Data preprocessing
After searching in Gene Expression Omnibus database with the inclusion criteria included: (1) patients with psoriatic arthritis or psoriasis; (2) whole blood, 2 datasets (GSE61281 and GSE55201) were obtained. For details, 52 blood samples from GSE61281 (20 from patients with psoriatic arthritis,20 from psoriasis and 12 normal person); 44 blood samples of psoriasis from GSE145725 were used only for validation.
After obtaining the datasets, we firstly performed background correction and data normalization, and the result is presented in box plots (Fig. 2).
Identification of DEGS
After data preprocessing, DEGs were identified with the setting of cutoff at FDR < 0.05 and |log2 (FC)| >1. As shown in the volcano plot, totally 86 DEGs in the comparison of PsA and Ps, and 273 DEGs in the comparison of PsA and normal person were obtained (Fig. 3A, B). The top 50 DEGs of two sets are illustrated in two heatmap plots (Fig. 3C, D). The 28 overlapped DEGs between the two sets are presented in a Venn diagram (Fig. 3E).
Functional Correlation Analysis
Enriched GO terms were divided into three categories: BP, CC, and MF. As shown in Fig. 4C, the DEGs merged from the comparison of PsA, Ps and CL were mainly enriched in the ‘protein targeting to ER’, ‘cotranslational protein targeting to membrane’ and ‘SRP-dependent cotranslational protein targeting to membrane’ in the BP group. CC analysis indicated that the DEGs were mainly enriched in ‘cytosolic large ribosomal subunit’, ‘ribosomal subunit’ and ‘cytosolic ribosome’. In terms of MF, DEGs were most enriched in ‘mRNA 5’-UTR binding’ and ‘structural constituent of ribosome’. KEGG analysis of DEGs revealed that they were mainly enriched in ‘Ribosome’. Other results of DEGS from different comparison were shown in Fig. 4A and B.
GSEA and DO results are shown in Fig. 5. In the set of PsA and Ps, the enriched pathways mainly involved ‘signaling by receptor tyrosine kinases’ and ‘VEGFavegfr2 signaling pathway’ (Fig. 5B). While in the set of PsA and CL, ‘metabolism of amino acids and derivatives’ and ‘translation’ were mainly involved (Fig. 5A). The DO analysis results are shown in Fig. 5C. Diseases enriched by overlapped DEGs mainly included Osteoarthritis of hip and Renal cancer metastatic.
PPI Network Construction
The PPI network was constructed using STRING to investigate the interaction of overlapped DEGs, and then the result was visualized in Cytoscape. Subsequently, cytoHubba plugin was used to characterize the hub genes. The top 10 genes based on the filtering algorithm (degree) were shown in Fig. 6.
Expression Profile of Selected Ferroptosis Regulators
Differences in the expression levels of 24 selected ferroptosis regulators were compared between PsA, Ps and CL. Compared with expression in Ps, ACSL4, CISD1, EMC2 and SAT1 expression levels were increased, while CARS was decreased expressed in PsA group (Fig. 7B). While compared with CL, the expression level of CISD1, CS, EMC2 were significantly increased (Fig. 7A), but CARS, FDFT1 and HSPA5 were decreased. These results revealed that ferroptosis may correlated with psoriasis and psoriasis arthritis development, and the underlying mechanism needs to be further explored in the future.
Association between key ferroptosis regulators and hub genes
We then tried to investigate the correlation between top 10 hub genes and the key ferroptosis regulators. Limited by the platform raw data, totally 3 ferroptosis regulators and 7 hub genes participated this analysis. As shown in Fig.7 C, in PsA group, among the 3 key ferroptosis regulators, SAT1 showed negative association with all 7 hub genes, ACSL4 also showed a negative association with all 7 hub genes, but only 4 significantly. CARS showed positive association with 2 hub genes (CLEC12B AND LPAR6). And the association analysis result in Ps group was shown in Fig.7D.
Association analysis validation
As shown in Fig. 8, similar negative associations between 3 ferroptosis regulators and 7 hub genes were found, which further clarify the potential role of ferroptosis in the increased risk of cancer in patients with psoriasis and psoriasis arthritis.