Through experiments in gnotobiotic mice that employed isogenic mutants of bacterial strains that produce (Bifidobacterium breve) and utilize (Lactobacillus reuteri) 1,2-propanediol, this study provides mechanistic insight into the ecological ramifications of a trophic interaction between gut symbionts. The findings improve our understanding on how cross-feeding influences the competitive fitness of L. reuteri in the vertebrate gut and revealed a putative selective force that shaped the evolution of the species. The findings are relevant since they provide a basis to design rational microbial-based strategies to modulate gut ecosystems, which could employ mixtures of bacterial strains that establish trophic interactions or a personalized approach based on the ability of a resident microbiota to provide resources for the incoming microbe.
KEYWORDS: 1,2-propanediol; cross-feeding; gut microbiome; Lactobacillus; trophic interactions; bifidobacteria; competition; fitness; metabolism; microbial ecology
ABSTRACT
Cross-feeding based on the metabolite 1,2-propanediol has been proposed to have an important role in the establishment of trophic interactions among gut symbionts, but its ecological importance has not been empirically established. Here, we show that in vitro growth of Lactobacillus reuteri (syn. Limosilactobacillus reuteri) ATCC PTA 6475 is enhanced through 1,2-propanediol produced by Bifidobacterium breve UCC2003 and Escherichia coli MG1655 from the metabolization of fucose and rhamnose, respectively. Work with isogenic mutants showed that the trophic interaction is dependent on the pduCDE operon in L. reuteri, which encodes the ability to use 1,2-propanediol, and the l-fucose permease (fucP) gene in B. breve, which is required for 1,2-propanediol formation from fucose. Experiments in gnotobiotic mice revealed that, although the pduCDE operon bestows a fitness burden on L. reuteri ATCC PTA 6475 in the mouse digestive tract, the ecological performance of the strain was enhanced in the presence of B. breve UCC2003 and the mucus-degrading species Bifidobacterium bifidum. The use of the respective pduCDE and fucP mutants of L. reuteri and B. breve in the mouse experiments indicated that the trophic interaction was specifically based on 1,2-propanediol. Overall, our work established the ecological importance of cross-feeding relationships based on 1,2-propanediol for the fitness of a bacterial symbiont in the vertebrate gut.
IMPORTANCE Through experiments in gnotobiotic mice that employed isogenic mutants of bacterial strains that produce (Bifidobacterium breve) and utilize (Lactobacillus reuteri) 1,2-propanediol, this study provides mechanistic insight into the ecological ramifications of a trophic interaction between gut symbionts. The findings improve our understanding on how cross-feeding influences the competitive fitness of L. reuteri in the vertebrate gut and revealed a putative selective force that shaped the evolution of the species. The findings are relevant since they provide a basis to design rational microbial-based strategies to modulate gut ecosystems, which could employ mixtures of bacterial strains that establish trophic interactions or a personalized approach based on the ability of a resident microbiota to provide resources for the incoming microbe.
INTRODUCTION
The gut microbiota is a complex microbial community whose members form interdependent trophic relationships that determine the ecology of bacterial populations and their interplay with the host (1). One such interaction involves the exchange of products between microbes, otherwise known as cross-feeding (2). Cross-feeding of nutrients is central to the fermentative degradation of nondigestible dietary fibers and host-derived substrates, such as mucin or human milk oligosaccharides (HMOs) (3–10). For example, glycans released from the partial degradation of HMOs are utilized by other bifidobacterial species (11), and certain bifidobacteria and Akkermansia muciniphila liberate glycans from host mucins that are utilized by other inhabitants of the gut (6, 12). Similarly, Ruminococcus bromii releases carbohydrates from resistant starch fermentations into the gut environment (13). These interactions ultimately lead to the formation of dynamic metabolic networks essential in the ecology of the gut microbiome and the production of short-chain fatty acids (SCFA), which represent a key ecosystem service that benefits the host (14, 15).
Trophic interactions can also result from the exchange of metabolic end products derived from fermentation processes. Hydrogenotrophic microbes utilize molecular hydrogen produced by several fermentative organisms, a process important to maintain energy flux (16). In addition, cross-feeding based on intermediary metabolites such as lactate, acetate, succinate, and 1,2-propanediol is important for the production of SCFA (5). In the gut, 1,2-propanediol results from the microbial fermentation of deoxyhexose sugars that originate from the hydrolyzation of dietary fiber, fucosylated HMOs, and host mucins, such as rhamnose and fucose (17, 18). Several gut commensal bacteria (i.e., Bacteroides thetaiotaomicron, Roseburia inulinivorans, Escherichia coli, and Bifidobacterium breve) can produce 1,2-propanediol from the fermentation of either rhamnose or fucose (3, 5, 17, 18). Other bacteria, such as Eubacterium hallii and Lactobacillus reuteri (syn. Limosilactobacillus reuteri [55]), do not utilize fucose or rhamnose directly but metabolize 1,2-propanediol (19–21) and convert it to propanol and propionate (21, 22). Cross-feeding of 1,2-propanediol between Eu. hallii and Bifidobacterium species based on metabolism of mucin and HMOs has been studied in vitro (3, 4, 6). Although cross-feeding based on 1,2-propanediol is considered important in gut ecosystems (5) and has been shown to confer a fitness advantage in vivo for pathogens such as Salmonella spp. (23), its ecological importance for bacterial gut symbionts and the underlying mechanisms by which it impacts bacterial performance in the gut have not been empirically established in vivo in a tractable experimental model.
In both Eu. hallii and L. reuteri, 1,2-propanediol is metabolized through the cobalamin-dependent glycerol/diol dehydratase encoded in the pduCDE operon, found in the pdu-cbi-cob-hem gene (pdu) cluster (19–21). The glycerol/diol dehydratase converts 1,2-propanediol to propionaldehyde and further to propanol and propionate and has been shown to increase growth rates for L. reuteri in vitro (19, 24). In L. reuteri, the pdu cluster is observed predominantly in two L. reuteri phylogenetic lineages that are dominated by strains from human, herbivore, and chicken origin but is rarer in strains from other hosts such as mice and rats (25). This distribution suggests that the cluster constitutes an adaptation to the characteristics of the gastrointestinal tract of specific hosts (26, 27). In the murine forestomach, which is densely colonized by L. reuteri (25), fermentable mono- and disaccharides are in ample supply (28). In contrast, distal portions of the human intestinal tract are characterized by low concentrations of mono- and disaccharides. The pdu cluster may therefore constitute a colonization factor in the distal regions of the gut to take advantage of 1,2-propanediol produced by other microbes (19, 25).
Most studies on diol metabolism in L. reuteri have focused on the synthesis of the antimicrobial intermediate β-hydroxypropionaldehyde (reuterin) from glycerol, which is also mediated by the pdu operon (29). It remains unclear whether L. reuteri uses the pdu operon to engage in cross-feeding interrelationships with other gut bacterial species that produce 1,2-propanediol in its gastrointestinal habitat and how this interaction contributes to its ecological competitiveness. It was therefore the aim of this study to characterize the ecological importance of 1,2-propanediol-cross-feeding between L. reuteri and the 1,2-propanediol producers Bifidobacterium breve and Escherichia coli, both in vitro and in the digestive tracts of mice.
RESULTS
1,2-Propanediol enhances growth of L. reuteri in vitro.
