Siderophore Immunization Restricted Colonization of Adherent-Invasive Escherichia coli and Ameliorated Experimental Colitis

Abstract

Inflammatory bowel diseases (IBD) are characterized by chronic inflammation of the gastrointestinal tract and profound alterations to the gut microbiome. Adherent-invasive Escherichia coli (AIEC) is a mucosa-associated pathobiont that colonizes the gut of patients with Crohn's disease, a form of IBD. Because AIEC exacerbates gut inflammation, strategies to reduce the AIEC bloom during colitis are highly desirable. To thrive in the inflamed gut, Enterobacteriaceae acquire the essential metal nutrient iron by producing and releasing siderophores. Here, we implemented an immunization-based strategy to target the siderophores enterobactin and its glucosylated derivative salmochelin to reduce the AIEC bloom in the inflamed gut. Using chemical (dextran sulfate sodium) and genetic (Il10^-/- mice) IBD mouse models, we showed that immunization with enterobactin conjugated to the mucosal adjuvant cholera toxin subunit B potently elicited mucosal and serum antibodies against these siderophores. Siderophore-immunized mice exhibited lower AIEC gut colonization, diminished AIEC association with the gut mucosa, and reduced colitis severity. Moreover, Peyer's patches and the colonic lamina propria harbored enterobactin-specific B cells that could be identified by flow cytometry. The beneficial effect of siderophore immunization was primarily B cell-dependent because immunized muMT^-/- mice, which lack mature B lymphocytes, were not protected during AIEC infection. Collectively, our study identified siderophores as a potential therapeutic target to reduce AIEC colonization and its association with the gut mucosa, which ultimately may reduce colitis exacerbation. Moreover, this work provides the foundation for developing monoclonal antibodies against siderophores, which could provide a narrow-spectrum strategy to target the AIEC bloom in Crohn's disease patients. IMPORTANCE Adherent-invasive Escherichia coli (AIEC) is abnormally prevalent in patients with ileal Crohn's disease and exacerbates intestinal inflammation, but treatment strategies that selectively target AIEC are unavailable. Iron is an essential micronutrient for most living organisms, and bacterial pathogens have evolved sophisticated strategies to capture iron from the host environment. AIEC produces siderophores, small, secreted molecules with a high affinity for iron. Here, we showed that immunization to elicit antibodies against siderophores promoted a reduction of the AIEC bloom, interfered with AIEC association with the mucosa, and mitigated colitis in experimental mouse models. We also established a flow cytometry-based approach to visualize and isolate siderophore-specific B cells, a prerequisite for engineering monoclonal antibodies against these molecules. Together, this work could lead to a more selective and antibiotic-sparing strategy to target AIEC in Crohn's disease patients.

Keywords: Escherichia coli; gut inflammation; immunization; inflammatory bowel disease; iron utilization; mucosal immunity; siderophores.

FIG 1

CTB-Ent immunization protected mice from AIEC infection in the absence of Lcn2. (A) Experimental timeline (schematic created with BioRender) and weight loss during AIEC infection in Lcn2^−/− mice. Mice received 100 μg intranasal CTB or CTB-Ent at day 0, followed by a booster 14 days later. Animals were considered fully immunized 28 days after the initial immunization. (B) Fecal AIEC shedding was monitored in mice immunized with CTB (gray circles) or CTB-Ent (green circles) over time. Each circle represents a mouse. The dashed horizontal line indicates the limit of detection (LOD). (C) Colon lengths and colitis scores were assessed on day 5 post-AIEC infection. Representative pictures from H&E-stained colonic sections are shown. 100× magnification; Scale bars, 200 μm. (D) Optical density 450 nm (OD₄₅₀) values of fecal anti-Ent IgA and blood anti-Ent Ig (120 h pi, dilutions indicated in the figure) are shown. Fecal OD₄₅₀ values are shown from undiluted samples (pre-AIEC) or 1:5 dilutions (120 h pi). (E) OD₄₅₀ values for anti-GlcEnt Ig in feces (undiluted) and serum (120 h pi, dilutions indicated in the figure). Assay cutoffs are indicated by dashed lines. (F) FISH scores represent epithelial attachment and epithelial invasion of AIEC in proximal colon sections from FISH experiments. Representative confocal images of each group are shown. The colonic epithelium is highlighted by a white dashed line. AIEC are indicated with white arrowheads. 200× magnification; Scale bars, 25 μm. LOD, limit of detection; pi, postinfection. Data represent mean ± SEM (A, C, and F), geometric mean (B), or geometric mean ± SD (D and E); ns, not significant; **, P ≤ 0.01; ****, P ≤ 0.0001.

