Neural cell adhesion molecule regulates chondrocyte hypertrophy in chondrogenic differentiation and experimental osteoarthritis

Abstract Chondrocyte hypertrophy‐like change is an important pathological process of osteoarthritis (OA), but the mechanism remains largely unknown. Neural cell adhesion molecule (NCAM) is highly expressed and involved in the chondrocyte differentiation of mesenchymal stem cells (MSCs). In this study, we found that NCAM deficiency accelerates chondrocyte hypertrophy in articular cartilage and growth plate of OA mice. NCAM deficiency leads to hypertrophic chondrocyte differentiation in both murine MSCs and chondrogenic cells, in which extracellular signal‐regulated kinase (ERK) signaling plays an important role. Moreover, NCAM expression is downregulated in an interleukin‐1β‐stimulated OA cellular model and monosodium iodoacetate‐induced OA rats. Overexpression of NCAM substantially inhibits hypertrophic differentiation in the OA cellular model. In conclusion, NCAM could inhibit hypertrophic chondrocyte differentiation of MSCs by inhibiting ERK signaling and reduce chondrocyte hypertrophy in experimental OA model, suggesting the potential utility of NCAM as a novel therapeutic target for alleviating chondrocyte hypertrophy of OA.


| INTRODUCTION
Osteoarthritis (OA) is the most common joint disorder worldwide, 1 showing a progressive increase in the last two decades as the leading cause of large social and economic burden. 2,3 The study of OA pathophysiology mainly focuses on the mechanisms of cartilage degeneration and chondrocyte biology. 4 Since chondrocyte is poorly regenerated, osteoarthritic cartilage injury is irreversible and difficult to repair. In recent years, mesenchymal stem cells (MSCs) have become an attractive tool for cartilage regenerative therapy. 5 MSCs are capable of differentiating into multiple lineages of cells, including chondrocytes. 6 The process of chondrogenic differentiation of MSCs can be divided into distinct phases including cell condensation, chondrocyte differentiation/proliferation, and chondrocyte hypertrophy. 7 Chondrocyte hypertrophy is an important physiological process involved in the development of long bones from the cartilage anlagen. This stage is marked by increase in cell volume, 8

extracellular matrix
(ECM) remodeling, and expression of hypertrophic chondrocyte markers such as type X collagen (Col X), matrix metalloproteinase(MMP)-13, and runt-related transcription factor 2 (RunX2). 9 Chondrocyte hypertrophy was shown to be involved in OA pathogenesis. Chondrocytes in healthy cartilage resist proliferation and terminal differentiation whereas chondrocytes in OA resemble hypertrophic differentiation with characteristics of cartilage matrix remodeling, cartilage calcification, and expression of hypertrophy markers including RunX2 and Col X. 10,11 Understanding the molecular mechanisms regulating chondrocyte hypertrophy is important for clinical MSCs application and drug development to repair cartilage defects of OA.
Neural cell adhesion molecule (NCAM), a member of immunoglobulin superfamily that mediates cell-cell and matrix interactions, 12 plays a key role in regulation of neurite outgrowth, synaptic plasticity, neuronal development, learning, and memory. [13][14][15] NCAM is also expressed in MSCs, 16,17 but its function remains largely unknown. We have previously demonstrated that NCAM promotes adipocyte differentiation of murine MSCs via PI3K-Akt pathways. 18 NCAM is also involved in chondrogenesis. In the process of chondrogenic differentiation, NCAM is expressed in prechondrogenic cells and increased during cell condensation, 19 but it becomes undetectable in hypertrophic chondrocytes. 20,21 Previous studies showed that NCAM initiated chondrogenesis by promoting and stabilizing the condensation step while may not contribute to chondrocyte differentiation directly. 22,23 However, the role of NCAM in chondrocyte hypertrophy is still poorly understood.
In the present work, we evaluated the effects of NCAM on chondrocyte hypertrophy in murine MSCs, in chondrogenic ATDC5 cells, and in experimental murine OA. We also investigated the underlying signaling pathway in MSCs involved in hypertrophic differentiation of chondrocytes.

| NCAM-deficient mice and cell culture
The Ncam −/− (knockout; KO) mice were generated on a C57/BL6 background as previously described. 24 Wild-type (WT) and KO MSCs were obtained from 8-week-old male mice as previously described. 18 Briefly, cells were harvested from mouse bone marrow and cultured in low glucose Dulbecco's modified Eagle's medium (DMEM-LG) containing 10% FBS, 100 IU/mL penicillin and 100 g/mL streptomycin. Nonadherent hematopoietic cells were discarded after incubation for 7 days, and the adhered MSCs were purified by repeated passaging. MSCs with fibroblast-like morphology from passage 6 to 15 were used in this study.

| Quantitative real-time PCR
Total RNA was extracted with TRIzol (Invitrogen) according to the manufacturer's instructions, and cDNA was synthesized using a High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, California). The mRNA levels were measured using an ABI Prism  The qPCR primers were designed and shown in Table S1.

