Gut Microbiota and Bacterial DNA Suppress Autoimmunity by Stimulating Regulatory B Cells in a Murine Model of Lupus

Abstract

Autoimmune diseases, such as systemic lupus erythematosus, are characterized by excessive inflammation in response to self-antigens. Loss of appropriate immunoregulatory mechanisms contribute to disease exacerbation. We previously showed the suppressive effect of vancomycin treatment during the "active-disease" stage of lupus. In this study, we sought to understand the effect of the same treatment given before disease onset. To develop a model in which to test the regulatory role of the gut microbiota in modifying autoimmunity, we treated lupus-prone mice with vancomycin in the period before disease development (3-8 weeks of age). We found that administration of vancomycin to female MRL/lpr mice early, only during the pre-disease period but not from 3 to 15 weeks of age, led to disease exacerbation. Early vancomycin administration also reduced splenic regulatory B (Breg) cell numbers, as well as reduced circulating IL-10 and IL-35 in 8-week old mice. Further, we found that during the pre-disease period, administration of activated IL-10 producing Breg cells to mice treated with vancomycin suppressed lupus initiation, and that bacterial DNA from the gut microbiota was an inducer of Breg function. Oral gavage of bacterial DNA to mice treated with vancomycin increased Breg cells in the spleen and mesenteric lymph node at 8 weeks of age and reduced autoimmune disease severity at 15 weeks. This work suggests that a form of oral tolerance induced by bacterial DNA-mediated expansion of Breg cells suppress disease onset in the autoimmune-prone MRL/lpr mouse model. Future studies are warranted to further define the mechanism behind bacterial DNA promoting Breg cells.

Keywords: autoimmunity; bacterial DNA; gut microbiota; immunoregulation; systemic lupus erythematosus.

Figure 1

Oral vancomycin given at the pre-disease stage exacerbated lupus. Female MRL/lpr mice were given vancomycin (2 g/L) in the drinking water for the indicated time periods (in weeks). Analyses were performed at 15 weeks of age unless specified to be 8 weeks of age. (A) Spleen weight to body weight ratio (left) and total number of splenocytes (right). (B) The ratio of anti-dsDNA IgG to total IgG in the mouse serum (n ≥ 10). (C) Level of proteinuria over time (n ≥ 10). (D) RLN weight. (E) Renal histopathology. Left: representative PAS-stained kidney sections; bar equals 400 µm. Middle: glomerular score. Right: tubulointerstitial (TI) score. (F) Serum level of IL-6 (n ≥ 7). (G) Treg to Th17 ratio in the spleen (n ≥ 5). (H) Percentage and absolute number of DN-T cells in splenocytes (n ≥ 5). (I) Transcript level of Il17a gene in the spleen and kidney (n ≥ 7). (J) Serum level of IFNγ (n ≥ 5). (K) Absolute number of IFNγ producing CD4⁺ T cells in the spleen (n ≥ 5). (L) FACS analysis of IL-10⁺ Breg cells in the spleen at 8 weeks of age (n = 10). (M) Serum IL-10 at 8 weeks of age (n = 10). (N) Serum IL-35 at 8 weeks of age (n = 10). Data shown is representative of at least two independent experiments. ^# p < 0.1, *p < 0.05, **p < 0.01, ***p < 0.001. n.s., not statistically significant.

