The human bowel contains a large and biodiverse bacterial community known as the microbiota or microbiome. It seems likely that the microbiota, fractions of the microbiota, or specific species comprising the microbiota provide the antigenic fuel that drives the chronic immune inflammation of the bowel mucosa that is characteristic of Crohn's disease and ulcerative colitis. At least twenty years of microbiological research have been expended on analysis of the composition of the bowel microbiota of inflammatory bowel disease patients in comparison to that of control subjects. Despite extensive speculations about the aetiological role of dysbiosis in inflammatory bowel diseases, knowledge that can be easily translated into effective remedies for patients has not eventuated. The causes of this failure may be due to poorly defined and executed bacteriological studies, as well as the overwhelming complexity of a biome that contains hundreds of bacterial species and trillions of bacterial cells.

The ability to predictably engineer the composition of bowel microbial communities (microbiota) using dietary components is important because of the reported associations of altered microbiota composition with medical conditions. In a synecological study, weanling conventional Sprague-Dawley rats (21 days old) were fed a basal diet (BD) or a diet supplemented with resistant starch (RS) at 5%, 2.5%, or 1.25% for 28 days. Pyrosequencing of 16S rRNA genes and temporal temperature gradient electrophoresis (TTGE) profiles in the colonic digesta showed that rats fed RS had altered microbiota compositions due to blooms of Bacteroidetes and Actinobacteria. The altered microbiota was associated with changes in colonic short-chain fatty acid (SCFA) concentrations, colonic-tissue gene expression (Gsta2 and Ela1), and host physiology (serum metabolite profiles and colonic goblet cell numbers). Comparisons between germ-free and conventional rats showed that transcriptional and serum metabolite differences were mediated by the microbiota and were not the direct result of diet composition. Altered transcriptomic and physiological responses may reflect the young host's attempts to maintain homeostasis as a consequence of exposure to a new collection of bacteria and their associated biochemistry.

Autoinducer-2 (AI-2)-mediated quorum sensing has been extensively studied in relation to the regulation of microbial behavior. There are, however, two potential roles for the AI-2 synthase (LuxS). The first is in the production of AI-2 and the second is as an enzyme in the activated methyl cycle, where it catalyzes the conversion of S-ribosylhomocysteine to homocysteine. The by-product of the reaction catalyzed by LuxS is (S)-4,5-dihydroxy-2,3-pentanedione, which spontaneously forms the furanones known collectively as AI-2. The mammalian gut contains a complex collection of bacterial species so a method of interspecies communication might influence community structure and function. Lactobacillus reuteri 100-23 is an autochthonous inhabitant of the rodent forestomach, where it adheres to the nonsecretory epithelium, forming a biofilm. Microarray comparisons of gene expression profiles of the L. reuteri 100-23 wild type and a luxS mutant under different culture conditions revealed altered transcription of genes encoding proteins associated with cysteine biosynthesis/oxidative stress response, urease activity, and sortase-dependent proteins. Metabolomic analysis showed that the luxS mutation affected cellular levels of fermentation products, fatty acids and amino acids. Cell density-dependent changes (log phase versus stationary phase growth) in gene transcription were not detected, indicating that AI-2 was unlikely to be involved in gene regulation mediated by quorum sensing in L. reuteri 100-23.

Lactobacillus reuteri strain 100-23 together with a Lactobacillus-free mouse model, provides a system with which the molecular traits underpinning bacterial commensalism in vertebrates can be studied. A polysaccharide was extracted from sucrose-containing liquid cultures of strain 100-23. Chemical analysis showed that this exopolysaccharide was a levan (β-2, 6-linked fructan). Mutation of the fructosyl transferase (ftf) gene resulted in loss of exopolysaccharide production. The ftf mutant was able to colonise the murine gastrointestinal tract in the absence of competition, but colonisation was impaired in competition with the wild type. Biofilm formation by the mutant on the forestomach epithelial surface was not impaired and the matrix between cells was indistinguishable from that of the wild type in electron micrographs. Colonisation of the mouse gut by the wild-type strain led to increased proportions of regulatory T cells (Foxp3+) in the spleen, whereas colonisation by the ftf mutant did not. Survival of the mutant in sucrose-containing medium was markedly reduced relative to the wild type. Comparison of the genomic ftf loci of strain 100-23 with other L. reuteri strains suggested that the ftf gene was acquired by lateral gene transfer early in the evolution of the species and subsequently diversified at accelerated rates. Levan production by L. reuteri 100-23 may represent a function acquired by the bacterial species for life in moderate to high-sucrose extra-gastrointestinal environments that has subsequently been diverted to novel uses, including immunomodulation, that aid in colonisation of the murine gut.

