Tankyrase inhibition preserves osteoarthritic cartilage by coordinating cartilage matrix anabolism via effects on SOX9 PARylation

Kim, Sukyeong; Han, Sangbin; Kim, Yeongjae; Kim, Hyeon-Seop; Gu, Young-Ran; Kang, Donghyun; Cho, Yongsik; Kim, Hyeonkyeong; Lee, Jeeyeon; Seo, Yeyoung; Chang, Moon Jong; Chang, Chong Bum; Kang, Seung-Baik; Kim, Jin-Hong

doi:10.1038/s41467-019-12910-2

Download PDF

Article
Open access
Published: 25 October 2019

Tankyrase inhibition preserves osteoarthritic cartilage by coordinating cartilage matrix anabolism via effects on SOX9 PARylation

Nature Communications volume 10, Article number: 4898 (2019) Cite this article

9385 Accesses
28 Citations
6 Altmetric
Metrics details

Subjects

Abstract

Osteoarthritis (OA) is a prevalent degenerative disease, which involves progressive and irreversible destruction of cartilage matrix. Despite efforts to reconstruct cartilage matrix in osteoarthritic joints, it has been a difficult task as adult cartilage exhibits marginal repair capacity. Here we report the identification of tankyrase as a regulator of the cartilage anabolism axis based on systems-level factor analysis of mouse reference populations. Tankyrase inhibition drives the expression of a cartilage-signature matrisome and elicits a transcriptomic pattern that is inversely correlated with OA progression. Furthermore, tankyrase inhibitors ameliorate surgically induced OA in mice, and stem cell transplantation coupled with tankyrase knockdown results in superior regeneration of cartilage lesions. Mechanistically, the pro-regenerative features of tankyrase inhibition are mainly triggered by uncoupling SOX9 from a poly(ADP-ribosyl)ation (PARylation)-dependent protein degradation pathway. Our findings provide insights into the development of future OA therapies aimed at reconstruction of articular cartilage.

Targeting Fascin1 maintains chondrocytes phenotype and attenuates osteoarthritis development

Article Open access 04 September 2024

Kindlin-2 preserves integrity of the articular cartilage to protect against osteoarthritis

Article 10 February 2022

Dual functions of microRNA-17 in maintaining cartilage homeostasis and protection against osteoarthritis

Article Open access 04 May 2022

Introduction

Osteoarthritis (OA) is the most common of all arthritis and one of the leading causes of chronic disability in elderly populations¹. The prevalence of OA is increasing worldwide, imposing a tremendous medical and socioeconomic burden. OA is primarily characterized with the loss of proteoglycan content and degradation of collagen in the cartilage matrix². The degeneration of cartilage matrix during OA development eventually causes failure in the load-bearing functions of articular cartilage.

Disease-modifying OA therapy essentially aims at prolonged and functional repair of articular cartilage^3,4. Cartilage repair is steered by chondrogenic differentiation of resident progenitor cells and extracellular matrix (ECM) anabolism by the differentiated chondrocytes⁵; however, these abilities generally decline with aging and disease progression, leaving a marginal repair capacity in osteoarthritic joints². To compensate for the lack of intrinsic repairing abilities, stem cell-based therapies relying on the enrichment of mesenchymal stem cell (MSC) population via marrow stimulation or direct transplantation into the damaged region have been widely attempted^6,7,8. These cells, however, tend to rapidly lose chondrogenic features after their differentiation, accompanied with the cessation of cartilage matrix molecule synthesis⁹.

The extensive molecular exploration of chondrogenesis has led to the identification of several key factors involved in the regulation of chondrocytic phenotypes^10,11,12. In particular, SOX9 was identified as a master transcription factor of chondrogenesis expressed in the developing cartilage^13,14,15. SOX9 is most actively expressed during chondrogenic differentiation of multipotent mesenchymal progenitors and its activity is repressed during the hypertrophic maturation of chondrocytes in growth plates^16,17. SOX9 remains expressed in chondrocytes of articular cartilage and regulates ECM homeostasis through the regulation of the expression of major constituents of the cartilage-specific matrix, including type II, type IX, and type XI collagen as well as aggrecan¹⁸. However, SOX9 transcriptional activity is downregulated in osteoarthritic chondrocytes, thereby resulting in the decrease in the matrix anabolism of articular cartilage¹⁹. In fact, forced overexpression of SOX9 in MSCs prior to its transplantation in cartilage defects has demonstrated its beneficial roles in promoting chondrogenic differentiation^20,21. Despite the disease-modifying potentials of SOX9, no effective ways have been described to fully activate endogenous SOX9 for the purpose of cartilage repair.

Tankyrase (encoded by TNKS or TNKS2) belongs to the poly(ADP-ribose) polymerase (PARP) superfamily^22,23. Tankyrase interacts with target proteins through their unique ankyrin repeat protein-interaction domain and catalyzes the addition of ADP-ribose moieties onto target proteins by using NAD⁺ as the substrate²⁴. This post-translational modification process, termed as poly(ADP-ribosyl)ation (PARylation), is known to regulate the activity and stability of substrate proteins²². At present, several highly specific and potent tankyrase inhibitors are available that may provide opportunities to develop therapies to modulate tankyrase activity²⁴.

Here, we demonstrate that tankyrase inhibition substantially stimulates pro-anabolic pathways in articular cartilage. Tankyrase inhibition augments cartilage-specific ECM synthesis by chondrocytes and enhances the chondrogenic differentiation of MSCs. The controlled delivery of tankyrase inhibitors to knee joints ameliorates surgically induced OA in mice. MSCs with tankyrase knockdown exhibit superior regeneration capacity with robust ECM accumulation upon their transplantation to cartilage defects. The pro-anabolic effect of tankyrase suppression in cartilage is mainly mediated by the activating cartilage-signature transcriptional network via decoupling of SOX9 from PARylation-dependent degradation.

Results

Tankyrase is a regulator of cartilage matrix anabolism

To screen for a key regulatory factor that could be targeted to stimulate cartilage matrix anabolism, we conducted genetic analysis on transcriptomes of mouse reference populations by using post hoc factor analysis. First, we assessed the transcriptional variance in the cartilage tissues of 16 strains of BXD mice²⁵. We noted that among 21 cartilage matrix genes listed up by Heinegard and Saxne²⁶, 14 cartilage matrix genes showed strong positive correlation in their transcript abundance (Fig. 1a). These high correlations were absent in organs without cartilaginous functions, such as bone femur, kidney, lung, and brain (Supplementary Fig. 1a). A genome-wide co-expression analysis by using BXD cartilage transcriptome datasets similarly resulted in the identification of a highly correlated cluster enriched with cartilage matrix genes (Supplementary Fig. 1b). We then attempted to extract a common axis underlying cartilage anabolism by performing a principal component analysis on 14 highly intercorrelated cartilage matrix genes (see black box in Fig. 1a). The first axis identified (Factor 1) essentially reflects the state of cartilage matrix anabolism (Fig. 1b). Next, we used the Gene Ontology (GO) Resource to annotate the 16,074 genes included in the cartilage transcriptome, of which 6824 belonged to the GO terms: “catalytic activity”, “DNA-binding transcription factor activity”, or “regulation of signaling”, which can potentially have regulatory functions on the collective expression of a set of genes. We then computed Pearson’s correlation coefficients between Factor 1 and these 6824 genes with potential regulatory functions. Tankyrase showed striking negative correlations with the anabolic axis and with individual cartilage matrix genes and was therefore investigated further (Fig. 1b, c and Supplementary Fig. 1c, d).

We examined the potential regulatory role of tankyrase in cartilage anabolism. Knockdown of both Tnks and Tnks2 collectively induced the expression of cartilage-specific matrix genes in primary cultured mouse chondrocytes (Fig. 1d–f). On the other hand, the individual knockdown of Tnks or Tnks2 failed to increase the cartilage matrix anabolism, suggestive of the redundant roles of TNKS and TNKS2 in this regulation (Fig. 1e). Treatment with XAV939 or IWR-1, highly specific and potent TNKS/2 inhibitors²⁷, also increased the expression of cartilage-specific matrix genes in chondrocytes (Fig. 1g–j). However, the PARP1/2 inhibitor, ABT-888, failed to increase their expressions (Fig. 1i, j).

To comprehensively elucidate the effect of tankyrase inhibition at the whole transcriptome level, we performed RNA sequencing for chondrocytes treated with siRNAs targeting Tnks and Tnks2, XAV939, or IWR-1 (Supplementary Fig. 2a, b). Of the 711 differentially expressed genes (DEGs) identified in this analysis, the 527 genes (74.12%) were consistently upregulated or downregulated in all three tankyrase inhibition groups compared with their respective control groups (Supplementary Fig. 2c). GO analysis of the commonly upregulated genes in response to tankyrase inhibition revealed a strong association with terms related to cartilage development (Supplementary Fig. 2d). Next, we generated a comprehensive list of cartilage-signature genes by utilizing public transcriptome datasets¹⁸. Tankyrase inhibition induced strong transcription of key cartilage-identity genes (Supplementary Fig. 2e). In addition, gene set enrichment analysis (GSEA) revealed that cartilage-signature genes were positively enriched in the whole transcriptome obtained from chondrocytes treated with siTnks and siTnks2 or tankyrase inhibitors, XAV939 and IWR-1 (Fig. 1k, l). Thus, tankyrase inhibition promotes cartilage matrix anabolism and strengthens the overall chondrogenic features in chondrocytes.

SOX9 interacts with tankyrase

To understand the molecular mechanism underlying the effect of tankyrase inhibition on cartilage anabolism, we aimed to identify tankyrase substrates responsible for the regulation of cartilage matrix genes. Axin, a well-established target of tankyrase, is subjected to proteasomal degradation upon PARylation-dependent ubiquitination²⁷. In fact, Tnks2 was the 12th highest among the 431 genes enriched for the GO term “Wnt signaling pathway” in terms of Pearson’s correlation coefficients with the cartilage anabolic axis (Supplementary Table 1). Tankyrase inhibition caused a minor reduction in the β-catenin protein level in chondrocytes (Fig. 2a); the effects of tankyrase inhibition on β-catenin stability and activity were more pronounced in chondrocytes treated with exogenous Wnt ligands (Fig. 2b–e). Therefore, we sought to determine whether the effect of tankyrase inhibition on cartilage anabolism was associated with its effect on Wnt/β-catenin signaling inhibition. siRNA-mediated knockdown of β-catenin had no significant effect on the expression of cartilage matrisome (Fig. 2f). Similarly, treatment with Dkk-1, which antagonizes canonical Wnt ligands²⁸, or IWP-2, a porcupine inhibitor that blocks the secretion of canonical and noncanonical Wnts²⁹, did not affect the expression of Col2a1 and Acan (Fig. 2g, h). Moreover, the upregulation of Col2a1 and Acan induced by XAV939 treatment was not affected by siRNA-mediated knockdown of β-catenin in chondrocytes (Fig. 2i). Therefore, in the absence of exogenous Wnt ligands, the effect of tankyrase inhibition on cartilage anabolism was independent of Wnt/β-catenin signaling.

