents, whereas the cecal tissues (~0.2 g) were placed in an ice bath, homogenized in 1.5 mL of 50 g/L TCA, sonicated three times for 30 s (ultrasonic processor 200 W, Hielscher GmbH, Stahnsdorf, Germany) and centrifuged as described above. The polyamine contents of the cecal content and tissue were determined by HPLC (GYNKOTEK, Germering, Germany). The polyamines were separated on a cation-exchange column with potassium bromide buffers (0.7 and 1.6 mol/L , pH 3.7 and 7.8, respectively), and quantified after postcolumn derivatization with ninhydrin. With this method, amino acids and small peptides were eluted with the buffer front, whereas all polyamines were well separated from one another. The retention times for putrescine, cadaverine, spermidine and spermine were 18.6, 21.8, 33.0 and 56.3 min, respectively. The detection limit of each polyamine was 20-35 pmol/20 µL injection volume. The reproducibilty of the assay was ± 1.3%.

Undigested pectin and guar gum in cecal contents of germ-free rats were estimated by the m-hydroxyphenyl method and the anthron method, respectively (Blumenkrantz and Asboe-Hansen 1973, Kunerth and Youngs 1984). All results obtained with germ-free rats were related to pectin- or guar-free dry mass. The total nitrogen of cecal contents was determined with the Kjeldahl method.

Microbial studies.  Approximately 0.5 g of cecal contents was immediately placed into preweighed tubes with 2.0 mL of a prereduced brain-heart infusion broth (DIFCO, Augsburg, Germany). After homogenization, the specimens were subjected to a series of 10-fold dilutions (10² to10⁸) in a prereduced saline buffer, and duplicate samples of 0.05 mL of each dilution were plated on nonselective and selective media (Kleessen et al. 1995). All manipulations of the anaerobic media were performed in an anaerobic chamber (MK3 anaerobic workstation, dW Scientific, West Yorkshire, England). The inoculated media were incubated at 37°C for 4 d (anaerobic bacteria) or for 1-2 d (aerobic microorganisms). Total colony counts were determined on Columbia agar plates supplemented with 5% sheep blood (BioMerieux, Nürtingen, Germany). Total colony counts represent the sum of the anaerobic and aerobic colony counts determined after growth on Columbia agar plates. Total anaerobic counts were corrected for facultative anaerobes by evaluating aerotolerance of the different colony types. Enumeration of the specific bacterial groups was done by plating on selective media.³ Selected colonies were further purified for identification. Bacterial reference strains from the laboratory strain collection of the American Type Culture Collection (ATCC, Rockville, MD) were used for typing the different bacterial groups. The criteria used for identification of anaerobic isolates were those outlined by the Anaerobic Laboratory at Virginia Polytechnic Institute and State University (Holdemann et al. 1977). The viable counts are expressed as log₁₀ of colony-forming units (CFU)/g dry cecal contents.

In vitro experiments.  Microorganisms used in this study were obtained from the ATCC, the German Collection of Microorganisms and Cell Cultures (DSMZ, Braunschweig, Germany) the Laboratory strain collection (DIfE Rehbrücke) and from BioMerieux. All collection specimens used in our study originated from the human intestine (adult or infant). To examine gut bacteria for polyamine production, the representative species of the numerically predominant organisms from the human intestinal microflora (bacteroides, fusobacteria, gram-positive anaerobic cocci and bifidobacteria) were grown anaerobically in polyamine-free culture medium⁴ at 37°C for 24 h. The bacterial cells were harvested by centrifuging at 3270 × g for 10 min at 4°C and washing once with PBS, pH 7.2. For analysis of protein and intracellular polyamine concentrations, cells were disrupted by freezing the samples in liquid nitrogen and subsequently heating them for 8 min at 100°C. The chilled samples were suspended in 50 g/L TCA and sonicated for 2 min (4 × 0.5-min periods, power setting 100%, cycle 0.3). Throughout sonic treatment, the samples were chilled in an ice bath.

The protein content in bacterial cells was determined by the modified micro-Lowry method (determination kit from Sigma-Aldrich Fine Chemicals, Steinheim, Germany). Bovine serum albumin served as the standard.