We first sought to confirm the importance of 1,2-propanediol metabolism for L. reuteri ATCC PTA 6475 (referred to as L. reuteri PTA 6475) growth in vitro. The presence of 1,2-propanediol in the medium containing glucose improved growth rates and increased the final cell density of the strain compared to the growth solely on glucose or to growth of a pduCDE deletion mutant (referred to as L. reuteri ΔpduCDE) (Fig. 1A). Neither L. reuteri PTA 6475 nor L. reuteri ΔpduCDE is able to use 1,2-propanediol as the sole carbon source for growth (Fig. 1A). To confirm that the enhanced growth of L. reuteri PTA 6475 was due to 1,2-propanediol metabolism, the metabolic end products in the supernatant were measured using high-performance liquid chromatography (HPLC). As shown in Fig. 1B and C, L. reuteri PTA 6475, but not L. reuteri ΔpduCDE, converted 1,2-propanediol to propanol and low concentrations of propionate. For L. reuteri PTA 6475, but not L. reuteri ΔpduCDE, utilization of 1,2-propanediol resulted in acetate production and decreased production of ethanol (Fig. 1D and E; see also Fig. S1A and B in the supplemental material). These findings confirm that L. reuteri PTA 6475 is able to disproportionate 1,2-propanediol to propanol and propionate (22). However, 1,2-propanediol is not used as the sole substrate but is cometabolized with glucose. The predominant production of propanol demonstrates that the reducing branch of the propanediol pathway is preferred over the oxidizing branch to regenerate electron acceptors, thus enhancing acetate formation and ATP production and, therefore, growth (30).
FIG 1.
Impact of 1,2-propanediol on growth and metabolism of L. reuteri PTA 6475 and L. reuteri ΔpduCDE. (A) L. reuteri strains were grown in half-strength mMRS supplemented with either glucose (Glc; 25 mM), 1,2-propanediol (50 mM), or a mixture of the two. Asterisks indicate a significant difference (two-way ANOVA; P < 0.001) in the growth of L. reuteri PTA 6475 on glucose plus 1,2-propanediol compared to the other conditions. (B and C) Utilization of 1,2-propanediol and production of propanol and propionate by L. reuteri PTA 6475 (B) and L. reuteri ΔpduCDE (C) during growth on glucose in the presence of 1,2-propanediol. (D and E) Production of acetate (D) and ethanol (E) by the two strains during growth on glucose in the presence of 1,2-propanediol.
Interspecies cross-feeding of 1,2-propanediol enhances growth of L. reuteri in vitro.
We developed an experimental system to study cross-feeding between L. reuteri and gut bacteria that produce 1,2-propanediol. We chose B. breve and E. coli, which are known to produce 1,2-propanediol from fucose and rhamnose, respectively, which are substrates not utilized by L. reuteri. Since growth rates and growth conditions differ among L. reuteri, B. breve, and E. coli, cross-feeding was not studied in cocultures. Instead, B. breve and E. coli were first grown under their respective optimal conditions and on the specific substrates that result in the production of 1,2-propanediol. Spent supernatant obtained from these fermentations was then supplemented with glucose and half-strength mMRS (conditioned media; see Materials and Methods) and used for analyzing growth kinetics and metabolite production of the L. reuteri strains (Table 1). Conditioned medium from an isogenic mutant of B. breve with an insertion mutation in fucP (encoding the l-fucose transporter), which was unable to metabolize fucose into 1,2-propanediol, served as a control (Table 1).
TABLE 1.
Media used for in vitro cross-feeding experiments
Fermenting strain | Deoxyhexose sugar, concn (mM) | Other carbohydrate, concn (mM) | Abbreviation | Purpose |
---|---|---|---|---|
B. breve UCC2003 | NAa | Cellobiose, 30 | BM (C) | Control for L. reuteri growth absent of 1,2-propanediol production |
Fucose, 30 | Cellobiose, 30 | BM (CF) | For the study of effect of 1,2-propanediol produced from fucose fermentation on L. reuteri | |
B. breve UCC2003-fucP | NA | Cellobiose, 30 | B-fucP M (C) | Control for L. reuteri growth absent of 1,2-propanediol production |
Fucose, 30 | Cellobiose, 30 | B-fucP M (CF) | Control for L. reuteri growth absent of 1,2-propanediol production | |
E. coli MG1655 | NA | Glucose, 25 | EM (G) | Control for L. reuteri growth absent of 1,2-propanediol production |
Rhamnose, 30 | NA | EM (R) | For the study of effect of 1,2-propanediol produced from rhamnose fermentation on L. reuteri |
NA, not applicable.
B. breve UCC2003 and its fucP mutant B. breve UCC2003-fucP were grown in a medium containing cellobiose with or without fucose. Fucose does not support the growth of B. breve UCC2003 and yet is coutilized with cellobiose to produce 1,2-propanediol. Importantly, neither fucose nor cellobiose is metabolized by L. reuteri (20). In the fucose/cellobiose-containing medium, B. breve UCC2003 and B. breve UCC2003-fucP reached similar cell density after 24 h of growth (see Fig. S2 in the supplemental material). L. reuteri PTA 6475 reached a significantly higher optical density at 600 nm (OD600) and showed an elevated growth rate when grown in conditioned medium with the supernatant of B. breve UCC2003 grown with fucose and cellobiose compared to L. reuteri ΔpduCDE or L. reuteri PTA 6475 grown in conditioned medium of B. breve UCC2003 grown in the absence of fucose (Fig. 2A). The growth advantage of L. reuteri PTA 6475 was not observed in conditioned media from the B. breve UCC2003-fucP, even if grown in the presence of fucose (Fig. 2B). HPLC analysis confirmed the presence of 1,2-propanediol in the conditioned medium of B. breve UCC2003 but not B. breve UCC2003-fucP grown with fucose (Fig. S3A) and also showed that enhanced growth of L. reuteri PTA 6475 was linked to the conversion of 1,2-propanediol to propanol, which was not detected in L. reuteri ΔpduCDE cultures (Fig. 2C and D). Propionate, acetate, and ethanol could not be quantified since unknown compounds in the conditioned media interfered with the metabolite analysis. Fucose and cellobiose did not alter the growth kinetics of the L. reuteri strains when grown in glucose (Fig. S4A), and they did not serve as growth substrates on their own (Fig. S4B), confirming that the enhanced growth in L. reuteri PTA 6475 was not due to a direct effect of residual concentrations of these sugars in the spent supernatant.
FIG 2.
Growth and metabolites of L. reuteri PTA 6475 and L. reuteri ΔpduCDE in conditioned media of B. breve UCC2003 and B. breve UCC2003-fucP grown with cellobiose alone or with the addition of fucose. (A and B) Growth curves of L. reuteri strains in B. breve UCC2003 conditioned media (A) and L. reuteri in B. breve UCC2003-fucP conditioned media (B). Asterisks indicate a significant difference (two-way ANOVA; P < 0.001) in growth of L. reuteri PTA 6475 grown in B. breve UCC2003 conditioned medium that had fermented cellobiose and fucose together compared to growth of L. reuteri PTA 6475 in B. breve UCC2003 conditioned medium without fucose or to L. reuteri ΔpduCDE grown in all conditioned media derived from B. breve UCC2003. (C and D) Utilization of B. breve derived 1,2-propanediol present in the conditioned media and production of propanol in cultures of L. reuteri PTA 6475 (C) and L. reuteri ΔpduCDE (D) grown in the conditioned medium of B. breve UCC2003 grown in the presence of fucose. Propionate, acetate, and ethanol concentrations could not be determined due to interference of unknown compounds in the medium. Abbreviations: BM, B. breve UCC2003 conditioned media; B-fucP M, B. breve UCC2003-fucP conditioned media; (C), preculture fermentations of cellobiose only; (CF), preculture fermentations of cellobiose with added fucose (see Table 1 for more details about the media used in this study).