We then examined the role of β-catenin in regulating cartilage anabolism in the presence of exogenous Wnt ligands. Treatment of chondrocytes with Wnt-3a, a canonical Wnt ligand, significantly reduced the expression of Col2a1 and Acan (Fig. 2g, h). The addition of recombinant Dkk-1 effectively abolished the Wnt-3a-mediated suppression of these chondrogenic genes. Meanwhile, IWP-2 did not rescue the expression of Col2a1 and Acan, which is consistent with its inability to block exogenously added Wnt ligands. Therefore, in the presence of abundant Wnt ligands, tankyrase inhibition may exert an additional effect to promote cartilage matrix anabolism in a Wnt/β-catenin signaling-dependent manner.

Nonetheless, because our study was mainly conducted in a context where β-catenin signaling plays a marginal role in cartilage matrix gene regulation, we hypothesized that there may be an alternative mechanism, whereby tankyrase plays a regulatory role in cartilage-specific anabolic signaling. To find tankyrase-binding substrates that regulate cartilage matrix anabolism, we performed liquid chromatography–tandem mass spectrometry (LC–MS/MS) analysis for the proteome co-immunoprecipitated with the endogenous tankyrase in chondrocytes (Fig. 3a and Supplementary Data 1). We considered proteins that are detected in more than one biological replicate as putative tankyrase-interacting proteins (Fig. 3b). Among these binding partners, candidate substrates were further screened by using a tankyrase-targeting score (TTS) system (Supplementary Data 2)³⁰. Ingenuity-pathway analysis (IPA) revealed four candidate proteins above the TTS cutoff (≥0.385) that fell into the chondrogenesis category (Fig. 3c). IUPred disorder score³¹ was used to filter unlikely targets, wherein tankyrase-binding motifs are positioned in a highly structured region (Fig. 3d). SOX9, which is known as the master transcription factor of chondrogenesis³², was highly predicted as a substrate candidate, exhibiting both high TTS and disorder scores. Endogenous interactions between tankyrase and SOX9 in chondrocytes were confirmed by co-immunoprecipitation assay and in situ proximity ligation assay (PLA) (Fig. 3e, f). Moreover, our cell-based assay indicated that SOX9 binds to both TNKS and TNKS2 (Fig. 3g). The two tankyrase-binding domains (TBDs) of SOX9, designated as TBD1 and TBD2, are highly conserved among vertebrates (Fig. 3h). Based on structural simulations, TBD1 and TBD2 peptides fit into the binding pocket located central to the ankyrin repeat cluster (ARC) IV domain of tankyrase where known substrates, SH3 domain-binding protein (3BP2) and myeloid cell leukemia sequence 1 protein (MCL1), are aligned (Fig. 3i). The deletion of either TBD1 or TBD2 resulted in the reduction in the binding affinity of SOX9 for tankyrase (Fig. 3j), while simultaneous deletion of both TBDs nearly abolished this association (Fig. 3k).

Tankyrase PARylates SOX9 and promotes its degradation

We investigated whether tankyrase binding to SOX9 is coupled to PARylation of SOX9. Wild-type SOX9 underwent extensive PARylation, whereas SOX9 mutant missing both TBDs exhibited a markedly reduced PARylation level (Fig. 3l). Consistently, the extent of SOX9 PARylation was also reduced by the pharmacological inhibition of tankyrase activity through treatment with XAV939 (Fig. 3m). Tankyrase-dependent PARylation is generally linked to the degradation of substrate proteins²². In fact, tankyrase inhibition promoted SOX9 protein expression in chondrocytes (Fig. 3n, o). The SOX9 TBD mutant showed augmented stability compared with wild-type SOX9 (Fig. 3p, q). Taken together, the disruption of the physical interactions between tankyrase and SOX9 and the consequent abolishment of SOX9 PARylation results in the stabilization of SOX9.

We then investigated the mechanism whereby tankyrase-mediated PARylation of SOX9 affects its polyubiquitination and subsequent proteasomal degradation. HEK293 cells expressing FLAG-tagged human SOX9 were treated with XAV939 or vehicle. Overexpressed SOX9 underwent extensive polyubiquitination, whereas XAV939 treatment nearly abolished the ubiquitination of SOX9 in HEK293 cells (Fig. 3r), indicating that tankyrase-mediated PARylation is an essential modification that leads to SOX9 ubiquitination. To provide further evidence for a functional interaction between SOX9 and tankyrase in a physiological context, we examined the ubiquitination of endogenous SOX9 in primary chondrocytes, with or without tankyrase inhibition. Using a tandem ubiquitin-binding entity (TUBE) directed against polyubiquitin moieties, endogenous proteins with polyubiquitin chains were immunoprecipitated. Western blotting analysis revealed that endogenous SOX9 undergoes polyubiquitination and is degraded by the proteasome in chondrocytes, as MG132 treatment led to the accumulation of SOX9 and its polyubiquitinated forms (Fig. 3s). XAV939 treatment effectively reduced the polyubiquitinated forms of SOX9 to levels comparable with those observed in the absence of MG132. Together, our results show that SOX9 undergoes degradation through the ubiquitin–proteasome system and this can be effectively reversed by the inhibition of tankyrase activity.

To date, RNF146 is the only known E3 ubiquitin ligase that mediates PARylation-dependent ubiquitination and degradation of substrates^33,34,35. In particular, RNF146 is best known to regulate tankyrase-dependent Axin degradation and hence β-catenin stabilization³³. Consistent with this notion, shRNA or siRNA-mediated knockdown of Rnf146 effectively reduced TOPFlash activity and β-catenin level (Fig. 4a, b). However, unlike Tnks and Tnks2 double knockdown, Rnf146 knockdown in chondrocytes failed to increase SOX9 transcriptional activity, the expression of cartilage matrix genes, or SOX9 protein level (Fig. 4c–f). Our experimental findings were further supported by factor analysis results based on mouse reference populations. A total of 14 intercorrelated cartilage matrix genes exhibited insignificant correlation with Rnf146 (r = −0.12; P = 0.66; Fig. 4g). Our data suggest an intriguing possibility that PAR-dependent E3 ligases other than RNF146 may exist and regulate PARylation-dependent SOX9 regulation.

Tankyrase inhibition enhances SOX9 transcriptional activity

Next, we used a 4 × 48-p89 SOX9-dependent Col2a1 enhancer reporter³⁶ to investigate whether the increase in SOX9 levels with tankyrase inhibition enhances the overall transcriptional activity of SOX9. Double knockdown of Tnks and Tnks2, and nine different tankyrase-specific inhibitors, specifically increased the transcriptional activity of SOX9 in chondrocytes (Fig. 5a–c). Moreover, the overexpression of wild-type TNKS2 resulted in a marked reduction in the transcriptional activity of SOX9, while the catalytically inactive form of TNKS2 (TNKS2 M1054V) suppressed SOX9 activity to a moderate extent (Fig. 5d). SOX9 target genes³⁷ were overall upregulated upon tankyrase knockdown or inhibition at the whole transcriptome level (Fig. 5e). Meanwhile, SOX9 is known to bind to its own enhancer and autoregulate its expression³⁸. Indeed, Sox9 transcript levels also increased upon simultaneous knockdown of Tnks and Tnks2 in mouse chondrocytes (Supplementary Fig. 3). Pharmacological inhibition of tankyrase by XAV939 treatment did not have a statistically significant effect on Sox9 mRNA levels. To further confirm the proposed mechanism that tankyrase regulates SOX9 activity post-transcriptionally, we repeated the luciferase reporter assays by using the SOX9-dependent Col2a1 enhancer construct in cells constitutively expressing SOX9 mRNA under the control of a cytomegalovirus (CMV) promoter. Tankyrase inhibition by using siRNAs or drugs increased the transcriptional activity of exogenously expressed SOX9 in HEK293T cells (Fig. 5f, g). Furthermore, point mutations of Arg in the first position to Ala in both TBD1 and TBD2 of SOX9 synergistically enhanced the transcriptional activity of SOX9 (Fig. 5h), suggesting that disruption of the interaction between tankyrase and SOX9 is sufficient to enhance the transcriptional activity of SOX9. Cartilage matrix gene expression induced by tankyrase inhibition was completely abolished by SOX9 knockdown (Fig. 5i, j). Taken together, SOX9 serves as an essential target of tankyrase for the role of tankyrase as an anabolic regulator in chondrocytes.

Tankyrase inhibition ameliorates OA progression in mice

Our results suggest that tankyrase may perform a physiological role in the regulation of cartilage matrix homeostasis. As cartilage homeostasis is disrupted during OA development, we first characterized the expression of tankyrase in articular cartilage of knee OA patients. Tankyrase proteins were markedly upregulated in OA-affected cartilage, but were barely detectable in undamaged regions of osteoarthritic cartilage (Fig. 6a, b). In contrast, type II collagen and aggrecan expression levels were significantly reduced in the damaged regions of human OA cartilage, inversely correlating with the expression pattern of tankyrase (Fig. 6a). We then performed destabilization of the medial meniscus (DMM) surgery in mice as a model of post-traumatic OA. Two weeks after DMM, we observed no distinct damage in the cartilage, but detected robust expression of SOX9 and aggrecan (Supplementary Fig. 4a), consistent with studies that have reported a compensatory synthesis of matrix molecules during the early stage of OA^39,40. At this stage, tankyrase expression was not detected in the knee joint cartilage of the DMM-operated mice. However, at 8 weeks after DMM surgery, significant damage in the articular cartilage and loss of SOX9 and aggrecan expression were apparent, indicating a mid-to-late stage of OA (Supplementary Fig. 4a, b). A marked increase in tankyrase expression was detected in this condition, consistent with our observations on human OA cartilage.

The increased expression of tankyrase in OA cartilage suggested its possible involvement in OA pathogenesis. Thus, we investigated how tankyrase inhibition affects the expression of OA-associated genes (Supplementary Fig. 5). By utilizing public transcriptome datasets⁴¹, we generated a comprehensive list of OA-associated genes that are upregulated and downregulated in OA patients. Notably, OA-associated genes upregulated in patients were overall repressed in chondrocytes upon tankyrase inhibition (Supplementary Fig. 5a). In contrast, OA-associated genes suppressed in patients were strongly transactivated by tankyrase inhibition (Supplementary Fig. 5b). This inverted pattern of gene expression profiling was evident even at the whole transcriptome level (Fig. 6c, d).

Next, we assessed the in vivo effects of tankyrase inhibition on cartilage matrix homeostasis in surgically induced OA mouse model. For the stable and prolonged delivery of tankyrase inhibitors to mouse knee joints, we used injectable hydrogels made of ascorbyl palmitate⁴². Intra-articular (IA) injection of this hydrogel-based drug delivery system allowed controlled local release of the loaded small molecule to articular cartilage over 9 days (Supplementary Fig. 6a, b). IA administration of hydrogel-mediated XAV939 or IWR-1, the two representative tankyrase inhibitors with different modes of actions, resulted in a significant reduction in the degeneration of cartilage matrix caused by the DMM (Fig. 6e–g). A concomitant increase in type II collagen and aggrecan was observed (Fig. 6h), and the expression of SOX9 was retained in the cartilage treated with tankyrase inhibitors (Fig. 6i). In addition, we observed that IA delivery of tankyrase inhibitors effectively reduced the production of matrix metalloproteinase 13 (MMP13) and the specific expression of β-catenin in the superficial zone of articular cartilage^43,44 (Supplementary Fig. 6c). These experimental results are in line with the correlation analysis based on mouse reference populations, indicating that tankyrase exhibits a negative and positive correlation with cartilage matrix genes (Fig. 1b, c) and catabolic regulators (Supplementary Fig. 5c, d), respectively.

