Bioinformatics identification based on causal association inference using multi-omics reveals the underlying mechanism of Gui-Zhi-Shao-Yao-Zhi-Mu decoction in modulating rheumatoid arthritis

Abstract

Object: Rheumatoid arthritis (RA) is a prevalent and currently incurable autoimmune disease. Existing conventional medical treatments are limited in their efficacy, prolonged disease may lead to bone destruction, joint deformity, and loss of related functions, which places a huge burden on RA patients and their families. For millennia, the use of traditional Chinese medicine (TCM), exemplified by the Gui-Zhi-Shao-Yao-Zhi-Mu decoction (GZSYZM), has been demonstrated to offer distinct therapeutic advantages in the management of RA. Exploring the potential mechanism of GZSYZM in the treatment of RA is a hot topic in the field of TCM.

Method: High-throughput sequencing data of RA at bulk level and single cell level and Chinese Materia Medica target-related databases were used as data sources. Ultra-high-performance liquid chromatography coupled with high-resolution mass spectrometry was employed for the identification of the most relevant compounds to the active ingredients present in the GZSYZM granules. Potential disease genes were identified using a combination of differential expression analysis and weighted gene co-expression network analysis, and the "Chinese Materia Medica-Ingredient-Target" network was constructed to obtain candidate drug target genes. The GZSYZM-RA hub genes were then identified based on Molecular Complex Detection algorithm. To explore the associations and potential mechanisms between the GZSYZM-RA hub gene set and RA, Mendelian randomization (MR) analysis and Bayesian co-localization analysis were used to further identify the GZSYZM-RA core genes that were causally associated with RA. A nomogram was constructed based on a multifactorial logistic regression model using the GZSYZM-RA core genes as predictors of RA. To evaluate its diagnostic value, receiver operating characteristic (ROC) curves, calibration curves, and decision curves were plotted. The potential downstream regulatory mechanisms of the gene of interest in GZSYZM in RA therapy were finally investigated using single- gene set enrichment analysis and molecular docking. The aim was to model the optimal conformation of its target protein receptor binding to the small molecule ligand in GZSYZM to identify the key constituents.

Result: Functional enrichment analysis revealed that the GZSYZM-RA hub gene set is enriched in several autoimmune-related mechanistic pathways, with a particular emphasis on the phosphoinositide 3 kinase (PI3K)‑serine/threonine kinase (AKT) signaling pathway. AUCell scores demonstrated active expression of the GZSYZM-RA hub gene set with the PI3K-AKT signaling pathway on monocytes, especially non-classical monocytes. Immunol infiltration analysis based on the CIBERSORT algorithm also showed a strong correlation between several genes in the GZSYZM-RA hub gene set and monocytes by calculating Spearman's rank correlation coefficients. MR analysis with co-localization analysis further identified seven core genes (CASP8, PPARG, IKBKB, PPARA, IFNG, MYC, and STAT3) causally associated with RA. Diagnostic value for clinical decision making was demonstrated by a multivariable logistic regression model constructed with GZSYZM-RA core genes. Molecular docking analysis indicates that CASP8 and GZSYZM have high docking scores, with three key constituents (quercetin, kaempferol, and diosmetin) exhibiting strong binding affinities.

Conclusion: GZSYZM may regulate the abnormal over-proliferation and apoptotic imbalance of fibroblast-like synoviocytes in RA patients by inhibiting signaling of the PI3K-AKT signaling pathway while activating CASP8-mediated pro-apoptotic effects. And it may be effective in directly or indirectly inhibiting monocyte-to-osteoclast differentiation, ultimately improving the poor prognosis of joint destruction in RA patients.

Keywords: Bioinformatics techniques; CASP8; Gui-Zhi-Shao-Yao-Zhi-Mu decoction; Multi-omics; PI3K-AKT signaling pathway; Rheumatoid arthritis.