Microbiota-derived indole limits Campylobacter jejuni colonization by inhibiting respiration and metabolism Authors: Ritam Sinha, Barsha Bhattarai, Cristina Kraemer Zimpel, Elizabeth Ottosen, Rhiannon M. LeVeque, Pallavi Singh, Victor J. DiRita Published: 2026 ABSTRACT Campylobacter jejuni is a major enteric pathogen whose ability to grow in the inflamed gut remains poorly understood, limiting the development of effective therapeutic strategies. Our prior study in ferrets suggested that intestinal inflammation promotes C. jejuni expansion during infection. However, conventional mice are naturally resistant to C. jejuni colonization unless the microbiota or host inflammatory pathways are altered. To directly investigate the role of inflammation in C. jejuni infection, transient colitis was induced in conventional mice using short-term dextran sodium sulfate (DSS) treatment. DSS-mediated inflammation disrupted colonization resistance and enabled rapid C. jejuni growth in the colon within three days of infection, accompanied by aggravated intestinal inflammation. Microbiota analysis revealed enrichment of mucin-degrading bacteria and depletion of taxa associated with short-chain fatty acid and indole production. Metabolomic profiling further demonstrated significantly reduced colonic indole levels in DSS-treated and infected mice. In vitro studies showed that physiological concentrations of indole inhibited C. jejuni growth and downregulated genes involved in major energy-generating pathways, including nitrate respiration (napA), aerobic respiration (ccoN), lactate utilization (lctP), and the acetate switch (ackA/ptaA). Consistent with these findings, mutations in these pathways reduced bacterial fitness in DSS-treated mice, highlighting their importance for colonization in the inflamed intestine. Furthermore, administration of indole or the indole-producing probiotic Escherichia coli Nissle 1917 significantly decreased C. jejuni colonization in vivo. Collectively, these findings demonstrate that intestinal inflammation facilitates C. jejuni colonization while microbiota-derived metabolites, particularly indole, play a critical role in suppressing pathogen growth and pathogenicity. INTRODUCTION Campylobacter jejuni, a Gram-negative foodborne pathogen, is a leading cause of gastroenteritis in humans, primarily transmitted via contaminated food products (1). The emergence of multidrug-resistant strains highlights the urgent need to understand the molecular mechanisms underlying C. jejuni pathogenesis (2). A central question in this field is how C. jejuni adapts and proliferates in the face of host inflammation. We demonstrated that the ferret is a physiologically relevant model of C. jejuni disease and determined that the pathogen exploits intestinal inflammation to grow in the gut (3). During acute infection in this model, C. jejuni triggered pronounced crypt hyperplasia in the colon, consistent with activation of epithelial repair programs. This response was accompanied by goblet cell depletion and a marked increase in undifferentiated, proliferating (Ki67+) epithelial cells. We also observed increased epithelial oxygenation and metabolic remodeling, including elevated colonic luminal L-lactate, which supported bacterial growth (3). Although highly informative, the ferret model is constrained by high cost, limited tractability, and the lack of robust genetic tools to dissect host determinants of C. jejuni proliferation. Mice, in contrast, offer a well-characterized immune system, genetic tractability, and affordability. However, to support C. jejuni colonization and growth, conventional mouse models typically require genetic modifications, such as IL-10 deficiency, or manipulation of the microbiota, complicating studies of host-pathogen interactions (4–6). Based on our observations using ferrets to study C. jejuni infection, we hypothesized that inducing an intestinal inflammatory environment in mice would permit C. jejuni colonization and growth without genetic or antibiotic interventions. In this study, we applied short-term treatment with dextran sodium sulfate (DSS), a widely used inflammatory agent in murine models of colitis and irritable bowel syndrome, leading to increases in epithelial oxygenation and luminal lactate levels (7–9). Following DSS treatment, mice were orally infected with C. jejuni, resulting in successful colonization and growth, demonstrating that inflammation alone is sufficient to facilitate this outcome. We further characterized the DSS-treated mouse model by profiling gut microbiota and metabolites. Notably, DSS treatment reduced levels of indole-producing commensals, leading to a significant decrease in luminal indole. Further analysis revealed that indole directly inhibits C. jejuni growth and colonization by disrupting bacterial respiration and metabolic pathways, thereby highlighting a previously unappreciated microbiota-mediated mechanism of colonization resistance. Together, these findings establish a physiologically relevant and genetically tractable mouse model of C. jejuni colonization driven by inflammation. This model provides a robust platform to dissect host, microbial, and metabolic factors that regulate pathogen growth and offers insights into how C. jejuni exploits inflammatory niches to overcome microbiota-mediated resistance. RESULTS Intestinal inflammation promotes robust C. jejuni growth in the murine colon Intestinal inflammation promotes the outgrowth of several facultative enteric bacteria, including Salmonella Typhimurium, E. coli, and Citrobacter rodentium, in the murine gut (10–12). However, whether inflammation similarly enhances colonization by the microaerophilic pathogen C. jejuni remains poorly understood and our work with a ferret model led us to consider this possibility. We performed an infection experiment using four groups of female mice (n = 5 per group). Two groups were treated with 3% DSS in drinking water for three days to induce intestinal inflammation, while the remaining two groups received normal drinking water as controls. On the following day, one DSS-treated group and one control group were orally infected with C. jejuni WT 11168 strain (108 CFU/mL). After infection, DSS was withdrawn and all mice were returned to normal drinking water (Fig. 1A). Intestinal inflammation promotes Campylobacter jejuni growth in the mouse gut. (A) Schematic diagram of the experimental design for the DSS-induced colitis and C. jejuni infection mouse model. (B) C. jejuni 11168 WT bacterial loads in stool samples were quantified by colony-forming unit (CFU) counts on days 1, 2, and 3 postinfection following oral inoculation with 108 CFU/mL (n = 5). (C) C. jejuni 11168 WT loads in colon tissues were determined by CFU counts on day 3 postinfection (n = 5). (D) Immunohistochemistry (IHC) using a C. jejuni-specific antibody was performed to determine bacterial localization in colon tissue on day 3 postinfection. Black arrows indicate the presence of C. jejuni in infected colonic tissue. Representative images are shown from two independent experiments. (E) Colonic inflammatory responses were evaluated by histological analysis and histological scoring on day 3 postinfection. Representative hematoxylin and eosin (H&E)–stained sections are shown. Yellow arrows indicate epithelial cells, red arrows indicate intestinal crypts, green arrows indicate goblet cells, and blue arrows indicate infiltrating leukocytes in the lamina propria. Data are presented as mean ± SE. Statistical analysis was performed using one-way ANOVA. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.000. At day one postinfection, fecal loads of C. jejuni were approximately 108 CFU/g in both DSS-treated and untreated infected groups. However, by day three, fecal C. jejuni levels gradually declined in DSS-untreated infected mice, whereas DSS-treated infected mice showed an increase to approximately 109–1010 CFU/g (Fig. 1B). Because peak colonization in human infection is typically observed two-to-three days postinfection in the large intestine, we next quantified C. jejuni loads in the colon at day three. We detected approximately 109–1010 CFU C. jejuni/g of colon tissue in DSS-treated infected mice (Fig. 1C). In contrast, C. jejuni was undetectable or below the limit of detection in DSS-untreated infected mice, as well as in both uninfected control groups. These results indicate that DSS-induced inflammation promotes C. jejuni growth in the mouse colon by day three postinfection. The highest colonization was observed in the colon and cecum (∼109 CFU/g tissue), whereas the small intestine showed lower levels (∼106 CFU/g tissue) at day three postinfection (fig. S1A). Because C. jejuni levels were highest in the colon, we examined its localization within colonic tissue. Immunohistochemistry with an anti-Campylobacter antibody revealed sparse clusters of C. jejuni in the colonic lumen, closely