Salmonella Typhimurium expansion in the inflamed murine gut is dependent on aspartate derived from ROS-mediated microbiota lysis

Abstract

Inflammation boosts the availability of electron acceptors in the intestinal lumen, creating a favorable niche for pathogenic Enterobacteriaceae. However, the mechanisms linking intestinal inflammation-mediated changes in luminal metabolites and pathogen expansion remain unclear. Here, we show that mucosal inflammation induced by Salmonella enterica serovar Typhimurium (S. Tm) infection increases intestinal levels of the amino acid aspartate. S. Tm used aspartate-ammonia lyase (aspA)-dependent fumarate respiration for growth in the murine gut only during inflammation. AspA-dependent growth advantage was abolished in the gut of germ-free mice and restored in gnotobiotic mice colonized with members of the classes Bacteroidia and Clostridia. Reactive oxygen species (ROS) produced during the host response caused lysis of commensal microbes, resulting in the release of microbiota-derived aspartate that was used by S. Tm, in concert with nitrate-dependent anaerobic respiration, to outcompete commensal Enterobacteriaceae. Our findings demonstrate the role of microbiota-derived amino acids in driving respiration-dependent S. Tm expansion during colitis.

Keywords: Enterobacteriaceae; Salmonella Typhimurium; amino acids; colonization resistance; intestinal inflammation; microbe-microbe interactions; microbial metabolism; microbiome.

Figure 1.. Aspartate utilization via aspA fuels Salmonella Typhimurium expansion during colitis.

(A) Fecal aspartate levels measured from CBA/J mice infected intragastrically with 10⁹ CFU of Salmonella enterica serovar Typhimurium (S. Tm) for ten days. (B) Fecal aspartate levels measured from C57BL/6J mice treated with 2.5% Dextran Sodium Sulfate (DSS) in their drinking water for seven days. (C) aspA gene expression in S. Tm grown in NCE minimal media containing glycerol (40mM) and/or L-aspartate (0.1–10mM) for 4 hours under anaerobic conditions. (D) Wild-type S. Tm and ΔaspA mutant were grown in NCE minimal media supplemented with glycerol (40mM) and/or L-aspartate (30mM). Bacterial numbers were determined at 18 hours post-inoculation. (E) Female CBA/J mice were intragastrically infected with 10⁹ CFU of a 1:1 ratio of wild-type S. Tm (WT) and isogenic ΔaspA mutant for ten days. Competitive index in fecal samples was determined on days one, three, seven, and ten post-infection. (F-G) Combined histopathology score of pathological lesions in the cecum of mice from (D) at three (F) and ten (G) days post-infection (n=5). (H) Representative histopathological image from (G). Bar = 200 μm. For in vitro experiments, each dot represents one biological replicate (average of triplicate technical replicate per biological replicate). For mouse experiments, each dot represents data from one animal (biological replicate). Bars represent mean ± SEM. The boxes in the whisker plot represent the first to third quartiles, and a line indicates the mean value of the gross pathology scores. **, p < 0.01; ***, p < 0.001; ****, p < 0.0001. See also Figure S1.

Figure 2.. Intestinal inflammation is required for aspartate utilization by S. Tm in the gut.

(A) CBA/J mice were intragastrically infected with 10⁹ CFU of a 1:1 ratio of wild-type S. Tm (WT) and ΔaspA mutant or S. Tm ΔinvA ΔspiB and ΔinvA ΔspiB ΔaspA mutant for ten days. Competitive index in feces was determined on days one, three, seven, and ten post-infection. (B) Combined histopathology score of pathological lesions in the cecum of mice from (B) at ten days post-infection (n=5). (C) Fecal aspartate levels from mice in (A) measured at day ten post-infection. (D) CBA/J mice were intragastrically infected with 10⁹ CFU of a 1:1 ratio of S. Tm ΔinvAΔspiB and ΔinvAΔspiBΔaspA mutant for ten days. Some mice received L-aspartate (0.4%) supplementation in their drinking water for the experiment duration (Aspartate +). (E) Combined histopathology score of pathological lesions in the cecum of mice from (B) at ten days post-infection (n=5). (F) Fecal aspartate levels from mice in (A) measured at day ten post-infection. Each dot represents data from one animal (biological replicate). Bars represent mean ± SEM. The boxes in the whisker plot represent the first to third quartiles, and a line indicates the mean value of the gross pathology scores. *, p<0.05; **, p < 0.01; ***, p < 0.001.