Recent research has provided mechanistic insight into the important contributions of the gut microbiota to vertebrate biology, but questions remain about the evolutionary processes that have shaped this symbiosis. In the present study, we showed in experiments with gnotobiotic mice that the evolution of Lactobacillus reuteri with rodents resulted in the emergence of host specialization. To identify genomic events marking adaptations to the murine host, we compared the genome of the rodent isolate L. reuteri 100-23 with that of the human isolate L. reuteri F275, and we identified hundreds of genes that were specific to each strain. In order to differentiate true host-specific genome content from strain-level differences, comparative genome hybridizations were performed to query 57 L. reuteri strains originating from six different vertebrate hosts in combination with genome sequence comparisons of nine strains encompassing five phylogenetic lineages of the species. This approach revealed that rodent strains, although showing a high degree of genomic plasticity, possessed a specific genome inventory that was rare or absent in strains from other vertebrate hosts. The distinct genome content of L. reuteri lineages reflected the niche characteristics in the gastrointestinal tracts of their respective hosts, and inactivation of seven out of eight representative rodent-specific genes in L. reuteri 100-23 resulted in impaired ecological performance in the gut of mice. The comparative genomic analyses suggested fundamentally different trends of genome evolution in rodent and human L. reuteri populations, with the former possessing a large and adaptable pan-genome while the latter being subjected to a process of reductive evolution. In conclusion, this study provided experimental evidence and a molecular basis for the evolution of host specificity in a vertebrate gut symbiont, and it identified genomic events that have shaped this process.

Author Summary

The gastrointestinal microbiota of vertebrates is important for nutrient utilization, resistance against pathogens, and immune maturation of its host, but little is known about the evolutionary relationships between vertebrates and individual bacterial members of these communities. Here we provide robust evidence that the evolution of the gut symbiont Lactobacillus reuteri with vertebrates resulted in the emergence of host specialization. Genomic approaches using a combination of genome sequence comparisons and microarray analysis were used to identify the host-specific genome content in rodent and human strains and the evolutionary events that resulted in host adaptation. The study revealed divergent patterns of genome evolution in rodent and human lineages and a distinct genome inventory in host-restricted sub-populations of L. reuteri that reflected the niche characteristics in the gut of their particular vertebrate hosts. The ecological significance of representative rodent-specific genes was demonstrated in gnotobiotic mice. In conclusion, this work provided evidence that the vertebrate gut symbiont Lactobacillus reuteri, despite the likelihood of horizontal transmission, has remained stably associated with related groups of vertebrate hosts over evolutionary time and has evolved a lifestyle specialized to these host animals.

Background

We recently showed that Bifidobacterium animalis is more prevalent within the colons of IL-10 deficient (−/−) mice than in wild type (WT) animals colonized with the same specific pathogen free (SPF) fecal contents. Here we tested the ability of this organism to cause T cell-mediated intestinal inflammation by introducing it into germ-free (GF) IL-10−/− mice.

Methods

GF IL-10−/− or WT mice were monoassociated with Bifidobacterium animalis subsp. animalis ATCC 25527T or with Bifidobacterium infantis ATCC 15697T. Inflammation was measured by blinded histologic scores of the duodenum, cecum and colon and by spontaneous secretion of IL-12/IL-23 p40 from colonic explants. Bacterial antigen-specific CD4+ mesenteric lymph node (MLN) T cell recall responses were measured in response to antigen presenting cells (APC) pulsed with bacterial lysates.