Based on the pro-anabolic effect of tankyrase inhibitors, we tested the potential of XAV939 to treat late-stage OA cartilage. In the mouse DMM model, early osteoarthritic lesions were observed 2 weeks after surgery⁴⁵, while 70% of mice had reached late-stage OA after 6 weeks from DMM surgery. XAV939 administration for additional 6 weeks resulted in the reduction in the cartilage destruction as compared with the vehicle-treated mice, which experienced further OA progression (Supplementary Fig. 6d–f). Taken together, our results indicate that tankyrase inhibitors effectively ameliorate cartilage destruction in mice through the attenuation of the imbalance between matrix anabolism and catabolism.

Tankyrase inhibition stimulates chondrogenesis of MSCs

As mesenchymal progenitor cells are responsible for the regenerative capacity of damaged cartilage^5,46, we investigated the role of tankyrase in the chondrogenic differentiation of MSCs. First, the expression pattern of Tnks and Tnks2 mRNA during chondrogenesis was examined by using a micromass culture of mesenchymal cells isolated from the limb buds of E11.5 mouse embryos. Along with the progression of in vitro chondrogenesis, Tnks and Tnks2 expression were markedly decreased, inversely correlating with the expression of Acan, a marker of chondrogenesis (Fig. 7a). We then analyzed a publicly available RNA-Seq dataset⁴⁷ generated from pellet cultures of human bone marrow-derived MSCs (hMSCs). TNKS and TNKS2 expression were markedly reduced over the course of the chondrogenic differentiation of hMSCs (Fig. 7b).

Next, we examined how tankyrase inhibition influences the chondrogenic differentiation of MSCs. The tankyrase inhibitors, XAV939 and IWR-1, effectively induced chondrogenic nodule formation in micromass cultures of mouse limb-bud mesenchymal cells (Fig. 7c), enhancing the amount of Alcian blue bound to sulfated glycosaminoglycans (Fig. 7d). The two tankyrase inhibitors significantly promoted Col2a1 and Acan expression, whereas the PARP1/2 inhibitor, ABT-888, failed to increase their expression (Fig. 7e), suggesting that the chondrogenic effect is specific to tankyrase inhibition and not to other PARP family member inhibition. Similarly, both pharmacological inhibition and double knockdown of TNKS and TNKS2 effectively enhanced the chondrogenic differentiation of hMSCs (Fig. 7f–h), with a concomitant upregulation of cartilage-specific proteins, such as type II collagen, aggrecan, and SOX9 (Supplementary Fig. 7a, b).

We next evaluated the effect of tankyrase inhibition on stem cell-based restoration of hyaline cartilage. A full-thickness osteochondral lesion was filled with a fibrin gel containing hMSCs transduced with control or TNKS and TNKS2 shRNAs. After 8 weeks, no positive signals for human-specific mitochondria were detected in osteochondral lesions filled with fibrin gel only, whereas defects transplanted with hMSC-control shRNA or hMSC-TNKS/2 shRNA displayed strong signals localized in the cytoplasm of transplanted cells (Fig. 7i and Supplementary Fig. 8a). Defects transplanted with hMSC-control shRNA failed to fully recover the organization of hyaline cartilage and exhibited features of fibrocartilage (Fig. 7i, j and Supplementary Fig. 8b, c). However, lesions implanted with hMSC-TNKS/2 shRNA showed regenerated hyaline cartilage, similar to the articular cartilage. The regenerative tissues derived from hMSC-TNKS/2 shRNA maintained stable knockdown of tankyrase protein (Fig. 7k) and exhibited robust expression of SOX9 and cartilage-specific matrix proteins (Fig. 7l), compared with the fibrocartilage originated from hMSC-control shRNA. Finally, β-catenin expression was minimal in lesions implanted with hMSC-control shRNA and in those implanted with hMSC-TNKS/2 shRNA (Fig. 7l).

Discussion

Recent progress in the understanding of the pathophysiology of OA has led to the identification of promising therapeutic targets, highlighting the potential for the development of disease-modifying osteoarthritis drugs (DMOADs)^{48,49,50,51,52,53,54,55,56,57}. The identification of MMP13⁵⁸ and a disintegrin and metalloproteinase with thrombospondin motifs 5 (ADAMTS5)^55,56 as destructive enzymes for type II collagen and aggrecan led to the active investigation of drugs that may delay OA advancement through the erosion of cartilage matrix^59,60. However, agents targeting matrix-degrading processes may be insufficient to restore cartilage ECM and accomplish functional repair of articular cartilage.

We demonstrated here that tankyrase is a key target molecule whose inhibition enhances the expression of cartilage-specific matrisome and contributes to reconstruction of cartilage matrix. Although no functional characterization of tankyrase was previously made in cartilage tissues, there has been an evidence pointing tankyrase as a genetic factor that may potentially affect OA susceptibility. A genome-wide association study indicates that copy-number deletion of tankyrase is associated with a decreased risk for OA⁶¹. Furthermore, Tnks is located within quantitative trait loci associated with spontaneous OA in STR/ORT mice⁶². Our study with quantitative genetic analysis of articular cartilage in BXD mouse reference population illustrates a significant degree of negative correlation between the expression of tankyrase and cartilage-signature genes responsible for ECM construction.

We revealed a previously unrecognized role of tankyrase in the regulation of the stability of SOX9—a master regulator of chondrogenic program^32,63. In addition to its key role in the regulation of the expression of cartilage matrix genes through its binding to their cis-regulatory enhancer elements, SOX9 was shown to be essential for establishing super-enhancer at chondrocyte-identity genes¹⁸. SOX9 plays an essential role in chondrocyte survival, as SOX9 knockout in chondrocytes triggered extensive apoptosis⁶⁴. SOX9 is downregulated in human osteoarthritic cartilage, and this event is thought to be associated with the cessation of cartilage matrix synthesis and chondrocyte dysfunctions in osteoarthritic cartilage^65,66. Our findings suggest that tankyrase inhibition may serve as an effective trigger in such OA cartilage to sustain the transcriptional activity of SOX9 and reinforce cartilage-signature gene expression.

Tankyrase associates with multiple target proteins and regulates various signaling pathways, suggestive of the involvement of factors other than SOX9 in the overall effect of tankyrase inhibition on OA pathogenesis. In fact, tankyrase interacts with TRF1⁶⁷, Axin²⁷, NUMA⁶⁸, PTEN⁶⁹, 3BP2⁷⁰, and AMOTs⁷¹, which are implicated in controlling telomere length, Wnt signaling, mitosis, AKT pathway, osteoclastogenesis, and Hippo pathway, respectively. High-throughput proteomic analysis recently revealed extensive tankyrase interactomes, including both enzymatic substrates and nonenzymatic effectors^72,73.

Consistent with the notion that Axin is a well-established tankyrase substrate, tankyrase inhibition effectively reduced β-catenin stability and activity in chondrocytes. Increased levels of β-catenin in OA cartilage have been thought to elicit catabolic profiles in chondrocytes and contribute to cartilage degeneration^74,75. Various β-catenin inhibitory therapeutics have been used at the preclinical^76,77,78 and clinical stage⁷⁹ (Phase 1: NCT02095548; Phase 2: NCT02536833, NCT03122860; Phase 3: NCT03928184), with promising outcomes in delaying cartilage destruction. The inhibitory role in Wnt/β-catenin signaling may contribute to the chondro-protective effects of tankyrase inhibitors. This notion is further supported by our findings highlighting the overall repression of catabolic factors such as MMP13 upon tankyrase inhibition. Therefore, our study suggests that tankyrase inhibition may serve as a potential therapeutic strategy to simultaneously modulate the two key signaling pathways in chondrocytes, SOX9 and β-catenin, in a direction that enhances cartilage matrix anabolism and inhibits matrix degeneration processes.

Tankyrase-mediated PARylation serves as a signal cue for the ubiquitin conjugation system, which couples to protein degradation²². RNF146 is the most prominent E3 ubiquitin ligase responsible for the recognition and ubiquitination of various PARylated substrates, including Axin³³, PTEN⁶⁹, and AMOTs⁷¹. RNF146 forms a complex with tankyrase and directly interacts with PARylated Axin through its WWE domain, thereby promoting its degradation^33,34. Although the knockdown of RNF146 effectively reduced β-catenin stability and activity, it affected neither SOX9 stability and activity nor cartilage matrix gene expression. Our findings suggest the potential for the presence of another PAR-dependent E3 ligase responsible for tankyrase-mediated SOX9 regulation. Our analysis of endogenous tankyrase interactomes reveals that thyroid hormone receptor interactor 12 (TRIP12) and HECT, UBA, and WWE domain-containing 1 (HUWE1) are the additional E3 ubiquitin ligases that form complexes with tankyrase in chondrocytes. Intriguingly, both TRIP12 and HUWE1 contain the WWE domain that is predicted to mediate interactions with PARylated substrates^33,80,81.

Although we demonstrated that tankyrase-mediated PARylation of SOX9 plays an essential role in regulating SOX9 ubiquitination and degradation, inhibiting tankyrase activity or mutating the tankyrase-binding site in SOX9 did not completely abolish the PARylation of SOX9. We speculate that the residual PARylation of SOX9 may involve modifications mediated by other PARP family members. Extensive molecular-level investigation is warranted to illustrate the role of other forms of SOX9 PARylation that are not mediated by tankyrase.

Taken together, our results have direct implications for the development of regenerative therapy for osteoarthritic cartilage lacking its load-bearing function as a result of cartilage matrix loss. We demonstrate that tankyrase inhibition stimulates cartilage-specific anabolism, illustrating an intriguing possibility for the reconstruction of cartilage at both molecular and organismal levels. The reparative capacity of tankyrase inhibition may be attributable to the combinatorial effects that invert OA-associated gene expression landscape in chondrocytes and stimulate chondrogenesis of MSCs. Our findings may provide fundamental insights into the role of tankyrase-mediated PARylation in fine-tuning the activity of transcription factors associated with development processes and disease pathogenesis. We expect our findings to provide a guideline for future OA therapeutic strategies circumventing surgical intervention and aiming at cartilage regeneration.

Methods

In silico analysis of multi-tissue BXD transcriptomes

Cartilage (GN208)²⁵, bone femur (GN411)⁸², kidney (GN118), lung (GN160)⁸³, and brain (GN123)⁸⁴ datasets were obtained from GeneNetwork (www.genenetwork.org). illuminaMousev1.db (http://bioconductor.org/packages/illuminaMousev1.db/) and illuminaMousev1p1.db (http://bioconductor.org/packages/illumina Mousev1p1.db/) were used for probe annotation in the cartilage and bone femur datasets, respectively. These two databases provided information on the quality grade of each probe, and whether a probe overlaps with any known single-nucleotide polymorphisms (SNPs). A probe that did not overlap with any known SNPs, perfectly and uniquely matched the target transcript, and also had the highest expression used for each transcript. The processed cartilage dataset with the expression profile of 16,074 genes across 16 BXD mouse strains was used to calculate the Pearson’s correlation coefficients between gene expression levels. The resulting correlation matrix was hierarchically clustered by using fastcluster package (https://cran.r-project.org/package=fastcluster) in R (complete linkage and Euclidean distance). For the data presented in Supplementary Fig. 1c, principal component analysis was performed by using the R stats package. For the kidney, lung, and brain datasets, we used the mouse4302.db (http://bioconductor.org/packages/mouse4302.db/) for probe annotation, and the probe with the highest expression was used for each transcript. For the data presented in Fig. 1a and Supplementary Fig. 1a, gene co-expression matrices were clustered by using the hierarchical clustering algorithm (complete linkage and uncentered correlation distance) in Cluster (http://bonsai.hgc.jp/~mdehoon/software/cluster/software.htm), and correlation heatmaps were drawn with Gitools⁸⁵. For the data presented in Fig. 1b, factor analysis was performed by using IBM SPSS Statistics (www.ibm.com/analytics/us/en/technology/spss/).