Statistical methods.  Data are expressed as means ± SEM. The rat served as the experimental unit. Both germ-free and conventional rats were randomly assigned to one of three dietary treatments (n = 10, germ-free and n = 6, conventional rats). The question to be answered was whether the guar gum and pectin content of the diet affected the polyamine concentrations, total-N, dry mass and pH in the cecum of any of the rat models. Differences between groups were established using two-way (diet and rat model) ANOVA ( SPSS software for WINDOWS, Version 6.1.2., SPSS, Chicago, IL). Differences among treatment groups were determined using the least significant difference (LSD) test (SPSS). To analyze the effect of dietary treatment on microbial counts in cecum contents of conventional rats, logarithmic data transformation to log₁₀ CFU was applied. Differences among treatment groups were estimated by one-way ANOVA and by the nonparametric Kruskal-Wallis test (for small numbers of observations). To compare mean values where appropriate, the LSD and Nemenyi tests (Wilcoxon and Roberts 1964) were used. Analysis by both tests resulted in similar output for the effect of treatment. Statistical significance of differences was established at P < 0.05.

	RESULTS

Abstract Introduction Methods Results Discussion References

Intake of food and polyamines.  The food intake was 19.0 ± 0.6, 21.5 ± 0.8 and 21.0 ± 1.0 g/d for all rats fed fiber-free, guar gum and pectin diets, respectively. Intake did not differ in all feeding groups of germ-free and conventional rats when calculated on the basis of the guar- or pectin-free proportion of the diet. The polyamine intakes that were calculated from the polyamine contents of the fiber-free diet and the daily consumption of the fiber-free proportion of the experimental diets were 59 ± 4, 995 ± 42 and 443 ± 25 nmol/d putrescine, spermidine and spermine, respectively, for all feeding groups of germ-free and conventional rats.

Cecal total nitrogen, dry mass and pH.  In germ-free rats, the total nitrogen content in the cecum decreased significantly (P < 0.01) in response to consumption of the guar gum or the pectin diet compared with the fiber-free control diet (Table 2). In conventional rats, in contrast, the total nitrogen content increased upon consumption of the polysaccharide-containing diets (P < 0.05). The opposite trend was observed with respect to the dry mass, i.e., guar gum or pectin led to an increase in the dry mass of cecal contents of germ-free rats (P < 0.01) but to a decrease in dry mass in conventional rats.The proportion of undigested guar and pectin in the cecum of germ-free rats was 51 and 50% of dry mass, respectively. The cecal pH in germ-free rats consuming the guar gum diet increased by 0.72 pH units (P < 0.01), compared with rats fed the fiber-free control diet.

Polyamine concentrations in cecal contents and tissue.  The predominant polyamine detected in cecal contents of germ-free rats was putrescine. Guar or pectin supplementation did not affect the polyamine concentrations in these rats (Table 3). Cadaverine, which is produced exclusively by bacteria, was not detected in germ-free rats. In conventional rats, in contrast, the polyamine concentrations in cecal contents were strongly affected by the polysaccharides supplemented to the diet, i.e., when conventional rats were fed the fiber-free control diet, spermidine predominated and cadaverine was not present. Consumption of the guar- or pectin-containing diet led to 7- and 11-fold elevation of putrescine concentrations (P < 0.01), respectively, and to the appearence of high cadaverine concentrations in the cecal contents. Nevertheless, in guar gum-fed conventional rats, spermidine was the predominant polyamine. There were no differences in the spermidine concentrations of the cecal content of rats fed the guar gum diet or the fiber-free control diet. It should be noted, however, that after consumption of the pectin diet, cadaverine became the predominant polyamine in conventional rats.

The concentrations of the individual polyamines in cecal tissue of both germ-free and conventional rats consuming the fiber-free control diet were characterized by low concentrations of putrescine and high but nearly equal concentrations of spermidine and spermine (putrescine 0.03 ± 0.01 vs. 0.11 ± 0.08; spermidine 1.51 ± 0.24 vs. 1.55 ± 0.10; spermine 1.84 ± 0.17 vs. 1.28 ± 0.09 µmol/g dry weight, germ-free vs. conventional rats, respectively). In germ-free rats, none of the experimental diets induced any alterations in the polyamine concentrations of cecal tissue. In contrast, the cecal tissue of conventional rats contained cadaverine when rats consumed the guar- or pectin-supplemented diet (0.27 ± 0.07 or 0.45 ± 0.05 µmol/g dry weight, respectively).