associated with the colonic epithelium and detectable within crypts. This localization pattern resembles what has been reported during both human and ferret infection (Fig. 1D) (3, 13). C. jejuni colonized across the colon, with no significant difference between proximal and distal segments (∼107 CFU/g tissue) (fig. S1B). We also examined whether colonization was sex-dependent by comparing male and female mice and found no significant difference between sexes (fig. S1C). We evaluated the inflammatory response following DSS treatment and C. jejuni infection. H&E staining and histopathological scoring showed a moderate inflammatory response from DSS treatment alone, characterized by disruption of the epithelial cell layer, leukocyte infiltration into the superficial lamina propria, cryptic edema and cryptic hyperplasia in colonic tissue compared to the two control groups. Infection with C. jejuni in DSS-treated mice triggered a more severe inflammatory response, with significantly higher pathological scores and more pronounced tissue damage compared to DSS treatment alone (Fig. 1D). To further characterize the inflammatory environment, we measured colonic mRNA levels of key cytokines using RT-PCR. are Established proinflammatory mediators in colitis are IL-6, TNF-α, and KC: IL-6 promotes immune cell activation and sustains chronic inflammation, TNF-α drives epithelial damage and leukocyte recruitment, and KC (CXCL1) recruits neutrophils that contribute to mucosal injury (14, 15). In contrast, IL-10 is a potent anti-inflammatory cytokine that limits excessive immune responses in the gut by suppressing proinflammatory signaling, preserving barrier function, and maintaining immune tolerance to microbial antigens; loss of IL-10 leads to more severe colitis in animal models (16). Transcript levels were compared among DSS-treated uninfected mice, DSS-treated infected mice, and untreated uninfected controls. As expected, DSS treatment alone induced a strong inflammatory response, with ∼five-to-15-fold increased transcription of IL-6, TNF-α, and KC, along with an ∼five-fold reduction in IL-10 expression compared to untreated uninfected mice (Fig. 2A). C. jejuni infection in DSS-treated mice amplified this response. Relative to DSS-treated uninfected mice, infected mice showed elevated expression of all three pro-inflammatory cytokines, while IL-10 expression was further reduced (Fig. 2A). Inflammatory and pathophysiological changes during Campylobacter jejuni infection in DSS-treated mice. (A) Relative expression of proinflammatory cytokine genes (IL-6, TNF-α, and KC) and the anti-inflammatory cytokine gene IL-10 was measured by qRT-PCR in colonic tissues from DSS-treated uninfected and C. jejuni-infected mice relative to PBS-treated controls (n = 5). Fold changes were calculated using the 2−ΔΔCT method. (B) Representative Alcian blue-stained colon sections and quantification of goblet cells per crypt, together with representative Ki67+ immunostaining and quantification of Ki67+ proliferating cells per crypt, on day 3 postinfection (n = 5). (C) Colonic hypoxia was evaluated by pimonidazole staining. Pimonidazole binding was detected by immunohistochemistry using an anti-pimonidazole antibody in colon sections collected on day 3 postinfection. Hypoxia was quantified by blinded scoring of staining intensity. (D) Competitive infection between the wild-type (WT) strain and the motAB::kan mutant. DSS-treated mice were infected with a 1:1 mixture of WT and mutant bacteria. On day 3 postinfection, the mutant-to-WT ratio was determined and expressed as the competitive index (CI; n = 5). Data are presented as mean ± SD. Statistical significance in panels [(A) to (C)] was determined by one-way ANOVA. The CI in panel D was analyzed using a one-sample t-test against a hypothetical value of 1. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. Infection with C. jejuni in the ferret induces significant epithelial differentiation and proliferation (3). To determine whether DSS-induced inflammation and C. jejuni colonization results in similar hyperplasia in the mouse, we quantified goblet cells and Ki67+ proliferative cells in colonic tissue at day three postinfection. DSS treatment alone significantly reduced the number of Alcian blue–positive goblet cells per crypt, with an approximately two-fold reduction relative to untreated uninfected controls (Fig. 2B). This effect was enhanced after C. jejuni infection, with the DSS-treated, infected mice exhibiting approximately three-fold reduction in goblet cells per crypt compared with controls, indicating a significant loss of mucus-producing, terminally differentiated cells during infection (Fig. 2B). We next assessed epithelial proliferation using Ki67 immunohistochemistry. DSS treatment alone increased by almost two-fold the number of Ki67+ cells per crypt compared to untreated uninfected mice, consistent with activation of epithelial repair. The Ki67+ cell numbers were further increased to approximately three-fold after C. jejuni infection of DSS-treated mice, demonstrating that infection amplifies crypt hyperplasia and expansion of proliferating epithelial cells (Fig. 2B). This phenotype closely mirrors the epithelial remodeling response we previously observed during C. jejuni infection in ferrets (3). A major consequence of epithelial differentiation state is its impact on oxygen consumption and oxygen availability at the mucosal surface (10). Mature colonocytes metabolize microbially derived butyrate through β-oxidation, a process that consumes oxygen and contributes to maintenance of physiological hypoxia in the intestinal lumen (10). In contrast, undifferentiated proliferating epithelial cells, enriched at the base of crypts, lack robust β-oxidation and rely on aerobic glycolysis, producing lactate from glucose. Because these proliferating cells consume less oxygen, crypt bases are relatively more oxygenated than the upper crypt regions, and expansion of undifferentiated epithelial populations is predicted to increase epithelial oxygenation overall (10). Because DSS treatment and C. jejuni infection both increased Ki67+ proliferating cells and reduced terminally differentiated goblet cells, we hypothesized that these conditions disrupt physiological epithelial hypoxia and lead to elevated epithelial oxygenation during acute infection. To test this, we used pimonidazole staining, a well-established in vivo hypoxia probe. Under hypoxic conditions, pimonidazole is reduced to reactive intermediates that form stable adducts with proteins and DNA, allowing hypoxic regions to be detected by immunohistochemistry using anti-pimonidazole antibodies. Mice were injected intraperitoneally with pimonidazole one hour before euthanasia. Mice without either DSS treatment or C. jejuni infection exhibited strong pimonidazole retention in colonic epithelial cells, consistent with maintenance of a hypoxic epithelial environment under steady-state conditions. Uninfected, DSS-treated mice showed reduced pimonidazole staining, indicating partial loss of epithelial hypoxia during DSS-induced colitis. C. jejuni-infected, DSS-treated mice showed near-complete loss of pimonidazole signal, consistent with severe disruption of epithelial hypoxia and increased epithelial oxygenation (Fig. 2C). These results demonstrate that DSS-induced colitis partially disrupts epithelial hypoxia, and that C. jejuni infection further exacerbates this loss, consistent with our previous observations linking C. jejuni infection to increased epithelial oxygenation (3). We observed that C. jejuni can reproducibly colonize and grow in the murine intestinal tract after DSS treatment. As a proof of principle to determine whether this DSS inflammation model can be used to identify bacterial fitness factors, we performed an in vivo competition assay using the wild-type strain and a mutant deficient for the genes (motAB) encoding the established fitness determinant motAB, which regulates motility and is critical for colonization in other models (17). Following DSS treatment, mice were orally inoculated with a 1:1 mixture of wild-type C. jejuni 11168 and a motAB::kan derivative at a total dose of 108 CFU. Colonization was assessed in the colon at day three postinfection. In this competitive setting, the motAB::kan mutant exhibited a dramatic colonization defect compared with the wild-type strain, demonstrating that motility provides a strong fitness advantage in the inflamed gut environment (Fig. 2D). These data suggest that DSS-treated mice can serve as a suitable model for identifying and studying fitness factors required for C. jejuni colonization. Overall, our findings demonstrate that DSS treatment disrupts intestinal homeostasis by inducing epithelial injury and mucosal inflammation, overcoming colonization resistance and creating a permissive niche for C. jejuni colonization and growth. This model shows that transient