Primary culture of mouse articular chondrocytes

For the primary culture of mouse articular chondrocytes, cells were isolated from femoral condyles and tibial plateaus of 4–5-day-old ICR mice by 0.2% collagenase digestion⁸⁶. Chondrocytes were maintained in DMEM supplemented with 10% fetal bovine serum (FBS), 100 units/ml penicillin, and 100 μg/ml streptomycin, and cells were treated as indicated in each experiment. Transfection was performed with METAFECTENE PRO (Biontex) according to the manufacturer’s protocol. Small interfering RNAs (siRNAs) used for RNA interference (RNAi) in mouse articular chondrocytes are listed in Supplementary Table 2. All siRNAs, including negative control siRNA, were purchased from Bioneer. Recombinant mouse Wnt-3a (315-20) was purchased from PeproTech, and recombinant mouse Dkk-1 (5897-DK) was purchased from R&D Systems.

Reverse transcription polymerase chain reaction and quantitative PCR

Total RNAs were extracted by using TRI reagent (Molecular Research Center, Inc.). RNAs were reverse transcribed by using EasyScript Reverse Transcriptase (Transgen Biotech). Then, cDNA was amplified by PCR or qPCR with the primers listed in Supplementary Table 3. qPCR was performed with SYBR TOPreal qPCR 2× preMIX (Enzynomics) to determine transcript abundance. Transcript quantity was calculated by using the ∆∆C_t method, and Hprt or HPRT1 levels were used as housekeeping controls.

Whole-cell lysate preparation

Whole-cell lysates were prepared in radioimmunoprecipitation assay (RIPA) buffer (150 mM NaCl, 1% NP-40, 50 mM Tris, pH 8.0, 0.5% sodium deoxycholate, and 0.1% sodium dodecyl sulfate (SDS)) supplemented with a protease inhibitor cocktail (Sigma-Aldrich). The lysates were quantified by using a BCA assay and analyzed by SDS polyacrylamide gel electrophoresis (PAGE). Protein band intensity was quantified by densitometric analysis by using ImageJ⁸⁷.

Antibodies

Anti-FLAG tag antibody (3165) was purchased from Sigma-Aldrich. Antibodies against green fluorescent protein (GFP) (sc-9996), Sox9 (sc-20095), Sox9 (sc-166505), Tankyrase-1/2 (sc-8337), Tankyrase-1/2 (sc-365897), Ubiquitin (sc-8017), Actin (sc-1615), α-Tubulin (sc-23948), and Histone H3 (sc-517576) were obtained from Santa Cruz Biotechnology. Sox9 (sc-20095) antibody was used only in Fig. 3e, n and Sox9 (sc-166505) antibody was used only in Fig. 3f. Tankyrase-1/2 (sc-365897) antibody was used only in Fig. 6a, b, 7k and Supplementary Fig. 4. Antibodies against aggrecan (AB1031), type II collagen (MAB8887), and human mitochondria (MAB1273) were purchased from Millipore, and antibodies against Myc tag (2276) and Sox9 (82630) were purchased from Cell Signaling Technology. Prior to detection of aggrecan, samples were treated with chondroitinase ABC (C3667) from Sigma-Aldrich. Antibodies against HA tag (ab9110) and MMP13 (ab51072) were purchased from Abcam. Anti-β-catenin antibody (610154) was obtained from BD Biosciences. Anti-Poly(ADP-ribose) antibody (AG-20T-0001) was purchased from AdipoGen and was used only in Fig. 3l. Anti-Poly(ADP-ribose) antibody (4335-AMC) was purchased from Trevigen. All antibodies were used according to the manufacturer’s protocol.

Chemical Inhibitors

XAV939 (X3004), IWR-1 (I0161), JW55 (SML0630), WIKI4 (SML0760), IWP-2 (I0536), and cycloheximide (C7698) were obtained from Sigma-Aldrich. G007-LK (B5830) and MG132 (A2585) were purchased from Apexbio, G244-LM (1563007-08-8) was from AOBIOUS, MN-64 (HY19351) from MedChem Express, AZ6102 (S7767) from SelleckChem, and TC-E 5001 (5049) from Tocris. Tankyrase inhibitors were classified into three different classes depending on their mode of action^24,88. ABT-888 (11505) was purchased from Cayman, and ADP–HPD (118415) was purchased from Calbiochem.

RNA sequencing (RNA-seq)

Primary cultured mouse articular chondrocytes were treated with DMSO or 10 μM of XAV939 or IWR-1 for 108 h or transfected with control siRNA or Tnks and Tnks2 siRNAs. Three biological replicates were used for each group. One microgram of high-quality RNA samples (RIN > 7.0) were used to construct RNA-seq libraries with the TruSeq Stranded mRNA Library Prep kit (Illumina). Libraries were validated with an Agilent 2100 Bioanalyzer. RNA-seq was performed on an Illumina HiSeq 2500 sequencer at Macrogen. The sequence reads were trimmed with Trimmomatic⁸⁹ and mapped against the mouse reference genome (mm10) by using TopHat⁹⁰. Read counts per gene were calculated by using HTSeq⁹¹. Differential expression analysis was conducted by using the DESeq2 R package⁹². DEGs were selected by using a |fold change| cutoff of >3 and a FDR q cutoff of <1 × 10⁻⁵. DEGs in at least one condition were clustered with hierarchical clustering algorithm (ward.D linkage with euclidean distance) by using gplots R package. GO analysis was conducted by using Enrichr⁹³. Heatmaps of DEGs that are in the cartilage-signature gene set or the OA-signature gene sets were drawn with Gitools. Venn diagram for the upregulated or downregulated DEGs of the three different tankyrase inhibition groups (siTnks + siTnks2 vs. siControl, XAV939 vs. DMSO, and IWR-1 vs. DMSO) was drawn by using eulerAPE⁹⁴.

GSEA analysis

Genes were ranked according to the shrunken log₂ fold change calculated via DESeq2. GSEA⁹⁵ was performed in pre-ranked mode, with all default parameters, for the cartilage-signature gene set or the OA-signature gene sets. A thousand permutations were used to calculate P values.

Generation of a cartilage-signature gene set

Microarray data for nasal chondrocytes at embryonic day 17.5 and rib chondrocytes at postnatal day 1 were obtained from GSE69108¹⁸. Microarray data for mouse embryonic fibroblasts (MEFs) were obtained from GSM577694, GSM577695, and GSM577696 of GSE23547⁹⁶. The limma R package⁹⁷ was used to compute differential expression between nasal chondrocytes and MEFs or between rib chondrocytes and MEFs. The probe with the highest expression was used for each transcript. Genes with a fold change of >5 and a FDR q of <1 × 10⁻⁵ in both nasal chondrocytes and rib chondrocytes compared with MEFs were selected as cartilage-signature genes. The cartilage-signature genes are listed in Supplementary Table 9.

Subcellular fractionation

Cells were lysed with hypotonic buffer (20 mM Tris-HCl, pH 7.4, 10 mM NaCl, and 3 mM MgCl₂) supplemented with the protease inhibitor cocktail. After incubation in ice, 10% NP-40 was added to the cell lysate. The homogenate was centrifuged at 720 × g for 10 min, and the supernatant was collected as the cytoplasmic fraction. The pellet was washed twice with hypotonic buffer and resuspended in cell extraction buffer (10 mM Tris, pH 7.4, 2 mM Na₃VO₄, 100 mM NaCl, 1% Triton X-100, 1 mM EDTA, 10% glycerol, 1 mM EGTA, 0.1% SDS, 1 mM NaF, 0.5% deoxycholate, and 20 mM Na₄P₂O₇) supplemented with the protease inhibitor cocktail. The resuspended pellet was centrifuged at 14,000 × g for 30 min, and the supernatant was kept as the nuclear fraction. The cytoplasmic and nuclear fractions were quantified by using the BCA assay. Then 10 µg of proteins per fraction sample were loaded and analyzed by SDS-PAGE. Protein band intensity was quantified by densitometric analysis by using ImageJ.

Lentiviral infection and selection of HEK293 cells

psPAX2, pMD2.G, and pLVX-Puro-TK-FLAG-SOX9-WT or pLVX-Puro-TK-FLAG-SOX9-ΔTBD1/2 were transfected into HEK293T cells. After 2 days, cell supernatants were harvested and filtered through a 0.45-μm filter. HEK293 cells were treated with 8 μg/ml polybrene and transduced with the indicated lentiviruses. HEK293-FLAG-SOX9-WT and HEK293-FLAG-SOX9-ΔTBD1/2 were then selected by using 1 μg/ml puromycin for 2 days.

Immunoprecipitation

Cell lysates were prepared by using EBC200 buffer (50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 0.5% NP-40, and 1 mM EDTA) supplemented with the protease inhibitor cocktail. Only for Fig. 3m, q, cell lysates were prepared by using RIPA buffer supplemented with the protease inhibitor cocktail and diluted with EBC200 buffer supplemented with the protease inhibitor cocktail. Cell lysates were used for pull-down with the indicated antibodies and protein A/G-Sepharose beads (GE Healthcare), anti-FLAG M2 magnetic beads (M8823; Sigma-Aldrich), or Anti-Ub Agarose-TUBE2 (UM402; LifeSensors). The bound proteins were subjected to SDS-PAGE or LC–MS/MS analysis.

Endogenous tankyrase pulldown and mass spectrometry

Primary cultured mouse articular chondrocytes were grown for 4 days. Cells were lysed, and lysates were incubated with normal rabbit IgG or anti-tankyrase antibody. The bound proteins were eluted with 8 M urea in 50 mM NH₄HCO₃ buffer, pH 8.2 for 1 h at 37 °C, and in-solution digestion was performed⁹⁸. Peptide sequencing was carried out by LC–MS/MS on a Thermo Ultimate 3000 RSLCnano high-pressure liquid chromatography system coupled to a Thermo Q-Exactive Hybrid Quadrupole-Orbitrap mass spectrometer. LC–MS/MS raw data were converted into .mzML files by using ProteoWizard MSConvert⁹⁹, and the MS-GF+ algorithm¹⁰⁰ with a parameter file consisting of no enzyme criteria and static cysteine modification ( + 57.022 Da) was used for comparison of all MS/MS spectra against the mouse Uniprot database. The final peptide identifications had <1% false-discovery rate (FDR) q, at the unique peptide level. Only fully tryptic and semitryptic peptides were considered. For each biological replicate, proteins that were detected only once and proteins that were co-immunoprecipitated with normal rabbit IgG were not considered. For proteins detected in more than one biological replicate, the peptides and proteins are listed in Supplementary Data 1. The Venn diagram was drawn with eulerAPE.

