This study aimed to investigate the molecular mechanism of systemic vasculitis via bioinformatics analysis. Gene express profile of E-GEOD-16945 (13 Takayasu arteritis samples and 13 control samples) was downloaded from European Bioinformatics Institute (EBI) database. Differentially expressed genes (DEGs) were screened between Takayasu arteritis and normal controls (|log FC| > 1). Basic local alignment search tool (BLASTX) was used for the Clusters of Orthologous Groups (COG) classification of DEGs. Gene ontology analysis was performed for the DEGs (P < 0.05). A gene expression network was built with DEGs. Mcode in Cytoscape software was used to extract modules from the network (degree ≥ 2, K-core ≥ 2 and adjusted P-value < 0.05) followed by pathway analysis using GenMAPP (false discovery rate < 0.05). A total of 747 DEGs were identified. There were 16 significant GO function terms enriched with DEGs, of which immune and defence response was the most significant GO term. Totally, three modules were extracted from gene expression network, including one module constituted with upregulated genes and two modules constituted with downregulated genes. Furthermore, human leucocyte antigen (HLA)-DRB1, HLA-DPA1, HLA-DPB1, HLA-DOA and HLA-DRA in the downregulated modules were significantly linked to immune-related pathways (intestinal immune network for IgA production and systemic lupus erythematosus pathways), while ribosomal protein L 31 (RPL31), RPS3A and RPL9 in the upregulated module were enriched in ribosome pathway. The immune-related pathways, ribosome pathway, immune-related genes including (HLA-DRB1, HLA-DPA1, HLA-DPB1, HLA-DOA and HLA-DRA) and ribosome-related genes (RPL31, RPS3A and RPL9) might be involved in systemic vasculitis.

Background Glucocorticoid-induced osteonecrosis of the femoral head (GIONFH) is a common disease in osteoarticular surgery, with a high disability rate, which brings great physical and mental pain and economic burden to patients. Its specific pathogenesis has not been fully demonstrated, and there is a lack of recognized effective biomarkers for earlier detection and prompt treatment. This has become an urgent clinical problem for orthopedic scholars. Materials and methods We downloaded the gene expression profile dataset GSE123568 from the Gene Expression Omnibus database, used STRING and Cytoscape to carry out module analysis and built a gene interaction network. The four core genes most related to GIONFH in this network were ultimately found out by precise analysis and animal experiment were then conducted for verification. In this verification process, thirty-six New Zealand white rabbits were randomly divided into blank control group, model group and drug group. Except for the blank control group, the animal model of GIONFH was established by lipopolysaccharide and methylprednisolone, while the drug group was given the lipid-lowering drugs for intervention as planned. The rabbits were taken for magnetic resonance imaging at different stages, and their femoral head specimens were taken for pathological examination, then the expression of target genes in the femoral head specimens of corresponding groups was detected. Validation methods included RT-PCR and pathological examination. Results A total of 679 differential genes were selected at first, including 276 up-regulated genes and 403 down-regulated genes. Finally, four genes with the highest degree of correlation were screened. Animal experiment results showed that ASXL1 and BNIP3L were in low expression, while FCGR2A and TYROBP were highly expressed. Conclusion Through animal experiments, it was confirmed that ASXL1, BNIP3L, FCGR2A and TYROBP screened from the comparative analysis of multiple genes in the database were closely related to GIONFH, which is important for early diagnosis of Glucocorticoid-induced osteonecrosis of the femoral head.