NLRP3 inflammasome activation contributes to the pathogenesis of rheumatoid arthritis

Abstract

Nucleotide-binding, oligomerization domain (NOD)-like receptor family, pyrin domain containing 3 (NLRP3) gene polymorphism was reported to be associated with susceptibility, disease activity or anti-tumour necrosis factor (TNF) treatment response in rheumatoid arthritis (RA). However, the roles of NLRP3 inflammasome in the development of RA have not yet been elucidated fully. The present study aimed to study the role of NLRP3 inflammasome in RA. NLRP3 inflammasome activation in synovial tissues from RA and osteoarthritis (OA) patients were assessed by Western blot. Active caspase-1 in synovia was stained by a FAM-FLICA caspase-1 probe. Mice with collagen-induced arthritis (CIA) were treated with MCC950, a selective NLRP3 inhibitor, or vehicle for 2 weeks. The clinical score of arthritis, synovial inflammation and cartilage erosion were assessed. Proinflammatory cytokines were measured by enzyme-linked immunosorbent assay (ELISA). The results showed that NLRP3 inflammasome was highly activated in both synovia from RA patients and CIA mice. Activation of NLRP3 inflammasome occurred mainly in the infiltrating monocyte/macrophages in synovia, but not in fibroblast-like synoviocytes. Treatment with MCC950 resulted in significantly less severe joints inflammation and bone destruction. NLRP3 inflammasome activation in the synovia was inhibited significantly by MCC950 with reduced production of interleukin (IL)-1β. The inhibition of NLRP3 inflammasome activation by MCC950 was confirmed further in a human monocytic cell line, THP-1. In conclusion, NLRP3 inflammasome is involved in the pathogenesis of RA. Targeting NLRP3 inflammasome with a small molecule inhibitor might be a novel therapeutic strategy for RA.

Keywords: NLRP3 inflammasome; NLRP3 inhibitor; rheumatoid arthritis.

Figure 1

Nucleotide‐binding, oligomerization domain (NOD)‐like receptor family, pyrin domain containing 3 (NLRP3) inflammasome pathway was activated in synovia of rheumatoid arthritis (RA) patients. Human synovial tissues from RA or osteoarthritis (OA) patients were prepared and subjected to Western blot. (a) Representative Western blot bands of NLRP3 and caspase‐1 p20 expression in synovia of RA patients compared to OA patients. (b) Statistical analysis of relative target protein expression/internal control ratio. Data were recorded as mean ± standard deviation (s.d.), *P < 0·05, **P < 0·01, six patients per group.

Figure 2

CD14 co‐localized with active caspase‐1 in synovia. Synovia from rheumatoid arthritis (RA) or osteoarthritis (OA) patients were collected for immunofluorescence staining. Slides were stained for active caspase‐1 (green) and CD14 (red) (a), CD3 (red) (b) or cadherin‐11 (red) (c). Original magnification ×20.

Figure 3

MCC950 treatment alleviated severity of collagen‐induced arthritis (CIA). Established CIA mice were randomized into control or treatment groups. Mice were injected intraperitoneally with MCC950 (10 mg/kg) or vehicle once every 2 days from day 6 after boosted immunization for 2 weeks. (a) Mean clinical scores of CIA mice treated with MCC950 (nine mice) and control group (eight mice). Blue arrows indicate the time when treatments were given. (b) Macroscopic observation and microcomputed tomography analysis of joints under anesthesia using a Siemens Inveon CT/PET multi‐modality system. Data were recorded as mean ± standard deviation (s.d.), **P < 0·01; normal group, eight mice.

Figure 4

MCC950 treatment attenuated histological change in CIA mice. (a) (Up) Representative haematoxylin and eosin (H&E) staining images and amplification of respective typical lesion (×100). Black arrow, infiltration of inflammatory cells or hyperplasia of synovia. (Low) Representative safranin O‐fast green staining images (×40). Red arrow, bone erosion; blue arrow, cartilage loss. (b) Semiquantitative score of synovial inflammation and cartilage erosion. Data were recorded as mean ± standard deviation (s.d.), **P < 0·01.

Figure 5

MCC950 treatment inhibited Nucleotide‐binding, oligomerization domain (NOD)‐like receptor family, pyrin domain containing 3 (NLRP3) inflammasome pathway activation. Following MCC950 treatment, knee joint synovial tissues were subjected to Western blot to evaluate NLRP3 inflammasome activation. (a) Representative Western blot bands of NLRP3 and caspase‐1 p20 expression in vehicle‐treated, MCC950‐treated and normal group. (b) Statistical analysis of relative target protein expression/internal control ratio. Data were recorded as mean ± standard deviation (s.d.), *P < 0·05.

Figure 6

Effects of MCC950 treatment on proinflammatory cytokine production. (a) Interleukin (IL)‐1β levels in knee joint synovia and sera. (b) Tumour necrosis factor (TNF) levels in knee joint synovia and sera. (c) IL‐6 levels in knee joint synovia and sera. Data were recorded as mean ± standard deviation (s.d.), **P < 0·01, NS = not significant.

Figure 7

MCC950 treatment affected Nucleotide‐binding, oligomerization domain (NOD)‐like receptor family, pyrin domain containing 3 (NLRP3) inflammasome pathway activation. Cell lysates and supernatants were prepared and subjected to Western blot. (a) Representative Western blot bands of NLRP3, caspase‐1 p20, interleukin (IL)‐1β and pro‐IL‐1β in normal, lipopolysaccharide + adenosine triphosphate (LPS + ATP) and MCC950‐LPS + ATP groups. (b) Statistical analysis of relative target protein expression/internal control ratio. Data were recorded as mean ± standard deviation (s.d.), **P < 0·01, NS = not significant.