1,2-Propanediol results also from the fermentation of rhamnose by E. coli MG1655 (Fig. S3B). Cross-feeding experiments revealed that L. reuteri PTA 6475 had a higher growth rate and reached a significantly higher cell density when grown in conditioned medium of E. coli grown on rhamnose compared to L. reuteri ΔpduCDE or L. reuteri PTA 6475 grown in conditioned medium from E. coli that did not contain rhamnose (Fig. 3A). Importantly, growth experiments of L. reuteri strains in media with rhamnose with or without glucose confirmed that rhamnose could neither be used as a carbon source nor alter growth (see Fig. S4C and D). Metabolite analysis of fermentations conducted in the conditioned medium of E. coli grown on rhamnose revealed that L. reuteri PTA 6475, and not L. reuteri ΔpduCDE, could metabolize 1,2-propanediol produced by E. coli and form propanol, propionate, and acetate (Fig. 3B to D; see also Fig. S5). Metabolite interference in the conditioned media did not allow the quantification of ethanol by HPLC.
FIG 3.
Growth curves and metabolites of L. reuteri PTA 6475 and L. reuteri ΔpduCDE in conditioned media of E. coli grown with glucose or rhamnose. (A) Growth of L. reuteri strains. Asterisks indicate a significant difference (two-way ANOVA; P < 0.01) in growth of L. reuteri PTA 6475 grown in conditioned medium from E. coli grown on rhamnose compared to conditioned medium from E. coli grown on glucose as well as growth of L. reuteri ΔpduCDE grown in conditioned medium from E. coli grown with either glucose or rhamnose. (B and C) Utilization of E. coli-derived 1,2-propanediol and production of propanol and propionate by L. reuteri PTA 6475 (B) and L. reuteri ΔpduCDE (C) grown in conditioned medium from E. coli grown with rhamnose. (D) Comparison of acetate production by the two strains grown in conditioned medium from E. coli grown with rhamnose. Ethanol concentrations could not be determined due to interference of an unknown compound in the medium. Abbreviations: EM, E. coli conditioned media; (G), preculture fermentation of glucose only by E. coli; (R), preculture fermentation of rhamnose only by E. coli (see Table 1 for more details about the media used in the study).
Together, these findings demonstrate that L. reuteri PTA 6475 utilizes 1,2-propanediol produced by B. breve and E. coli from the fermentation of deoxyhexose sugars as an electron acceptor, enhancing its growth capabilities.
Importance of 1,2-propanediol cross-feeding in the gastrointestinal tract.
The ecological relevance of cross-feeding based on 1,2-propanediol in the gastrointestinal tract was investigated with a series of colonization experiments in gnotobiotic mice (Fig. 4; see Tables S1 and S2). Germ-free Swiss-Webster mice (6 to 16 weeks old; male and female) were housed in individually ventilated cages (in groups of two or three). A fat-free diet was used in order to avoid possible confounding effects from the hydrolysis of dietary triglyceride fats, which is a source of glycerol (also utilized via the pdu cluster to enhance growth). Fecal cell numbers of colonizing strains were determined by selective plating (see Materials and Methods).
FIG 4.
Graphical illustration of hypothesized trophic interactions of 1,2-propanediol in gnotobiotic mice. (A) In triple-species associated gnotobiotic mice (colonized by B. bifidum, B. breve, and L. reuteri), B. bifidum liberates fucose from the degradation of host mucin, which is metabolized by B. breve UCC2003 producing 1,2-propanediol, that is subsequently utilized by L. reuteri PTA 6475. (B) In dual-species (E. coli and L. reuteri) associated mice whose diet has been supplemented with rhamnose added through the drinking water, E. coli metabolizes rhamnose, producing 1,2-propanediol that is subsequently utilized by L. reuteri PTA 6475.
As described earlier, B. breve produces 1,2-propanediol from fucose. Host mucins are an intrinsic source of fucose in the gastrointestinal tract, but B. breve does not possess glycosidases required for mucin degradation (9). 1,2-Propanediol cross-feeding between B. breve and L. reuteri was therefore studied in a triple-species associated mouse in the presence of the mucinolytic strain B. bifidum PRL2010, which is capable of degrading mucin and releasing fucose without producing 1,2-propanediol (Fig. 4A; see Table S1) (9). We first tested wild-type and mutant L. reuteri strains separately. Mice were gavaged with an inoculum that contained B. bifidum PRL2010, B. breve UCC2003, or its fucP mutant and also either L. reuteri PTA 6475 or L. reuteri ΔpduCDE (Table S1, inocula A to D). In these experiments, the two Bifidobacterium strains formed stable populations of ca. 108 to 109 CFU/g (see Fig. S6A to H), while the L. reuteri strains colonized at ca. 106 to 108 CFU/g (see Fig. S7A). When colonizing with wild-type B. breve UCC2003, L. reuteri PTA 6475 formed higher populations than L. reuteri ΔpduCDE over the duration of the experiment. In contrast, the fecal cell numbers of L. reuteri PTA 6475 were consistently lower than those of L. reuteri ΔpduCDE when colonizing with B. breve UCC2003-fucP (but differences did not reach statistical significance due to the high variation between mice) (Fig. S7A). However, normalized ratios (see Materials and Methods) between fecal cell numbers of L. reuteri PTA 6475 and L. reuteri ΔpduCDE were significantly higher in mice colonized with B. breve UCC2003 compared to mice colonized with B. breve UCC2003-fucP (Fig. 5A). Interestingly, while the ratio between L. reuteri PTA 6475 and L. reuteri ΔpduCDE was greater than 1 when the L. reuteri strains were cocolonized with B. breve UCC2003, it was substantially lower than 1 when cocolonized with B. breve UCC2003-fucP (Fig. 5A). These finding suggests that the pdu cluster confers a fitness burden on L. reuteri unless 1,2-propanediol is provided, under which the cluster becomes advantageous.
FIG 5.
Populations of L. reuteri PTA 6475 and L. reuteri ΔpduCDE in the gastrointestinal tract of triple-species and double-species associated gnotobiotic mice. (A) Normalized ratios between L. reuteri PTA 6475 to L. reuteri ΔpduCDE obtained from Bifidobacterium-L. reuteri triple-species associated gnotobiotic mice in which colonization by L. reuteri PTA 6475 and L. reuteri ΔpduCDE was tested separately. A “+” indicates mice colonized with B. breve UCC2003, and a “–” indicates mice colonized with B. breve UCC2003-fucP. (B) Percent CFU for L. reuteri PTA 6475 and L. reuteri ΔpduCDE as measured in triple-species associated gnotobiotic mice in which the two L. reuteri strains were tested in competition. (C) Normalized ratios between L. reuteri PTA 6475 and L. reuteri ΔpduCDE in E. coli-L. reuteri double-species associated gnotobiotic mice in which colonization of L. reuteri PTA 6475 and L. reuteri ΔpduCDE was tested separately. (D) Percent CFU for L. reuteri PTA 6475 and L. reuteri ΔpduCDE mutant in double-species associated gnotobiotic mice in which the two L. reuteri strains were tested in competition. A “+” indicates the presence of rhamnose (Rha) in the diet, while a “–” indicates absence of rhamnose in the diet. Statistical significance for ratios and percent abundance (% CFU) was determined using a Mann-Whitney test and a Fisher exact test, respectively.
In a parallel set of experiments, we tested the importance of 1,2-propanediol cross-feeding in gnotobiotic mice that were gavaged with B. bifidum PRL2010, either B. breve UCC2003 or B. breve UCC2003-fucP, and both L. reuteri PTA 6475 and L. reuteri ΔpduCDE (wild-type and mutant) in direct competition (Table S1, inocula E and F). In agreement with the experiments with only one L. reuteri strain, Bifidobacterium species formed stable populations that were comparable among groups (ca. 108 to 109 CFU/g; Fig. S6I to L). L. reuteri formed stable populations (ca. 106 to 108 CFU/g) and, concordant with the experiments using single strains of L. reuteri, the cell counts of L. reuteri PTA 6475 were higher in mice colonized with B. breve UCC2003 than in mice colonized with B. breve UCC2003-fucP while counts of L. reuteri ΔpduCDE were higher in mice colonized with B. breve UCC2003-fucP (but the differences were not statistically significant) (Fig. S7B). However, significant differences were observed between the relative proportions of the two L. reuteri strains. Specifically, L. reuteri PTA 6475 reached significantly higher proportions (>75%) in mice colonized with B. breve UCC2003 compared to B. breve UCC2003-fucP (Fig. 5B). Interestingly, L. reuteri ΔpduCDE reached ∼75% of the total Lactobacillus population in mice colonized with B. breve UCC2003-fucP (Fig. 5B), supporting the notion that the pdu cluster is a fitness burden to L. reuteri unless 1,2-propanediol is provided. Overall, these observations demonstrated that B. breve UCC2003 can provide 1,2-propanediol as the result of a trophic chain from the degradation of mucin by B. bifidum PRL2010 that facilitates colonization of L. reuteri in the gastrointestinal tract (9).