inflammatory perturbation alone is sufficient to render conventional mice susceptible to robust C. jejuni growth without the need for prior antibiotic treatment or genetic manipulation of the host. We also observed that C. jejuni further exacerbates intestinal inflammation following DSS exposure, amplifying mucosal pathogenicity and pro-inflammatory responses. Composition of the DSS-treated microbiota and its associated metabolites promote C. jejuni colonization The resident microbiota of conventional mice is a key determinant of colonization resistance against C. jejuni (18), which readily colonizes germ-free and antibiotic-treated mice (18, 19). Given this established barrier function of the microbiota, we investigated whether DSS-induced inflammation alters the intestinal microbial community in a manner that could reduce colonization resistance and facilitate C. jejuni growth. To assess whether DSS treatment changes the overall microbial community structure and whether C. jejuni colonization further modifies this environment, we performed 16S rRNA gene sequencing of the V3-V4 region from colonic contents collected at day three postinfection. Three experimental groups were analyzed: DSS-treated C. jejuni–infected mice (n = 5), DSS-treated uninfected mice (n = 5), and DSS-untreated uninfected controls (n = 3). We first evaluated within-sample microbial diversity using alpha diversity metrics, including observed richness, Shannon diversity index, and Chao1. Across all three measures, we observed no significant changes (fig. S2A), indicating that DSS treatment and C. jejuni colonization do not substantially alter overall microbial richness or evenness at day three. Suggesting that the ability of DSS-treated mice to support C. jejuni growth is not driven by a global loss of microbial diversity. Analysis of beta diversity using Bray–Curtis dissimilarity, which accounts for relative taxon abundance, and Jaccard distance, which is based on presence or absence of taxons, revealed significant differences in microbial community composition among experimental groups. Principal coordinate analysis (PCoA) derived from both metrics demonstrated clear separation of DSS-treated mice, whether infected or uninfected, from untreated controls (Fig. 3, A and B and fig. S2, B and C). This indicates that DSS treatment is the primary driver of microbiota restructuring, whereas C. jejuni colonization does not significantly alter overall community composition beyond DSS-induced changes. Microbial and metabolic alterations during Campylobacter jejuni infection in DSS-treated mice. Microbial composition of colon contents from control, DSS-treated uninfected, and C. jejuni-infected mice was analyzed by 16S rRNA gene sequencing. (A) Principal coordinate analysis (PCoA) based on Bray–Curtis dissimilarity showing distinct clustering of uninfected controls (n = 3) compared with DSS-treated uninfected (n = 5) and infected groups (n = 5). (B) Boxplots of Bray–Curtis distances demonstrating significant differences in microbial community structure between uninfected controls and DSS-treated uninfected and infected groups. Statistical significance was determined by PERMANOVA (999 permutations) with pairwise post hoc testing. (C) Differential abundance analysis of bacterial genera in DSS-treated groups relative to uninfected controls. Log fold change (LFC) analysis highlights genera significantly enriched (red bars) or depleted (purple bars) in DSS-treated infected groups compared with controls. (D) Concentrations of indole and tryptophan in colon contents from control (n = 4), DSS-treated uninfected (n = 5), and infected mice (n = 6) were determined by mass spectrometry on day 3 postinfection. Statistical analysis was performed using one-way ANOVA. (D) Relative fold changes in AhR gene expression were determined by qRT-PCR analysis of colon tissues from DSS-treated uninfected and C. jejuni-infected mice compared with PBS controls (n = 5). Gene expression changes were calculated using the 2−ΔΔCT method. Data are presented as mean ± SD. Statistical analysis was performed using one-way ANOVA unless otherwise indicated. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.000. Differential abundance analysis (ANCOM-BC) using untreated controls as the reference revealed pronounced microbiota changes following DSS treatment, again indicating that DSS-induced dysbiosis is the primary driver of microbial remodeling in this model. In DSS-treated mice, Akkermansia and Meridibacter were significantly enriched, likely accounting for increased mucin degradation and altered gut barrier interactions (fig. S2D) (20, 21). DSS-treated C. jejuni infected groups showed a similar pattern, with enrichment of Akkermansia and Meridibacter and deeper depletion of SCFA producers. Although most of the changes we identified were correlated to DSS treatment, we note that Ruminiclostridium_E and members of the Muribaculaceae family were significantly enriched only in the DSS-treated infection group, suggesting infection-specific taxonomic shifts (Fig. 3C) (22, 23). The DSS-treated groups were depleted for a broad set of obligate anaerobic commensals, including Anaerotruncus, Roseburia_B, Lachnoclostridium_B, multiple Eubacterium clades, Peptoclostridium_A, Faecenecus, and members of the Oscillospiraceae family (all log-fold change < −2, with false discovery rates of <0.05) (fig. S2D and Fig. 3C). Many of these taxa are major producers of short-chain fatty acids (SCFA), particularly butyrate, essential for maintaining epithelial barrier integrity, mucus production, and immunoregulatory signaling (23, 24). In addition to SCFA production, several Clostridia-associated taxa, including Anaerotruncus, Oscillibacter, Eubacterium, and Lachnoclostridium, participate in amino acid fermentation pathways that generate indole and related metabolites (Fig. 3C) (25, 26). Certain anaerobes, including Peptostreptococcus species, convert dietary tryptophan into bioactive indole derivatives such as indoleacrylic acid (IA) and indole-3-propionic acid (IPA) (27). The positive association between Eubacterium abundance and levels of indole metabolites further underscores the contribution of these taxa to microbial tryptophan metabolism (25). Indole and its derivatives support gut homeostasis by strengthening epithelial barrier function and regulating mucosal immunity as ligands of the Aryl hydrocarbon Receptor (AhR), which mediates anti-inflammatory signaling (28). We conclude that DSS treatment induced a dysbiotic microbiota marked by expanding populations of mucin-degrading organisms and depletion of SCFA- and indole-producing obligate anaerobes, thereby creating a gut environment permissive for C. jejuni colonization, with reduced microbial support for intestinal homeostasis, while infection primarily amplified changes in select taxa rather than generating a distinct community configuration. Because several indole-producing bacterial taxa were significantly reduced by DSS in both infected and uninfected groups, we hypothesized that luminal indole levels would be correspondingly diminished in these mice. To test this, we quantified indole and L-tryptophan concentrations in colonic samples by LC/MS mass spectrometry, as gut microbiota convert L-tryptophan into indole and related derivatives through tryptophan metabolism. Consistent with the loss of indole-producing taxa, indole levels were significantly decreased in both DSS-treated uninfected and infected groups compared with untreated controls, whereas tryptophan levels showed a relative accumulation, suggesting impaired microbial conversion of tryptophan to indole (Fig. 3, D and E). The aryl hydrocarbon receptor (AhR) is a key transcription factor that regulates epithelial barrier integrity, IL-22, IL-10 expression, and mucosal immune homeostasis (29–31). Given reduced levels of the AhR ligand indole, we assessed AhR transcriptional expression in colonic tissues by RT PCR. Ahr mRNA levels were significantly reduced in both DSS-treated groups, with an approximately four-fold reduction in DSS-treated uninfected mice and an approximately eight-fold reduction in DSS-treated infected mice compared with untreated controls (Fig. 3F). Overall, DSS treatment reshaped the gut microbiota, resulting in depletion of SCFA- and indole-producing taxa, consistent with reduced butyrate availability and increased epithelial oxygenation in both DSS-treated groups. Indole levels were significantly decreased, accompanied by reduced expression of AhR and IL-10. Because AhR activation promotes IL-10–mediated mucosal tolerance (31), and IL-10 impacts C. jejuni colonization, diminished indole availability may facilitate C. jejuni pathogenicity through impaired immune regulation. It may also result in loss of direct antimicrobial pressure, as indole exhibits antibacterial and virulence activity against multiple enteric pathogens (32–34). Microbiota derived indole