Results

B. animalis caused marked duodenal inflammation and mild colitis in monoassociated IL-10−/− mice, whereas the intestinal tracts of WT animals remained free of inflammation. B. infantis colonization resulted in mild inflammation in the duodena of IL-10−/− mice. CD4+ MLN T cells from B. animalis monoassociated IL-10−/− mice secreted high levels of IFN-γ and IL-17 in response to B. animalis lysate. B. animalis equally colonized the different intestinal regions of WT and IL-10−/− mice.

Conclusions

B. animalis, a traditional probiotic species that is expanded in experimental colitis in this model, induces marked duodenal and mild colonic inflammation and TH1/TH17 immune responses when introduced alone into GF IL-10−/− mice. This suggests a potential pathogenic role for this commensal bacterial species in a susceptible host.

One hundred forty-one Campylobacter jejuni isolates from humans with diarrhea and 100 isolates from retailed poultry meat were differentiated by flaA typing. The bacteria were isolated in a specific geographical area (Dunedin) in New Zealand over a common time period. Twenty nine flaA types were detected, one of which (flaA restriction fragment length polymorphism type 15 [flaA-15]) predominated among isolates from humans (∼30% of isolates). This strain was of low prevalence (5% of isolates) among poultry isolates. flaA-15 strains were five to six times more invasive of HEp2 cells in an in vitro assay than a flaA type (flaA-3) that was commonly encountered on poultry meat (23% of isolates) but was seldom associated with human illness (5%). Competitive-exclusion experiments with chickens, utilizing real-time quantitative PCR to measure the population sizes of specific strains representing flaA-15 (T1016) and flaA-3 (Pstau) in digesta, were carried out. These experiments showed that T1016 always outcompeted Pstau in the chicken intestine. Genomic comparisons of T1016 and Pstau were made using DNA microarrays representing the genome of C. jejuni NCTC 11168. These comparisons revealed differences between the strains in the gene content of the Cj1417c-to-Cj1442c region of the genome, which is associated with the formation of capsular polysaccharide. The strains differed in Penner type (T1016, O42; Pstau, O53). It was concluded that poultry meat was at least one source of human infection with C. jejuni, that some Campylobacter strains detected in poultry meat are of higher virulence for humans than others, and that bacterial attributes affecting strain virulence and commensal colonization ability may be linked.

A real-time quantitative PCR assay targeting a 16S-23S intergenic spacer region sequence was devised to measure the sizes of populations of Lactobacillus salivarius present in ileal digesta collected from broiler chickens. This species has been associated with deconjugation of bile salts in the small bowel and reduced broiler productivity. The assay was tested as a means of monitoring the sizes of L. salivarius populations from broilers fed diets with different compositions, maintained at different stocking densities, or given the antimicrobial drugs bacitracin and monensin in the feed. Stocking densities did not influence the numbers of L. salivarius cells in the ileum. A diet containing meat and bone meal reduced the size of the L. salivarius population relative to that of chickens given the control diet, as did administration of bacitracin and monensin in the feed. These changes in the target bacterial population were associated with improved broiler weight gain.

BBn (BioBreeding) rats were fed casein-based diets supplemented with barley flour, oatmeal flour, cellulose, or barley β-glucans of high [HV] or low viscosity [LV] in order to measure the prebiotic effects of these different sources of dietary fiber. The dietary impact on the composition of the cecal microbiota was determined by the generation of denaturing gradient gel electrophoresis (DGGE) profiles of PCR-amplified 16S rRNA gene sequences. The DGGE profiles produced from the cecal microbiota of rats within each dietary group were similar, but consensus profiles generated from pooled bacterial DNAs showed differences between rat groups. Animals fed HV glucans (HV-fed rats) had DGGE consensus profiles that were 30% dissimilar from those of the other rat groups. A 16S rRNA gene fragment that was more conspicuous in the profiles of HV-fed animals than in those of cellulose-fed rats had sequence identity with Lactobacillus acidophilus. Measurements of L. acidophilus rRNA abundance (DNA-RNA hybridization), the preparation of cloned 16S rRNA gene libraries, and the enumeration of Lactobacillus cells (fluorescent in situ hybridization) showed that lactobacilli formed a greater proportion of the cecal microbiota in HV-fed rats. In vitro experiments confirmed that some lactobacilli utilize oligosaccharides (degree of polymerization, 3 or 4) present in β-glucan hydrolysates. The results of this study have relevance to the use of purified β-glucan products as dietary supplements for human consumption.

