Estrogen-related receptor γ causes osteoarthritis by upregulating extracellular matrix-degrading enzymes

Abstract

The estrogen-related receptor (ERR) family of orphan nuclear receptor is composed of ERRα, ERRβ, and ERRγ, which are known to regulate various isoform-specific functions under normal and pathophysiological conditions. Here, we investigate the involvement of ERRs in the pathogenesis of osteoarthritis (OA) in mice. Among ERR family members, ERRγ is markedly upregulated in cartilage from human OA patients and various mouse models of OA. Adenovirus-mediated overexpression of ERRγ in mouse knee joint or transgenic expression of ERRγ in cartilage leads to OA. ERRγ overexpression in chondrocytes directly upregulates matrix metalloproteinase (MMP)-3 and MMP13, which are known to play crucial roles in cartilage destruction in OA. In contrast, genetic ablation of Esrrg or shRNA-mediated downregulation of Esrrg in joint tissues abrogates experimental OA in mice. Our results collectively indicate that ERRγ is a novel catabolic regulator of OA pathogenesis.

Fig. 1

ERRγ is upregulated in OA cartilage from humans and mouse models. a–c qRT-PCR analysis (n = 11) was used to assess the mRNA expression levels of ERR isoforms in undamaged (UD) and damaged (D) regions of human OA cartilage (a). Representative images (n = 10 patients) show alcian blue staining of sulfated proteoglycans and immunostaining of ERRγ in UD and D regions of human OA cartilage (b). ICRS (International Cartilage Repair Society) grade of human OA cartilage sourced from individuals undergoing arthroplasty (c). d–f ERR mRNA levels were quantified by qRT-PCR (n = 10 mice per group) in cartilage tissues from mice subjected to sham operation or DMM surgery (d). Cartilage destruction was scored by OARSI grading (e). Representative images (n = 12 mice per group) of safranin-O-stained and ERRγ-immunostained OA cartilage from DMM-operated mice or cartilage from sham-operated mice (f). Values are presented as means ± s.e.m. (*P < 0.01 and **P < 0.001; ns, not significant). Two-tailed t test (a,d) and Mann–Whitney U test (e). Scale bar: 50 μm

Fig. 2

Signaling upstream of ERRγ expression. a qRT-PCR analysis (n ≥ 8) of IL-6 in chondrocytes treated with IL-1β and infected with 800 MOI of control virus (Ad-C) or the indicated MOI of Ad-Epas1 or Ad-Slc3918. b, c Detection of IL-6 in chondrocytes treated with IL-1β (1 ng ml^–1) by immunofluorescence microscopy (b) and Western blotting (c). d, e Western blot (d) and qRT-PCR (e) analysis of ERRγ, MMP3, and MMP13 in chondrocytes treated with IL-6 (n = 10). MMP3 and MMP13 proteins in conditioned medium of chondrocyte cultures were detected by Western blotting. f qRT-PCR analysis (n ≥ 8) of ERRγ mRNA levels in chondrocytes treated with IL-1β (left panel). qRT-PCR analysis (n ≥ 8) of ERRγ mRNA levels in chondrocytes treated with IL-6 siRNA and exposed to IL-1β (middle panel). qRT-PCR analysis (n ≥ 8) of ERRγ mRNA levels in chondrocytes treated with AG490 and exposed to IL-6 (right panel). Values are presented as means ± s.e.m. (*P < 0.05, **P < 0.005, and ***P < 0.0005. One-way ANOVA)

Fig. 3

Overexpression of ERRγ in joint tissue causes OA in mice. a, b WT mice were IA-injected with Ad-C (control) or Ad-Esrrg to overexpress ERRγ in the joint tissues. Three weeks after the first injection, sections of mouse joint tissues were immunostained for ERRγ (a). Cartilage destruction and synovial inflammation were determined by safranin-O staining and H&E staining, respectively (b). c, d WT mice were IA-injected with Ad-C (control) or Ad-Esrrg to overexpress ERRγ in mouse knee joint tissues. Mice were killed 8 weeks after the first injection. Representative images of safranin-O staining, showing the whole joint (40×), subchondral bone sclerosis and osteophyte formation (200×), and cartilage destruction (400×) (c). OARSI grade (for cartilage destruction), subchondral bone plate thickness (for subchondral bone sclerosis), and osteophyte size were quantified (d; n ≥ 5 mice per group). Values are presented as means ± s.e.m. (*P < 0.005). Two-tailed t test and Mann–Whitney U test. Scale bar: 50 μm. C, cartilage; S, subchondral bone; O, osteophyte

