Gut Microbiota in Systemic Lupus Erythematosus and Correlation With Diet and Clinical Manifestations

Abstract

Despite the existing studies relating systemic lupus erythematosus (SLE) to changes in gut microbiota, the latter is affected by external factors such as diet and living environment. Herein, we compared the diversity and composition of gut microbiota in SLE patients and in their healthy family members who share the same household, to link gut microbiota, diet and SLE clinical manifestations. The study cohort included 19 patients with SLE and 19 of their healthy family members. Daily nutrition was assessed using a food frequency questionnaire (FFQ). Microbiota was analyzed using amplicons from the V4 regions of the 16S rRNA gene, to obtain microbiota diversity, taxa relative abundances and network analysis. The gut microbiota in the SLE group had lower alpha diversity and higher heterogeneity than the control group. SLE patients had decreased Acidobacteria, Gemmatimonadetes, Nitrospirae and Planctomycetes at the phylum level, and increased Streptococcus, Veillonella, Clostridium_XI, and Rothia at the genus level. Streptococcus was extremely enriched among patients with lupus nephritis. Lactobacillus, Clostridium_XlVa, Lachnospiracea_incertae_sedis and Parasutterella OTUs were associated with diet and clinical features of SLE. Finally, the gut microbiota of SLE patients remained different from that in healthy controls even after accounting for living conditions and diet.

Keywords: Diets; Gut microbiota; Network analysis; Streptococcus; Systemic lupus erythematosus.

Figure 1

The gut microbiota alpha diversity comparison between the SLE and HC groups. Boxplot visualized five summary statistics (the median, two hinges and two whiskers). Notches represented as median. The lower and upper hinges corresponded to the first and third quartiles (the 25th and 75th percentiles). The whisker extended from the hinge to the largest value no further than 1.5 ^* IQR (the distance between the first and third quartiles) from the hinge. Data beyond the end of the whiskers were outlying points. (A) observed OTU number, (B) richness index of Chao1, (C) equability, and (D) Shannon_e index.

Figure 2

The NMDS plot of gut microbiota from the SLE and HC groups. (A) The NMDS coordinates of SLE (red dot) and HC (blue dot) were plotted in 3D review. (B) The NMDS coordinates of SLE (red circle) and HC (blue circle) were re-plotted with samples from each disease-control pair connected.

Figure 3

Phylum abundance comparison of gut microbiota between the SLE and HC groups. Three major phyla: the Firmicute (A), Bacteroides (B), Proteobacteria (C), and were compared and the ratio of Firmicute/Bacteroides (D) were plotted. The percentage of Acidobacterium (E), Gemmatimonadetes (F), Nitrospirae (G), and Planctomycetes (H) in SLE fecal were significantly reduced after FDR adjustment (q < 0.05). F/B: the relative abundance of Firmicute to Bacteroides.

Figure 4

Volcano plot comparing SLE and HC gut microbiota. The genus abundances of all SLE recipients (A) or lupus nephritis (LN) subgroup (B) were compared with healthy controls. Wilcoxon p-values were calculated and adjusted by the FDR method. Each dot represented a genus. Significantly increased genera of SLE group were plotted and annotated in red, while decreased in blue. The X axis represents the log2 values fold change and the Y axis represents the –log10 value of the adjusted p-values.

Figure 5

Network analysis of dietary elements, gut microbiota and manifestations of SLE patients. Each blue dot represented for an OTU and its diameter was proportional to the square root of its abundance. Red triangles represented for disease manifestations and green squares for dietary factors. Solid line represented for the significant positive correlations (Spearman r > 0 and FDR adjusted p < 0.05) and dashed for negative correlations (Spearman r <0 and FDR adjusted p < 0.05). PLT, platelets; Flare: the state of SLE disease flare; PAP, pulmonary artery pressure.