Cecal microbial population groups.  Bacteroides-fusobacteria, eubacteria, lactobacilli and gram-positive anaerobic cocci were the numerically predominant bacterial population groups of the cecal microflora of rats fed the fiber-free control diet (Table 4). Elevated total bacterial counts (P < 0.05) were found as a result of guar or pectin consumption compared with consumption of the fiber-free control diet. The increase in total counts was due to elevated numbers of bacteroides-fusobacteria and enterobacteria. The counts of these bacterial groups did not differ for guar gum- and pectin-fed rats. The most important finding was the detection of high numbers of bifidobacteria as a result of consumption of guar gum.

Intracellular polyamine contents of intestinal microorganisms.  In polyamine-free medium, members of the genus Bacteroides synthesized mainly spermidine but no putrescine or cadaverine (Table 5). In contrast, fusobacteria had high contents of putrescine. Only one of the strains tested, Fusobacterium varium, contained low amounts of spermidine, corresponding to ~2% of the molar concentration of putrescine. Peptostreptococcus productus, a representative of the anaerobic cocci, synthesized primarily putrescine and some spermidine. The bifidobacteria tested did not synthesize any of these polyamines under the growth conditions used.

	DISCUSSION

Abstract Introduction Methods Results Discussion References

In this study, germ-free and conventional rats were fed purified diets with low but constant polyamine contents. Our results demonstrated that the polyamine metabolism in germ-free and conventional rats is fundamentally different. When the fiber-free control diet was fed, putrescine was the predominant polyamine in the cecal contents of germ-free rats, whereas the cecal contents of conventional rats contained spermidine instead of putrescine as the major polyamine. With this diet, cadaverine was not detected in the cecal contents of any of the rat models. It should be noted that the intraluminal polyamine concentrations in germ-free rats are not influenced by the dietary polyamine intake (Noack et al. 1996). Therefore, the polyamines detected in the cecum of germ-free rats must have originated from endogenous sources. Our results suggest that putrescine is the main endogenous polyamine secreted into the gut lumen. Thus the different concentrations and patterns of polyamines in the cecal contents of conventional rats compared with those of germ-free rats cannot be attributed to the dietary or endogenous sources, but rather to the intestinal microflora, which form primarily spermidine.

Bacteria synthesize polyamines by decarboxylation of the amino acids ornithine, arginine and lysine. The bacterial population in the large intestine has an enormous enzymatic capacity for degradation of proteins (Macfarlane and Macfarlane 1997), resulting in peptides and amino acids that serve as precursors of biosynthetic reactions or as energy substrates for the microorganisms. Polyamines are formed as by-products in these activities. In healthy humans and animals, the majority of proteins reaching the large intestine originate from endogenous sources such as pancreatic secretions and desquamated intestinal cells, whereas dietary protein degradation products are largely and rapidly absorbed in the small intestine. But it also has been shown that a small proportion of dietary protein, in particular after consumption of high levels of protein, may result in the accumulation of undigested or partially digested proteins in the lower part of the gut, which in turn may serve as substrates for the microbial polyamine synthesis. The effects of high dietary levels of protein on the formation of polyamines by gut bacteria have not yet been studied.

Guar- or pectin-supplemented diets provide additional energy substrates for the gut flora. Guar gum and pectin are soluble branched polysaccharides that escape pancreatic digestion in the small intestine but are easily degraded by the resident microbes in the large bowel (Macfarlane and Macfarlane 1993, Nyman and Asp 1982).

In germ-free rats, both fibers resulted in an elevated pH in cecal contents. The 50% decrease in total nitrogen in cecal contents of germ-free rats observed upon supplementation of the fiber-free diet with guar gum or pectin may be explained by the finding that ~50 % of the solids in cecal contents were undigested guar gum or pectin. Both the rise in pH and the undigested guar gum or pectin indicate a lack of microbial fermentation in the cecum of germ-free rats. In conventional rats, the feeding of guar gum or pectin effected a significant increase in total nitrogen in cecal contents. The guar- or pectin-induced elevated levels of total nitrogen coincided with an increase in total microbial counts. The proliferation of bacteria in the large intestine, reflected by a rise in biomass and total nitrogen, depends on the availability of suitable energy sources such as fermentable polysaccharides (Cummings and Macfarlane 1997, Macfarlane and Cummings 1991). The decrease in pH in the cecal contents of both fiber-fed groups observed in this study is consistent with previous studies (Edwards and Eastwood 1995, Lupton and Kurtz 1993).

