Th17 responses and natural IgM antibodies are related to gut microbiota composition in systemic lupus erythematosus patients

Abstract

Intestinal dysbiosis, characterized by a reduced Firmicutes/Bacteroidetes ratio, has been reported in systemic lupus erythematosus (SLE) patients. In this study, in vitro cultures revealed that microbiota isolated from SLE patient stool samples (SLE-M) promoted lymphocyte activation and Th17 differentiation from naïve CD4(+) lymphocytes to a greater extent than healthy control-microbiota. Enrichment of SLE-M with Treg-inducing bacteria showed that a mixture of two Clostridia strains significantly reduced the Th17/Th1 balance, whereas Bifidobacterium bifidum supplementation prevented CD4(+) lymphocyte over-activation, thus supporting a possible therapeutic benefit of probiotics containing Treg-inducer strains in order to restore the Treg/Th17/Th1 imbalance present in SLE. In fact, ex vivo analyses of patient samples showed enlarged Th17 and Foxp3(+) IL-17(+) populations, suggesting a possible Treg-Th17 trans-differentiation. Moreover, analyses of fecal microbiota revealed a negative correlation between IL-17(+) populations and Firmicutes in healthy controls, whereas in SLE this phylum correlated directly with serum levels of IFNγ, a Th1 cytokine slightly reduced in patients. Finally, the frequency of Synergistetes, positively correlated with the Firmicutes/Bacteroidetes ratio in healthy controls, tended to be reduced in patients when anti-dsDNA titers were increased and showed a strong negative correlation with IL-6 serum levels and correlated positively with protective natural IgM antibodies against phosphorylcholine.

Figure 1. Fecal microbiota isolated from SLE patients influences Treg/Th differentiation.

Naïve CD4⁺CD45RA⁺ lymphocytes were co-cultured for 12 days with DC previously maturated with LPS (maturation control), fecal microbiota isolated from healthy controls (HC-M) or SLE patients (SLE-M) as well as with Bifidobacterium bifidum LMG13195 (Bb), Clostridia strains (Cl) or SLE-M containing 5, 10 or 30% Bb or Cl. Cultured CD4⁺ T cells were recovered, stained for Treg markers, IL-17 and IFNγ and analyzed by flow cytometry. (A) Sequential gating strategy used to identify Treg cells (CD4⁺CD25^highCD127^lowFoxp3⁺ ). Positive cells for each marker were determined using fluorescence of cells labelled with the corresponding isotype-matched conjugated irrelevant MAb as a negative control. CD4⁺ T cells showing the highest CD25 expression were identified as CD4⁺CD25^high. Then, CD25^high population was analyzed to determine the proportion of cells expressing Foxp3 (Foxp3⁺ within CD25^high) as well as the amount of Treg cells (CD25^highCD127^lowFoxp3⁺). Density plots correspond to a representative experiment. Analyses of (B) CD25^highpopulation, (C) Foxp3⁺ cells within CD25^high population, and (C) IL-17/IFNγ expression. Bars represent the mean and SEM of seven independent experiments performed with different blood donors. Statistical differences between SLE-M and the different treatments were evaluated by the Wilcoxon test for paired data. *p < 0.1; **p < 0.05.

Figure 2. Increased IL-17 producing cells in SLE patients with anti-dsDNA antibodies.

Foxp3, CD25, CD127, IL-17 and IFNγ expression was analyzed in fresh peripheral blood CD4⁺ lymphocytes from SLE patients and HC. (A) Dot-plots show cells positive for IL-17 or IFNγ expression, determined attending to the fluorescence of cells labelled with the corresponding isotype-matched conjugated irrelevant MAb as a negative control. Scatter plots represent the percentage of IL-17⁺ (Th17) and IFNγ⁺ (Th1) CD4⁺ cells in HC and SLE patients presenting (pos) or not (neg) anti-dsDNA antibodies (aDNA). Horizontal bars show the median. (B) Treg cells were sequentially identified as CD4⁺CD25^highCD127^lowFoxp3⁺ cells. Scatter plots represent the quantity of Treg, Foxp3⁺ IL-17⁺ and Foxp3⁺ IFNγ⁺ cells in HC and SLE patients in function of their anti-dsDNA status, and horizontal bars show the median. Statistical differences among groups were evaluated by Kruskal-Wallis test and Dunn’s post test was conducted to determine which groups’ pairs had different means. **p < 0.01; ***p < 0.001.

Figure 3. IFNγ serum levels in SLE patients were associated with Firmicutes to Bacteroidetes ratio.

(A) Box and whiskers represent median and interquartile range of circulating amounts of cytokines in SLE patients compared to controls. Differences between both groups were assessed by the non-parametric Mann-Whitney U test. (B) Correlations between serum levels of IFNγ and the frequency of Firmicutes, Bacteroidetes or the Firmicutes to Bacteroidetes ratio (F/B) in SLE patients were evaluated using Spearman test.