FAM96A Protects Mice From Dextran Sulfate Sodium (DSS)-Induced Colitis by Preventing Microbial Dysbiosis

Abstract

Family with sequence similarity 96 member A (FAM96A) is an evolutionarily conserved intracellular protein that is involved in the maturation of the Fe/S protein, iron regulatory protein 1 (IRP1), and the mitochondria-related apoptosis of gastrointestinal stromal tumor cells. In this study, we used a mouse model of chemically induced colitis to investigate the physiological role of FAM96A in intestinal homeostasis and inflammation. At baseline, colons from Fam96a^-/- mice exhibited microbial dysbiosis, dysregulated epithelial cell turnover, an increased number of goblet cells, and disordered tight junctions with functional deficits affecting intestinal permeability. After cohousing, the differences between wild-type and Fam96a^-/- colons were abrogated, suggesting that FAM96A affects colonic epithelial cells in a microbiota-dependent manner. Fam96a deficiency in mice resulted in increased susceptibility to dextran sulfate sodium (DSS)-induced colitis. Importantly, the colitogenic activity of Fam96a^-/- intestinal microbiota was transferable to wild-type littermate mice via fecal microbial transplantation (FMT), leading to exacerbation of DSS-induced colitis. Taken together, our data indicate that FAM96A helps to maintain colonic homeostasis and protect against DSS-induced colitis by preventing gut microbial dysbiosis. This study used gene knockout animals to help to understand the in vivo effects of the Fam96a gene for the first time and provides new evidence regarding host-microbiota interactions.

Keywords: DSS; FAM96A; gut microbiota; inflammatory bowel diseases; intestinal mucosa.

Figure 1

Fam96a^−/− mice harbor altered and transferable gut microbiota communities. (A–I) Real-time PCR analysis of relative 16S rDNA of several major bacterial groups in WT and Fam96a^−/− feces before and after cohousing for 8 weeks. Erec, Eubacterium rectale/Clostridium coccoides; Lac, Lactobacillus sp.; Bact, Bacterioides sp.; MIB, mouse intestinal Bacterioides; SFB, Segmented Filamentous Bacteria; Ent, Enterobacteriaceae; C. perf, Clostridium perfringens; H. pylori, Helicobacter pylori; C. lept, Clostridium leptum. SH, single housed; CH, cohoused; Data are expressed as mean ± SEM. *P < 0.05; ***P < 0.001. n.s., not significant. n = 6. Data are representative of three independent experiments.

Figure 2

FAM96A regulates colonic antimicrobial peptide (AMP) genes in a microbiota-dependent manner. (A) mRNA levels of several AMPs in the colons of WT and Fam96a^−/− mice. (B) Mice were treated with a mixture of antibiotics for 2 weeks to deplete the intestinal microbiota, and the mRNA level in WT and Fam96a^−/− colons were then analyzed. (C,D) mRNA levels of indicated AMPs (Ang4, Reg3g, Defa2, Defa3, Defa24) in WT and Fam96a^−/− colons before (C) and after (D) cohousing. Data are expressed as mean ± SEM. *P < 0.05; **P < 0.005; ***P < 0.001. For (A,B), n = 6. For (C–G), n = 5. Data are representative of three independent experiments.

Figure 3

The altered colonic epithelial phenotype is affected by housing conditions. WT and Fam96a^−/− mice were cohoused for 8 weeks, and the colons from the cohoused mice and the gender-/age-matched single-housed mice were then collected and analyzed. (A,B) Representative HE staining of sections of WT and Fam96a^−/− colons. (C) Quantification of crypt height. (D,E) Representative Ki67 immunohistochemistry staining. (F) Quantification of proliferative colonic epithelial cells (Ki67⁺ cells/crypts). (G,H) Representative TUNEL staining. (I) Quantification of apoptotic colonic epithelial cells (TUNEL⁺ cells/crypts). (J,K) Representative Alcian blue/periodic acid–Schiff (AB-PAS) staining of goblet cells. (L) AB-PAS-positive cells in each crypt were enumerated. Scale bars, for (A,D,G,J), 50 μm, for (B,E,H,K), 25 μm. Data are expressed as mean ± SEM. *P < 0.05; **P < 0.005; ***P < 0.001. n.s., not significant. n = 4–5. Data are representative of three independent experiments.