The consumption of guar- or pectin-supplemented diets did not alter the polyamine contents and patterns in the cecum of germ-free rats. Putrescine remained the predominant polyamine. This result was not surprising because guar gum and pectin are not digested by pancreatic or small-intestinal enzymes. However, when guar- or pectin-containing diets were fed to conventional rats, the cecal polyamine patterns in the cecal contents changed substantially. It is interesting to note that the expression of bacterial amino acid decarboxylases is influenced by the pH of the medium (Olson 1993). Moreover, the production of amines by Bacteroides fragilis is maximal under acidic conditions (Allison and Macfarlane 1989). The elevated putrescine concentrations and the substantial amounts of cadaverine that we found in the cecal contents of guar- and especially pectin-fed rats suggest an alteration of either the bacterial metabolic activities or the microbial population in response to these polysaccharides. The presence of cadaverine, an exclusive bacterial polyamine, in the cecal tissue of guar- or pectin-fed conventional rats indicates that the bacterial flora make a large contribution to the polyamine pool in cecal tissue.

The enumeration of the major intestinal bacterial groups in the cecum of guar- or pectin-fed rats documented a considerable change in the composition of the bacterial flora. Our results show a significant increase in total counts, specifically in the counts of bacteroides, fusobacteria and enterobacteria. Bacteroides and fusobacteria were the numerically dominant organisms in the cecal flora of guar gum- or pectin-fed rats. Interestingly, bifidobacteria were detected only when the diet was supplemented with guar gum. Such a selective stimulation of bifidobacteria in the human colon was also reported for oligofructose or inulin ingestion (Campbell et al. 1997, Kleessen et al. 1997). The results obtained with the guar- and pectin-supplemented diets confirmed our previous observation that the polyamine contents in the guts of conventional rats are highly influenced by the metabolic activity of intestinal bacteria. This view is also supported by experiments that demonstrated that rats infected with Salmonella typhimurium and Salmonella enteritidis responded with increased polyamine contents in the small intestine (Naughton et al. 1995). It is not clear, however, whether these effects are direct or indirect.

The results obtained from animal experiments prompted us to examine the representative species of the numerically dominant population groups of the human gut flora for their ability to synthesize polyamines. Members of the genus Bacteroides were identified as spermidine producers, whereas the fusobacteria tested and to some extent Peptostreptococcus productus, representing the gram-positive anaerobic cocci, synthesized mainly putrescine when grown in a defined polyamine-free medium. The predominance of bacteroides in the cecal flora of conventional rats would explain the high spermidine concentrations in the gut contents observed in this study. The elevated concentrations of putrescine in the cecal contents of guar- or pectin-fed conventional rats, compared with those of control rats, coincided with the observed increase in the counts of the bacteroides and fusobacteria groups. Although the differentiation of bacteroides and fusobacteria in gut contents with classical microbiological methods was not performed, it is reasonable to assume that the increase in the number of fusobacteria was responsible for the concomitant rise in putrescine. Surprisingly, bifidobacteria, which were detected exclusively in cecal contents of guar-fed rats, neither synthesized any polyamines nor did they grow in polyamine-free medium. Obviously, bifidobacteria depend on exogenous polyamines for cell growth and maintenance. It is tempting to speculate that the selective stimulation of bifidobacteria by guar gum, but not by pectin, is based on a highly efficient enzyme system for the utilization of guar gum, which enables the bifidobacteria to compete successfully with other bacterial groups in this ecosystem.

None of the polyamine-forming microorganisms under study synthesized cadaverine under the in vitro growth conditions used. Enterobacteria such as Escherichia coli and Klebsiella are known for their ability to produce cadaverine (Tabor and Tabor 1985). However, because enterobacteria are only a minor proportion (<1%) of the intestinal flora, it is unlikely that enterobacteria were the only source of the cadaverine detected in response to pectin or guar gum. The factors that induce cadaverine synthesis have not yet been identified.

To estimate the quantitative significance of microbial polyamine formation in the human gut, we calculated the total amounts of polyamines that might be synthesized by predominant organisms of the intestinal flora in the entire colon content (on the basis of in vitro results) and compared it with the dietary polyamine intake. Our calculations are based on the following assumptions:

Even though the calculation of polyamines formed by the bacteria in the large bowel is necessarily imprecise, we feel that the intestinal microflora synthesize polyamines in substantially larger quantities in vivo than we are able to determine in vitro with selected organisms. Although the intake of dietary polyamines is considerable, it can be assumed that they are almost completely absorbed in the small intestine (Bardocz et al. 1993, Benamouzig et al. 1997). Therefore, the proportion of polyamines that may be available for the metabolic demand of the large intestinal mucosal tissue will be primarily of microbial rather than of dietary origin.