In silico prediction of tankyrase substrate proteins

The 8 × 20 position-specific scoring matrix generated in Guettler et al.³⁰ was used to calculate a TTS for each octapeptide in the proteins identified by LC–MS/MS.

$${\mathrm{TTS}} = \frac{{\mathop {\sum }\nolimits_{\rm{pos.} = 0}^8\; \rm{PSSM}_{\rm{pos.}}}}{{\max \left( {\mathop {\sum }\nolimits_{\rm{pos.} = 0}^8 \;\rm{PSSM}_{\rm{pos.}}} \right)}}.$$

Only those proteins having at least one octapeptide with a TTS of ≥0.385 were considered. This cutoff is the TTS of the tankyrase-binding motifs of mouse AXIN1 and AXIN2. AXIN1 and AXIN2, the known tankyrase substrates²⁷, have the lowest maximum TTS among the known tankyrase substrates, due to the suboptimal amino acids at the fourth and fifth positions³⁰. For further screening, the chondrogenesis category in IPA (https://www.qiagenbioinformatics.com/products/ingenuity-pathway-analysis/) was used. The mouse proteins in the IPA chondrogenesis category are listed in Supplementary Table 7. For the candidate proteins, IUPred disorder scores were calculated for the octapeptides with a TTS of ≥0.385. The heatmap of TTS and IUPred disorder scores for candidate proteins was drawn with Gitools.

Cell line culture

HEK293 and HEK293T cells were cultured in DMEM containing 10% FBS, 100 units/ml penicillin, and 100 μg/ml streptomycin. Transfection was performed with METAFECTENE PRO (Biontex) or PEI transfection reagent (Sigma-Aldrich) according to the manufacturer’s protocol. The siRNAs used in HEK293T are listed in Supplementary Table 2. The siRNA sequences targeting TNKS or TNKS2 were described previously²⁷.

Plasmids

Human SOX9 cDNA (hMU008919) was purchased from Korea Human Gene Bank and subcloned into a pcDNA3-HA plasmid or a p3xFLAG-CMV10 plasmid. To express human SOX9 under TK promoter, Renilla luciferase gene in a pRL-TK plasmid was replaced by 3xFLAG-SOX9. TK-3xFLAG-SOX9 was subcloned into a pLVX-Puro plasmid. To generate mutant constructs, PCR-mediated mutagenesis was conducted. The GFP-tagged human TNKS plasmid was a gift from Dr. Chang-Woo Lee¹⁰¹, and the Myc-tagged human TNKS2 plasmid was a gift from Dr. Junjie Chen^69,71. The FLAG-tagged human TNKS2 plasmid and the FLAG-tagged human TNKS2 M1054V plasmid were gifts from Dr. Nai-Wen Chi¹⁰². The 4 × 48-p89 SOX9-dependent Col2a1 luciferase reporter construct was a gift from Dr. Veronique Lefebvre³⁶. Human TNKS2 cDNA was subcloned into a pEGFP-C1 plasmid to construct a GFP-tagged human TNKS2 plasmid. A control shRNA sequence was inserted into the pLKO.1 puro and pLKO.1 hygro plasmids. Human TNKS and TNKS2 shRNA sequences were inserted into the pLKO.1 puro and pLKO.1 hygro plasmids, respectively. The shRNA sequences targeting human TNKS or TNKS2 were as described previously²⁷. Mouse Tnks and Tnks2 shRNA sequences were inserted into the pLKO.1 puro and pLKO.1 hygro plasmids, respectively. Mouse Rnf146 shRNA sequence was inserted into the pLKO.1 puro plasmid. The shRNA sequence targeting mouse Tnks was as described previously⁶⁸. The primers used to generate the above plasmids are listed in Supplementary Tables 4–6.

In situ PLA

Primary cultured mouse articular chondrocytes were used for in situ PLA. Duolink^® PLA was performed according to the manufacturer’s protocol (Sigma-Aldrich). Antibodies against Sox9 (sc-166505) and Tankyrase-1/2 (sc-8337) were used to recognize endogenous mouse SOX9 and endogenous mouse tankyrase, respectively.

Sequence alignment of SOX9–TBD1 and TBD2 among vertebrates

For the sequence alignment of TBD1 and TBD2 of SOX9 among vertebrates, NP_000337.1 (Homo sapiens SOX9), NP_035578.3 (Mus musculus SOX9), NP_989612.1 (Gallus gallus SOX9), NP_001016853.1 (Xenopus tropicalis SOX9), and NP_571718.1 (Danio rerio SOX9) were used.

Structural modeling of protein–peptide interactions

GalaxyPepDock¹⁰³ was used for modeling of the ARC4 domain of human TNKS2 in complex with the TBD1 or TBD2 peptide of human SOX9. The structures of ARC4:3BP2 (PDB ID: 3TWR) and ARC4:MCL1 (PDB ID: 3TWU) were obtained from Guettler et al.³⁰. The ARC4 domain of human TNKS2 (PDB ID: 3TWU_A) and MCL1 peptide (PDB ID: 3TWU_B) were used as templates. The MCL1 peptide was substituted by the TBD1 (255–266 aa) or TBD2 (269–280 aa) peptide of human SOX9 and docked into a complex. The best predicted model for each of ARC4:SOX9–TBD1 and ARC4:SOX9 TBD2 was selected. The model structures were superimposed with ARC4:3BP2 and ARC4:MCL1 and visualized by using the BIOVIA Discovery Studio Visualizer (http://accelrys.com/products/collaborative-science/biovia-discovery-studio/visualization.html).

Cycloheximide chase analysis

HEK293 cells were treated with 50 μM of cycloheximide for the indicated number of hours before lysis. Protein samples were subjected to SDS-PAGE to analyze protein stability.

Reporter gene assay

A firefly luciferase reporter plasmid with SOX9-dependent Col2a1 enhancer elements was used to quantify the transcriptional activity of SOX9. To quantify β-catenin transcriptional activity, the TOPFlash reporter plasmid¹⁰⁴ was used. Primary mouse articular chondrocytes or HEK293T cells were transfected with both a reporter plasmid and a constitutive Renilla luciferase plasmid. Cells were also treated with siRNAs or drugs as indicated. Renilla and firefly luciferase activity were sequentially measured by using a Dual Luciferase Assay Kit (Promega). Renilla luciferase was used as a control.

List of SOX9 target genes in chondrocytes

Based on Oh et al.³⁷, genes with a log₂(fold change) of <−2 after Sox9 deletion in mouse rib chondrocytes and associated with SOX9 ChIP-Seq peaks in mouse rib chondrocytes were selected as SOX9 target genes in chondrocytes. The SOX9 target genes in chondrocytes are listed in Supplementary Table 8.

Generation of OA-signature gene sets

Based on Dunn et al.⁴¹, genes with a |fold change| of >2 and a FDR q of <1 × 10⁻⁵ in damaged sites of articular cartilage compared with intact sites within the same patients with OA were selected, and converted into mouse nomenclature by using the biomaRt R package¹⁰⁵. Genes that are upregulated and downregulated in osteoarthritic cartilage are listed in Supplementary Tables 10 and 11, respectively.

Preparation and in vivo evaluation of hydrogels

6-O-Palmitoyl-l-ascorbic acid (76183) was purchased from Sigma-Aldrich. Hydrogels (final concentration: 4% m/v) were prepared by dissolving 6-O-Palmitoyl-l-ascorbic acid with or without drugs in DMSO and mixing with sterilized distilled water⁴². DiD percholate (5702) purchased from Tocris was loaded into the hydrogels and used for imaging of controlled release in mouse knee joints. PBS-suspended hydrogel (10 μl, PBS:hydrogel = 1:1) containing 50 pmol DiD was administered intra-articularly, and at 1–9 days post injection, light-emitting diode and fluorescence images of knee joints were obtained. LuminoGraph II (Atto) was used to acquire the images.

Experimental OA in mice

Eight-week-old male ICR mice were used for experimental OA. Post-traumatic OA was induced by DMM surgery; sham operation was used as a control¹⁰⁶. In all, 10 μl of PBS-suspended hydrogel (PBS:hydrogel = 1:1) containing vehicle or 10 nmol drugs was administered intra-articularly. To generate the data presented in Supplementary Fig. 4, experimental OA was induced by DMM surgery in 12-week-old male C57BL/6 mice and their knee joints were processed for histological and immunostaining analysis at 2 and 8 weeks after surgery.

Human OA cartilage

Human OA cartilage specimens were sourced from OA patients undergoing total knee replacement at SNU Boramae Medical Center. The Boramae Medical Center Institutional Review Board (IRB No. 30-2017-48) approved the collection of the human specimens and the Seoul National University Institutional Review Board (IRB No. E1803/003-009) approved the use of these specimens. Written informed consent was obtained from all participants before the total knee replacement arthroplasty operative procedure.

Histology and immunohistochemistry

Mouse and rat knee joint samples and human cartilage samples were fixed with 4% paraformaldehyde overnight at 4 °C. Samples used in Supplementary Fig. 4 were decalcified in 0.5 M EDTA, pH 7.4, for 2–4 weeks at 4 °C. Samples used in Supplementary Fig. 5e–i, Fig. 7i–l, and Supplementary Figs. 6c–f, 8 were decalcified with 8% nitric acid for 8 h at room temperature. The human cartilage samples were not decalcified. All samples were embedded in paraffin. Mouse and rat paraffin blocks were sectioned at a thickness of 6 μm, and human paraffin blocks were sectioned to a thickness of 5 μm. For Safranin O staining, Alcian blue/Fast red staining, or immunostaining, sections were deparaffinized in xylene and hydrated by using a graded ethanol series. For immunohistochemistry, type II collagen antibody was used at 1:100–1:200 dilutions, and the other primary antibodies were used at a 1:50 dilution. All mouse histology images were acquired from medial tibial plateau except β-catenin immunostaining images where medial femoral condyle was used for imaging. To assess cartilage destruction in DMM mouse model, Safranin O stained samples were graded based on the Osteoarthritis Research Society International (OARSI)¹⁰⁷ by three blinded observers. On the basis of OARSI grading system, we primarily conducted integrative evaluation focusing on structural changes and proteoglycan loss in articular cartilage as a measure of cartilage destruction. Cartilage regeneration in osteochondral defect model was scored according to the International Cartilage Repair Society (ICRS) scoring system^108,109 by three blinded observers. Image-Pro Premier (http://www.mediacy.com/support/imagepropremier) was used for the quantification of immunostaining images. Signal-positive regions or cells were defined based on the chosen threshold. For intracellular proteins, the number of immunopositive cells or the percentage of immunopositive cells were calculated. For extracellular proteins, the relative chromogen intensity was calculated.

Mouse limb-bud micromass culture

For the micromass culture of mesenchymal cells, limb-bud cells were isolated from E11.5 ICR mouse embryos. In total, 2.0 × 10⁷ cells/ml were suspended in DMEM supplemented with 10% FBS, 100 units/ml penicillin, and 100 μg/ml streptomycin, and 15-µl drops were spotted on culture dishes. After 24 h, cells were treated as indicated for 3 days and subjected to Alcian blue staining or RNA extraction. For Fig. 7a, cells were harvested at days 1, 3, and 5 and subjected to RNA extraction.