A set of dual-associated gnotobiotic mouse experiments were also conducted to test whether the production of 1,2-propanediol, from the metabolism of rhamnose by E. coli, influences the fitness of L. reuteri in the gastrointestinal tract. Mice were colonized with E. coli and either (i) L. reuteri PTA 6475 or L. reuteri ΔpduCDE alone or (ii) the two strains in competition. Rhamnose was provided to mice in the drinking water (Fig. 4B; Table S2). Stable populations of E. coli were reached in all mice (∼109 CFU/g), and the provision of rhamnose led to an ∼10-fold increase in fecal cell numbers of E. coli (Fig. S8). In contrast to the findings with gnotobiotic mice colonized by bifidobacteria, L. reuteri ΔpduCDE colonized with higher cell numbers and outcompeted L. reuteri PTA 6475 under all conditions tested (Fig. 5C and D). This was indicated by ratios of <1 between L. reuteri PTA 6475 and L. reuteri ΔpduCDE in mice colonized with single L. reuteri strains (Fig. 5C) and an enrichment of the mutant to ∼60% of the total Lactobacillus population in competition experiments (Fig. 5D). The latter results resemble those obtained with the triple-species Bifidobacterium experiments in mice colonized with B. breve UCC2003-fucP (Fig. 5A and B), confirming the fitness burden of the pdu cluster.
DISCUSSION
Cross-feeding between members of the gut microbiota based on 1,2-propanediol is considered an important ecological process that shapes the gut microbiome and its metabolism. However, conclusions have so far been derived from in vitro experiments and metagenomic predictions. In this study, we demonstrate that L. reuteri engages in trophic interactions with bacteria that are common in the human gut and generate 1,2-propanediol, which is used predominantly to regenerate reduced metabolic cofactors. Using isogenic mutants in both the bacterium that produces 1,2-propanediol and L. reuteri, we ensured that this cross-feeding interaction is in fact based on this metabolic intermediate. We further established that the pduCDE genes constitute a fitness burden for L. reuteri in the gut unless 1,2-propanediol is provided, which makes the cluster ecologically advantageous. Our findings therefore provide insight into both the ecological role of the pdu cluster in L. reuteri and how it evolved.
Our results demonstrate that in vitro, L. reuteri can obtain a growth advantage by cross-feeding from 1,2-propanediol derived from the fermentation of fucose and rhamnose by B. breve and E. coli, respectively. These findings extend previous work showing that pdu cluster-containing L. reuteri strains grow at a higher rate and to a higher cell yield in the presence of glycerol or 1,2-propanediol (19). In contrast to L. reuteri DSM 20016, which produces propanol and propionate in equimolar concentrations through disproportionation when utilizing 1,2-propanendiol as a sole substrate (22), L. reuteri PTA 6475 produces propanol in excess over propionate (Fig. 1B) and enhances acetate formation (Fig. 1D) when utilizing 1,2-propanendiol together with glucose. These metabolic patterns suggest that when L. reuteri PTA 6475 grows with glucose, 1,2-propanediol functions mainly as an electron acceptor, allowing the conversion of acetyl-phosphate to acetate and to generate an extra ATP (30). Apart from enhancing bacterial growth, 1,2-propanendiol metabolism by L. reuteri can therefore potentially contribute to SCFA formation in the gut, which might have health implications as propionate and acetate impact host physiology, e.g., by contributing to gluconeogenesis in the liver, reducing cholesterol, and promoting satiety (5, 18).
Our experiments in gnotobiotic mice provide empirical evidence for the ecological relevance of 1,2-propanendiol cross-feeding for the ecological performance of a gut symbiont in the gastrointestinal tract. The use of the respective pduCDE and fucP mutants of L. reuteri and B. breve in the mouse experiments indicated that this trophic interaction was specifically based on 1,2-propanediol produced through the fermentation of fucose. Since B. breve cannot degrade mucins and the mouse feed was devoid of fucose, the findings suggest that L. reuteri benefits from 1,2-propanediol resulting from interspecies trophic interactions between bifidobacteria, in which B. bifidum PRL2010 degrades host mucin and provides fucose for B. breve UCC2003 to produce 1,2-propanediol (9). The hydrolysis of mucin is suggested to play a key role in the facilitation of bacterial species in the gut microbiota and has been demonstrated in vitro (7, 31, 32) and with bacterial pathogens in vivo (e.g., Salmonella spp. and Clostridium difficile) (23, 33). Our data suggest that it also plays a role in cross-feeding interactions among symbionts or commensals, ultimately conferring a fitness advantage to L. reuteri in the murine gut through the provision of 1,2-propanendiol.
Although our experiments demonstrated 1,2-propanediol cross-feeding between bifidobacteria and L. reuteri, equivalent findings were not observed in mouse experiments with L. reuteri and E. coli. This observation can potentially be attributed to a phenomenon called carbon catabolite repression or the “all-or-none” effect in E. coli, in which a hierarchy-based regulatory system controls the sequential uptake of carbon sources (34). The mouse diet was highly saturated with glucose (34.4% [wt/wt]) and, in vitro, the presence of glucose suppresses the uptake of other carbohydrates in E. coli (35), which is in agreement with our finding that E. coli MG1655 did not produce 1,2-propanendiol from rhamnose while in the presence of glucose (Fig. S3B). Hence, it is possible that the uptake and metabolism of rhamnose into 1,2-propanediol by E. coli was suppressed in the murine gut.
Interestingly, the mouse experiments consistently revealed a fitness burden of the pduCDE genes when 1,2-propendiol was absent. Fitness trade-offs are well understood in the evolution of antibiotic resistance in bacteria, where resistance genes are costly and lead to a reduction of growth (36) and are therefore often lost in the absence of antibiotic pressure (37). Our findings indicate that genes that facilitate cross-feeding interactions are also subjected to fitness trade-offs in that they are only beneficial when the metabolite is provided. Such trade-offs have also been shown in cross-feeding based on the exchange of carbohydrates. Bacteroides ovatus possesses an enzyme system dedicated to the digestion of polysaccharides that does not directly benefit itself, but rather cooperative members of the gut microbiota through reciprocal cross-feeding. This enzyme system is energetically unfavorable and in the absence of a reciprocating species, a knockout mutant strain can outcompete the enzyme-encoding wild-type (38). In L. reuteri, the fitness burden of the pdu genes provides a possible explanation for the evolution of the pdu cluster (26). Although likely ancestral to currently known lineages of L. reuteri, there is evidence of a deletion of the pdu cluster from most rodent strains (25, 26). Growth substrates are abundant in the forestomach and bacteria providing 1,2-propanediol are absent (25, 26), which might have led to loss of the pdu cluster since it was evolutionarily advantageous, in agreement with the Black Queen Hypothesis (39). In the human distal gut, 1,2-propanediol is likely provided through other microbiota members, which may explain why the pdu cluster is conserved among these strains (26, 27).