In conclusion, dietary supplementation with the soluble indigestible polysaccharides guar gum or pectin affects the composition and metabolic activity of the intestinal microbial flora. Bifidobacteria were found exclusively in guar-fed rats. Both carbohydrates stimulated bacterial formation of putrescine, spermidine and cadaverine in the cecal contents of conventional rats, but cadaverine production was more pronounced in pectin- than in guar-fed rats.

In general, polyamines are known to stimulate cell proliferation and differentiation, processes of high physiologic importance. Enhanced bacterial polyamine formation in the large bowel due to the presence of fermentable carbohydrates may contribute to maintaining the integrity of the gut mucosa in a healthy organism. In the case of a damaged intestinal mucosa, as is encountered in patients with inflammatory bowel diseases or after surgical intervention (massive bowel resection), the large intestinal microflora may help to meet the demand for polyamines for proliferation and renewal of the mucosal tissue. Whether polyamines of bacterial origin enhance colon tumorigenesis cannot be determined satisfactorily at present.

	FOOTNOTES

Manuscript received 10 February 1998. Initial reviews completed 24 March 1998. Revision accepted 6 May 1998.

	ACKNOWLEDGMENTS

We thank Sabine Zimmermann and Sabine Schmidt for technical assistance.

	LITERATURE CITED

Abstract Introduction Methods Results Discussion References

• Allison C., Macfarlane G. T. Influence of pH, nutrient availablity, and growth rate on amine production by Bacteroides fragilis and Clostridium perfringens. Appl. Environ. Microbiol. 1989; 55:2894-2898[Abstract/Free Full Text]
• Bardocz S., Duguid T. J., Brown D. S., Grant G., Pusztai A., White A., Ralph A. The importance of dietary polyamines in cell regeneration and growth. Br. J. Nutr. 1995; 73:819-828[Medline]
• Bardocz S., Grant G., Brown D. S., Ralph A., Pusztai A. Polyamines in food implications for growth and health. J. Nutr. Biochem. 1993; 4:66-71
• Benamouzig R., Mahe S., Luengo C., Rautureau J., Tome D. Fasting and postprandial polyamine concentrations in the human digestive lumen. Am. J. Clin. Nutr. 1997; 65:766-770[Abstract/Free Full Text]
• Blumenkrantz N., Asboe-Hansen G. New method for quantitative determination of uronic acids. Anal. Biochem. 1973; 54:484-489[Medline]
• Campbell J. M., Fahey G. C., Wolf B. W. Selected indigestible oligosaccharides affect large bowel mass, cecal and fecal short-chain fatty acids, pH and microflora in rats. J. Nutr. 1997; 127:130-136[Abstract/Free Full Text]
• Capano G., Bloch K. J., Schiffrin E. J., Dascoli J. A., Israel E. J., Harmatz P. R. Influence of the polyamine, spermidine, on intestinal maturation and dietary antigen uptake in the neonatal rat. J. Pediatr. Gastroenterol. Nutr. 1994; 19:34-42[Medline]
• Cummings J. H., Macfarlane G. T. Role of intestinal bacteria in nutrient metabolism. Clin. Nutr. 1997; 16:3-11
• Edwards C. A., Eastwood M. A. Caecal and faecal short-chain fatty acids and stool output in rats fed on diets containing non-starch polysaccharides. Br. J. Nutr. 1995; 73:773-781[Medline]
• Hessels J., Kingma A. W., Ferwerda H., Keij J., Van den Berg G. A., Muskiet F.A.J. Microbial flora in the gastrointestinal tract abolishes cytostatic effect of a-difluoromethylornithine in vivo. Int. J. Cancer 1989; 43:1155-1164[Medline]
• Holdemann, L. V., Cato, E. P. & Moore, W.E.C. (1977) Anaerobic Laboratory Manual, 4th ed. Virginia Polytechnic Institute and State University, Blacksburg VA.
• Kleessen B., Bunke H., Tovar K., Noack J., Sawatzki G. Influence of two infant formulas and human milk on the development of the faecal flora in newborn infants. Acta Paediatr. 1995; 84:1347-1356[Medline]