The luxS gene of Lactobacillus reuteri 100-23C was amplified by PCR, cloned, and then sequenced. To define a physiological and ecological role for the luxS gene in L. reuteri 100-23C, a luxS mutant was constructed by insertional mutagenesis. The luxS mutant did not produce autoinducers AI-2 or AI-3. Complementation of the luxS mutation by a plasmid construct containing luxS restored AI-2 and AI-3 synthesis. In vitro experiments revealed that neither the growth rate, nor the cell yield, nor cell survival in the stationary phase were compromised in the luxS mutant relative to the wild type and complemented mutant. The ATP content of exponentially growing cells of the luxS mutant was, however, 65% of that of wild-type cells. Biofilms formed by the luxS mutant on plastic surfaces in a bioreactor were thicker than those formed by the wild type. Biofilm thickness was not restored to wild-type values by the addition of purified AI-2 to the culture medium. In vivo experiments, conducted with ex-Lactobacillus-free mice, showed that biofilms formed by the mutant strain on the epithelial surface of the forestomach were approximately twice as thick as those formed by the wild type. The ecological performance of the luxS mutant, when in competition with L. reuteri strain 100-93 in the mouse cecum, was reduced compared to that of a xylA mutant of 100-23C. These results demonstrate that LuxS influences important ecological attributes of L. reuteri 100-23C, the consequences of which are niche specific.

Thirty-eight isolates of Lactobacillus gallinarum cultured from the crops of broiler chickens were screened for the presence of genes encoding S-layer proteins. All of the isolates had two S-protein genes, which were designated Lactobacillus gallinarum S-protein (lgs) genes. One gene in each isolate was either lgsA or lgsB. The Lactobacillus isolates were further characterized by pulsed-field gel electrophoresis of DNA digests, which grouped the isolates into 17 genotypes (strains). The second gene in each of eight representative strains was sequenced and shown to differ among strains (lgsC, lgsD, lgsE, lgsF, lgsG, lgsH, and lgsI). The genome of each strain thus encoded a common S-protein (encoded by either lgsA or lgsB) and a strain-specific S-protein. The extraction of cell surface proteins from cultures of the eight strains showed that each strain produced a single S-protein that was always encoded by the strain-specific lgs gene. Two of the strains were used to inoculate chickens maintained in a protected environment which were Lactobacillus-free prior to inoculation. DNAs and RNAs extracted from the digesta of the chickens were used for PCR and reverse transcription-PCR, respectively, to demonstrate the presence and transcription of lgs genes in vivo. In both cases, only the strain-specific gene was transcribed. Both of the strains adhered to the crop epithelium, consistent with published data predicting that S-proteins of lactobacilli are adhesins. The results of this study provide a basis for the investigation of gene duplication and sequence variation as mechanisms by which bacterial strains of the same species can share the same habitat.

A metagenomic (community genomic) library consisting of 5,760 bacterial artificial chromosome clones was prepared in Escherichia coli DH10B from DNA extracted from the large-bowel microbiota of BALB/c mice. DNA inserts detected in 61 randomly chosen clones averaged 55 kbp (range, 8 to 150 kbp) in size. A functional screen of the library for β-glucanase activity was conducted using lichenin agar plates and Congo red solution. Three clones with β-glucanase activity were detected. The inserts of these three clones were sequenced and annotated. Open reading frames (ORF) that encoded putative proteins with identity to glucanolytic enzymes (lichenases and laminarinases) were detected by reference to databases. Other putative genes were detected, some of which might have a role in environmental sensing, nutrient acquisition, or coaggregation. The insert DNA from two clones probably originated from uncultivated bacteria because the ORF had low sequence identity with database entries, but the genes associated with the remaining clone resembled sequences reported in Bacteroides species.