FIG 2

CTB-Ent immunization conferred protection from AIEC-mediated colitis in WT mice (Lcn2 proficient). (A) Experimental timeline (schematic created with BioRender) and weight loss during AIEC infection in WT mice. Mice received 100 μg intranasal CTB or CTB-Ent at day 0 followed by a booster 14 days later. Animals were considered fully immunized 28 days after the initial immunization. (B) Fecal AIEC shedding was monitored in mice immunized with CTB (gray circles) or CTB-Ent (blue circles) over time. Each circle represents a mouse. The dashed horizontal line indicates the limit of detection (LOD). (C) Colon lengths, histopathology colitis scores, and representative H&E-stained colon sections are shown. 100× magnification; Scale bars, 200 μm. (D) Ent-specific IgA levels in feces or Ig in serum are shown as the OD₄₅₀ values. (E) OD₄₅₀ values of fecal and systemic anti-GlcEnt Ig. Fecal anti-Ent OD₄₅₀ values are shown from 1:2 dilutions (at 120 h pi) or undiluted samples (pre-AIEC and GlcEnt). Dashed lines indicate assay cutoffs. (F) Epithelial attachment and epithelial invasion of AIEC in the proximal colon of CTB or CTB-Ent immunized mice were assessed on confocal FISH images and converted into a FISH score. 200× magnification; Scale bars, 25 μm. WT, wild-type; LOD, limit of detection; pi, postinfection. Data represent mean ± SEM (A, C, and F), geometric mean (B), or geometric mean ± SD (D and E); ns, not significant; *, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001.

FIG 3

CTB-Ent immunization protected Il10^−/− mice from severe colitis and reduced the AIEC bloom. (A) Experimental timeline (schematic created with BioRender) along with weight loss in piroxicam-treated Il10^−/− mice during AIEC infection. Mice were previously immunized with 100 μg intranasal CTB (gray circles) or CTB-Ent (blue circles) at day 0, followed by a booster 14 days later, and were considered fully immunized after 28 days. (B) AIEC in fecal samples throughout the infection. Each circle represents a mouse. (C) Colon lengths and histological colitis scores were assessed at the end of the experiments. Representative pictures from H&E-stained colonic sections are shown. 100× magnification; Scale bars, 200 μm. (D) Colonic mRNA expression of indicated genes was determined in mice immunized with CTB or CTB-Ent 120 h pi. Data are expressed as fold change in comparison to uninfected mice. (E) OD₄₅₀ values of fecal (undiluted) and systemic (dilutions are indicated in the figure) anti-Ent levels are shown. Assay cutoffs are indicated by dashed lines. (F) FISH scores represent the degree of epithelial attachment and invasion. Representative FISH images of the proximal colon of mice are shown. The epithelium is highlighted by a white dashed line. AIEC are indicated with white arrowheads. 200× magnification; Scale bars, 25 μm. LOD, limit of detection; pi, postinfection. Data represent mean ± SEM (A, C, D, and F), geometric mean (B) or geometric mean ± SD (E); ns, not significant; *, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001; ****, P ≤ 0.0001.

FIG 4

Protection from AIEC infection in mice immunized with CTB-Ent was dependent on B cells. (A) Fecal AIEC of wild-type (WT) and muMT^−/− littermate mice that were either immunized with CTB or CTB-Ent before AIEC infection according to the timeline shown in (B). Each symbol represents an individual mouse. (B) Experimental timeline (schematic created with BioRender) along with weight loss during AIEC infection. (C) Colon lengths and histological colitis scores were determined at the end of the experiments. (D) Fecal anti-Ent titers and (E) serum anti-Ent levels of respective groups are shown. Dashed lines indicate the assay cutoffs. WT, wild-type; LOD, limit of detection; pi, postinfection. Data represent geometric mean (A), mean ± SEM (B and C), or geometric mean ± SD (D and E); ns, not significant; *, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001.

FIG 5

Ent-specific B cells were identified by flow cytometry. (A) WT mice were immunized and infected with AIEC as shown in the schematic (created with BioRender). At 96 h pi, colon lamina propria mononuclear cells (LPMC) and Peyer’s patches (PP) were prepared for flow cytometry (FC) staining. (B) Schematic showing how Ent-specific B cells that recognize biotinylated Ent are detected by incubation with both PE- and APC-conjugated streptavidin to increase sensitivity. Created with BioRender. (C) Flow cytometry gating strategy of cells isolated from PP is shown. After identifying the lymphocyte population and excluding doublets, cells were gated on live CD45⁺ cells followed by identification of CD19⁺/B220⁺ double-positive B cells. After exclusion of IgM⁺ B cells, target B cells (preincubated with Ent-biotin) were identified by PE⁺-Streptavidin/APC⁺-Streptavidin double positivity. (D) Representative plots show Ent-specific B cells in PP and colon LPMC from 4 pooled mice, respectively. LPMC, lamina propria mononuclear cells; PP, Peyer’s patches; FC, flow cytometry; pi, postinfection.