inhibits C. jejuni growth in vitro and in vivo Indole is microbiota-derived tryptophan metabolite that accumulates in the colon at physiological concentrations (∼0.25–1 mM) (32). At these levels, it supports intestinal homeostasis by strengthening epithelial barrier function as a ligand of AhR, regulating mucosal immunity, and exerting antimicrobial activity by disrupting membrane potential and respiration (32, 35). Given that indole levels were markedly reduced following DSS treatment and during C. jejuni infection, we investigated whether indole directly influences C. jejuni growth, as its effect on this pathogen remains poorly defined. For this, wild-type C. jejuni was cultured in Mueller–Hinton broth supplemented with physiologically relevant concentrations of indole (0.25–1 mM), with DMSO serving as the vehicle control, and bacterial growth was assessed by CFU enumeration over a 30-hour time course. There was little effect on C. jejuni growth in 0.25 mM indole, while significant growth reduction was detected as early as six hours in 1 mM indole, and by nine hours in 0.5 mM indole (Fig. 4A). In these latter two concentrations, bacterial viability progressively declined so that by 30 hour of cultivation in 1 mM indole C. jejuni exhibited near-total loss of viability, an observation that was not limited to strain 11168 but was also observed in three additional C. jejuni isolates from different sources tested under the same conditions (fig. S3A). These findings demonstrate that indole at physiological concentrations inhibits C. jejuni growth in vitro. Indole inhibits Campylobacter jejuni growth by suppressing respiratory and metabolic pathways. (A) C. jejuni growth was determined by colony-forming unit (CFU) counts at different time intervals up to 30 h in the presence or absence of indole at physiological gut concentrations (0.25–1 mM) under in vitro conditions. (B) Transcriptomic analysis of C. jejuni gene expression after 9 h of growth in the presence of indole (0.5 mM) or vehicle control (DMSO) was performed by RNA-seq (n = 3). A volcano plot shows significantly upregulated and downregulated genes (>2-fold change) in response to indole treatment. (C to D) Representative pathway diagrams highlighting significantly downregulated respiratory and metabolic genes of C. jejuni (shown in red) in response to indole treatment. (E) Quantitative RT-PCR analysis of genes involved in respiration and metabolism, including nitrate reductase (napA), cytochrome c oxidase (ccoN), lactate transporter (lctP), and acetogenesis pathway genes (ackA and pta), following 9 h growth in the presence or absence of indole (0.5 mM) under anaerobic and microaerobic conditions. Gene expression changes were calculated using the method. (F) Competitive infection analysis between the WT strain and metabolic mutants (napA::cm, lctP::kan, and ackA::pta::cm). DSS-treated mice were infected with an equal ratio (1:1) of WT and mutant strains. On day 3 postinfection, the ratio of mutant to WT bacteria was determined and expressed as the competitive index (CI). Statistical analysis was performed using a one-sample t-test against a hypothetical value of 1 (n = 5 per group). Data are presented as mean ± SD from three independent experiments unless otherwise indicated. Statistical analysis was performed using one-way ANOVA or an unpaired two-tailed t test as appropriate. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.000. To investigate the molecular mechanisms underlying indole-mediated growth inhibition, we performed transcriptomic analysis using RNA sequencing (RNA-seq). Based on our growth kinetics data, we selected the nine-hour time point for analysis, as a significant growth defect was first observed at this stage in the presence of 0.5 mM indole, without loss of viability (Fig. 4A). Wild-type C. jejuni was cultured in Mueller–Hinton broth with 0.5 mM indole or DMSO control, and total RNA was isolated at nine hours for transcriptomic profiling. We identified 231 transcripts with lower abundance and 192 genes with greater abundance in the presence of indole (log 2 > 1, P < 0.05) (Fig. 4B). Several respiratory regulatory gene mRNAs were reduced by more than twofold, including those involved in aerobic respiration (e.g., ccoNPOQ, encoding a terminal oxidase), nitrate respiration (napAGHBLD), hydrogenase (hydAB), and formate hydrogenase (fdhA and fdhB) (Fig. 4C and table S1). C. jejuni lacks a complete glycolytic pathway but possesses a complete citric acid (TCA) cycle, which serves as its primary metabolic pathway (36). Our data show that transcripts associated with TCA cycle intermediates (pycA, gltA, acnB, oorC, fumC, mqo, sucC, frdC, and frdB) were significantly less abundant (> 2 fold) in the presence of indole (Fig. 4D and table S1). The acetate switch in C. jejuni, which converts acetyl-CoA to acetate and represents a major ATP-generating pathway, was also affected: transcripts from both pta and ackA were less abundant by more than two- to four-fold in the presence of indole (37). In addition, those of the lactate transporter lctP and the lutABC lactate oxidoreductase complex, which converts lactate to pyruvate, were reduced by three- to six-fold (Fig. 4D and table S1). Lactate serves as a major carbon source supporting C. jejuni growth during inflammation in the ferret (3). To validate the RNA-seq results, we measured napA, ccoN, ackA, pta, and lctP expression under the same conditions by RT-PCR. We observed an approximately sixfold decrease in napA, a fourfold reduction in ccoN, and nearly eightfold decreases in both ackA and pta in the presence of indole, confirming the RNASeq findings (Fig. 4E). Collectively, these data indicate that indole exposure induces a coordinated reduction of respiratory capacity and central metabolic pathways, providing a mechanistic basis for the observed growth inhibition of C. jejuni. To validate the transcriptomic findings further, we assessed bacterial respiration using an assay that measures reduction of 2,3,5-triphenyl-tetrazolium chloride (TTC). Wild-type C. jejuni was grown for nine hours in the presence of indole (0.5 mM) or DMSO. Cells were washed with PBS, normalized to OD600 = 1 to ensure equal bacterial numbers, and TTC reduction was quantified by measuring optical density at 496 nm. Treatment with 0.5 mM indole resulted in approximately 10-fold reduction in TTC reduction relative to DMSO-treated bacteria, indicating marked impairment of respiratory activity (fig. S3B). These data support the transcriptomic findings and demonstrate that indole inhibits C. jejuni growth, at least in part, by suppressing cellular respiration. In addition to its effects on respiratory gene transcript abundance, indole resulted lower abundance of transcripts from genes of the TCA cycle genes and the ackA-pta acetate switch pathway, which are important for ATP generation and metabolic flux in C. jejuni. To determine whether indole affects cellular energy status, we measured intracellular ATP levels. Wild type bacteria were grown for nine hours in different concentrations of indole, normalized to OD600 = 1, and intracellular ATP was quantified using a commercial luminescence-based assay. Consistent with impaired respiration, 0.5 mM indole caused an approximately three-fold reduction in intracellular ATP levels compared with controls (fig. S3C), consistent with previous findings showing that indole reduces intracellular ATP levels in Pseudomonas putida (38). This suggests a metabolic mechanism underlying indole-mediated growth inhibition. In our transcriptomic analysis, abundance of mRNA from several oxidative stress regulatory genes was significantly elevated in the presence of indole, including perR (peroxide-sensing regulator), katA (catalase), aphC (alkyl hydroperoxide reductase), and sodB (Superoxide dismutase B), showing approximately two- to nine-fold increase in abundance. In addition, transcripts from iron-regulated genes, cj0175c (cfbpA), cj0174c (cfbpB), cj1659 (p19), and the associated ABC transporter genes cj1660–cj1663, and the inorganic phosphate transporter operon pstSCAB, were elevated by roughly six- to eight-fold following indole exposure. These genes are induced under acidic conditions in C. jejuni. Furthermore, transcripts from cj0414 and cj0415, which encode gluconate dehydrogenase subunits, exhibited a six- to sevenfold increase in expression in the presence of indole. These genes promote C. jejuni survival under acidic conditions (table S2). Consistent with these observations, our transcriptomic data reveal greatly increased abundance of mRNA from acid-adaptation and stress-response genes in C. jejuni following indole exposure, suggesting induction of a response associated with intracellular acidification. We therefore hypothesize that indole inhibits C. jejuni growth by reducing cytoplasmic pH. To test this, wild-type C. jejuni was grown for nine hours in the presence or absence of