Members of the genus Lactobacillus are common inhabitants of the gut, yet little is known about the traits that contribute to their ecological performance in gastrointestinal ecosystems. Lactobacillus reuteri 100-23 persists in the gut of the reconstituted Lactobacillus-free mouse after a single oral inoculation. Recently, three genes of this strain that were specifically induced (in vivo induced) in the murine gut were identified (38). We report here the detection of a gene of L. reuteri 100-23 that encodes a high-molecular-mass surface protein (Lsp) that shows homology to proteins involved in the adherence of other bacteria to epithelial cells and in biofilm formation. The three in vivo-induced genes and lsp of L. reuteri 100-23 were inactivated by insertional mutagenesis in order to study their biological importance in the murine gastrointestinal tract. Competition experiments showed that mutation of lsp and a gene encoding methionine sulfoxide reductase (MsrB) reduced ecological performance. Mutation of lsp impaired the adherence of the bacteria to the epithelium of the mouse forestomach and altered colonization dynamics. Homologues of lsp and msrB are present in the genomes of several strains of Lactobacillus and may play an important role in the maintenance of these bacteria in gut ecosystems.

The gut microbiota may be important in the postnatal development of the immune system and hence may influence the prevalence of atopic diseases. Bifidobacteria are the most numerous bacteria in the guts of infants, and the presence or absence of certain species could be important in determining the geographic incidence of atopic diseases. We compared the fecal populations of bifidobacteria from children aged 25 to 35 days in Ghana (which has a low prevalence of atopy), New Zealand, and the United Kingdom (high-prevalence countries). Natal origin influenced the detection of bifidobacterial species in that fecal samples from Ghana almost all contained Bifidobacterium infantis whereas those of the other children did not. Choosing species on the basis of our bacteriological results, we tested bifidobacterial preparations for their effects on cell surface markers and cytokine production by dendritic cells harvested from cord blood. Species-specific effects on the expression of the dendritic-cell activation marker CD83 and the production of interleukin-10 (IL-10) were observed. Whereas CD83 expression was increased and IL-10 production was induced by Bifidobacterium bifidum, Bifidobacterium longum, and Bifidobacterium pseudocatenulatum, B. infantis failed to produce these effects. We concluded that B. infantis does not trigger the activation of dendritic cells to the degree necessary to initiate an immune response but that B. bifidum, B. longum, and B. pseudocatenulatum induce a Th2-driven immune response. A hypothesis is presented to link our observations to the prevalence of atopic diseases in different countries.

Human subjects consumed biscuits containing either galacto-oligosaccharides or fructo-oligosaccharides in a double-blinded, crossover study. The impact of supplementing the diet with three biscuits per day on the fecal microbiota was evaluated by selective culture of particular bacterial groups, measurement of β-galactosidase activity, and nucleic acid-based analytical methods (PCR-denaturing gradient gel electrophoresis [PCR-DGGE] and fluorescent in situ hybridization). The composition of the bifidobacterial populations was monitored at the level of species (PCR-DGGE) and strains (pulsed-field gel electrophoresis of DNA digests), and representative cultures were tested quantitatively for their ability to use galacto-oligosaccharides. Technical improvements to DGGE analysis of the microbiota were made by the use of an internal standard that allowed valid comparisons of fragment staining intensities to be made between profiles, the use of S1 nuclease digestion to remove single-stranded DNA to facilitate cloning of DNA sequences cut from gels, and the extraction of RNA to be used as the template in reverse transcription-PCR-DGGE. RNA-DGGE profiles were markedly different (Dice's similarity coefficient, 58.5%) from those generated by DNA-DGGE. Neither the sizes of the bacterial populations nor the DNA-DGGE profiles of the microbiota were altered by the consumption of the biscuits, but the RNA-DGGE profiles were altered by the detection or increased staining intensity of 16S rRNA gene sequences originating from Bifidobacterium adolescentis and/or Colinsella aerofaciens in the feces of 11 of 15 subjects. β-Galactosidase activity was elevated in the feces of some subjects as a result of biscuit consumption. Subjects differed in the ability of the bifidobacterial strains harbored in their feces to use galacto-oligosaccharides. Our observations suggest that a phylogenetic approach to analysis of the gut ecosystem may not always be optimal and that a more physiological (biochemical) method might be more informative.