Figure 4

FAM96A plays a role in protection against epithelial barrier permeabilization. (A) Relative concentration of FITC-dextran in WT and Fam96a^−/− mice serum before and after cohousing for 8 weeks. (B,C) Real-time PCR analysis of TJ proteins in IECs from single-housed (B) and cohoused (C) mice. SH, single housed; CH, cohoused. Data are expressed as mean ± SEM. *P < 0.05; **P < 0.005; ***P < 0.001; ****P < 0.0001. For (A), n = 4–5. For (B,C), n = 5–6. Data are representative of three independent experiments.

Figure 5

FAM96A protects mice from DSS-induced colitis. (A) Body weights of the two groups of mice. (B) Colon length (at day 7 post DSS treatment). (C) DAI (assessed every day). (D) HE staining of colon sections and (E) Histopathologic scores based on a double-blind method. (F) DSS-treated mice and gender-/age-matched controls were gavaged with FITC-dextran, and the FITC-dextran concentration in the serum was then determined 4 h later. Relative concentrations are shown. (G) Cytokine expression was assessed using a LegendPlex Mouse Inflammation Panel kit. Mice were given 2.5% DSS for 5 days followed by normal drinking water; survival rate (H) and body weight (I) were monitored every day. Scale bars, 50 μm. Data are expressed as mean ± SEM. *P < 0.05; **P < 0.005; ***P < 0.001. n = 6–7. For (A–E), data are representative of five independent experiments. For (F–H), data are representative of three independent experiments.

Figure 6

Fam96a^−/− mice have a transmissible and colitogenic gut microbiota. (A–E) WT and Fam96a^−/− mice were pretreated with a mixture of ampicillin, neomycin, metronidazole, and vancomycin for 2 weeks followed by 2.5% DSS treatment. The body weight (A), DAI (B), representative HE staining images (C), histopathologic score (D), and colon length (E) are shown. (F–J) WT mice were pretreated with the antibiotics mixture and then received either WT or Fam96a^−/− feces. After 2 weeks of FMT, the two groups of mice were challenged with 2.5% DSS. Colitis severity is indicated by body weight (F), DAI (G), histopathologic score (I), HE staining (H), and colon length (J). Scale bars, 50 μm. Data are expressed as mean ± SEM. *P < 0.05; **P < 0.005; ***P < 0.001. For (A–E), n = 5–6. For (F–J), n = 4–5. Data are representative of three independent experiments.

Figure 7

Relative gut bacteria levels in WT and Fam96a^−/− mice at baseline and at day 5 after 2.5% DSS treatment. (A–F) α- and β-diversity of the fecal microbiota in Group_WT, Group_KO, Group_DSS_WT, and Group_DSS_KO. (A) Chao 1. (B) Simpson index. (C) Shannon index. (D) Principal coordinate analysis (PCoA) of weighted UniFrac distances. (E) PCoA analysis of Bray-Curtis dissimilarity. (F) Principal component analysis (PCA) of gut bacteria. α-diversity data are expressed as mean ± SEM. n = 5. (G–J) Analysis of commensal bacteria at the phylum (G,I) and family (H,J) levels at baseline. (I) Relative abundance of the phylum Cyanobacteria. (J) Relative abundance of the families Alcaligenaceae, Bacteroidaceae, Bifidobacteriaceae, Clostridiales_vadinBB60_group, Erysipelotrichaceae, Porphyromonadaceae, ratAN060301C, Rikenellaceae, and Ruminococcaceae. (K–N) Quantification of bacterial phyla (K,M) and families (L,N) with respect to genotype (Fam96a^−/− and WT) during colitis. (M) Relative abundance of the phyla Tenericutes, Verrucomicrobia, Fusobacteria, Actinobacteria, and Cyanobacteria during colitis. (N) Relative abundance of the families Bifidobacteriaceae, Porphyromonadaceae, Alcaligenaceae, Verrucomicrobiaceae, and Streptococcaceae during colitis. Data are expressed as mean ± SEM. *P < 0.05; **P < 0.005 (Mann–Whitney U test). n = 5.