Chondrogenic differentiation of hMSCs

hMSCs were purchased from Lonza and Thermo Scientific. hMSCs were cultured in α-MEM supplemented with 20% FBS, 100 units/ml penicillin, 100 μg/ml streptomycin, and 250 ng/ml amphotericin B. To induce chondrogenesis, 2.5 × 10⁵ hMSCs were centrifuged to form a pellet in α-MEM supplemented with 20% FBS, 100 units/ml penicillin, 100 μg/ml streptomycin, and 250 ng/ml amphotericin B. After 3 days, the medium was changed to chondrogenic medium consisting of DMEM/F-12 supplemented with 100 units/ml penicillin, 100 μg/ml streptomycin, 250 ng/ml amphotericin B, 1.25 mg/ml BSA, 1% insulin–transferrin–selenium, 1 mM sodium pyruvate, 50 μM l-aspartic acid, 50 μM l-proline, 100 nM dexamethasone, and 10 ng/ml of TGF-β1 with or without indicated drugs. On day 21 (for drug treatment) or day 28 (for siRNA treatment), cells were harvested and subjected to Alcian blue/Fast red staining. RNA-Seq data of human MSCs (GSE109503⁴⁷) during chondrogenesis at days 0, 1, 3, 7, 14, and 21 were obtained from gene expression omnibus repository at the National Center for Biotechnology Information (NCBI). Differential expression analysis was conducted by using the DESeq2 R package.

shRNA-mediated knockdown in hMSCs

psPAX2, pMD2.G, and the indicated shRNAs were transfected to HEK293T cells. After 3 days, cell supernatants were harvested and filtered through a 0.45-μm filter. hMSCs were treated with 8 μg/ml polybrene and infected with the indicated lentiviruses. Twenty-four hours after infection, hMSCs were selected with 5 μg/ml puromycin and 200 μg/ml hygromycin for 4 days.

Rat osteochondral defect model

Twelve-week-old male Sprague Dawley rats were used for an osteochondral defect model. To expose the articular cartilage in the knee joints, a medial parapatellar incision was made, and the patella was slightly displaced toward the medial condyle. A full-thickness cartilage defect (3 mm × 1 mm × 1 mm) was created by using a 1-mm-diameter spherical drill at the surface of the femoral patellar groove. At the same time, hMSCs were suspended in 10 µl of fibrin glue (TISSEEL) by tapping and implanted on the defect. To avoid immune rejection, cyclosporine A (C988900) from Toronto Research Chemicals was injected intraperitoneally every day. At 8 weeks, rats were killed for histological analyses.

Statistics

All experiments were carried out independently at least three times. All images are representative of at least three independent trials. For parametric tests, two-tailed Student’s t test or one-way analysis of variance followed by Fisher’s least significant difference post hoc test was used. For nonparametric tests, Mann–Whitney test or Kruskal–Wallis test followed by Mann–Whitney test were used. All statistical analysis was performed by using IBM SPSS Statistics. A P value < 0.05 was considered statistically significant. Cell culture samples and animals were randomly assigned to each group with about equivalent numbers in each group, and all samples were analyzed in a blinded manner. The sample sizes were determined on the basis of previous study⁵⁰, and no exclusion criteria were applied.

Study approval

All animal studies were conducted with the approval of the Seoul National University Institutional Animal Care and Use Committee (IACUC). We conformed to the ARRIVE guidelines for reporting animal experiments¹¹⁰.

Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.

Data availability

RNA-seq datasets are deposited in the Gene Expression Omnibus (GSE137899).