Figure 3.. Increased aspartate availability supports fumarate respiration by Salmonella Typhimurium in the inflamed gut.

(A) Wild-type S. Tm and isogenic Δfrd mutants were grown in NCE minimal media supplemented with glycerol (40 mM) and/or L-aspartate (30 mM) for 18 hours under anaerobic conditions. (B) CBA/J mice were intragastrically infected with 10⁹ CFU of a 1:1 ratio of wild-type S. Tm (WT) and ΔaspA mutant or S. Tm Δfrd and ΔfrdΔaspA mutant for ten days. Competitive index in fecal samples was determined at day ten post-infection. (C) Combined histopathology score of pathological lesions in the cecum of mice from (B) at ten days post-infection (n=4). (D) CBA/J mice were intragastrically infected with 10⁹ CFU of wild-type S. Tm (WT) or Δfrd mutant for ten days. Bacterial numbers in feces were determined at day ten post-infection. (E) Fecal aspartate levels from mice in (D) measured at day ten post-infection. (F) Female CBA/J mice were intragastrically infected with 10⁹ CFU of a 1:1 ratio of S. Tm Δfrd and ΔfrdΔaspA isogenic mutant for ten days. Some mice received L-aspartate (0.4%) supplementation in their drinking water for the experiment duration (Aspartate +). (G) Fecal aspartate levels from mice in (F) measured at day ten post-infection. For in vitro experiments, each dot represents one biological replicate (average of triplicate technical replicate per biological replicate). For mouse experiments, each dot represents data from one animal (biological replicate). Bars represent mean ± SEM. The boxes in the whisker plot represent the first to third quartiles, and a line indicates the mean value of the gross pathology scores. **, p < 0.01; ***, p < 0.001. See also Figure S2 and S3.

Figure 4.. Dietary aspartate is not a major source of aspartate for Salmonella Typhimurium in the inflamed gut.

(A) CBA/J mice received a control (Asp +) or an aspartate-free diet (Asp −) for the duration of the experiment. Mice were intragastrically infected with 10⁹ CFU of a 1:1 ratio of wild-type S. Tm (WT) and ΔaspA for ten days. Competitive index in fecal samples was determined at day ten post-infection. (B) Fecal aspartate levels from mice in (A) measured at day ten post-infection. (C) Combined histopathology score of pathological lesions in the cecum of mice from (A) at ten days post-infection (n=5). (D) C57Bl6/J mice received a control (Asp +) or an aspartate-free diet (Asp −) for the duration of the experiment. Mice were treated with 2.5% Dextran Sodium Sulfate (DSS) in their drinking water for seven days. Fecal aspartate levels were measured on day eight post-DSS treatment. (E) Combined histopathology score of pathological lesions in the cecum of mice from (D) at eight days post-DSS treatment (n=5). (F) Representative histopathological image from (E). Bar = 200 μm. Each dot represents data from one animal (biological replicate). Bars represent mean ± SEM. The boxes in the whisker plot represent the first to third quartiles, and a line indicates the mean value of the gross pathology scores. **, p < 0.01.

Figure 5.. Microbiota-derived aspartate supports Salmonella Typhimurium expansion during intestinal inflammation.

(A) Conventional CBA/J and germ-free Swiss Webster (SW) mice were intragastrically infected with 10⁹ CFU of wild-type S. Tm. Conventional C57Bl/6J mice and germ-free SW mice were given 2.5% DSS in the drinking water for seven days. Fecal aspartate levels were measured at day ten post-infection for conventional CBA/J mice, day three post-infection for germ-free SW mice, and day eight after the beginning of treatment for DSS-treated mice. (B-D) Swiss-Wester (SW) germ-free mice were pre-colonized with individual species or a combination of gut commensal microbes for five days. After five days, mice were intragastrically infected with 10⁹ CFU of a 1:1 ratio of wild-type S. Tm (WT) and ΔaspA for three days. (B) Fecal aspartate levels measured at three days after S. Tm infection. (C) Competitive index of S. Tm strains in fecal samples was determined on day three post-infection. (D) Combined histopathology score of pathological lesions in the cecum of mice from (C) at three days post-infection (n=5–10). (E-F) SW germ-free mice were pre-colonized with individual species or a combination of gut commensal microbes for five days. After five days, mice were treated with 2.5% Dextran Sodium Sulfate (DSS) in their drinking water for seven days. (E) Fecal aspartate levels measured at seven days after DSS treatment. (F) Combined histopathology score of pathological lesions in the cecum of mice from (E) at eight days post-DSS treatment (n=4). (B-F) Bact: Bacteroides caecimuris; Clost: a combination of Blautia coccoides, Clostridium cochlearium, and Clostridium sporogenes; Lact: Lactobacillus reuteri. Each dot represents data from one animal (biological replicate). Bars represent mean ± SEM. The boxes in the whisker plot represent the first to third quartiles, and a line indicates the mean value of the gross pathology scores. *, p<0.05; **, p < 0.01. See also Figure S4.

