Perturbed Mucosal Immunity and Dysbiosis Accompany Clinical Disease in a Rat Model of Spondyloarthritis

Abstract

Objective: The HLA-B27/β2 -microglobulin (β2 m)-transgenic (Tg) rat is a leading model of B27-associated spondyloarthritis (SpA), and the disease is dependent on the presence of intestinal bacteria. Previous studies have shown that adult HLA-B27/β2 m-Tg rats have an altered intestinal microbiota. This study sought to better define the age-dependent changes to both mucosal immune function and dysbiosis in this rat model of SpA.

Methods: Intestinal contents were collected from wild-type and HLA-B27/β2 m-Tg rats postweaning (ages 3 and 6 weeks), at disease onset (age 10 weeks), and after the establishment of disease (ages ≥16 weeks). The microbial community structure was determined by 16S ribosomal RNA sequencing and quantitative reverse transcription-polymerase chain reaction (qRT-PCR). Mucosal and systemic Th1, Th17, and Treg cell responses were analyzed by flow cytometry, as was the frequency of IgA-coated intestinal bacteria. Intestinal expression of inflammatory cytokines and antimicrobial peptides (AMPs) was determined by qRT-PCR.

Results: An inflammatory cytokine signature and elevated AMP expression during the postweaning period preceded the development of clinical bowel inflammation and dysbiosis in HLA-B27/β2 m-Tg rats. An early and sustained expansion of the Th17 cell pool was specifically observed in the cecal and colonic mucosa of HLA-B27/β2 m-Tg rats. Strongly elevated intestinal colonization of Akkermansia muciniphila and an increased frequency of IgA-coated fecal bacteria were significantly associated with expression of HLA-B27 and arthritis development.

Conclusion: HLA-B27/β2 m expression in this rat model renders the host hyperresponsive to microbial antigens from infancy. Early activation of innate immunity and expansion of a mucosal Th17 signature are soon followed by dysbiosis in HLA-B27/β2 m-Tg animals. The pathologic processes of perturbed mucosal immunity and dysbiosis strongly merit further study in both prediseased and diseased populations of patients with SpA.

Figure 1. Development of intestinal inflammation and an altered intestinal microbiota in HLA-B27/β2m rats

Fischer 344 (33-3) HLA/B27/β2m transgenic and WT littermate control mice were sacrificed at the indicated time points and cecal mucosal colonization determined by RT-qPCR using phylum- or species-specific primers (A–C). Abundance of Firmicutes (A), Proteobacteria (B) and Akkermansia muciphila (C) were calculated relative to 16s (arbitrary units). (D) Cecal and colonic inflammation was assessed histologically. Each symbol represents a single animal, and the data were pooled from two to three independent experiments per time point (n=7–14 per age/genotype). Horizontal lines represent group means. *p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 2. Cecal cytokine expression in WT and HLA-B27/β2m rats

Fischer 344 (33-3) HLA/B27/β2m transgenic and WT littermate control mice were sacrificed at the indicated time points and cecal cytokine mRNA expression determined by RT-qPCR. Expression level of (A) IFNγ, IL-17A, IL-23, IL-1β, TNFα,IL-8 and IL-6 was calculated relative to housekeeping gene HPRT. Each symbol represents a single animal, and the data were pooled from two to three independent experiments per time point (n=7–14 per age/genotype). Horizontal lines represent group means. *p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 3. Cecal antimicrobial peptide (AMP) expression in WT and HLA-B27/β2m rats

Fischer 344 (33-3) HLA/B27/β2m transgenic and WT littermate control mice were sacrificed at the indicated time points and cecal AMP mRNA expression determined by RT-qPCR. Expression level of (A) RegIIIγ, (B) S100A8 (C) cRAMP was calculated relative to housekeeping gene HPRT. Each symbol represents a single animal, and the data were pooled from two to three independent experiments per time point (n=7–14 per age/genotype). Horizontal lines represent group means. *p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 4. Disruption of the Th17/Treg compartment in HLA-B27/β2m rats

Fischer 344 (33-3) HLA/B27/β2m transgenic and WT littermate control mice were sacrificed at the indicated time points. Lymphocytes were isolated from cecal and colonic lamina propria, mesenteric lymph node (MLN) and Spleen and the frequency of (A) CD4+IL-17+ T cells and (B) CD4+FoxP3+ T cells was determined by flow cytometry (n=5-12 animals/genotype/time point). Data represents two to three pooled experiments per time point. Bars represent group means +/− SEM. *p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 5. HLA-B27/β2m rats harbor an increased frequency of IgA-coated microbes

Feces were collected from Fischer 344 (33-3) HLA/B27/β2m transgenic and WT littermate control mice at the indicated time points. To enumerate the frequency of IgA coated bacteria, fecal contents were stained with SYTO BC nuclear stain and with anti-rat IgA. (A–B) Representative flow cytometric plots of IgA staining in A) WT and B) HLA/B27/β2m transgenic rats (16weeks). (C) The percentage of IgA+ve fecal bacteria at the indicated time points. (D) Feces from adult (16–20wk) arthritic and non-arthritic H;A-B27/β2M rats were analyzed for IgA coating (as in C). Each symbol represents a single animal, with data pooled from two independent experiments. ** p < 0.01.

Figure 6. Am colonization correlates with the severity of bowel inflammation and development of arthritis in HLA-B27/β2m rats

(A–C) Linear regression was performed on log-transformed cecal mucosal Akkermansia muciniphila colonization (relative to 16s) and cecal mRNA expression of either (A) IFNγ, (B) IL-17 or (C) IL-23 (calculated relative to HPRT). F and p values shown on figure. Each symbol represents a single animal. (D) Longitudinal Am colonization (relative to 16s) was determined at the indicated time points in WT (red line) animals or HLA-B27/β2m animals that either progressed to arthritis (blue line) or remained non-arthritic (green line). Bars represent group means +/− SEM, n = 5–6 per group. *p < 0.05.