Fig 2: PRC2 core components contributed to Utx depletion-mediated anabolic chondrocytic activity.Gene track illustrating limited H3K27me3 chromatin occupation at regulator loci and promoter regions of Ezh2, Eed, and Suz12 genes in UtxKO chondrocytes (red) as compared to WT cells (blue) (a). Utx loss repressed the H3K27me3 enrichment at the Ezh2 promoter but increased that at the Eed and Suz12 promoters (b). It increased abundance of Ezh2, but reduced that of Eed, Suz12, and H3K27me3 (c). Evident Ezh2 immunostaining and weak Eed2 and Suz12 immunoactivity in UtxKO cartilage (d), scale bar, 10 µm. Forced Eed or Suz12 expression increased H3K27me3 levels (e, f) and H3K27me3 occupancy at the Sox9 promoter (g). Chondrocytic marker gene expression (h) and ECM production (i, j scale bar, 500 µm) in UtxKO chondrocytes were suppressed by forced expression of both components. Cells were transfected with Eed or Suz12 expression vectors or empty vectors. RT-PCR was conducted upon transfection for 24 h. Micromass for Alcian blue staining were incubated for 7 days. Cell cultures were harvested from three mice and experiments were repeated three times. Data are expressed as mean ± standard errors. *P < 0.05; **P < 0.001.

Fig 3: Utx knockout promoted ECM synthesis and protected articular cartilage integrity.Schematic drawing of the generation of chondrocyte-specific Utx knockout mice; exon 24 of Utx, containing the JmjC domain of the enzyme, was floxed and cleavage of these sites resulted in the transcription/translation of inactive Utx. Through mating Utx mice with Col2-Cre mice, its expression is further under control of the cartilage-specific collagen type II gene promoter (a). Confirmation of proper genotypes of UtxKO and WT mice; heterozygous UtxKO mice carried flox constructs corresponding to 430 and 249 bp, whereas homozygous UtxKO mice only expressed amplicons corresponding to 430 bp. Absence of a 100 bp PCR amplicon in WT mice, corresponding to the Cre construct, but presence in UtxKO mice (b). Appearance and hair color of UtxKO mice were similar to WT mice (c). mRNA expression (d) and protein abundance (e), respectively, confirming the absence of Utx and reduced H3K27me3 levels in UtxKO mice. Very faint Utx and H3K27me3 immunostaining in articular cartilage of KO mice (f); scale bar, 10 µm. Utx loss promoted Col2a1 and Acan expression (g). Safranin-O staining of articular cartilage (h; scale bar, 100 µm) and increased relative thickness of articular cartilage, uncalcified portion (i), calcified zone (j), and density of articular chondrocytes (k), respectively, in UtxKO mice. Articular cartilage degeneration and decreased OARSI scores (l) together with osteophyte formation (arrows) in 9-month-old mice (m). Data are expressed as mean ± standard errors calculated from five to six mice. *P < 0.05; **P < 0.001.

Fig 4: UTX-dependent reduction in chondrocytic activities.Safranin-O staining of macroscopically normal human articular cartilage next to severely osteoarthritic tissue. Quantification of UTX mRNA abundance in both tissues revealed elevated UTX expression in gonarthrotic cartilage (a); scale bar, 200 µm. Strong UTX (b) and H3K27me3 (c) immunostaining in osteoarthritic chondrocytes, next to quantified protein expression from 34 donors; scale bar, 20 µm (low magnification), 10 µm (high magnification). Forced UTX expression increased H3K27me3 levels and reduced Sox9 abundance in articular chondrocytes (d). Utx gain-of-function upregulated H3K27me3 enrichment at the Sox9 promoter (e), but reduced Sox9, Col2a1 and Acan expression (f), respectively, and glycosaminoglycan synthesis (g) quantified by Alcian blue staining; scale bar, 500 µm. Utx knockdown repressed H3K27me3 levels and promoted chondrocytic activity. Cells were transfected with Utx RNAi or cDNA or scramble control for 24 hours. RT-PCR was conducted upon transfection for 24 h. Micromass for Alcian blue staining were incubated for 7 days. Culture experiments were conducted from three to five mice and data are expressed as mean ± standard error; *P < 0.05; **P < 0.001. SC scrambled control.

Fig 5: UtxKO mice showed few signs of collagenase or destabilized medial meniscus-induced gonarthrosis.Upon OA induction, UtxKO mice developed weaker symptoms, including articular cartilage degradation (a; scale bar, 20 µm) and quantitative OARSI scores (b) together with synovitis (c; scale bar, 100 µm) and synovitis scores (d). Utx loss compromised osteophyte formation (e scale bar, 500 µm), osteophyte volume (f) and loss in subchondral BMD and BV/TV (g) of UtxKO mice. Eight weeks DMM postoperatively, UtxKO mice showed less cartilage damage and synovitis (h scale bar, 20 µm) together with decreased OARSI scores (i) and synovitis scores (j) than WT mice. Osteophyte formation (k), osteophyte volume (l), and subchondral bone BMD (m) were compromised in UtxKO mice. Data are expressed as mean ± standard errors calculated from five to six mice. *P < 0.05; **P < 0.001.