The microflora of the crop was investigated throughout the broiler production period (0 to 42 days) using PCR combined with denaturing gradient gel electrophoresis (PCR-DGGE) and selective bacteriological culture of lactobacilli followed by amplified ribosomal DNA restriction analysis (ARDRA). The birds were raised under conditions similar to those used in commercial broiler production. Lactobacilli predominated and attained populations of 108 to 109 CFU per gram of crop contents. Many of the lactobacilli present in the crop (61.9% of isolates) belonged to species of the Lactobacillus acidophilus group and could not be differentiated by PCR-DGGE. A rapid and simple ARDRA method was developed to distinguish between the members of the L. acidophilus group. HaeIII-ARDRA was used for preliminary identification of isolates in the L. acidophilus group and to identify Lactobacillus reuteri and Lactobacillus salivarius. MseI-ARDRA generated unique patterns for all species of the L. acidophilus group, identifying Lactobacillus crispatus, Lactobacillus johnsonii, and Lactobacillus gallinarum among crop isolates. The results of our study provide comprehensive knowledge of the Lactobacillus microflora in the crops of birds of different ages using nucleic acid-based methods of detection and identification based on current taxonomic criteria.

Lactobacilli are common inhabitants of the gastrointestinal tracts of mammals and have received considerable attention due to their putative health-promoting properties. Little is known about the traits that enhance the ability of these bacteria to inhabit the gastrointestinal tract. In this paper we describe the development and application of a strategy based on in vivo expression technology (IVET) that enables detection of Lactobacillus reuteri genes specifically induced in the murine gut. A plasmid-based system was constructed containing ′ermGT (which confers lincomycin resistance) as the primary reporter gene for selection of promoters active in the gastrointestinal tract of mice treated with lincomycin. A second reporter gene, ′bglM (β-glucanase), allowed differentiation between constitutive and in vivo inducible promoters. The system was successfully tested in vitro and in vivo by using a constitutive promoter. Application of the IVET system with chromosomal DNA of L. reuteri 100-23 and reconstituted lactobacillus-free mice revealed three genes induced specifically during colonization. Two of the sequences showed homology to genes encoding xylose isomerase (xylA) and peptide methionine sulfoxide reductase (msrB), which are involved in nutrient acquisition and stress responses, respectively. The third locus showed homology to the gene encoding a protein whose function is not known. Our IVET system has the potential to identify genes of lactobacilli that have not previously been functionally characterized but which may be essential for growth of these bacteria in the gastrointestinal ecosystem.

The effect of dietary fat source (soy oil or a mixture of lard and tallow) and dietary supplementation with antibiotics (a combination of avilamycin at 10 mg kg of feed−1 and salinomycin at 40 mg kg of feed−1) on the bacterial community in the ileum of broiler chickens at different ages (7, 14, 21, and 35 days) was studied using PCR with denaturing gradient gel electrophoresis (DGGE) analysis and bacteriological culture. The bacterial origin of fragments in DGGE profiles was identified by sequencing. Bacterial enumeration results, together with PCR-DGGE profiles, showed that the composition of the microflora was age dependent and influenced by dietary fat source and antibiotic supplementation. An increased incidence of streptococci, enterobacteria, and Clostridium perfringens with age of the chickens was demonstrated. Lactobacilli and C. perfringens were the bacterial groups most strongly affected by the dietary treatments. Moreover, different strains (clonal variants of the alpha-toxin gene) of C. perfringens type A were detected in response to age, dietary fat source, and dietary supplementation with antibiotics.