indole, washed with PBS, and normalized to OD600 = 1. Cells were stained with the pH-sensitive fluorescent dye CMFDA, and fluorescence emission was measured at 520 nm with dual excitation at 450 nm and 490 nm. Intracellular pH was calculated using a standard calibration curve generated under defined pH conditions. Indole treatment resulted in a significant reduction in intracellular cytoplasmic pH compared with untreated controls. While control cells-maintained a near-neutral cytoplasmic pH approximately 7.2, indole-exposed bacteria exhibited a decreased intracellular pH of approximately 6 (fig. S3D). Thus, indole disrupts intracellular pH homeostasis in C. jejuni, providing an additional mechanistic explanation for indole-mediated growth inhibition, likely through metabolic and respiratory stress associated with cytoplasmic acidification. To test the requirement for lactate-dependent growth of C. jejuni in the DSS-treated mouse, we performed a competitive colonization assay in DSS-treated mice using a 1:1 mixture (108 CFU total) of wild-type and lctP::kan mutant strain. Colonization was quantified in colonic tissue at day three postinfection. The lctP::kan mutant exhibited greater than 100-fold fitness defect relative to the wild-type strain, demonstrating a marked competitive disadvantage in vivo (Fig. 4F). These findings indicate that lactate uptake via LctP is critical for C. jejuni fitness during DSS-induced colitis. Together, our data support a model in which inflammation-associated oxygenation and lactate accumulation create a metabolic niche that favors C. jejuni growth through oxygen-enhanced lactate metabolism, while indole acts to suppress this pathway. Nitrate respiration, fueled by host-derived nitrate produced through iNOS-dependent nitric oxide synthesis, provides a growth advantage to several enteric pathogens in the inflamed gut. Unlike E. coli and Salmonella enterica serovar Typhimurium, which encode multiple nitrate reductases (12, 39), C. jejuni possesses a single periplasmic nitrate reductase encoded by the napAGHBLD operon (39), and indole negatively impacts transcript abundance from this operon. Nitrate respiration in C. jejuni is required for efficient commensal colonization in the chicken intestine, while the contribution of nitrate respiration to C. jejuni fitness during inflammatory infection remains poorly defined (40). We generated a nitrate reductase–deficient mutant (napA::cm), which is impaired in nitrate respiration, and performed a competitive colonization assay in DSS-treated mice using a 1:1 mixture (108 CFU total) of wild-type and napA::cm bacteria. Bacterial burdens were quantified in colonic tissue at day three postinfection. The napA::cm mutant exhibited an approximately 100-fold fitness defect relative to the wild-type strain, indicating a marked competitive disadvantage in vivo (Fig. 4F). We observed a ∼ five-fold increase in colonic iNOS transcription following DSS treatment and an approximately 10-fold increase during infection (fig. S3E), with elevated expression of iNOS detected with immunohistochemistry; this is observed mainly in colonic epithelial cells in DSS uninfected and infected tissue (fig. S3F). These findings suggest that inflammation enhances the availability of alternative electron acceptors in the inflamed gut and that C. jejuni relies on nitrate respiration to maximize fitness in this environment. We also investigated the ackA/pta pathway of ATP generation and its role in C. jejuni growth in the inflamed gut., as its requirement for commensal, non-inflammatory colonization of poultry has already been demonstrated (37). After inoculating DSS-treated mice with 1:1 mixture (108 CFU total) of wild-type and ackA/pta::cm bacteria, the mutant demonstrated approximately 100-fold competitive defect after three days (Fig. 4F). Given our in vitro finding that indole inhibits C. jejuni growth, and that several genes whose expression is negatively affected by indole are required for fitness in the inflamed gut, we next evaluated whether indole administration could limit bacterial colonization in vivo. DSS-treated mice were orally infected with C. jejuni (108 CFU) as described above and, following infection, one group received oral indole (150 mg/kg of body weight, prepared in DMSO) once each of the two days postinfection, while the control group received DMSO alone (Fig. 5A). Mice treated with indole exhibited an approximately 1000-fold reduction in C. jejuni colonization compared with control mice treated with DMSO alone (Fig. 5B). Indole treatment did not significantly alter histological scores or inflammatory parameters compared with DMSO-treated infected mice. (Fig. 5C). These data indicate that short-term indole treatment does not alter the overall inflammatory gut pathophysiology that favors C. jejuni growth. Rather, the findings suggest that indole directly inhibits bacterial growth without substantially affecting intestinal inflammation. Indole treatment reduces Campylobacter jejuni colonization in vivo. (A) Schematic diagram of the experimental design for indole or vehicle (control) treatment following C. jejuni infection. (B) Comparative colonization of C. jejuni in colon tissues between indole-treated and vehicle-treated groups on day 3 postinfection, determined by colony-forming unit (CFU) counts (n = 5). (C) Colonic inflammatory responses were evaluated by histological analysis and histological scoring of hematoxylin and eosin (H&E)-stained sections on day 3 postinfection. Representative images (H&E) and histological scores are shown. Data are presented as mean ± SE. Statistical analysis was performed using one-way ANOVA. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.000; ns, not significant. Probiotic indole production inhibits C. jejuni growth in vivo Among other bacteria, the probiotic Escherichia coli strain Nissle 1917 (EcN) produces indole from L-tryptophan via tryptophanase A (tnaA) (41). To examine the impact of bacterial indole production on C. jejuni fitness, we grew co-cultures of C. jejuni 11168 and either wild type or tnaA mutant EcN in Mueller–Hinton broth (MHB) supplemented with L-tryptophan (2 mM). After 24 hours, C. jejuni CFUs were quantified, and indole concentrations in culture supernatants were measured using the Kovac assay. Co-culture with wild-type EcN in the presence of tryptophan resulted in significant suppression of C. jejuni growth, coinciding with indole accumulation (∼1 mM) in the supernatant (Fig. 6A). In contrast, co-culture with the tnaA::cm mutant did not impair C. jejuni growth, and indole was undetectable in the supernatant. Growth of C. jejuni alone in tryptophan-containing medium was unaffected, confirming that growth inhibition was dependent on EcN-derived indole production rather than tryptophan supplementation itself (Fig. 6B). These findings demonstrate that microbially-derived indole production is sufficient to suppress C. jejuni growth in vitro. Indole produced by probiotic E. coli Nissle 1917 (EcN) inhibits Campylobacter jejuni colonization in vitro and in vivo. (A) Indole production and (B) C. jejuni growth were measured after 24 h in Mueller–Hinton broth (MHB) supplemented with and without tryptophan (2 mM). Cultures included C. jejuni alone, C. jejuni co-cultured with WT EcN, and C. jejuni co-cultured with the EcN tnaA::cm mutant. Indole levels were quantified, and C. jejuni growth was determined by colony-forming unit (CFU) counts. (C) Comparative colonization of C. jejuni in colon tissues was determined by CFU counts among mice infected with C. jejuni alone, co-infected with C. jejuni and WT EcN, or co-infected with C. jejuni and the EcN tnaA::cm mutant. (D) Colonic inflammatory responses were evaluated by histological analysis and histological scoring of hematoxylin and eosin (H&E)-stained sections on day 3 postinfection among three groups: control, C. jejuni + EcN, and C. jejuni + EcN tnaA::cm. Representative images (H&E) and histological scores are shown. Error bars represent SD. Statistical analysis was done by unpaired two-tailed t test. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.000. We also investigated whether indole-producing EcN could suppress C. jejuni colonization during infection. DSS-treated mice were orally infected with C. jejuni alone or co-infected at a 1:1 mixture (108 CFU total) with either wild-type EcN or the EcN tnaA::cm mutant incapable of producing indole. Colonic colonization was assessed at day three postinfection. Co-infection with wild-type EcN resulted in an approximately 100-fold reduction in C. jejuni colonization compared with co-infection with the tnaA::cm mutant. In contrast, C. jejuni grew to high levels when co-infected with the EcN tnaA::cm strain which cannot produce indole. Infection with C. jejuni alone yielded colonization levels comparable to those observed in the tnaA::cm co-infection group (Fig. 6C). We also measured colonization levels of EcN and the EcN tnaA::cm mutant in the colon within their respective infection groups and found no significant difference in colonization between EcN and EcN tnaA::cm at day three of post infection (fig. S4). Histological scores and other inflammatory parameters were not altered across the three infected groups suggesting that the impact of indole production was on the C. jejuni growth per se and not on the host (Fig. 6D). Our findings support the hypothesis that metabolic competition mediated by probiotic derived indole is a mechanism of colonization resistance during intestinal inflammation. DISCUSSION The gastrointestinal (GI) tract is densely populated by microbial communities that maintain colonization resistance and protect the host from enteric pathogens (42). Resident microbiota restrict pathogen growth by limiting nutrient availability and maintaining a hypoxic intestinal environment, partly through microbiota-derived short-chain fatty acids such as butyrate, metabolism of which preserves epithelial hypoxia (10). Pathogenic bacteria disrupt this barrier by inducing intestinal inflammation, creating conditions favorable for their growth. Inflammation remodels the intestinal ecosystem by altering epithelial metabolism, oxygen availability, nutrient flux, and microbial composition, a mechanism well described for facultative anaerobic Enterobacteriaceae such as S. Typhimurium and E. coli, which exploit inflammation-associated electron acceptors and increased oxygen to outcompete obligate anaerobic commensals (10). In previous work using a ferret infection model, we showed that C. jejuni infection induces broad host pathological changes in the colon, including inflammation and crypt hyperplasia associated with activation of tissue repair responses and accumulation of undifferentiated proliferative Ki67+ epithelial cells (3). These infection-induced changes support C. jejuni growth by increasing epithelial oxygenation and shifting host metabolism toward elevated lactate production, which the pathogen uses as a growth substrate (3). In the current study, we modeled a similar inflammatory and pathological intestinal environment in mice using DSS-induced colitis. We demonstrate that these inflammation-associated host changes are sufficient to overcome colonization resistance and support robust C. jejuni growth in conventional mice without the need for antibiotic treatment or host genetic modification. This outcome is consistent with findings from the widely used IL-10–deficient mouse model for studying C. jejuni pathogenesis. In this model, constitutive intestinal inflammation, together with the accumulation of undifferentiated proliferative Ki67+ cells in the colon, is associated with increased susceptibility to C. jejuni colonization (43). Together, these models provide further evidence that inflammation-associated host pathological changes are key determinants of C. jejuni fitness and disease progression. Inflammation not only facilitates bacterial growth but is also intensified by infection, establishing a feed-forward interaction between host inflammatory responses and pathogen fitness. These findings may help explain clinical observations that Campylobacter species are frequently isolated from patients with active colitis, and acute enteric infection is known to trigger disease flares in individuals with inflammatory bowel disease (IBD) (44). Additionally, epidemiological studies have reported that acute Campylobacter infection is associated with an increased long-term risk of developing IBD (45). Compared with the widely used genetically modified IL-10–deficient mouse model, which often requires antibiotic treatment to achieve high C. jejuni colonization and inflammatory responses within 2–5 days of infection, our DSS-treated mouse model provides a simple, reproducible, and cost-effective alternative for studying C. jejuni pathogenicity without the need for antibiotic treatment or host genetic manipulation (46). This model is well suited for identifying bacterial physiological traits and fitness determinants that are specifically required under inflammatory conditions, as well as host factors that contribute to C. jejuni pathogenesis in the gut. In patients with inflammatory bowel disease, particularly ulcerative colitis, epithelial metabolic changes like those observed in DSS-induced colitis have been reported (7). DSS treatment induces ER stress–mediated epithelial injury, leading to enhanced repair responses characterized by crypt elongation, proliferation of transit-amplifying cells, and reduced numbers of terminally differentiated cells such as goblet cells, resulting in thinning of the mucus layer (47). These histological changes were also observed in our model and were exacerbated following C jejuni infection. Terminally differentiated colonocytes typically rely on β-oxidation of fatty acids, consuming high levels of oxygen, whereas proliferating epithelial cells preferentially use glycolysis, producing lactate with lower oxygen consumption (10). Consistent with previous reports showing elevated luminal lactate during colitis, this metabolic shift reduces epithelial oxygen use and increases oxygen and lactate availability at the mucosal surface (8, 9). In our model, epithelial oxygenation increased following DSS treatment and was further elevated after C. jejuni infection, suggesting increased oxygen availability in the gut lumen. In addition, the lactate transporter mutant (lctP) exhibited a colonization defect, indicating that L-lactate serves as a major carbon source for C. jejuni. As an obligate microaerophile, C. jejuni requires low oxygen levels, with a partial oxygen tension (pO2) of 2–10%, for growth (1). Oxygen also functions as an electron acceptor that supports ATP production, growth, and motility in C. jejuni, which encodes the ccoNPOQ terminal oxidase (1). In our previous study, we showed that oxygen promotes expression of L-lactate uptake genes by C. jejuni and that the redox regulator LctR regulates lctP expression in the presence and absence of oxygen (3). During inflammatory responses to enteric pathogens, epithelial oxygenation increases to approximately 3–10%, creating a microaerobic niche near epithelial cells (10, 48). Together, our findings suggest that inflammation-associated epithelial oxygenation creates a favorable microaerobic niche while simultaneously increasing lactate availability. Under these conditions, C. jejuni uses oxygen as an electron acceptor through aerobic respiration and utilizes L-lactate as a carbon source to support its growth in the inflamed intestine. Host-derived nitrate serves as an alternative electron acceptor for facultative anaerobic bacteria such as E. coli and S. enterica Typhimurium during intestinal inflammation (10, 12, 39). During colitis, the host inflammatory response generates reactive nitrogen and oxygen species, and inducible nitric oxide synthase (iNOS) is elevated, leading to increased nitric oxide production in the gut. Nitric oxide reacts with superoxide to form peroxynitrite, which can subsequently generate nitrate, thereby increasing luminal nitrate availability (10, 12). We detected elevated iNOS expression in both DSS-treated and C. jejuni–infected colonic tissue, suggesting increased nitrate levels in the inflamed intestine. Additionally, the nitrate reductase mutant showed a competitive colonization defect, indicating that nitrate respiration contributes to C. jejuni fitness in the inflamed gut. Overall, our study suggests that luminal lactate availability, epithelial oxygenation, and nitrate levels are key factors supporting bacterial growth in the inflamed gut. Diet can strongly influence these parameters; for example, long-term high-fat diets increase epithelial oxygenation and nitrate availability, thereby promoting growth of Enterobacteriaceae such as E. coli and S Typhimurium (49). However, little is known about how dietary factors affect the pathogenesis of C. jejuni. Our findings suggest that diet-driven changes in gut metabolism and redox conditions may influence C. jejuni colonization and disease progression, highlighting an important area for future investigation. We also identified luminal indole as important in this model. DSS treatment and C. jejuni infection significantly reduced luminal indole levels, correlating with decreased abundance of indole-producing members of the microbiota, consistent with reports showing reduced indole concentrations during intestinal inflammation and in patients with inflammatory bowel disease (50). Indole treatment inhibited C. jejuni growth both in vitro and in vivo and although indole has anti-inflammatory effect treatmentwith transcriptomic analysis indicating that indole results in reduced abundance of transcripts for aerobic respiration, lactate utilization, nitrate respiration and acetate switch pathways, all critical determinants of bacterial fitness in the