Identification of SCRG1 as a Potential Therapeutic Target for Human Synovial Inflammation

Abstract

Synovial inflammation of joint tissue is the most important cause of tissue damage, joint destruction, and disability and is associated with higher morbidity or mortality. Therefore, this study aims to identify key genes in osteoarthritis synovitis tissue to increase our understanding of the underlying mechanisms of osteoarthritis and identify new therapeutic targets. Five GEO datasets with a total of 41 normal synovial membrane tissues and 45 osteoarthritis synovial membrane samples were used for analysis, and seven common differential genes were identified. The classification model constructed by LASSO analysis showed that six genes including CDKN1A, FOSB, STMN2, SLC2A3, TAC, and SCRG1 can be used as biomarkers of osteoarthritis, and the SCRG1 gene shows importance in osteoarthritis. Furthermore, drug database enrichment found that these six DEGs may be the drug targets of synovitis in osteoarthritis, and Valproic Acid CTD 00006977 may be a potential targeted therapeutic drug of SCRG1. Spearman correlation analysis was performed on the SCRG1 gene, and 27 genes with consistent expression were obtained. Functional analysis showed that 27 genes were mainly involved in metabolism, complement, antigen presentation, apoptosis, and regulation of immune pathways. The co-regulatory network of TFs-miRNA suggested that the SCRG1 gene may be regulated by hsa-miR-363-3p miRNA. In conclusion, SCRG1, as a diagnostic marker of osteoarthritis, co-regulates immune-related pathways through the interaction of related proteins, playing an important role in the occurrence and development of osteoarthritis, which may be a novel drug target.

Keywords: LASSO; SCRG1; osteoarthritis; pharmaceutical therapy; synovium.

Figure 1

Differential expression gene analysis. (A) The different analysis results of 5 synovium-related GEO data sets, upset and Venn diagram showed the relationship between the number of different genes in different GEO datasets. (B) KEGG function analysis of each gene sets. (C) Boxplot showed the expression levels of 7 genes which are commonly differentially expressed between synovial tissues and normal tissues in 5 GEO datasets. (D) A heatmap plot of 7 common differentially expressed genes in all synovial tissues and normal tissues.

Figure 2

A model for predicting OA and verification of gene differential expression. (A) CV statistical graph during the construction of the LASSO regression model, which shows that the minimum lambda at model construction is 0.00154252 (dotted line on the left). (B) the model regression coefficient diagram shows the change trend of the coefficient corresponding to each gene variable with the change of lambda value. The results showed that there were 6 prediction genes corresponding to the minimum lambda. (C) ROC curve predicts the identification effect of the above models in different datasets. The closer AUC value is to 1, the better of prediction effect on the model. The figure is shown that the AUC in the training set is 0.8896 and that in the test set is 0.9394, indicating that the model has good prediction accuracy. In RNA-seq validation data set, the AUC is 0.8713, which shows that the models constructed by the 6 genes can also have good accuracy in different types of data sets. (D–I) The expression levels of 6 genes affecting disease occurrence in the model are verified. The figure shows that 6 genes are also significantly differentially expressed in the RNA-SEQ data set.

Figure 3

PCA cluster plot before and after sample correction and protein-protein interaction analysis of spearman correlation genes with SCRG1. (A, B) Two-dimensional PCA cluster plot of the 5 microarray datasets before and after sample correction, the colors represent each osteoarthritis datasets. (C) The Spearman’s correlation coefficients of gene expression with SCRG1, 27 genes were found. (D) Heatmap plot of 27 associated genes, and showed 27 genes also differentially expressed in all synovial tissues and normal tissues. (E) KEGG pathway analysis of the 27 genes, and the enriched immune-related pathways were shown.

Figure 4

Protein-Protein interactions analysis with 27 genes associated expression with SCRG1. The PPI network showed 5 sub-functional clusters were identified and suggest that SCRG1 may be a protein complexes service with important biological function in OA. And the biological function is mainly related to metabolic pathways, complement and coagulation cascades, Antigen processing and presentation, Apoptosis, and immune-related signaling pathway.

Figure 5

TF-miRNA co-regulatory network of 27 genes associated with SCRG1. (A) The TF-miRNA co-regulatory network for 27 genes, SCRG1 was not found as the corresponding transcript factor in network, but 6 miRNAs can regulate SCRG1 gene expression. (B,C) volcano and heatmap plot of the miRNA expression in GSE143514 dataset, which represents the differential expression of values in synovium and normal tissues. Red color dot in Volcano plot represents up-regulated expression in synovium tissue. (D) hsa-miRA-363-3p was significantly differentially expressed in synovium tissues.