References

Bian, Q., Wang, Y. J., Liu, S. F. & Li, Y. P. Osteoarthritis: genetic factors, animal models, mechanisms, and therapies. Front Biosci. 4, 74–100 (2012).
Article Google Scholar
Loeser, R. F. Aging and osteoarthritis: the role of chondrocyte senescence and aging changes in the cartilage matrix. Osteoarthr. Cartil. 17, 971–979 (2009).
Article CAS Google Scholar
Karsdal, M. A. et al. Disease-modifying treatments for osteoarthritis (DMOADs) of the knee and hip: lessons learned from failures and opportunities for the future. Osteoarthr. Cartil. 24, 2013–2021 (2016).
Article CAS Google Scholar
Fajardo, M. & Di Cesare, P. E. Disease-modifying therapies for osteoarthritis: current status. Drugs Aging 22, 141–161 (2005).
Article CAS PubMed Google Scholar
Johnson, K. et al. A stem cell-based approach to cartilage repair. Science 336, 717–721 (2012).
Article ADS CAS PubMed Google Scholar
Dorotka, R., Bindreiter, U., Macfelda, K., Windberger, U. & Nehrer, S. Marrow stimulation and chondrocyte transplantation using a collagen matrix for cartilage repair. Osteoarthr. Cartil. 13, 655–664 (2005).
Article CAS Google Scholar
Yubo, M. et al. Clinical efficacy and safety of mesenchymal stem cell transplantation for osteoarthritis treatment: a meta-analysis. PLoS ONE 12, e0175449 (2017).
Article PubMed PubMed Central CAS Google Scholar
Diekman, B. O. & Guilak, F. Stem cell-based therapies for osteoarthritis: challenges and opportunities. Curr. Opin. Rheumatol. 25, 119–126 (2013).
Article CAS PubMed PubMed Central Google Scholar
Szychlinska, M. A. et al. Mesenchymal stem cell-based cartilage regeneration approach and cell senescence: can we manipulate cell aging and function? Tissue Eng. Part B Rev. 23, 529–539 (2017).
Article PubMed Google Scholar
Indrawattana, N. et al. Growth factor combination for chondrogenic induction from human mesenchymal stem cell. Biochem Biophys. Res. Commun. 320, 914–919 (2004).
Article CAS PubMed Google Scholar
Barry, F., Boynton, R. E., Liu, B. & Murphy, J. M. Chondrogenic differentiation of mesenchymal stem cells from bone marrow: differentiation-dependent gene expression of matrix components. Exp. Cell Res. 268, 189–200 (2001).
Article CAS PubMed Google Scholar
Kramer, J. et al. Embryonic stem cell-derived chondrogenic differentiation in vitro: activation by BMP-2 and BMP-4. Mech. Dev. 92, 193–205 (2000).
Article CAS PubMed Google Scholar
Ng, L. J. et al. SOX9 binds DNA, activates transcription, and coexpresses with type II collagen during chondrogenesis in the mouse. Dev. Biol. 183, 108–121 (1997).
Article CAS PubMed Google Scholar
Lefebvre, V., Li, P. & de Crombrugghe, B. A new long form of Sox5 (L-Sox5), Sox6 and Sox9 are coexpressed in chondrogenesis and cooperatively activate the type II collagen gene. EMBO J. 17, 5718–5733 (1998).
Article CAS PubMed PubMed Central Google Scholar
Wright, E. et al. The Sry-related gene Sox9 is expressed during chondrogenesis in mouse embryos. Nat. Genet. 9, 15–20 (1995).
Article CAS PubMed Google Scholar
Dy, P. et al. Sox9 directs hypertrophic maturation and blocks osteoblast differentiation of growth plate chondrocytes. Dev. Cell 22, 597–609 (2012).
Article CAS PubMed PubMed Central Google Scholar
Zhao, Q., Eberspaecher, H., Lefebvre, V. & De Crombrugghe, B. Parallel expression of Sox9 and Col2a1 in cells undergoing chondrogenesis. Dev. Dyn. 209, 377–386 (1997).
Article CAS PubMed Google Scholar
Ohba, S., He, X., Hojo, H. & McMahon, A. P. Distinct transcriptional programs underlie sox9 regulation of the mammalian chondrocyte. Cell Rep. 12, 229–243 (2015).
Article CAS PubMed PubMed Central Google Scholar
Fukui, N. et al. Regional differences in chondrocyte metabolism in osteoarthritis: a detailed analysis by laser capture microdissection. Arthritis Rheum. 58, 154–163 (2008).
Article CAS PubMed Google Scholar
Lee, J. M. & Im, G. I. SOX trio-co-transduced adipose stem cells in fibrin gel to enhance cartilage repair and delay the progression of osteoarthritis in the rat. Biomaterials 33, 2016–2024 (2012).
Article CAS PubMed Google Scholar
Ikeda, T. et al. The combination of SOX5, SOX6, and SOX9 (the SOX trio) provides signals sufficient for induction of permanent cartilage. Arthritis Rheum. 50, 3561–3573 (2004).
Article CAS PubMed Google Scholar
Riffell, J. L., Lord, C. J. & Ashworth, A. Tankyrase-targeted therapeutics: expanding opportunities in the PARP family. Nat. Rev. Drug Discov. 11, 923–936 (2012).
Article CAS PubMed Google Scholar
Smith, S., Giriat, I., Schmitt, A. & de Lange, T. Tankyrase, a poly(ADP-ribose) polymerase at human telomeres. Science 282, 1484–1487 (1998).
Article CAS PubMed Google Scholar
Lehtio, L., Chi, N. W. & Krauss, S. Tankyrases as drug targets. FEBS J. 280, 3576–3593 (2013).
Article PubMed CAS Google Scholar
Suwanwela, J. et al. Systems genetics analysis of mouse chondrocyte differentiation. J. Bone Min. Res. 26, 747–760 (2011).
Article CAS Google Scholar
Heinegard, D. & Saxne, T. The role of the cartilage matrix in osteoarthritis. Nat. Rev. Rheumatol. 7, 50–56 (2011).
Article PubMed CAS Google Scholar
Huang, S. M. et al. Tankyrase inhibition stabilizes axin and antagonizes Wnt signalling. Nature 461, 614–620 (2009).
Article ADS CAS PubMed Google Scholar
Glinka, A. et al. Dickkopf-1 is a member of a new family of secreted proteins and functions in head induction. Nature 391, 357–362 (1998).
Article ADS CAS PubMed Google Scholar
Chen, B. et al. Small molecule-mediated disruption of Wnt-dependent signaling in tissue regeneration and cancer. Nat. Chem. Biol. 5, 100–107 (2009).
Article ADS CAS PubMed PubMed Central Google Scholar
Guettler, S. et al. Structural basis and sequence rules for substrate recognition by Tankyrase explain the basis for cherubism disease. Cell 147, 1340–1354 (2011).
Article CAS PubMed Google Scholar
Dosztanyi, Z., Csizmok, V., Tompa, P. & Simon, I. IUPred: web server for the prediction of intrinsically unstructured regions of proteins based on estimated energy content. Bioinformatics 21, 3433–3434 (2005).
Article CAS PubMed Google Scholar
Akiyama, H., Chaboissier, M. C., Martin, J. F., Schedl, A. & de Crombrugghe, B. The transcription factor Sox9 has essential roles in successive steps of the chondrocyte differentiation pathway and is required for expression of Sox5 and Sox6. Genes Dev. 16, 2813–2828 (2002).
Article CAS PubMed PubMed Central Google Scholar
Zhang, Y. et al. RNF146 is a poly(ADP-ribose)-directed E3 ligase that regulates axin degradation and Wnt signalling. Nat. Cell Biol. 13, 623–629 (2011).
Article CAS PubMed Google Scholar
DaRosa, P. A. et al. Allosteric activation of the RNF146 ubiquitin ligase by a poly(ADP-ribosyl)ation signal. Nature 517, 223–226 (2015).
Article ADS CAS PubMed Google Scholar
Andrabi, S. A. et al. Iduna protects the brain from glutamate excitotoxicity and stroke by interfering with poly(ADP-ribose) polymer-induced cell death. Nat. Med. 17, 692–699 (2011).
Article CAS PubMed PubMed Central Google Scholar
Murakami, S., Lefebvre, V. & de Crombrugghe, B. Potent inhibition of the master chondrogenic factor Sox9 gene by interleukin-1 and tumor necrosis factor-alpha. J. Biol. Chem. 275, 3687–3692 (2000).
Article CAS PubMed Google Scholar
Oh, C. D. et al. SOX9 regulates multiple genes in chondrocytes, including genes encoding ECM proteins, ECM modification enzymes, receptors, and transporters. PLoS ONE 9, e107577 (2014).
Article ADS PubMed PubMed Central CAS Google Scholar
Mead, T. J. et al. A far-upstream (-70 kb) enhancer mediates Sox9 auto-regulation in somatic tissues during development and adult regeneration. Nucleic Acids Res. 41, 4459–4469 (2013).
Article CAS PubMed PubMed Central Google Scholar
Goldring, M. B. & Goldring, S. R. Osteoarthritis. J. Cell Physiol. 213, 626–634 (2007).
Article CAS PubMed Google Scholar
Venkatesan, N. et al. Xylosyltransferase-I regulates glycosaminoglycan synthesis during the pathogenic process of human osteoarthritis. PLoS ONE 7, e34020 (2012).
Article ADS CAS PubMed PubMed Central Google Scholar
Dunn, S. L. et al. Gene expression changes in damaged osteoarthritic cartilage identify a signature of non-chondrogenic and mechanical responses. Osteoarthr. Cartil. 24, 1431–1440 (2016).
Article CAS Google Scholar
Zhang, S. et al. An inflammation-targeting hydrogel for local drug delivery in inflammatory bowel disease. Sci. Transl. Med. 7, 300ra128 (2015).
Article PubMed PubMed Central CAS Google Scholar
Koyama, E. et al. A distinct cohort of progenitor cells participates in synovial joint and articular cartilage formation during mouse limb skeletogenesis. Dev. Biol. 316, 62–73 (2008).
Article CAS PubMed PubMed Central Google Scholar
Yasuhara, R. et al. Roles of beta-catenin signaling in phenotypic expression and proliferation of articular cartilage superficial zone cells. Lab Invest. 91, 1739–1752 (2011).
Article CAS PubMed PubMed Central Google Scholar
Kim, J. H. et al. Matrix cross-linking-mediated mechanotransduction promotes posttraumatic osteoarthritis. Proc. Natl Acad. Sci. USA 112, 9424–9429 (2015).
Article ADS CAS PubMed PubMed Central Google Scholar
Jiang, Y. & Tuan, R. S. Origin and function of cartilage stem/progenitor cells in osteoarthritis. Nat. Rev. Rheumatol. 11, 206–212 (2015).
Article PubMed Google Scholar
Huynh, N. P. T., Zhang, B. & Guilak, F. High-depth transcriptomic profiling reveals the temporal gene signature of human mesenchymal stem cells during chondrogenesis. FASEB J. 33, 358–372 (2019).
Article CAS PubMed Google Scholar
Liu-Bryan, R. & Terkeltaub, R. Emerging regulators of the inflammatory process in osteoarthritis. Nat. Rev. Rheumatol. 11, 35–44 (2015).
Article CAS PubMed Google Scholar
Yang, S. et al. Hypoxia-inducible factor-2alpha is a catabolic regulator of osteoarthritic cartilage destruction. Nat. Med. 16, 687–693 (2010).
Article CAS PubMed Google Scholar
Kim, J. H. et al. Regulation of the catabolic cascade in osteoarthritis by the zinc-ZIP8-MTF1 axis. Cell 156, 730–743 (2014).
Article CAS PubMed Google Scholar
Saito, T. et al. Transcriptional regulation of endochondral ossification by HIF-2alpha during skeletal growth and osteoarthritis development. Nat. Med. 16, 678–686 (2010).
Article CAS PubMed Google Scholar
Echtermeyer, F. et al. Syndecan-4 regulates ADAMTS-5 activation and cartilage breakdown in osteoarthritis. Nat. Med. 15, 1072–1076 (2009).
Article CAS PubMed Google Scholar
Lin, A. C. et al. Modulating hedgehog signaling can attenuate the severity of osteoarthritis. Nat. Med. 15, 1421–1425 (2009).
Article CAS PubMed Google Scholar
Zhen, G. et al. Inhibition of TGF-beta signaling in mesenchymal stem cells of subchondral bone attenuates osteoarthritis. Nat. Med. 19, 704–712 (2013).
Article CAS PubMed PubMed Central Google Scholar
Stanton, H. et al. ADAMTS5 is the major aggrecanase in mouse cartilage in vivo and in vitro. Nature 434, 648–652 (2005).
Article ADS CAS PubMed Google Scholar
Glasson, S. S. et al. Deletion of active ADAMTS5 prevents cartilage degradation in a murine model of osteoarthritis. Nature 434, 644–648 (2005).
Article ADS CAS PubMed Google Scholar
Wang, Q. et al. Identification of a central role for complement in osteoarthritis. Nat. Med. 17, 1674–1679 (2011).
Article CAS PubMed PubMed Central Google Scholar
Neuhold, L. A. et al. Postnatal expression in hyaline cartilage of constitutively active human collagenase-3 (MMP-13) induces osteoarthritis in mice. J. Clin. Invest. 107, 35–44 (2001).
Article CAS PubMed PubMed Central Google Scholar
Dancevic, C. M. & McCulloch, D. R. Current and emerging therapeutic strategies for preventing inflammation and aggrecanase-mediated cartilage destruction in arthritis. Arthritis Res. Ther. 16, 429 (2014).
Article PubMed PubMed Central CAS Google Scholar
Wang, M. et al. MMP13 is a critical target gene during the progression of osteoarthritis. Arthritis Res. Ther. 15, R5 (2013).
Article MathSciNet CAS PubMed PubMed Central Google Scholar
Moon, S. et al. A genome-wide association study of copy-number variation identifies putative loci associated with osteoarthritis in Koreans. BMC Musculoskelet. Disord. 16, 76 (2015).
Article PubMed PubMed Central CAS Google Scholar
Jaeger, K. et al. The genetics of osteoarthritis in STR/ort mice. Osteoarthr. Cartil. 16, 607–614 (2008).
Article CAS Google Scholar
Bi, W., Deng, J. M., Zhang, Z., Behringer, R. R. & de Crombrugghe, B. Sox9 is required for cartilage formation. Nat. Genet. 22, 85–89 (1999).
Article CAS PubMed Google Scholar
Ikegami, D. et al. Sox9 sustains chondrocyte survival and hypertrophy in part through Pik3ca-Akt pathways. Development 138, 1507–1519 (2011).
Article CAS PubMed Google Scholar
Martinez-Sanchez, A., Dudek, K. A. & Murphy, C. L. Regulation of human chondrocyte function through direct inhibition of cartilage master regulator SOX9 by microRNA-145 (miRNA-145). J. Biol. Chem. 287, 916–924 (2012).
Article CAS PubMed Google Scholar
Haag, J., Gebhard, P. M. & Aigner, T. SOX gene expression in human osteoarthritic cartilage. Pathobiology 75, 195–199 (2008).
Article CAS PubMed Google Scholar
Smith, S. & de Lange, T. Tankyrase promotes telomere elongation in human cells. Curr. Biol. 10, 1299–1302 (2000).
Article CAS PubMed Google Scholar
Chang, P., Coughlin, M. & Mitchison, T. J. Tankyrase-1 polymerization of poly(ADP-ribose) is required for spindle structure and function. Nat. Cell Biol. 7, 1133–1139 (2005).
Article CAS PubMed Google Scholar
Li, N. et al. Poly-ADP ribosylation of PTEN by tankyrases promotes PTEN degradation and tumor growth. Genes Dev. 29, 157–170 (2015).
Article PubMed PubMed Central CAS Google Scholar
Levaot, N. et al. Loss of Tankyrase-mediated destruction of 3BP2 is the underlying pathogenic mechanism of cherubism. Cell 147, 1324–1339 (2011).
Article CAS PubMed PubMed Central Google Scholar
Wang, W. et al. Tankyrase Inhibitors Target YAP by Stabilizing Angiomotin Family Proteins. Cell Rep. 13, 524–532 (2015).
Article PubMed PubMed Central CAS Google Scholar
Bhardwaj, A., Yang, Y., Ueberheide, B. & Smith, S. Whole proteome analysis of human tankyrase knockout cells reveals targets of tankyrase-mediated degradation. Nat. Commun. 8, 2214 (2017).
Article ADS PubMed PubMed Central CAS Google Scholar
Li, X. et al. Proteomic analysis of the human tankyrase protein interaction network reveals its role in pexophagy. Cell Rep. 20, 737–749 (2017).
Article CAS PubMed Google Scholar
Corr, M. Wnt-beta-catenin signaling in the pathogenesis of osteoarthritis. Nat. Clin. Pract. Rheumatol. 4, 550–556 (2008).
Article CAS PubMed Google Scholar
Onuora, S. Osteoarthritis: Wnt inhibitor shows potential as a DMOAD. Nat. Rev. Rheumatol. 13, 634 (2017).
Article PubMed Google Scholar
Deshmukh, V. et al. A small-molecule inhibitor of the Wnt pathway (SM04690) as a potential disease modifying agent for the treatment of osteoarthritis of the knee. Osteoarthr. Cartil. 26, 18–27 (2018).
Article CAS Google Scholar
Lietman, C. et al. Inhibition of Wnt/beta-catenin signaling ameliorates osteoarthritis in a murine model of experimental osteoarthritis. JCI Insight 3, https://doi.org/10.1172/jci.insight.96308 (2018).
Bastakoty, D. et al. Inhibition of Wnt/beta-catenin pathway promotes regenerative repair of cutaneous and cartilage injury. FASEB J. 29, 4881–4892 (2015).
Article CAS PubMed PubMed Central Google Scholar
Yazici, Y. et al. A novel Wnt pathway inhibitor, SM04690, for the treatment of moderate to severe osteoarthritis of the knee: results of a 24-week, randomized, controlled, phase 1 study. Osteoarthr. Cartil. 25, 1598–1606 (2017).
Article CAS Google Scholar
Gudjonsson, T. et al. TRIP12 and UBR5 suppress spreading of chromatin ubiquitylation at damaged chromosomes. Cell 150, 697–709 (2012).
Article CAS PubMed Google Scholar
Chen, D., Shan, J., Zhu, W. G., Qin, J. & Gu, W. Transcription-independent ARF regulation in oncogenic stress-mediated p53 responses. Nature 464, 624–627 (2010).
Article ADS CAS PubMed PubMed Central Google Scholar
Farber, C. R. et al. An integrative genetics approach to identify candidate genes regulating BMD: combining linkage, gene expression, and association. J. Bone Miner. Res. 24, 105–116 (2009).
Article CAS PubMed Google Scholar
Alberts, R., Lu, L., Williams, R. W. & Schughart, K. Genome-wide analysis of the mouse lung transcriptome reveals novel molecular gene interaction networks and cell-specific expression signatures. Respir. Res. 12, 61 (2011).
Article CAS PubMed PubMed Central Google Scholar
Saba, L. et al. Candidate genes and their regulatory elements: alcohol preference and tolerance. Mamm. Genome 17, 669–688 (2006).
Article CAS PubMed Google Scholar
Perez-Llamas, C. & Lopez-Bigas, N. Gitools: analysis and visualisation of genomic data using interactive heat-maps. PLoS ONE 6, e19541 (2011).
Article ADS CAS PubMed PubMed Central Google Scholar
Gosset, M., Berenbaum, F., Thirion, S. & Jacques, C. Primary culture and phenotyping of murine chondrocytes. Nat. Protoc. 3, 1253–1260 (2008).
Article CAS PubMed Google Scholar
Schneider, C. A., Rasband, W. S. & Eliceiri, K. W. NIH Image to ImageJ: 25 years of image analysis. Nat. Methods 9, 671–675 (2012).
Article CAS PubMed PubMed Central Google Scholar
Haikarainen, T., Krauss, S. & Lehtio, L. Tankyrases: structure, function and therapeutic implications in cancer. Curr. Pharm. Des. 20, 6472–6488 (2014).
Article CAS PubMed PubMed Central Google Scholar
Bolger, A. M., Lohse, M. & Usadel, B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics 30, 2114–2120 (2014).
Article CAS PubMed PubMed Central Google Scholar
Kim, D. et al. TopHat2: accurate alignment of transcriptomes in the presence of insertions, deletions and gene fusions. Genome Biol. 14, R36 (2013).
Article PubMed PubMed Central CAS Google Scholar
Anders, S., Pyl, P. T. & Huber, W. HTSeq–a Python framework to work with high-throughput sequencing data. Bioinformatics 31, 166–169 (2015).
Article CAS PubMed Google Scholar
Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15, 550 (2014).
Article PubMed PubMed Central CAS Google Scholar
Kuleshov, M. V. et al. Enrichr: a comprehensive gene set enrichment analysis web server 2016 update. Nucleic Acids Res. 44, W90–W97 (2016).
Article CAS PubMed PubMed Central Google Scholar
Micallef, L. & Rodgers, P. eulerAPE: drawing area-proportional 3-Venn diagrams using ellipses. PLoS ONE 9, e101717 (2014).
Article ADS PubMed PubMed Central Google Scholar
Subramanian, A. et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc. Natl Acad. Sci. USA 102, 15545–15550 (2005).
Article ADS CAS PubMed PubMed Central Google Scholar
Brellier, F. et al. Tenascin-C triggers fibrin accumulation by downregulation of tissue plasminogen activator. FEBS Lett. 585, 913–920 (2011).
Article CAS PubMed Google Scholar
Ritchie, M. E. et al. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 43, e47 (2015).
Article PubMed PubMed Central CAS Google Scholar
Kim, J. S., Monroe, M. E., Camp, D. G. 2nd, Smith, R. D. & Qian, W. J. In-source fragmentation and the sources of partially tryptic peptides in shotgun proteomics. J. Proteome Res. 12, 910–916 (2013).
Article CAS PubMed PubMed Central Google Scholar
Chambers, M. C. et al. A cross-platform toolkit for mass spectrometry and proteomics. Nat. Biotechnol. 30, 918–920 (2012).
Article CAS PubMed PubMed Central Google Scholar
Kim, S. & Pevzner, P. A. MS-GF+ makes progress towards a universal database search tool for proteomics. Nat. Commun. 5, 5277 (2014).
Article ADS CAS PubMed Google Scholar
Ha, G. H. et al. Tankyrase-1 function at telomeres and during mitosis is regulated by Polo-like kinase-1-mediated phosphorylation. Cell Death Differ. 19, 321–332 (2012).
Article PubMed CAS Google Scholar
Sbodio, J. I., Lodish, H. F. & Chi, N. W. Tankyrase-2 oligomerizes with tankyrase-1 and binds to both TRF1 (telomere-repeat-binding factor 1) and IRAP (insulin-responsive aminopeptidase). Biochem J. 361, 451–459 (2002).
Article CAS PubMed PubMed Central Google Scholar
Lee, H., Heo, L., Lee, M. S. & Seok, C. GalaxyPepDock: a protein-peptide docking tool based on interaction similarity and energy optimization. Nucleic Acids Res. 43, W431–W435 (2015).
Article CAS PubMed PubMed Central Google Scholar
Veeman, M. T., Slusarski, D. C., Kaykas, A., Louie, S. H. & Moon, R. T. Zebrafish prickle, a modulator of noncanonical Wnt/Fz signaling, regulates gastrulation movements. Curr. Biol. 13, 680–685 (2003).
Article CAS PubMed Google Scholar
Durinck, S., Spellman, P. T., Birney, E. & Huber, W. Mapping identifiers for the integration of genomic datasets with the R/Bioconductor package biomaRt. Nat. Protoc. 4, 1184–1191 (2009).
Article CAS PubMed PubMed Central Google Scholar
Glasson, S. S., Blanchet, T. J. & Morris, E. A. The surgical destabilization of the medial meniscus (DMM) model of osteoarthritis in the 129/SvEv mouse. Osteoarthr. Cartil. 15, 1061–1069 (2007).
Article CAS Google Scholar
Glasson, S. S., Chambers, M. G., Van Den Berg, W. B. & Little, C. B. The OARSI histopathology initiative—recommendations for histological assessments of osteoarthritis in the mouse. Osteoarthr. Cartil. 18(Suppl 3), S17–S23 (2010).
Article Google Scholar
van den Borne, M. P. et al. International Cartilage Repair Society (ICRS) and Oswestry macroscopic cartilage evaluation scores validated for use in Autologous Chondrocyte Implantation (ACI) and microfracture. Osteoarthr. Cartil. 15, 1397–1402 (2007).
Article Google Scholar
Mainil-Varlet, P. et al. Histological assessment of cartilage repair: a report by the Histology Endpoint Committee of the International Cartilage Repair Society (ICRS). J. Bone Jt. Surg. Am. 85-A Suppl. 2, 45–57 (2003).
Article Google Scholar
Kilkenny, C., Browne, W. J., Cuthill, I. C., Emerson, M. & Altman, D. G. Improving bioscience research reporting: the ARRIVE guidelines for reporting animal research. PLoS Biol. 8, e1000412 (2010).
Article PubMed PubMed Central CAS Google Scholar

