Osteoarthritis (OA) is characterized by cartilage degradation, pain, and synovitis.1 Joint inflammation driven by cytokines has been demonstrated to cause cartilage degradation and pain.2 However, approaches to neutralize cytokines, such as IL-1 and TNF-α, known to be involved in OA have shown poor clinical efficacy.3 There is an unmet clinical need to find better anti-inflammatory and pain targets for OA therapy and to elucidate the role of other cytokines in OA pathogenesis. Previous studies have shown that IL-22 and its receptor IL-22R play central roles in inflammation and diseases such as psoriasis, ulcerative colitis, graft-versus-host disease, certain infections and tumors, as well as in liver and pancreas damage.4,5 The role of IL-22/IL-22R and the potential for therapeutic targeting of both proteins in OA remain largely unknown, which we sought to investigate.

We first examined human OA tissues to investigate whether IL-22/IL-22R expression levels change in disease. IL-22 was increased in the synovial fluid (SF) but not in the sera of OA patients compared to non-OA patients (Fig. 1a, b). Protein and mRNA expression of IL-22R was elevated in human chondrocytes isolated from OA patients (Fig. 1c–e). However, IL-22 protein and mRNA expression was only increased in fibroblast-like synoviocytes (FLS) isolated from OA patients (Fig. 1f–h). The increased concentration of IL-22 in the SF but not in the sera of OA patients seems to suggest that this cytokine plays a local role in OA joints with tissue-specific expression. It seems plausible that IL-22 produced by FLS in OA joints is important for disease progression. FLS-produced IL-22 may act mainly on IL-22R on chondrocytes, as indicated by the elevated expression levels of the receptor in human OA chondrocytes. Although IL-22/IL-22R have been reported to be increased in inflamed OA synovium and linked with increased protease expression,6,7 further studies investigating the precise downstream signaling of IL-22/IL-22R in chondrocytes need to be conducted.

Fig. 1
figure 1

Targeting IL-22 signaling protects against experimental osteoarthritis. IL-22 concentration (pg/mL) in a SF and b serum of non-OA and OA patients (n = 25). c IL-22 mRNA, d IL-22R mRNA and e IL-22 and IL-22R protein expression in isolated human chondrocytes from non-OA and OA patients (n = 18). f IL-22 mRNA, g IL-22R mRNA and h IL-22 and IL-22R protein expression in isolated FLS from non-OA and OA patients (n = 18). i, j OARSI scoring of cartilage and k von Frey pain assessment of sham- or DMM-operated IL-22Rfl/fl control mice and IL-22RAcan Cre-ERT2 mice (12 weeks post surgery end timepoint) (n = 20). l, m OARSI scoring of cartilage and n von Frey pain assessment from sham- or DMM-operated WT mice treated i.a. with either IgG1 (control; 50 µg per mouse; 3 times per week for 12 weeks post surgery) or αIL-22R (50 µg per mouse; 3 times per week for 12 weeks post surgery) (n = 20). All RT-qPCR gene expression levels were normalized to the endogenous level of 18S in the respective groups. Data are expressed as the mean ± SEM with two-tailed t-test or two-way analysis of variance followed by the Tukey-Kramer test or repeated measures 2-way ANOVA with Bonferroni’s post hoc tests. n indicates the number of human specimens or mice per group. NS nonsignificant. ***p < 0.001 or ****p < 0.0001 is represented in all figures

Full size image

Having observed the tissue-specific increase in IL-22 in chondrocytes and IL-22R in FLS from OA patients, we next investigated whether these results had an in vivo relevance during disease pathogenesis. We tested this hypothesis first by successfully generating inducible IL-22R chondrocyte-specific KO mice (IL-22RAcan Cre-ERT2) (Supplementary Fig. S1a, b). IL-22RAcan Cre-ERT2 mice displayed decreased cartilage degradation, synovitis, osteophyte maturity and pain compared to IL-22Rfl/fl control mice post experimental OA (surgical destabilization of the medial meniscus (DMM)) (Fig. 1i–k and Supplementary Fig. S1c–e). We also successfully generated inducible IL-22 FLS-specific KO mice (IL-22Col1a2 Cre-ERT2) (Supplementary Fig. S2a, b). IL-22Col1a2 Cre-ERT2 mice demonstrated reduced disease outcomes and pain compared to IL-22fl/fl control mice post DMM surgery (Supplementary Fig. S2c–h). To our knowledge, this is the first set of in vivo data that show, using tissue-specific KO mice, the pathogenic role of IL-22/IL-22R in both OA disease progression and OA-related pain. IL-22 in RA joints has both beneficial8 and pathogenic9 roles; together with our results, this may suggest that IL-22 is part of divergent inflammatory responses orchestrated by different joint cells in OA compared to RA.

Next, we wanted to investigate whether therapeutically neutralizing IL-22 and IL-22R may attenuate OA in vivo. WT mice treated with an IL-22R neutralizing antibody demonstrated decreased cartilage degradation, synovitis, osteophyte maturity and pain compared to IgG1-treated control mice post DMM surgery (Fig. 1l, m and Supplementary Fig. S3a–c). Similarly, WT mice treated with an anti-IL-22 antibody displayed reduced disease outcomes (Supplementary Fig. S4a–f). Our in vivo studies also showed the possible potential of using anti-IL-22/IL-22R antibodies to treat OA and its related pain. A number of basic studies and clinical trials have shown the benefits of targeting IL-22/IL-22R in systemic immune diseases.4 Our study indicates a further benefit of administering anti-IL-22/IL-22R in the joint to avoid any adverse systemic effects.10

Together, our data reveal the cell-specific pathogenic role of IL-22 (FLS specific) and IL-22R (chondrocyte specific) in OA. Targeting both IL-22 and IL-22R seems to be a plausible treatment option for OA and pain.