Identifying Immune Cell Infiltration and Effective Diagnostic Biomarkers in Rheumatoid Arthritis by Bioinformatics Analysis.

Sheng Zhou,Hongcheng Lu,Min Xiong

doi:10.3389/fimmu.2021.726747

Abstract

BackgroundRheumatoid arthritis (RA) is a chronic systemic autoimmune disorder characterized by inflammatory cell infiltration, leading to persistent synovitis and joint destruction. The pathogenesis of RA remains unclear. This study aims to explore the immune molecular mechanism of RA through bioinformatics analysis.MethodsFive microarray datasets and a high throughput sequencing dataset were downloaded. CIBERSORT algorithm was performed to evaluate immune cell infiltration in synovial tissues between RA and healthy control (HC). Wilcoxon test and Least Absolute Shrinkage and Selection Operator (LASSO) regression were conducted to identify the significantly different infiltrates of immune cells. Differentially expressed genes (DEGs) were screened by “Batch correction” and “RobustRankAggreg” methods. Functional correlation of DEGs were analyzed by Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG). Candidate biomarkers were identified by cytoHubba of Cytoscape, and their diagnostic effectiveness was predicted by Receiver Operator Characteristic Curve (ROC) analysis. The association of the identified biomarkers with infiltrating immune cells was explored using Spearman’s rank correlation analysis in R software.ResultsTen significantly different types of immune cells between RA and HC were identified. A total of 202 DEGs were obtained by intersection of DEGs screened by two methods. The function of DEGs were significantly associated with immune cells. Five hub genes (CXCR4, CCL5, CD8A, CD247, and GZMA) were screened by R package “UpSet”. CCL5+CXCR4 and GZMA+CD8A were verified to have the capability to diagnose RA and early RA with the most excellent specificity and sensitivity, respectively. The correlation between immune cells and biomarkers showed that CCL5 was positively correlated with M1 macrophages, CXCR4 was positively correlated with memory activated CD4+ T cells and follicular helper T (Tfh) cells, and GZMA was positively correlated with Tfh cells.ConclusionsCCL5, CXCR4, GZMA, and CD8A can be used as diagnostic biomarker for RA. GZMA-Tfh cells, CCL5-M1 macrophages, and CXCR4- memory activated CD4+ T cells/Tfh cells may participate in the occurrence and development of RA, especially GZMA-Tfh cells for the early pathogenesis of RA.

Highlights

Rheumatoid arthritis (RA) is a chronic systemic autoimmune disease [1]
The correlation between immune cells and biomarkers showed that CCL5 was positively correlated with M1 macrophages, CXCR4 was positively correlated with memory activated CD4+ T cells and follicular helper T (Tfh) cells, and GZMA was positively correlated with Tfh cells
CCL5, CXCR4, GZMA, and CD8A can be used as diagnostic biomarker for RA

Summary

Introduction

Rheumatoid arthritis (RA) is a chronic systemic autoimmune disease [1]. The main clinical manifestation of RA is chronic synovitis of the affected joint characterized by persistent synovitis, synovial hyperplasia, and pannus formation, which destroy the bone tissue and gradually lead to the damage of joint function [2, 3]. Immune cells in synovial membrane, including resident and infiltrating immune cells, play a vital role in the occurrence and development of RA, an autoimmune disorder [7]. Studies have shown that macrophages play an important role in promoting the development of RA. These cells secrete abundant cytokines, chemokines, and degrading enzymes, which lead to joint inflammation and bone destruction. These cells can cooperate with other immune cells to aggravate the formation of arthritis [8]. Rheumatoid arthritis (RA) is a chronic systemic autoimmune disorder characterized by inflammatory cell infiltration, leading to persistent synovitis and joint destruction. This study aims to explore the immune molecular mechanism of RA through bioinformatics analysis

Objectives

Methods

Results

Conclusion