inflamed intestine. Another possible mechanism is that indole treatment may reduce C. jejuni colonization by altering the gut microbiota in vivo. However, a recent study showed that seven days of treatment with the indole derivative indole-3-lactic acid did not alter microbiota composition in C. rodentium-infected DSS-treated mice (51). In our study, we used a short-term, two days indole treatment; therefore, we anticipate that indole is unlikely to have substantially altered the microbiota composition under these conditions. Although we generated inflammation experimentally leading to an altered, less-indole-producing microbiota, presumably C. jejuni must overcome indole-mediated colonization resistance during natural infection to establish disease. Microbiota from humans infected with Campylobacter have similar changes to what we observed with DSS-treatment in the mouse such as lower levels of Firmicutes, which include indole-producing Clostridia species (52). To draw solid conclusions about if and how C. jejuni alters the gut microbiota in a way that reduces indole-producing bacteria or otherwise relieves metabolic inhibition requires further study. In general, understanding how C. jejuni modulates the intestinal metabolic environment to overcome indole-dependent resistance is an important area for future investigation. With increasing emergence of multidrug-resistant C. jejuni strains, alternative therapeutic strategies are needed (2). Our results suggest that a metabolite-based approach to limit infection could be successful. Consistent with this idea, recent studies have reported that indole and synthetic indole-derived compounds exhibit antimicrobial or anti-virulence activity against a broad range of Gram-positive and Gram-negative pathogens, supporting this approach for therapeutic development against multidrug-resistant bacteria (53, 54). Collectively, this study reframes C. jejuni pathogenesis within the ecology of the inflamed gut. Less dependent on virulence determinants such as type III secretion effectors or exotoxins that commandeer host functions, C. jejuni exploits inflammation-induced oxygenation, nitrate availability, and glycolytic products to fuel respiratory metabolism. Concurrent depletion of indole-producing commensals removes a critical metabolic constraint on this growth. By integrating host metabolic remodeling, microbial community restructuring, and pathogen bioenergetics, our study provides a unified model explaining how inflammation converts the colon into a permissive niche for C. jejuni. These findings also highlight microbial metabolite restoration—particularly indole—as a rational therapeutic strategy to limit pathogen growth without broadly suppressing inflammation or disrupting the microbiota. MATERIALS AND METHODS Bacterial strains and media All strains and plasmids used in the study are listed in table S1. All C. jejuni strains were routinely grown under microaerobic conditions (85% N2,10% CO2, 5% O2) on Mueller-Hinton (MH) agar or in broth under necessary antibiotics at the following concentrations: kanamycin, 50 μg ml − 1; trimethoprim (TMP), 5 μg ml − 1. C. jejuni strains were stored in MH broth with 20% glycerol at −80°C. E. coli Nissle 1917 (EcN) were routinely grown under aerobic condition at 37°C in Luria Broth agar (LB). Construction of C. jejuni napA::cm, ackA::pta::cm and E. coli Nissle 1917 tnaA::cm insertion mutant The C. jejuni napA::cm and ackA::pta::cm mutants were constructed by inserting a chloramphenicol resistance cassette into the gene as previously described. Briefly, 500 bp of both upstream and downstream regions of the napA and ackA locus was amplified by PCR with primers containing SmaI restriction sites. The amplified upstream and downstream regions were joined by PCR. The 1,000-bp amplified PCR product was cloned into the plasmid pGEM-T Easy. The Campylobacter chloramphenicol resistance cassette was excised from pRY109 using SmaI and ligated between the upstream and downstream cloned regions of the pGEM-T construct. The plasmid was electroporated into the wild type strain and grown on MH agar overnight at 37°C under microaerobic conditions. Cells were then plated and grown on MH agar containing chloramphenicol for 2 to 3 days, and successful integration of napA::cm and ackA::pta::cm into the chromosome was confirmed by PCR (3). The E. coli Nissle 1917 tnaA::cm mutant was constructed by inserting a chloramphenicol resistance cassette into the open reading frame of tnaA using λ Red recombineering. Deletion–insertion alleles were generated by PCR amplification of the chloramphenicol resistance cassette from pKD3 (42) using primers containing 50-bp homology arms targeting the gene of interest. The first and last three codons of the open reading frame were retained. The resulting PCR products were introduced into EcN harboring pKD46 (43) by electroporation. Following electroporation, mutants were selected on LB agar containing chloramphenicol (15 μg/ml), and the mutation was confirmed by PCR (55). Growth analysis in MH broth C. jejuni wild-type strain was recovered from freezer stocks and cultured on Mueller–Hinton (MH) agar supplemented with appropriate antibiotics for 24 h under microaerobic conditions, followed by restreaking and incubation for an additional 16 h. Bacteria were then resuspended in MH broth and inoculated into 25 mL of Mueller–Hinton broth containing indole (0.25, 0.5, or 1 mM) or vehicle control (DMSO) at an initial OD600 of ∼0.05. Cultures were incubated at 37°C with shaking (100 rpm) under microaerobic conditions. Bacterial growth was monitored over 30 h by enumerating colony-forming units (CFUs) at indicated time points. For coculture assays, E. coli Nissle 1917 (EcN) and C. jejuni wild-type (WT) strains were inoculated individually or together into 25 mL Mueller–Hinton broth (MHB) supplemented with L-tryptophan (2 mM) or vehicle control (DMSO), with an initial OD600 of ∼0.05 for each strain. Cultures were incubated at 37°C with shaking (100 rpm) under microaerobic conditions for 24 h, after which C. jejuni growth was quantified by enumerating colony-forming units (CFUs) on selective Campylobacter agar plates (cefoperazone (40 μg/ml), cycloheximide (100 μg/ml), trimethoprim (5 μg/ml), and vancomycin (100 μg/ml). Indole concentrations in culture supernatants were measured using a modified Kovács assay (56). Briefly, cultures were centrifuged at 15,000 rpm for 15 min, and 100 μL of supernatant was incubated in triplicate with 150 μL Kovács reagent (Sigma-Aldrich) for up to 30 min at room temperature. The resulting chromophore was quantified spectrophotometrically at 530 nm, and indole concentrations were determined using standard curves prepared in MHB. Mouse infections All animal experiments were approved by the Michigan State University Institutional Animal Care and Use Committee (IACUC). Approval No. PROTO202400286). i) Specific pathogen-free male and female C57BL/6 J mice (7–8 weeks old; The Jackson Laboratory) were randomly assigned to four groups (n = 5 per group): (i) no DSS, no C. jejuni infection; (ii) no DSS, C. jejuni infection; (iii) DSS treatment only; and (iv) DSS treatment with C. jejuni infection. For DSS treatment, mice received 3% dextran sodium sulfate (DSS, MP Biomedicals Colitis grade (MW36,000–50,000 Da) in drinking water for 3 days. On day 4, mice were deprived of food and water for 2–3 h prior to infection and then orally gavaged with C. jejuni wild-type strain (108 CFU) in 5% sodium bicarbonate to neutralize gastric acid; DSS was subsequently replaced with regular drinking water. Control mice received PBS with sodium bicarbonate. At 1 and 3 days postinfection, mice were euthanized, and fecal samples and tissues (cecum, small intestine, and proximal and distal colon) were collected to assess bacterial burden. Samples were homogenized in sterile phosphate-buffered saline (PBS), serially diluted, and plated on selective Campylobacter agar plates. Plates were incubated for 48 h under microaerophilic conditions, and colonies were enumerated (3). ii) For in vivo competition assays, DSS-treated female mice were orally infected with a 1:1 mixture (total 108 CFU) of C. jejuni wild-type and mutant strains (lctP::kan, napA::cm, or ackA::pta::cm). At 3 days postinfection, colonic tissues were collected, homogenized, and serially diluted in sterile PBS. Samples were plated on Campylobacter-selective agar with or without selective antibiotics (kanamycin, 50 μg/mL; chloramphenicol, 15 μg/mL) to differentiate strains. Competitive indices were calculated as previously described. iii) For indole treatment on C. jejuni colonization in vivo, female mice were assigned to three groups: (i) no DSS, no infection; (ii) DSS treatment with C. jejuni infection plus indole; and (iii) DSS treatment with C. jejuni infection plus vehicle control. Indole (Sigma-Aldrich) was administered by oral gavage twice daily on days 2 and 3 