Although cross-feeding based on 1,2-propandiol is only one of many trophic interactions that establish interactive networks within the gut microbiota, this study provides important information as it establishes the importance and consequences of cross-feeding for the ecological performance of the involved members. Such knowledge has repercussions for our understanding on the ecological and evolutionary forces that shape gut ecosystems and determine how they function (38, 40–42). In addition, an understanding of mutualistic interactions has important implications since it can be translated into improved microbe-based strategies to modulate gut microbiomes (i.e., probiotics). A challenge encountered in the field of probiotics is that gut ecosystems are homeostatic, resilient to change, and thus difficult to modulate, and most probiotics do not persist or change the resident community (43–46). One solution to this problem is the adoption of an ecological framework for probiotic applications (44). A consideration of the mutualistic and facilitative interactions between community members can be used to design probiotic strain mixtures or personalized probiotic applications with the goal to achieve a more successful long-term persistence, which might be beneficial for certain applications. For example, 1,2-propanediol cross-feeding could be considered in generating probiotic products by pairing L. reuteri with Bifidobacterium species that release fucose from the degradation of host-derived substrates and convert it into 1,2-propanediol (9, 12). In addition, bifidobacteria are more prevalent in the infant gut (47, 48), and some strains only partly utilize HMOs, releasing fucose, possibly forming an effective synergistic combination with L. reuteri (3, 32, 40). Furthermore, cross-feeding of 1,2-propanediol derived from gut symbionts and S. enterica serovar Typhimurium has been demonstrated to play a role in promoting pathogen expansion in the gut (23). L. reuteri could play a therapeutic role in excluding pathogenic Salmonella during gastroenteritis by directly competing for the intermediary metabolite. Overall, this information could not only be used to formulate probiotic mixtures and synbiotic products but could also potentially personalize probiotic applications based on the baseline microbiome (43).
MATERIALS AND METHODS
Bacterial strains and culture conditions.
Bacterial strains used in this study are listed in Table 2. L. reuteri strains were grown in de Man, Rogosa, and Sharpe (MRS) medium (Difco) in an anaerobic chamber (gas mix of 5% CO2, 5% H2, and 90% N2). B. bifidum PRL2010, B. breve UCC2003, and B. breve UCC2003-fucP were grown anaerobically in MRS medium supplemented with 0.05% l-cysteine. E. coli were grown in Luria-Bertani (LB) broth with agitation. All incubations were performed at 37°C.
TABLE 2.
Strains used in this study
Species (strain code) | Origin | Relevant feature(s) | Source and/or reference |
---|---|---|---|
Lactobacillus reuteri | |||
ATCC PTA 6475 | Breast milk | 1,2-Propanediol utilizer with complete pdu cluster | BioGaia (53) |
ΔpduCDE | Isogenic mutant of PTA 6475 | Deletion mutant of glycerol/diol dehydratase genes (pduCDE) | 53 |
Escherichia coli | |||
MG1655 | Lab-derived strain | l-Rhamnose utilizer, 1,2-propanediol producer | CGSCa |
Bifidobacterium bifidum | |||
PRL2010 | Infant stool | Mucin degrader, l-fucose producer | 32 |
Bifidobacterium breve | |||
UCC2003 | Infant stool | l-Fucose utilizer, 1,2-propanediol producer | 54 |
UCC2003-fucP | Isogenic mutant of UCC2003 | Insertion mutant of the l-fucose transporter gene (fucP) | 9 |
CGSC, Coli Genetic Stock Center.
Evaluation of the impact of 1,2-propanediol on L. reuteri ATCC PTA 6475 growth.
Overnight cultures of L. reuteri PTA 6475 and L. reuteri ΔpduCDE were inoculated at 1% into 15 ml of half-strength mMRS (49) containing 25 mM glucose alone, 50 mM 1,2-propanediol (Sigma-Aldrich) alone, or 25 mM glucose plus 50 mM 1,2-propanediol. Growth of cell cultures were monitored based on an OD600 with a spectrophotometer every 3 h over the span of 12 h. Samples (1 ml) were collected for HPLC analysis every 3 h. All experiments were performed in triplicate at 37°C under anaerobic conditions.
In vitro production of 1,2-propanediol and cross-feeding assay development.
Precultures of B. breve strains and E. coli were prepared as follows. Full-strength mMRS supplemented with 30 mM cellobiose with or without 30 mM l-fucose were inoculated with 1% overnight cultures of B. breve UCC2003 or B. breve UCC2003-fucP. Full-strength mMRS containing 25 mM glucose or 30 mM l-rhamnose was inoculated with 1% overnight E. coli MG1655 cultures. These fermentations were conducted under anaerobic conditions for 24 h at 37°C. Conditioned media were prepared from precultures as follows. Cells were removed from precultures by centrifugation (5,000 × g for 10 min), and the supernatant was then collected. The supernatant was supplemented with half-strength mMRS from scratch (50% [wt/vol]; mMRS dry reagents) and glucose (25 mM; dry reagent) to the supernatant. These were further adjusted to pH 6.6 and filter sterilized (0.22 μm), stored at 4°C, and used within 48 h. Conditioned media are described in Table 1. For growth experiments, conditioned media were inoculated with L. reuteri strains (1% inoculation). Growth was monitored for 12 h by measuring the OD600 with a spectrophotometer, and 1-ml samples for HPLC analysis were collected every 3 h. All experiments were performed in triplicate under anaerobic conditions at 37°C.
Experiments in gnotobiotic mice.
All animal experiments were performed with the approval of the Animal Care and Use Committee (ACUC) of the University of Alberta (AUP 00002764). Germ-free Swiss-Webster mice (6 to 16 weeks of age, male and female) were bred and maintained in the Health Sciences Laboratory Animals Services (HSLAS) Facility at the University of Alberta. Mice were randomly selected and moved from a flexible-film isolator and housed in sterile, individually ventilated, positive-pressured biocontainment cages for the duration of the experiments (IsoCage P Biocontainment; Tecniplast). To avoid possible confounding effects of glycerol, which is also utilized by the pdu cluster-encoded diol/glycerol dehydratase (50), an irradiated fat-free diet (34.4% glucose and 34.4% cornstarch; Teklad TD.180765) was used in order to minimize possible interference from the hydrolysis of triglyceride fats. After transfer to the biocontainment cages, mice were fed with the new diet for 3 days before colonization with the bacteria.
To study the cross-feeding of 1,2-propanediol in the gastrointestinal tract, groups of mice (n = 5), two or three mice per cage, were assigned to receive either Bifidobacterium-L. reuteri triple-species mixtures (see Table S1) or E. coli-L. reuteri double-species mixtures (see Table S2). To test for cross-feeding of 1,2-propanediol produced from mucin-derived fucose, mice were gavaged with Bifidobacterium-L. reuteri triple-species mixtures containing B. bifidum PRL2010, either B. breve UCC2003 or B. breve UCC2003-fucP, and either L. reuteri PTA 6475 or L. reuteri ΔpduCDE (single L. reuteri strains), or both strains in competition (Table S1). To test for cross-feeding of dietary rhamnose, we gavaged mice with E. coli-L. reuteri double-species mixtures containing E. coli and L. reuteri PTA 6475 or L. reuteri ΔpduCDE (single L. reuteri strains), or both strains in competition (Table S2). Rhamnose was provided in the drinking water (2% [wt/vol]). Each mouse was gavaged with 200 μl of the corresponding bacterial cell mixtures containing ∼108 viable cells of each strain. Fecal pellets were collected from individual mice 1, 3, 5, and 7 days after inoculation and plated. Selective plating was used to enumerate bacterial cells in fecal samples as follows: modified Rogosa agar plates were used to quantify L. reuteri strains (51). L. reuteri PTA 6475 and L. reuteri ΔpduCDE were differentiated using a reuterin hydrazone detection assay (52). MacConkey agar was used for quantifying E. coli. Bifidobacterium was selected using Bifidobacterium selective iodoacetate mupirocin (BSIM) agar as previously described (48). B. bifidum PRL2010 and B. breve strains were differentiated based on colony morphology.
