Objective To reveal the therapeutic effects of moxibustion in collagen-induced arthritis (CIA) rat models using the combined analysis of plasma and synovial membrane lipidomic profiling and to enhance the understanding of how moxibustion affects lipid metabolism in rheumatoid arthritis (RA).

Methods A total of 32 male Sprague-Dawley (SD) rats were randomly assigned to four groups: control, moxibustion control (MC), model, and moxibustion model (MM) groups, with 8 rats in each group. CIA was induced in SD rats by two immunizations. The paw volume was measured before the induction of CIA. Following induction, after assessing paw volume and arthritis index (AI) scores, the MC and MM groups received treatment at bilateral Shenshu (BL23) and Zusanli (ST36) acupoints for 10 min per acupoint. The intervention included three treatment courses, each spanning 6 d and followed by a 1-d interval. Paw volume and AI scores were assessed after each treatment course. After the completion of the three treatment courses, serum, plasma, synovial tissue, and ankle joint samples were collected. Enzyme-linked immunosorbent assay (ELISA) was employed to quantify the levels of interleukin (IL)-6 and tumor necrosis factor (TNF)-α in serum. Hematoxylin and eosin (HE) staining was performed for histopathological examination of the ankle joint tissues. Meanwhile, ultra-high-performance liquid chromatography coupled with Q-Exactive Orbitrap mass spectrometry (UHPLC-Q-Exactive Orbitrap MS) was utilized to analyze the plasma and synovial tissue samples. In addition, multivariate statistical analysis was performed to identify differential lipid metabolites, and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis was applied to explore metabolic pathways modulated by moxibustion therapy.

Results No significant difference in hind paw volume and AI scores was observed among the groups (P > 0.05). After CIA induction, model group showed increased hind paw volume and AI scores compared with control group (P < 0.05), which were significantly reduced after moxibustion treatment in MM group compared with model group (P < 0.05). The levels of IL-6 and TNF-α were significantly higher in model and MM groups compared with control group (P < 0.05), but were lower in MM group than those in model group (P < 0.05). Histopathological analysis showed improved cartilage and reduced inflammation in MM group. A total of 33 differential lipid metabolites in the plasma and 24 in the synovial membranes of CIA rat models were identified when compared with control group. Among these lipid metabolites, 31 in the plasma and all 24 in the synovial membranes were regulated by moxibustion treatment. Pathological analysis revealed upregulation of diacylglycerol (DG) and fatty acid (FA) levels, alongside downregulation of lysophosphatidylcholine (LPC), phosphatidylcholine (PC), and phosphatidylethanolamine (PE). Under physiological conditions, the treatment specifically reduced LPC and PC levels. Pathway enrichment analysis revealed that moxibustion predominantly affected α-linolenic acid, glycerophospholipid, and sphingolipid metabolism under pathological conditions. Under physiological conditions, the regulation was centered around α-linolenic acid and glycerophospholipid metabolism.

Conclusion The RA rat models exhibited significant lipid metabolic disturbances. Moxibustion alleviated paw swelling, reduced AI scores, modulated inflammatory cytokine levels, and partially corrected the altered levels of multiple lipid metabolites. The potential metabolic pathways implicated in the regulation of lipid metabolism under both physiological and pathological conditions include α-linolenic acid, glycerophospholipid, and sphingolipid metabolism.