postinfection at a total dose of 150 mg·kg−1·day−1 in a vehicle consisting of DMSO/PEG400/5% citric acid (1:4.5:4.5) (28). At 3 days postinfection, colonic tissues were collected, homogenized in sterile PBS, and serially diluted. Bacterial burden was determined by plating on Campylobacter-selective Mueller–Hinton agar. iv) For invivo coinfection assays, DSS-treated female mice were orally inoculated with 1 × 108 CFU of a 1:1 mixture of C. jejuni and E. coli Nissle 1917 or its tnaA::cm mutant. At 3 days postinfection, colonic tissues were collected, homogenized in sterile PBS, and serially diluted. C. jejuni colonization levels were determined by plating on Campylobacter-selective Mueller–Hinton agar. Histology and Immunohistochemistry Colonic tissues were fixed in 10% neutral-buffered formalin and transferred to 60% ethanol prior to processing. Samples were sectioned at 5 μm, mounted on glass slides, and stained with hematoxylin and eosin (H&E) and Alcian blue to assess goblet cells. Histopathological evaluation was performed in a blinded manner using a semiquantitative scoring system based on epithelial damage, inflammatory cell infiltration, goblet cell depletion, crypt hyperplasia, and crypt abscess formation (0, none; 1, minimal; 2, mild; 3, moderate; 4, marked; 5, severe). Localization of C. jejuni in colonic tissues was assessed by immunohistochemistry using a polyclonal anti-C. jejuni primary antibody. Cell proliferation and inflammatory responses were evaluated by staining for Ki67+ and NOS2 using specific anti-mouse primary antibodies. All histological and immunohistochemical analyses were performed at the Investigative HistoPathology Laboratory, Department of Physiology, Michigan State University. Hypoxia staining Tissue hypoxia was assessed using pimonidazole labeling (8). Mice were administered pimonidazole HCl (60 mg/kg, i.p.; Hypoxyprobe-1 kit, Hypoxyprobe) 1 h prior to euthanasia. Colonic tissues were fixed in 10% phosphate-buffered formalin, paraffin embedded and processed for immunofluorescence staining. Sections were blocked using a mouse-on-mouse blocking reagent (Vector Laboratories) and incubated with a mouse anti-pimonidazole monoclonal antibody (MAb 4.3.11.3). Nuclei were counterstained with DAPI using SlowFade Gold mounting medium. Hypoxia was scored in a blinded manner based on epithelial staining intensity and distribution (0, none; 1, mild focal; 2, moderate multifocal; 3, intense diffuse) as previously described (8). Representative images were acquired using a microscope, and image brightness was adjusted uniformly across samples. RNA-seq and Real-time PCR C. jejuni strain 11168 was cultured in Mueller–Hinton broth supplemented with 0.5 mM indole or vehicle control (DMSO) for 9 h. Total RNA was extracted using TRIzol reagent followed by purification with the RNeasy Mini Kit (Qiagen) (57, 58). Residual genomic DNA was removed by TURBO DNase treatment (Invitrogen). Libraries were sequenced at SeqCenter, Pittsburgh, USA, generating 12 million paired-end reads (2 × 150 bp) per sample (n = 3). The data were analyzed using Proseq v2.0 software. Detailed analysis methods are available at the project repository: https://github.com/snandiDS/prokseq-v2.0. For RT-PCR analysis, first-strand cDNA was synthesized using SuperScript III First-Strand Synthesis SuperMix (Invitrogen). Quantitative real-time PCR was performed using SYBR Green chemistry, and gene expression levels were calculated using the 2−ΔΔCt method as previously described (58). Primer sequences for all genes analyzed are provided in table S2. Proinflammatory and anti-inflammatory cytokine expression in colonic tissues from DSS-treated, infected, and control mice was assessed as previously described. Flash-frozen tissues were homogenized in TRIzol reagent, and total RNA was extracted followed by purification using the RNeasy Mini Kit (Qiagen). Residual genomic DNA was removed by TURBO DNase treatment (Invitrogen). First-strand cDNA was synthesized, and quantitative real-time PCR was performed using SYBR Green chemistry to measure expression of Tnf-α, Il6, Cxcl1 (KC), Nos2, and Il10. Relative gene expression was calculated using the 2−ΔΔCt method as described above. Primer sequences are provided in table S2. 16S rRNA sequencing Colonic contents were collected and microbial DNA was extracted using the DNeasy PowerSoil Pro kit (Qiagen) according to the manufacturer’s instructions (3). Amplicon libraries targeting the V3–V4 regions of the 16S rRNA gene were generated using the Quick-16S Library Prep Kit (Zymo Research) with forward primers 341F (CCTACGGGDGGCWGCAG) and its variant (CCTAYGGGGYGCWGCAG), and reverse primer 806R (GACTACNVGGGTMTCTAATCC). Libraries were purified, normalized, and sequenced on an Illumina NextSeq 2000 platform (2 × 301 bp). Base calling, demultiplexing, and adapter trimming were performed using Illumina BCL Convert (v4.2.4), followed by quality filtering (minimum read length >150 bp; poly-N <10; poly-G <150). Sequence processing was conducted in QIIME 2 (v2024.5) using the DADA2 plugin for denoising, chimera removal, and amplicon sequence variant (ASV) inference, with reads truncated based on quality scores (Phred ≥24), yielding final lengths of ∼275–290 bp. Taxonomic classification was performed using a Naïve Bayes classifier trained on the Greengenes2 database (release 2024.9.2) trimmed to the V3–V4 region using experimental primers. Alpha diversity (Observed ASVs, Shannon index, Chao1) and beta diversity (Bray–Curtis and Jaccard distances) were calculated using the QIIME 2 core-metrics-phylogenetic pipeline, with group differences assessed by Kruskal-Wallis test and visualized by principal coordinates analysis (PCoA). Metabolite Analysis Indoles and tryptophan-derived metabolites were quantified from colonic contents collected from infected and uninfected groups at day 3 postinfection using liquid chromatography–tandem mass spectrometry (LC-MS/MS). Samples were extracted in 1 mL of 80:20 (vol/vol) methanol/water containing internal standards, followed by centrifugation to remove debris. Supernatants were subjected to LC-MS/MS analysis. A 2-μL aliquot of each sample was injected onto an Acquity HSS-T3 UPLC column (2.1 × 150 mm; Waters) maintained at 50°C and separated using an Acquity UPLC system (Waters) coupled to a Xevo TQ-XS triple quadrupole mass spectrometer (Waters). Chromatographic separation was performed at a flow rate of 0.3 mL/min using a 14-min gradient with mobile phase A (water containing 0.6% formic acid) and mobile phase B (methanol containing 0.6% formic acid). Initial conditions (98% A/2% B) were held for 1 min, followed by a linear gradient to 95% B over 10 min, maintained at 95% B until 11 min, and then returned to initial conditions (98% A) at 11.01 min and equilibrated until 14 min. Analytes were detected by electrospray ionization in both positive and negative ion modes using multiple reaction monitoring. Instrument parameters were optimized for each metabolite. Data acquisition and quantification were performed using TargetLynx software (Waters MassLynx platform). ATP measurement Intracellular ATP levels were measured using the BacTiter-Glo assay kit (Promega) according to the manufacturer’s instructions. C. jejuni wild-type strain was cultured for 9 h in 25 mL Mueller–Hinton broth supplemented with 0.5 mM indole or vehicle control (DMSO), starting at an initial OD600 of 0.05. Cultures were harvested by centrifugation, washed twice with PBS, and normalized to an OD600 of 1.0. ATP levels were then quantified using the BacTiter-Glo reagent, and concentrations were determined based on a standard curve (58, 59). Respiration Assays C. jejuni wild-type strain was cultured for 9 h in 25 mL Mueller–Hinton broth supplemented with 0.5 mM indole or vehicle control (DMSO), starting at an initial OD600 of 0.05. Cultures were harvested by centrifugation, washed twice with PBS, and normalized to an OD600 of 1.0. For respiration measurements, 135 μL of the cell suspension was mixed with 15 μL of 1% (wt/vol) 2,3,5-triphenyltetrazolium chloride (TTC). Control reactions lacking substrate (water only) were included as blanks. Samples were incubated for 1 h under microaerophilic conditions at 42°C, and TTC reduction was quantified by measuring absorbance at 496 nm. (54). Statistical Analysis Statistical differences across metadata categories, such as infection status, were evaluated using permutational multivariate analysis of variance (PERMANOVA) with 999 permutations, including pairwise post-hoc testing. Genus-level differential abundance testing was performed using Analysis of Compositions of Microbiomes with Bias Correction (ANCOM-BC3). Holm’s method corrected P values for multiple comparisons, and significant taxa (adjusted P < 0.05) were used for further analysis. All microbiota analysis graphs and plots were generated using R (version 4.5.14) with the packages ggplot2 (60) and phyloseq (61).