Untargeted metabolomics reveals immune-metabolic signatures in established cases of rheumatoid arthritis

Introduction Rheumatoid arthritis (RA) is a complex progressive autoimmune disorder wherein chronic inflammation is tightly coupled to metabolic reprogramming. The known diagnostic markers are not sensitive and specific enough to reflect disease activity. Finding a metabolomics-based biomarker specific for established cases of RA is important. This study aimed to investigate the metabolomic profiles of patients with established RA compared to those of controls. Methods An untargeted high resolution mass spectrometry (MS)-based metabolomics approach with bioinformatics analysis was used to analyze 122 plasma samples, patients (n = 60), and controls (n = 62). Results A total of 300 significantly dysregulated metabolites (unpaired t-test with FDR q value < 0.05, FC cut off 1.5) were identified between RA and controls, where 147 were upregulated and 153 downregulated. From among these, 182 metabolites were identified and annotated and after excluding the exogenous metabolites 60 endogenous metabolites were successfully identified. Results from the OPLSDA model showed a clear separation between patients with RA and controls (Q2 = 0.736, R2 = 0.988), indicating significant metabolic differences between the groups. The plasma metabolomics profile revealed statistically significant changes in metabolites belonging to different classes including those involved in lipid (including Succinyladenosine, CDP- DG (PGE 2 /i-19: 0), PGP (i-24:0/PGD2), Octadecenoylcarnitine), amino acid (including L-Isoleucine, Cysteinyl-Serine), and nucleotide (Inosine, N6-Methyladenosine), metabolisms in RA patients, consistent with immune–metabolic dysregulation. Bioinformatics and network pathway analysis using IPA showed interconnectedness between the metabolites centered around IL-6, IL-2, IL-1, MAPK, and kininogen. The pathways most affected between RA and controls included humoral immune response, inflammatory response, hematological system development, and function. The identified metabolites influenced eicosanoid/kinin signaling, nucleic acid–mediated innate immune activation, mitochondrial dysfunction, and altered glycosylation. Subgroup analysis based on stratification using erythrocyte sedimentation rate (ESR) above 35 mm/h identified 67 metabolites that differentiated patients with high versus low ESR. Among these, three metabolites, namely, cysteinyl-serine (upregulated), tyrosyl-arginine and N2-acetyl N6-methyllysine (downregulated) overlapped with the metabolites identified in the comparison between RA and controls, suggesting links between specific metabolic changes and systemic inflammation. Conclusion Our findings support the potential of plasma metabolomics for phenotyping and highlight potential candidate biomarkers for disease prognosis and monitoring in RA.