Download references

Acknowledgements

We thank the staff from the Department of Orthopedic Surgery, SNU College of Medicine, Boramae Hospital, and the patients for participating in the study. We would also like to acknowledge Jong Min Lee (Dongguk University Ilsan Hospital) for technical assistance on osteochondral defect model. We thank Dong Hoon Kang (Ewha Womans University) for helpful advice on PARylation and ubiquitination assays. We also thank all group members of J.-H.K. lab for helpful discussions. This work was supported by grants from the National Research Foundation of Korea (NRF-2015M3A9E6028674, NRF-2016R1A5A1010764, and NRF-2017M3A9D8064193), Korean Ministry for Health and Welfare (1520070), the Institute for Basic Science from the Ministry of Science, ICT, and Future Planning of Korea (IBS-R008-D1), Creative-Pioneering Researchers Program through SNU, and Suh Kyungbae Foundation.

Author information

These authors contributed equally: Sukyeong Kim, Sangbin Han.

Authors and Affiliations

Center for RNA Research, Institute for Basic Science, 08826, Seoul, South Korea
Sukyeong Kim, Sangbin Han, Yeongjae Kim, Hyeon-Seop Kim, Young-Ran Gu, Donghyun Kang, Yongsik Cho, Hyeonkyeong Kim, Jeeyeon Lee, Yeyoung Seo & Jin-Hong Kim
Department of Biological Sciences, College of Natural Sciences, Seoul National University, 08826, Seoul, South Korea
Sukyeong Kim, Sangbin Han, Yeongjae Kim, Hyeon-Seop Kim, Young-Ran Gu, Donghyun Kang, Yongsik Cho, Hyeonkyeong Kim, Jeeyeon Lee, Yeyoung Seo & Jin-Hong Kim
Department of Orthopaedic Surgery, Seoul National University College of Medicine, Boramae Hospital, 07061, Seoul, South Korea
Moon Jong Chang & Seung-Baik Kang
Department of Orthopaedic Surgery, Seoul National University Bundang Hospital, 13620, Seongnam, South Korea
Chong Bum Chang
Interdisciplinary Program in Bioinformatics, Seoul National University, 08826, Seoul, South Korea
Jin-Hong Kim

Authors

Sukyeong Kim
View author publications
You can also search for this author in PubMed Google Scholar
Sangbin Han
View author publications
You can also search for this author in PubMed Google Scholar
Yeongjae Kim
View author publications
You can also search for this author in PubMed Google Scholar
Hyeon-Seop Kim
View author publications
You can also search for this author in PubMed Google Scholar
Young-Ran Gu
View author publications
You can also search for this author in PubMed Google Scholar
Donghyun Kang
View author publications
You can also search for this author in PubMed Google Scholar
Yongsik Cho
View author publications
You can also search for this author in PubMed Google Scholar
Hyeonkyeong Kim
View author publications
You can also search for this author in PubMed Google Scholar
Jeeyeon Lee
View author publications
You can also search for this author in PubMed Google Scholar
Yeyoung Seo
View author publications
You can also search for this author in PubMed Google Scholar
Moon Jong Chang
View author publications
You can also search for this author in PubMed Google Scholar
Chong Bum Chang
View author publications
You can also search for this author in PubMed Google Scholar
Seung-Baik Kang
View author publications
You can also search for this author in PubMed Google Scholar
Jin-Hong Kim
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

S.K., S.H., and J.-H.K. designed the study. S.K. and S.H. performed the most in vitro and in vivo experiments. Y.K., H.-S.K., Y.-R.G., D.K., Y.C., J.L., and Y.S. conducted the cell-based assays. Y.K., H.K., and J.L. conducted the immunohistochemistry analysis. H.-S.K. performed the mass spectrometry analysis. M.J.C., C.B.C., and S.-B.K. collected and inspected the human patient samples. S.K., S.H., Y.K., H.-S.K., Y.-R.G., D.K., Y.C., H.K., J.L., and J.-H.K. analyzed the data. S.K., S.H., and J.-H.K. wrote the paper. All authors read and edited the paper. J.-H.K. supervised the study.

Corresponding author

Correspondence to Jin-Hong Kim.

Ethics declarations

Competing interests

S.K., S.H., Y.C., and J.-H.K. are the inventor authors on pending patent application, “Use of TNKS inhibitors for regeneration of cartilage”, Korea patent application no. 10-2018-0139120. The remaining authors declare no competing interests.

Additional information

Peer review information Nature Communications thanks Taku Saito, Yolande F. M. Ramos and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Peer reviewer reports are available.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Data 1

Supplementary Data 2

Supplementary Information

Reporting Summary

Description of Additional Supplementary Files

Peer Review File

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Kim, S., Han, S., Kim, Y. et al. Tankyrase inhibition preserves osteoarthritic cartilage by coordinating cartilage matrix anabolism via effects on SOX9 PARylation. Nat Commun 10, 4898 (2019). https://doi.org/10.1038/s41467-019-12910-2

Download citation

Received: 14 July 2018
Accepted: 07 October 2019
Published: 25 October 2019
DOI: https://doi.org/10.1038/s41467-019-12910-2

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.