Metabolite analysis of postfermentation.
1,2-Propanediol, propanol, propionate, acetate, and ethanol were measured using HPLC. A Bio-Rad Aminex HPX-87H column (300 mm by 7.8 mm) and a refractive index detector were used (HPLC-RI). Samples taken from fermentations were mixed with 70% HClO4 (0.005% [vol/vol]), stored at 4°C overnight to precipitate proteins, centrifuged (18,800 × g for 5 min), filtered (0.22 μm), and stored at –20°C before injection into HPLC. Then, 10 μl was injected and eluted with 5 mM H2SO4 at a flow rate of 0.4 ml/min at 70°C. 1,2-Propanediol, propanol, propionate, acetate, and ethanol were quantified using external standards.
Statistical analysis.
Statistical significance between L. reuteri growth curves were determined by two-way analysis of variance (ANOVA) with Bonferroni multiple-comparison test (α = 0.05). An unpaired two-tailed Student t test was used to analyze significance between B. breve growth in mMRS supplemented with cellobiose with or without fucose.
Comparisons between L. reuteri PTA 6475 and L. reuteri ΔpduCDE CFU recovered from fecal samples over the duration of the gnotobiotic mice experiments were performed by using an unpaired two-tailed Student t test. Tests were conducted between L. reuteri strains that were associated with (i) B. breve UCC2003 or (ii) B. breve UCC2003-fucP in the triple-species experiments and E. coli-L. reuteri double-species experiments with rhamnose present (iii) or absent (iv).
For the gnotobiotic mouse experiments inoculated with the “single L. reuteri strain” mixtures (see Tables S1 and S2), the CFU of L. reuteri PTA 6475 and L. reuteri ΔpduCDE recovered from mouse feces were used to produce normalized ratios. Ratios were generated using the formula (equation 1) below, where an is a CFU value for L. reuteri PTA 6475 (from a single mouse) used in the comparison, bn is the CFU value of L. reuteri ΔpduCDE from each mouse in the group, and nb is the total population of mice inoculated with the mutant strain used in the experiment.
(1) |
The formula was used to generate sets of ratios for the following comparisons from the “single L. reuteri strain” mouse experiments: (i) B. breve UCC2003 versus B. breve UCC2003-fucP and (ii) E. coli-L. reuteri double-associated mice in the presence versus the absence of rhamnose. The statistical significance between the sets of ratios was determined using the Mann-Whitney U test (P < 0.05).
The Fisher exact test was used to determine the statistical significance between L. reuteri population frequencies from murine groups inoculated with “L. reuteri strains in competition” mixtures (P < 0.05). This was performed between groups of mice from either (i) Bifidobacterium-L. reuteri triple-species associations, including B. breve UCC2003 versus B. breve UCC2003-fucP, or (ii) E. coli-L. reuteri double-species associations with a murine diet supplemented with rhamnose versus without rhamnose. Statistical analyses were performed using GraphPad Prism 6.07.
Supplementary Material
ACKNOWLEDGMENTS
J.W. acknowledges support from the Campus Alberta Innovation Program (CAIP) and the Natural Sciences and Engineering Research Council of Canada (NSERC) for a Discovery grant.
We thank Gautam Gaur for his technical advice concerning HPLC use and Nicole Coursen for her assistance throughout the mouse experiments. We also thank Marco Ventura (University of Parma, Parma, Italy) for generously providing us the B. bifidum PRL2010 strain.
C.C.C. designed and conducted the experiments, collected and analyzed data, and wrote the manuscript. R.M.D. contributed to conceiving the project, the development of experimental approaches, and supervision, provided intellectual leadership, and wrote the manuscript. X.L. contributed to experimental designs and edited the manuscript. M.E.P.-M. contributed to the design of animal experiments and contributed to the manuscript through editing and the generation of figures. S.T. was responsible for animal husbandry and oversaw animal experiments. J.-H.O. and J.-P.V.P. generated the mutant L. reuteri ATCC PTA 6475 ΔpduCDE strains and contributed to manuscript editing. F.L. contributed to supervision and manuscript editing. D.V.S. provided the Bifidobacterium bifidum PRL2010, Bifidobacterium breve UCC2003, Bifidobacterium breve UCC2003-fucP strains; contributed to the design of the animal experiments; and gave technical and conceptual advice. M.G.G. oversaw analytical analysis and data interpretation and contributed to supervision and manuscript editing. J.W. conceived the project, contributed to the conceptualization of the experiments, supervised data analysis and interpretation, and wrote the manuscript.
ADDENDUM IN PROOF
During the proof stage, a taxonomic note that proposed the reclassification of the genus Lactobacillus into 25 genera was published (55). According to that proposal, what is described in the present article as Lactobacillus reuteri is now Limosilactobacillus reuteri.
Footnotes
Supplemental material is available online only.
REFERENCES
- 1.Thursby E, Juge N. 2017. Introduction to the human gut microbiota. Biochem J 474:1823–1836. doi: 10.1042/BCJ20160510. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.D’Souza G, Shitut S, Preussger D, Yousif G, Waschina S, Kost C. 2018. Ecology and evolution of metabolic cross-feeding interactions in bacteria. Nat Prod Rep 35:455–488. doi: 10.1039/c8np00009c. [DOI] [PubMed] [Google Scholar]
- 3.Bunesova V, Lacroix C, Schwab C. 2018. Mucin cross-feeding of infant bifidobacteria and Eubacterium hallii. Microb Ecol 75:228–238. doi: 10.1007/s00248-017-1037-4. [DOI] [PubMed] [Google Scholar]
- 4.Schwab C, Ruscheweyh H, Bunesova V, Pham VT, Beerenwinkel N, Lacroix C. 2017. Trophic interactions of infant bifidobacteria and Eubacterium hallii during l-fucose and fucosyllactose degradation. Front Microbiol 8:95. doi: 10.3389/fmicb.2017.00095. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Louis P, Flint HJ. 2017. Formation of propionate and butyrate by the human colonic microbiota. Environ Microbiol 19:29–41. doi: 10.1111/1462-2920.13589. [DOI] [PubMed] [Google Scholar]
- 6.Belzer C, Chia LW, Aalvink S, Chamlagain B, Piironen V, Knol J, de Vos WM. 2017. Microbial metabolic networks at the mucus layer lead to diet-independent butyrate and vitamin B12 production by intestinal symbionts. mBio 8:e00770-17. doi: 10.1128/mBio.00770-17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Martens EC, Chiang HC, Gordon JI. 2008. Mucosal glycan foraging enhances fitness and transmission of a saccharolytic human gut bacterial symbiont. Cell Host Microbe 4:447–457. doi: 10.1016/j.chom.2008.09.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Egan M, O’Connell Motherway M, Ventura M, van Sinderen D. 2014. Metabolism of sialic acid by Bifidobacterium breve UCC2003. Appl Environ Microbiol 80:4414–4426. doi: 10.1128/AEM.01114-14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Egan M, O’Connell Motherway M, Kilcoyne M, Kane M, Joshi L, Ventura M, van Sinderen D. 2014. Cross-feeding by Bifidobacterium breve UCC2003 during cocultivation with Bifidobacterium bifidum PRL2010 in a mucin-based medium. BMC Microbiol 14:282. doi: 10.1186/s12866-014-0282-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Duranti S, Lugli GA, Milani C, James K, Mancabelli L, Turroni F, Alessandri G, Mangifesta M, Mancino W, Ossiprandi MC, Iori A, Rota C, Gargano G, Bernasconi S, Di Pierro F, van Sinderen D, Ventura M. 2019. Bifidobacterium bifidum and the infant gut microbiota: an intriguing case of microbe-host co-evolution. Environ Microbiol 21:3686–3695. doi: 10.1111/1462-2920.14705. [DOI] [PubMed] [Google Scholar]