Fig. 4

Col2a1-Esrrg Tg mice exhibit enhanced experimental OA. a Genotypes and ERRγ mRNA and protein levels were determined from primary cultured chondrocytes isolated from Col2a1-Esrrg Tg mice and WT littermates. GAPDH and extracellular signal-regulated kinase (ERK) were used as loading controls. b Immunostaining for ERRγ protein in cartilage of Col2a1-Esrrg Tg mice and WT littermates. c, d Col2a1-Esrrg Tg mice and WT littermates were subjected to sham operation or DMM surgery. OARSI grade, subchondral bone plate thickness, and osteophyte size were quantified in WT and Col2a1-Esrrg TG mice (n = 13 mice per group), mice 8 weeks after sham operation, or DMM surgery (c). Representative images of cartilage destruction, subchondral bone sclerosis, and osteophyte formation determined by safranin-O staining (d). Values are presented as means ± s.e.m. (*P < 0.001 and **P < 0.0001). Two-tailed t test and Mann–Whitney U test. Scale bar: 50 μm

Fig. 5

Genetic ablation of Esrrg abrogates OA pathogenesis in mice. a Immunostaining for ERRγ proteins in cartilage of Esrrg ^+/– and WT littermates. b, c Esrrg ^+/– mice and WT littermates were subjected to sham operation or DMM surgery. OARSI grade, osteophyte size, and subchondral bone plate thickness were quantified in WT and Esrrg ^+/– mice (n = 14 mice per group) (b). Representative images of safranin-O-stained joint sections showing the whole joint (40×), subchondral bone sclerosis and osteophyte size (200×), and cartilage (400×) (c). Values are presented as means ± s.e.m. Two-tailed t test and Mann–Whitney U test. Scale bar: 50 μm

Fig. 6

Silencing of Esrrg in joint tissues abrogates OA pathogenesis in mice. a Immunostaining for ERRγ in cartilage sections of DMM-operated mice given an IA injection of Ad-C or Ad-shEsrrg. b, c WT mice subjected to sham operation or DMM surgery were IA-injected with Ad-C (n = 11 mice) as a control or Ad-shEsrrg (n = 11 mice) to knockdown ERRγ in joint tissues. Cartilage sections were subjected to safranin-O staining (c), and determination of OARSI grade, osteophyte size, and subchondral bone plate thickness at 8 weeks after sham operation or DMM surgery (b). Values are presented as means ± s.e.m. Two-tailed t test and Mann–Whitney U test. Scale bar: 50 μm

Fig. 7

ERRγ directly upregulates MMP3 and MMP13 in chondrocytes. a The mRNA (left) and protein (right) levels of ERRγ, MMP3, and MMP13 in primary cultured chondrocytes infected with Ad-C (800 MOI) or the indicated MOI of Ad-Esrrg for 36 h (n = 12). b Representative images showing immunostaining for MMP3, MMP13, SOX9, and type II collagen (Coll-II) in cartilage sections of mice obtained after 3 weekly IA injections with Ad-C or Ad-Esrrg (n = 12 mice per group). c The ERRE sequences in the promoter regions of Mmp3 and Mmp13 are indicated. d Quantitative ChIP assays were performed, and the data are presented as fold-changes relative to each input (n ≥ 4). e qRT-PCR (n = 11) analyses of MMP3 and MMP13 expression in chondrocytes treated with IL-6 (100 ng ml^–1) in the presence of the indicated concentrations of GSK5182, an inverse agonist of ERRγ. f Sham-operated or DMM-operated mice were IP-injected with GSK5182 and killed 8 weeks after the operation. Representative images of safranin-O staining and scoring of OARSI grade, osteophyte size, and subchondral bone plate thickness (n = 10 mice per group). Values are presented as means ± s.e.m. (*P < 0.05, **P < 0.005, and ***P < 0.0005). One-way ANOVA (a,d,e), two-tailed t test (f), and Mann–Whitney U test (f). Scale bar: 50 μm