• Kleessen B., Sykura B., Zunft H.-J., Blaut M. Effect of inulin and lactose on fecal microflora, microbial activity, and bowel habit in elderly constipated persons. Am. J. Clin. Nutr. 1997; 65:1397-1402[Abstract/Free Full Text]
• Kummerlen C., Seiler N., Galluser M., Gosse F., Knodgen B., Hasselmann M., Raul F. Polyamines and the recovery of intestinal morphology and function after ischemic damage in rats. Digestion 1994; 55:168-174[Medline]
• Kunerth W. H., Youngs V. L. Modification of anthrone, carbazole, and orcinol reactions for quantification of monosaccharides. Cereal Chem. 1984; 61:344-349
• Luk, G. D. & Baylin, S. B. (1983) Polyamines and intestinal growthincreased polyamine biosynthesis after jejunectomy. Am. J. Physiol. 245 (Gastrointest. Liver Physiol. 8): G656-G660.
• Lupton J. R., Kurtz P. P. Relationship of colonic luminal short-chain fatty acids and pH to in vivo cell proliferation in rats. J. Nutr. 1993; 123:1522-1530
• Macfarlane, G. T. & Cummings, J. H. (1991) The colonic flora, fermentation, and large bowel digestive function. In: The Colonic Flora, Fermentation, and Large Bowel Digestive Function (Phillips, S.F., Pemberton, J. H. & Shorter, R. G., eds.), pp. 51-92. Mayo Foundation Raven Press, New York, NY.
• Macfarlane G. T., Macfarlane S. Factors affecting fermentation reactions in the large bowel. Proc. Nutr. Soc. 1993; 52:367-373[Medline]
• Macfarlane, G. T. & Macfarlane, S. (1997) Human colonic microbiota: ecology, physiology and metabolic potential of intestinal bacteria. Scand. J. Gastroenterol. 32 (suppl. 222): 3-9.
• McCormack, S. A. & Johnson, L. R. (1991) Role of polyamines in gastrointestinal mucosal growth. Am. J. Physiol. 260 (Gastrointest. Liver Physiol. 23): G795-G806.
• Mevissen-Verhage E.A.E., Marcellis J. H., de Vos, M .N., Harmsen-van Amerongen, W.C.M., Verhoef J. Bifidobacterium, Bacteroides, and Clostridium spp. in fecal samples from breast-fed infants with and without iron supplement. J. Clin. Microbiol. 1987; 25:285-289[Abstract/Free Full Text]
• Naughton P. J., Grant G., Ewen S. W., Spencer R. J., Brown D. S., Pusztai A., Bardocz S. Salmonella typhimurium and Salmonella enteritis induce gut growth and increase the polyamine content of the rat small intestine in vivo. FEMS Immunol. Med. Microbiol. 1995; 12:251-258[Medline]
• Noack J., Kleessen B., Lorenz A., Blaut M. The effect of alimentary polyamine depletion on germ-free and conventional rats. Nutr. Biochem. 1996; 7:560-566
• Nyman M., Asp N. G. Fermentation of dietary fibre components in the rat intestinal tract. Br. J. Nutr. 1982; 47:357-366[Medline]
• Olson E. R. Influence of pH on bacterial gene expression. Mol. Microbiol. 1993; 8:5-14[Medline]
• Satink H.P.W M., Hessels J., Kingma A. W., Van den Berg G. A., Muskiet F.A.J., Halie M. R. Microbial influences on urinary polyamine excretion. Clin. Chim. Acta 1989; 179:305-314[Medline]
• Seiler, N. (1990) Polyamine metabolism. Digestion 46 (suppl. 2): 319-330.
• Stephen A. M., Cummings J. H. The microbial contribution to human faecal mass. J. Med. Microbiol. 1980; 13:45-56[Abstract/Free Full Text]
• Tabor C. W., Tabor H. Polyamines in microorganisms. Microbiol. Rev. 1985; 49:81-99[Free Full Text]
• Wilcoxon, F. & Roberts, A. (1964) Some Rapid Approximate Statistical Procedures. Cyanamid Company, New York, NY.
• Wild G. E., Daly A. S., Sauriol N., Bennett G. Effect of exogenously administered polyamine on the structural maturation and enzyme ontogeny of the postnatal rat intestine. Biol. Neonate 1993; 63:246-257[Medline]

0022-3166/98 $3.00 ©1998 American Society for Nutritional Sciences