- 11.Gotoh A, Katoh T, Sakanaka M, Ling Y, Yamada C, Asakuma S, Urashima T, Tomabechi Y, Katayama-Ikegami A, Kurihara S, Yamamoto K, Harata G, He F, Hirose J, Kitaoka M, Okuda S, Katayama T. 2018. Sharing of human milk oligosaccharides degradants within bifidobacterial communities in faecal cultures supplemented with Bifidobacterium bifidum. Sci Rep 8:13958. doi: 10.1038/s41598-018-32080-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Tailford LE, Crost EH, Kavanaugh D, Juge N. 2015. Mucin glycan foraging in the human gut microbiome. Front Genet 6:81. doi: 10.3389/fgene.2015.00081. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Ze X, Duncan SH, Louis P, Flint HJ. 2012. Ruminococcus bromii is a keystone species for the degradation of resistant starch in the human colon. ISME J 6:1535–1543. doi: 10.1038/ismej.2012.4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Flint HJ, Scott KP, Duncan SH, Louis P, Forano E. 2012. Microbial degradation of complex carbohydrates in the gut. Gut Microbes 3:289–306. doi: 10.4161/gmic.19897. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Koh A, De Vadder F, Kovatcheva-Datchary P, Bäckhed F. 2016. From dietary fiber to host physiology: short-chain fatty acids as key bacterial metabolites. Cell 165:1332–1345. doi: 10.1016/j.cell.2016.05.041. [DOI] [PubMed] [Google Scholar]
- 16.Smith NW, Shorten PR, Altermann EH, Roy NC, McNabb WC. 2019. Hydrogen cross-feeders of the human gastrointestinal tract. Gut Microbes 10:270–288. doi: 10.1080/19490976.2018.1546522. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Saxena RK, Anand P, Saran S, Isar J, Agarwal L. 2010. Microbial production and applications of 1,2-propanediol. Indian J Microbiol 50:2–11. doi: 10.1007/s12088-010-0017-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Reichardt N, Duncan SH, Young P, Belenguer A, McWilliam Leitch C, Scott KP, Flint HJ, Louis P. 2014. Phylogenetic distribution of three pathways for propionate production within the human gut microbiota. ISME J 8:1323–1335. doi: 10.1038/ismej.2014.14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Rattanaprasert M, Roos S, Hutkins RW, Walter J. 2014. Quantitative evaluation of synbiotic strategies to improve persistence and metabolic activity of Lactobacillus reuteri DSM 17938 in the human gastrointestinal tract. J Funct Foods 10:85–94. doi: 10.1016/j.jff.2014.05.017. [DOI] [Google Scholar]
- 20.Schwab C, Gänzle M. 2011. Lactic acid bacteria fermentation of human milk oligosaccharide components, human milk oligosaccharides and galactooligosaccharides. FEMS Microbiol Lett 315:141–148. doi: 10.1111/j.1574-6968.2010.02185.x. [DOI] [PubMed] [Google Scholar]
- 21.Engels C, Ruscheweyh H, Beerenwinkel N, Lacroix C, Schwab C. 2016. The common gut microbe Eubacterium hallii also contributes to intestinal propionate formation. Front Microbiol 7:713. doi: 10.3389/fmicb.2016.00713. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Sriramulu DD, Liang M, Hernandez-Romero D, Raux-Deery E, Lunsdorf H, Parsons JB, Warren MJ, Prentice MB. Lactobacillus reuteri DSM 20016 produces cobalamin-dependent diol dehydratase in metabolosomes and metabolizes 1,2-propanediol by disproportionation. J Bacteriol 190:4559–4567. doi: 10.1128/JB.01535-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Faber F, Thiennimitr P, Spiga L, Byndloss MX, Litvak Y, Lawhon S, Andrews-Polymenis HL, Winter SE, Bäumler AJ. 2017. Respiration of microbiota-derived 1,2-propanediol drives Salmonella expansion during colitis. PLoS Pathog 13:e1006129. doi: 10.1371/journal.ppat.1006129. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Chen L, Hatti-Kaul R. 2017. Exploring Lactobacillus reuteri DSM20016 as a biocatalyst for transformation of longer chain 1,2-diols: limits with microcompartment. PLoS One 12:e0185734. doi: 10.1371/journal.pone.0185734. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Walter J, Britton RA, Roos S. 2011. Host-microbial symbiosis in the vertebrate gastrointestinal tract and the Lactobacillus reuteri paradigm. Proc Natl Acad Sci U S A 108:4645–4652. doi: 10.1073/pnas.1000099107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Frese SA, Benson AK, Tannock GW, Loach DM, Kim J, Zhang M, Oh PL, Heng NCK, Patil PB, Juge N, Mackenzie DA, Pearson BM, Lapidus A, Dalin E, Tice H, Goltsman E, Land M, Hauser L, Ivanova N, Kyrpides NC, Walter J. 2011. The evolution of host specialization in the vertebrate gut symbiont Lactobacillus reuteri. PLoS Genet 7:e1001314. doi: 10.1371/journal.pgen.1001314. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Oh PL, Benson AK, Peterson DA, Patil PB, Moriyama EN, Roos S, Walter J. 2010. Diversification of the gut symbiont Lactobacillus reuteri as a result of host-driven evolution. ISME J 4:377–387. doi: 10.1038/ismej.2009.123. [DOI] [PubMed] [Google Scholar]
- 28.Tannock GW, Wilson CM, Loach D, Cook GM, Eason J, O’Toole PW, Holtrop G, Lawley B. 2012. Resource partitioning in relation to cohabitation of Lactobacillus species in the mouse forestomach. ISME J 6:927–938. doi: 10.1038/ismej.2011.161. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Cleusix V, Lacroix C, Vollenweider S, Duboux M, Le Blay G. 2007. Inhibitory activity spectrum of reuterin produced by Lactobacillus reuteri against intestinal bacteria. BMC Microbiol 7:101. doi: 10.1186/1471-2180-7-101. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Gänzle MG. 2015. Lactic metabolism revisited: metabolism of lactic acid bacteria in food fermentations and food spoilage. Curr Opin Food Sci 2:106–117. doi: 10.1016/j.cofs.2015.03.001. [DOI] [Google Scholar]
- 31.Turroni F, Milani C, Duranti S, Mahony J, van Sinderen D, Ventura M. 2018. Glycan utilization and cross-feeding activities by bifidobacteria. Trends Microbiol 26:339–350. doi: 10.1016/j.tim.2017.10.001. [DOI] [PubMed] [Google Scholar]
- 32.Turroni F, Bottacini F, Foroni E, Mulder I, Kim J, Zomer A, Sánchez B, Bidossi A, Ferrarini A, Giubellini V, Delledonne M, Henrissat B, Coutinho P, Oggioni M, Fitzgerald GF, Mills D, Margolles A, Kelly D, van Sinderen D, Ventura M, Klaenhammer TR. 2010. Genome analysis of Bifidobacterium bifidum PRL2010 reveals metabolic pathways for host-derived glycan foraging. Proc Natl Acad Sci U S A 107:19514–19519. doi: 10.1073/pnas.1011100107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Ng KM, Ferreyra JA, Higginbottom SK, Lynch JB, Kashyap PC, Gopinath S, Naidu N, Choudhury B, Weimer BC, Monack DM, Sonnenburg JL. 2013. Microbiota-liberated host sugars facilitate post-antibiotic expansion of enteric pathogens. Nature 502:96–99. doi: 10.1038/nature12503. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Kremling A, Geiselmann J, Ropers D, de Jong H. 2015. Understanding carbon catabolite repression in Escherichia coli using quantitative models. Trends Microbiol 23:99–109. doi: 10.1016/j.tim.2014.11.002. [DOI] [PubMed] [Google Scholar]