Figure 6.. Host-derived ROS promotes microbiota-derived aspartate release during intestinal inflammation.

(A) Killing assay of B. theta in vitro cultures grown under anaerobic conditions and treated with increasing concentrations of H₂O₂ (ROS representative). n=3. (B) Competitive index (CI) between S. Tm wild-type and ΔaspA mutant grown in supernatant of B. theta cultures exposed to H2O2. (C-E) C57Bl/6 Wild-type and isogenic Cybb−/− mice were treated with streptomycin (5g/L) in drinking water for 48h. 24h after antibiotic removal, mice were infected intragastrically with 10⁹ CFU of a 1:1 ratio of wild-type S. Tm (WT) and ΔaspA for three days. (C) Fecal aspartate levels measured at three days after S. Tm infection. (D) CI of S. Tm strains in fecal samples was determined on day three post-infection. (E) Combined histopathology score of pathological lesions in the cecum of mice from (C) at three days post-infection (n=5). (F-G) SW germ-free mice were pre-colonized with B. theta 1XtetR: ssbfe-1. After five days, mice were provided treated with ATC (100 mg/mL) or vehicle control in the drinking water. 24h after the start of treatment, mice were infected with 10⁹ CFU of a 1:1 ratio of ΔinvAΔspiB S. Tm and ΔinvAΔspiB ΔaspA for two days. (F) a B. theta 1XtetR: ssbfe-1 abundance in fecal samples from SW GF mice before (black dots) and after (purple dots) treatment with ATC. (G) CI of S. Tm strains in feces was determined on day two after the ATC treatment. Each dot represents data from one animal (biological replicate). Bars represent mean ± SEM. The boxes in the whisker plot represent the first to third quartiles, and thea line indicates the mean value of the gross pathology scores*, p<0.05; **, p < 0.01; ***, p< 0.001. See also Figure S5.

Figure 7.. Aspartate-dependent fumarate respiration and nitrate-dependent respiration synergize to enable S. Tm to outcompete commensal Enterobacteriaceae.

(A) Competitive index (CI) of S. Tm wild-type and isogenic ΔaspA mutant at day 7 post-infection in CBA/J mice colonized or not with E. coli Mt1B1. (B) Wild-type S. Tm and isogenic ΔaspA mutant burden in fecal samples at 1,3, 5, and 7 d.p.i. from CBA/J mice colonized or not with E. coli Mt1B1. (C-D) Wild-type S. Tm (C) and wild-type E. coli (Mt1B1) (D) anaerobic growth in NCE media containing glycerol (10 mM) and supplemented with aspartate (30 mM) and/or nitrate (40mM). Bacterial growth measured by OD₆₀₀ at 18h post-inoculation. (E) Wild-type S. Tm and isogenic ΔaspA mutant anaerobic growth in media conditions described in (C). Bacterial growth measured by OD₆₀₀ at 18h post-inoculation. (F) CI for wild-type S. Tm and wild-type E. coli co-cultured in conditions described in (C). (G) CI for wild-type S. Tm or isogenic ΔaspA mutant and wild-type E. coli co-cultured in NCE media containing glycerol (40 mM) and supplemented with aspartate (30 mM) and nitrate (40mM). (H) CBA/J were colonized with E. coli Mt1B1 and were infected with 10⁹ CFU of a 1:1 ratio of wild-type S. Tm and ΔaspA mutant for ten days. CI of S. Tm strains in feces was determined at day ten post-infection. Each dot represents data from one animal (biological replicate). Bars represent mean ± SEM. *, p<0.05; **, p < 0.01; ***, p< 0.001. See also Figure S6.