Figure 2

Adoptive transfer of Breg cells at the pre-disease stage attenuated lupus in vancomycin-treated mice. Female MRL/lpr mice were treated with vancomycin (2 g/L) from 3 to 15 weeks of age. Adoptive transfer of cells (10⁶/mouse per dose) were performed at indicated times. Analyses were performed at 15 weeks of age. B10, early injection of activated IL-10⁺ Breg cells at 6 and 7 weeks of age. (A) Spleen and MLN weight. (B) Anti-dsDNA IgG level in the mouse serum (n ≥ 8). (C) Proteinuria level (left; n ≥ 8) and renal lymph node (RLN) weight (right). (D) Kidney glomerular score. (E) FACS plot showing enriched IL-10⁺ Breg cells and composition of the control cell population. (F) Percentage of IL-10 producing B cells in the spleen (n ≥ 7). The negative controls are adoptive transfer of non-Breg cells (Other) at 6 and 7 weeks of age, and late administration of activated IL-10⁺ Breg cells (LB10) at 11 and 12 weeks of age. (G) Spleen and MLN weight. (H) Anti-dsDNA IgG level in the serum (n ≥ 4). (I) Proteinuria level (n ≥ 4). (J) RLN weight. (K) Percentage of DN T cells (left) and Treg to Th17 ratio (right) in the mouse spleen (n ≥ 5). (L) Percentage of IFNγ producing CD4⁺ T cells in the mouse spleen (left) and serum IFNγ (right) (n ≥ 5). Data shown is representative of at least two independent experiments. #p < 0.1, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. n.s., not statistically significant.

Figure 3

Disease stage-specific effects of oral vancomycin on the gut microbiota. Fecal samples were collected at indicated times and analyzed with 16S rRNA sequencing. (A) Diversity and richness of gut microbiota at different ages. Top: Observed species; Middle: Shannon index; Bottom: Simpson index (n ≥ 6). (B) Weighted PCA analysis of gut microbiota at 15 weeks of age (n ≥ 10). (C) Relative abundance of phylum Firmicutes, Bacteroidetes and Proteobacteria at 3, 5, and 8 weeks of age (n ≥ 6). (D) Relative abundance of class Clostridia, Erysipelotrichia and Bacilli at 3, 5, and 8 weeks of age (n ≥ 6). (E) Relative abundance at the order level from 3 to 8 weeks of age (n ≥ 6). qPCR data (C, D) are representative of at least two independent experiments. **p < 0.01, ***p < 0.001.

Figure 4

Restoration of bacterial DNA attenuated lupus in vancomycin-treated mice. Female MRL/lpr mice were treated with vancomycin (2 g/L) starting 3 weeks of age. Endotoxin-free E. coli bacterial DNA (80 µg in PBS/mouse per dose) was orally gavaged to vancomycin-treated mice once a week for 4 consecutive weeks at 4, 5, 6, and 7 weeks of age. Analyses were performed at 15 weeks of age unless specified. The vancomycin-treated group was used as the control group except for panel A. (A) Left: bacterial DNA load in the feces at 7 weeks of age (n = 9). Right: bacterial DNA load in the circulation at 8 weeks of age (n = 5). (B) Percentage of IL-10 producing Breg cells in the spleen and MLN at 8 weeks of age (n = 5). (C) Spleen and MLN weight. (D) Serum anti-dsDNA IgG level at 12 and 15 weeks of age (n = 5). (E) Level of proteinuria over time (n = 5). (F) Kidney glomerular score. (G) Serum level of IL-6 (n = 5). (H) The ratio of Treg to Th17 cells in the spleen (n = 5). (I) The percentages of CD4⁺, CD8⁺ and DN T cells in the spleen (n = 5). (J) Transcript level of Il17a gene in the spleen and kidney (n = 5). (K) The percentage of IFNγ producing CD4⁺ T cells in the spleen (left) and IFNγ level in the serum (n = 5). (L) Representative CD44 FACS histogram (left) and percentage of CD44⁺ cells in CD4⁺, CD8⁺ and DN T cells (right) (n = 5). (M) Percentage of CD44⁺CD8⁺ T cells in the spleen (n = 5). (N) Transcript level of Has1, Has2, and Has3 genes in the kidney (n = 5). (O) Transcript level of IL1β, TNFα, and IL6 genes in the kidney (n = 5). #p < 0.1, *p < 0.05, **p < 0.01, ***p < 0.001. n.s., not statistically significant.