- 35.Aidelberg G, Towbin BD, Rothschild D, Dekel E, Bren A, Alon U. 2014. Hierarchy of non-glucose sugars in Escherichia coli. BMC Syst Biol 8:133. doi: 10.1186/s12918-014-0133-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Basra P, Alsaadi A, Bernal-Astrain G, O’Sullivan ML, Hazlett B, Clarke LM, Schoenrock A, Pitre S, Wong A. 2018. Fitness trade-offs of antibiotic resistance in extraintestinal pathogenic Escherichia coli. Genome Biol Evol 10:667–679. doi: 10.1093/gbe/evy030. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Andersson DI, Hughes D. 2010. Antibiotic resistance and its cost: is it possible to reverse resistance? Nat Rev Microbiol 8:260–271. doi: 10.1038/nrmicro2319. [DOI] [PubMed] [Google Scholar]
- 38.Rakoff-Nahoum S, Foster KR, Comstock LE. 2016. The evolution of cooperation within the gut microbiota. Nature 533:255–259. doi: 10.1038/nature17626. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Smith NW, Shorten PR, Altermann E, Roy NC, McNabb WC. 2019. The classification and evolution of bacterial cross-feeding. Front Ecol Evol 7:153. doi: 10.3389/fevo.2019.00153. [DOI] [Google Scholar]
- 40.O’Connell Motherway M, O’Brien F, O’Driscoll T, Casey PG, Shanahan F, van Sinderen D. 2018. Carbohydrate syntrophy enhances the establishment of Bifidobacterium breve UCC2003 in the neonatal gut. Sci Rep 8:10627. doi: 10.1038/s41598-018-29034-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Coyte KZ, Rakoff-Nahoum S. 2019. Understanding competition and cooperation within the mammalian gut microbiome. Curr Biol 29:R538–R544. doi: 10.1016/j.cub.2019.04.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Rakoff-Nahoum S, Coyne M, Comstock L. 2014. An ecological network of polysaccharide utilization among human intestinal symbionts. Curr Biol 24:40–49. doi: 10.1016/j.cub.2013.10.077. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Maldonado-Gómez MX, Martínez I, Bottacini F, O’Callaghan A, Ventura M, van Sinderen D, Hillmann B, Vangay P, Knights D, Hutkins RW, Walter J. 2016. Stable engraftment of Bifidobacterium longum AH1206 in the human gut depends on individualized features of the resident microbiome. Cell Host Microbe 20:515–526. doi: 10.1016/j.chom.2016.09.001. [DOI] [PubMed] [Google Scholar]
- 44.Walter J, Maldonado-Gómez MX, Martínez I. 2018. To engraft or not to engraft: an ecological framework for gut microbiome modulation with live microbes. Curr Opin Biotechnol 49:129–139. doi: 10.1016/j.copbio.2017.08.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Khalesi S, Bellissimo N, Vandelanotte C, Williams S, Stanley D, Irwin C. 2019. A review of probiotic supplementation in healthy adults: helpful or hype? Eur J Clin Nutr 73:24–37. doi: 10.1038/s41430-018-0135-9. [DOI] [PubMed] [Google Scholar]
- 46.Zmora N, Zilberman-Schapira G, Suez J, Mor U, Dori-Bachash M, Bashiardes S, Kotler E, Zur M, Regev-Lehavi D, Brik RB, Federici Sw, Cohen Y, Linevsky R, Rothschild D, Moor AE, Ben-Moshe S, Harmelin A, Itzkovitz S, Maharshak N, Shibolet O, Shapiro H, Pevsner-Fischer M, Sharon I, Halpern Z, Segal E, Elinav E. 2018. Personalized gut mucosal colonization resistance to empiric probiotics is associated with unique host and microbiome features. Cell 174:1388–1405.e21. doi: 10.1016/j.cell.2018.08.041. [DOI] [PubMed] [Google Scholar]
- 47.Ferretti P, Pasolli E, Tett A, Asnicar F, Gorfer V, Fedi S, Armanini F, Truong DT, Manara S, Zolfo M, Beghini F, Bertorelli R, De Sanctis V, Bariletti I, Canto R, Clementi R, Cologna M, Crifò T, Cusumano G, Gottardi S, Innamorati C, Masè C, Postai D, Savoi D, Duranti S, Lugli GA, Mancabelli L, Turroni F, Ferrario C, Milani C, Mangifesta M, Anzalone R, Viappiani A, Yassour M, Vlamakis H, Xavier R, Collado CM, Koren O, Tateo S, Soffiati M, Pedrotti A, Ventura M, Huttenhower C, Bork P, Segata N. 2018. Mother-to-Infant microbial transmission from different body sites shapes the developing infant gut microbiome. Cell Host Microbe 24:133–145.e5. doi: 10.1016/j.chom.2018.06.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Lewis ZT, Totten SM, Smilowitz JT, Popovic M, Parker E, Lemay DG, Tassell ML, Miller MJ, Jin Y, German JB, Lebrilla CB, Mills DA. 2015. Maternal fucosyltransferase 2 status affects the gut bifidobacterial communities of breastfed infants. Microbiome 3:13. doi: 10.1186/s40168-015-0071-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Stolz P, Vogel RF, Hammes WP. 1995. Utilization of electron acceptors by lactobacilli isolated from sourdough. Z Lebensm Unters Forch 201:402–410. doi: 10.1007/BF01192742. [DOI] [Google Scholar]
- 50.Talarico TL, Axelsson LT, Novotny J, Fiuzat M, Dobrogosz WJ. 1990. Utilization of glycerol as a hydrogen acceptor by Lactobacillus reuteri: purification of 1,3-propanediol:NAD+ oxidoreductase. Appl Environ Microbiol 56:943–948. doi: 10.1128/AEM.56.4.943-948.1990. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Duar RM, Frese SA, Lin XB, Fernando SC, Burkey TE, Tasseva G, Peterson DA, Blom J, Wenzel CQ, Szymanski CM, Walter J. 2017. Experimental evaluation of host adaptation of Lactobacillus reuteri to different vertebrate species. Appl Environ Microbiol 83:e00132-17. doi: 10.1128/AEM.00132-17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.Rosander A, Connolly E, Roos S. 2008. Removal of antibiotic resistance gene-carrying plasmids from Lactobacillus reuteri ATCC 55730 and characterization of the resulting daughter strain, L. reuteri DSM 17938. Appl Environ Microbiol 74:6032–6040. doi: 10.1128/AEM.00991-08. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53.Zhang S, Oh J, Alexander LM, Özçam M, van Pijkeren J. 2018. d-Alanyl-d-alanine ligase as a broad-host-range counterselection marker in vancomycin-resistant lactic acid bacteria. J Bacteriol 200:e00607-17. doi: 10.1128/JB.00607-17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 54.Mazé A, O’Connell-Motherway M, Fitzgerald GF, Deutscher J, van Sinderen D. 2007. Identification and characterization of a fructose phosphotransferase system in Bifidobacterium breve UCC2003. Appl Environ Microbiol 73:545–553. doi: 10.1128/AEM.01496-06. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55.Zheng J, Wittouck S, Salvetti E, Franz CMAP, Harris HMB, Mattarelli P, O’Toole PW, Pot B, Vandamme P, Walter J, Watanabe K, Wuyts S, Felis GE, Gänzle MG, Lebeer S. 15 April 2020. A taxonomic note on the genus Lactobacillus: description of 23 novel genera, emended description of the genus Lactobacillus Beijerinck 1901, and union of Lactobacillaceae and Leuconostocaceae. Int J Syst Evol Microbiol doi: 10.1099/ijsem.0.004107. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.