Gut Microbiota Structure and Metabolites, Before and After Treatment in Early Rheumatoid Arthritis Patients: A Pilot Study

Abstract

Rheumatoid Arthritis (RA) is a chronic systemic autoimmune disease. Modifications of gut microbiota seem to be associated with the disease, but the impact of gut microbiota on therapies' outcome remains unclear. A role of T cells in RA pathogenesis has been addressed, particularly on the Th17/Treg cells balance. Our study aimed to evaluate in early RA (ERA) patients compared to a control group, fecal gut microbiota composition, short-chain fatty acids concentrations, and the levels of circulating Th17/Treg and their own cytokines, before and after 3 months of standard treatment (Methotrexate (MTX) plus glucocorticoids). Fecal microbiota characterization was carried out on 19 ERA patients and 20 controls matched for sex and age. Significant decreased biodiversity levels, and a partition on the base of the microbiota composition, between the ERA patients at baseline compared to controls, were observed. The co-occurrent analysis of interactions revealed a characteristic clustered structure of the microbial network in controls that is lost in ERA patients where an altered connection between microbes and clinical parameters/metabolites has been reported. Microbial markers such as Acetanaerobacterium elongatum, Cristiansella massiliensis, and Gracilibacter thermotolerans resulted significantly enriched in control group while the species Blautia gnavus emerged to be more abundant in ERA patients. Our results showed an alteration in Th17/Treg balance with higher Th17 levels and lower Treg levels in ERA group respect to control at baseline, those data improved after therapy. Treatment administration and the achievement of a low disease activity/remission appear to exert a positive pressure on the structure of intestinal microbiota with the consequent restoration of biodiversity, of the structure of microbial network, and of the abundance of taxa that became closer to those presented by the subject without the disease. We also found an association between Blautia gnavus and ERA patients characterized by a significant reduction of propionic acid level. Furthermore significant differences highlighted at baseline among controls and ERA patients are no more evident after treatment. These data corroborate the role played by gut microbiota in the disease and suggest that therapy aimed to restore gut microbiota would improve treatment outcome.

Keywords: gut microbiota (GM); metabolomics; methotrexate; microbial network; rheumatoid arthritis.

FIGURE 1

Representative dot plots of peripheral blood lymphocytes from one patient showing the gating strategy to identify Th17 and Treg cells. (A) Live cells, gated for cell death and debris exclusion, were then gated as lymphocytes (R1) based on morphological parameters (SSC, side scatter; FSC, forward scatter). (B) Identification of CD4+ lymphocytes (R2) within the R1 population. (C) Identification of Treg cells and (D) Th17 cells within the R2 population.

FIGURE 2

Microbiota diversity analysis. (A) Color coded rarefaction curves. For each group, the average values of α-diversity indexes with 95% confidence intervals were reported at different sequencing depth. (B) Color-coded box and whisker plots showing the distribution of the considered α-diversity estimators among groups. The presence of statistically significant difference between group of data were also reported. (C) PCoA plot of bacterial α-diversity based on Bray-Curtis dissimilarity and Unweighted UniFrac distance according to considered groups of subjects. For each group, the 95% confidence interval has been drawn. Numbers between parentheses represents the percentage of the total variance explained by the principal coordinates. (D) Color-coded box and whisker plots showing the distribution of principal coordinates for the ERA group between different time points. (E,F) Differential abundance analysis of taxa at genus and species levels with a mean relative abundance ≥1% determined in at least one group. (G) Distribution of SCFAs levels among groups. The Mann-Whitney U test and was performed to determine statistically significant differences between the control group and the RA one at both considered time-points. The Wilcoxon Signed-Ranks test was performed to assess significant differences between the different time-point considered for the ERA group. The * and ** difference is statistically significant.

FIGURE 3

Percentage of Th17 and Treg cells in the peripheral blood of patients with rheumatoid arthritis at T0 and T1. (A) Significant increase in the percentage of Th17 in RA patients compared to controls and a reduction in the percentage of Th17 cells at T1 compared to T0. (B) Significant increase in the percentage of Treg cells in the control group compared to RA and a significant increase in Treg cells in RA patients at T1 compared to T0. * significant difference at alpha value 0.05.

FIGURE 4

Levels (pg/ml) of pro-inflammatory cytokines measured in the serum of patients with RA at T0 and T1. (A) Significant reduction in serum TNF-α levels and (B) IL-6 levels at T1 compared to T0 in patients with rheumatoid arthritis. * significant difference at alpha value 0.05.

FIGURE 5

Co-occurrence network analysis taking into account bacterial species presenting a mean relative abundance ≥ 0.01% across the whole population of samples. Node size is proportional to the number of edges departing from the node, while the edge thickness is proportional to the strength of correlations. Node name size is proportional to the betweenness centrality, meaning the bridging/key importance of that node within the network. Different shapes are used for nodes representing species (circles), clinical variables (rhombus) and metabolic variables (exagons). Continuous lines represent correlations between species while dotted ones indicate correlations between species and clinical and metabolic variables. Blue and red edges represent positive and negative correlation, respectively. Nodes are colored in agreement with the modularity class detected by the GLay algorithm in the network relative to the control group.