Methods that enabled the identification, detection, and enumeration of Bifidobacterium species by PCR targeting the transaldolase gene were tested. Bifidobacterial species isolated from the feces of human adults and babies were identified by PCR amplification of a 301-bp transaldolase gene sequence and comparison of the relative migrations of the DNA fragments in denaturing gradient gel electrophoresis (DGGE). Two subtypes of Bifidobacterium longum, five subtypes of Bifidobacterium adolescentis, and two subtypes of Bifidobacterium pseudocatenulatum could be differentiated using PCR-DGGE. Bifidobacterium angulatum and B. catenulatum type cultures could not be differentiated from each other. Bifidobacterial species were also detected directly in fecal samples by this combination of PCR and DGGE. The number of species detected was less than that detected by PCR using species-specific primers targeting 16S ribosomal DNA (rDNA). Real-time quantitative PCR targeting a 110-bp transaldolase gene sequence was used to enumerate bifidobacteria in fecal samples. Real-time quantitative PCR measurements of bifidobacteria in fecal samples from adults correlated well with results obtained by culture when either a 16S rDNA sequence or the transaldolase gene sequence was targeted. In the case of samples from infants, 16S rDNA-targeted PCR was superior to PCR targeting the transaldolase gene for the quantification of bifidobacterial populations.

Denaturing gradient gel electrophoresis (DGGE) of DNA fragments generated by PCR with 16S ribosomal DNA-targeted group-specific primers was used to detect lactic acid bacteria (LAB) of the genera Lactobacillus, Pediococcus, Leuconostoc, and Weissella in human feces. Analysis of fecal samples of four subjects revealed individual profiles of DNA fragments originating not only from species that have been described as intestinal inhabitants but also from characteristically food-associated bacteria such as Lactobacillus sakei, Lactobacillus curvatus, Leuconostoc mesenteroides, and Pediococcus pentosaceus. Comparison of PCR-DGGE results with those of bacteriological culture showed that the food-associated species could not be cultured from the fecal samples by plating on Rogosa agar. On the other hand, all of the LAB species cultured from feces were detected in the DGGE profile. We also detected changes in the types of LAB present in human feces during consumption of a milk product containing the probiotic strain Lactobacillus rhamnosus DR20. The analysis of fecal samples from two subjects taken before, during, and after administration of the probiotic revealed that L. rhamnosus was detectable by PCR-DGGE during the test period in the feces of both subjects, whereas it was detectable by culture in only one of the subjects.

Many lactobacilli from various origins were found to apparently lack cholic acid extrusion activity. Cholic acid was accumulated spontaneously, driven by the transmembrane proton gradient. Accumulation is a newly identified kind of interaction between intestinal microbes and unconjugated bile acids and is different from extrusion and modification, which have been described previously.

Plasmid pGT232 (5.1 kb), an indigenous plasmid of Lactobacillus reuteri 100-23, was determined, on the basis of nucleotide and deduced protein sequence data, to belong to the pC194-pUB110 family of plasmids that replicate via the rolling-circle mechanism. The minimal replicon of pGT232 was located on a 1.7-kb sequence consisting of a double-strand origin of replication and a gene encoding the replication initiation protein, repA. An erythromycin-selectable recombinant plasmid containing this minimal replicon was stably maintained (>97% erythromycin-resistant cells) without antibiotic selection in an L. reuteri population under laboratory growth conditions but was poorly maintained (<33% resistant cells) in the L. reuteri population inhabiting the murine gastrointestinal tract. Stable maintenance (>90% resistant cells) of pGT232-derived plasmids in the lactobacillus population in vivo required an additional 1.0-kb sequence which contained a putative single-strand replication origin (SSO). The SSO of pGT232 is believed to be novel and functions in an orientation-specific manner.