| Western blotting
Cells and tissue samples were lysed with radio immunoprecipitation (RIPA) buffer and lysed on ice supplemented with proteinase and phosphatase inhibitors cocktail (Sigma-Aldrich). Protein concentrations were determined using the BCA Protein Assay Kit (Pierce Biotechnology Inc, Rockford, Illinois). Equal amounts of protein were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred to polyvinylidene difluoride (PVDF) membrane. After blocking with 3% bovine serum albumin (BSA; Sigma-Aldrich) for 1 hour, the membranes were incubated with primary antibodies against NCAM, Col X, RunX2, phospho-ERK, ERK2, c-Myc, or β-actin overnight at 4 C. Subsequently, the membranes were washed three times with TBST and incubated with the secondary antibody for 40 minutes. Membranes were scanned by the ImageQuant LAS 4000 system (GE Healthcare) and images, in some cases, were analyzed using the ImageJ software.

| Alizarin Red staining and quantification
Cells were fixed in 4% paraformaldehyde (PFA) for 30 minutes followed by staining with 40 mM Alizarin Red S (ARS; pH 4.1) for 30 minutes. Quantitative analysis of ARS was performed by incubating the stained wells with 10% cetyl pyridinium chloride monohydrate for 30 minutes at room temperature before centrifugation. The extracted supernatant was measured at 562 nm using a Molecular Devices microplate reader.

| Immunofluorescence staining
Cells were fixed with 4% paraformaldehyde for 30 minutes and then permeabilized in 0.5% Triton X-100 (Sigma-Aldrich) for 30 minutes at room temperature. Permeabilized cells were blocked with 3% BSA for 1 hour at room temperature and then stained overnight with primary antibody against Col X at 4 C. Subsequently, cells were stained with corresponding secondary antibody (Cell Signaling Technology Inc) for 45 minutes at 4 C in the dark and incubated for 30 minutes in dimethylsulfoxide for nuclei visualization. Confocal imaging was carried out using a fluorescence microscope (Leica DMIL LED). Images represent the z-stack projection of confocal sections. Sigma-Aldrich) induced OA (n = 7 per group), rats received intra-articular knee injections with MIA (20 mg/mL, 30 μL) or sterile saline as a control.

| Induction of experimental OA rats
The rats were sacrificed on day 21, cartilage samples were obtained for analysis of NCAM by Western blotting. Serum was obtained for detection of interleukin (IL)-1β and tumor necrosis factor (TNF)-α using enzymelinked immunosorbent assay (ELISA).

| Induction of experimental OA mice
The KO and WT mice (male and female, 8-12 weeks, 20-30 g, n = 5 per group) were anaesthetized with chloralic hydrate (400 mg/kg, i.p.). OA mice model was induced by an injection of MIA (20 mg/mL, 15 μL) in sterile saline into the left knee joint cavity through the patellar ligament. On day 10 or 21, the whole knee joints were obtained for histological or immunohistochemical study. Serum was obtained at day 21 for detection of IL-1β and TNF-α using ELISA. All animal experiments were approved by the Ethics Committee of Xinxiang Medical University.

| Histological analysis and immunohistochemical staining
For histological analysis, the whole knee joints were fixed in 4% PFA, decalcified, embedded in paraffin, sectioned, and stained with Safranin O/Fast Green or hematoxylin and eosin (H&E) using standard protocols. For immunohistochemical staining, sections were pretreated with trypsin, incubated with primary antibody against RunX2 at 4 C overnight, then incubated with secondary antibody, and stained using standard protocols.

| Plasmid constructs and transfection
siRNA vectors silencing NCAM and plasmids expressing full-length mouse NCAM were designed and constructed as previously described. 18 The expression plasmid containing constitutive active form of MEK was a gift from Sheng-Cai Lin (Xiamen University, China). Transfection was conducted using Lipofectamine 2000 following manufacturer's instructions. To obtain stable mixed cell lines, cells were selected with zeocin at 150 μg/mL or neomycin at 800 μg/mL for 10-14 days. Gene silencing or overexpression of NCAM was validated by Western blotting with anti-NCAM or anti-c-Myc tag antibodies.

| Statistical analysis
All data are presented as mean ± SEM unless otherwise specified.
Statistical analysis was determined with GraphPad Prism (GraphPad Software, La Jolla, California). Comparisons between two groups were performed using a two-tailed unpaired Student's t test, comparisons between three or more groups were analyzed using one-way ANOVA followed by Tukey's post hoc test, Welch correction was used to protect against heteroscedastic data sets. A P value <.05 was regarded as statistically significant. Histologic results showed more hypertrophic chondrocytes in growth plate cartilage area in sham Ncam −/− mice as compared to sham WT mice ( Figure 1B). The data suggest that NCAM deficiency enhances chondrocyte hypertrophy in chondrogenic differentiation in vivo. Chondrocyte hypertrophy-like changes play a crucial role in OA cartilage degeneration. 25 To elucidate the role of NCAM in OA, experimental OA mice were induced by MIA in WT and Ncam −/− mice. Cell clusters, reduced matrix staining, and severe chondrocyte hypertrophy in articular cartilage and growth plate of Ncam −/− mice were notably accelerated, early observed in 10 days ( Figure 1A,B), whereas similar degree of destruction emerged in cartilage of WT mice until 21 days ( Figure S1). The growth plate thickness was also increased with more hypertrophic chondrocytes ( Figure 1E). The hypertrophy-characteristic marker RunX2, vital for pro-  MSCs as compared to WT cells ( Figure 2C), whereas the mRNA levels of chondrocyte differentiation genes Sox9 and Col 2a (collagen II) were comparable between the two groups ( Figure S3). To further confirm the effect of NCAM deficiency on hypertrophic differentiation, the protein levels of hypertrophic marker RunX2 were detected by Western blotting. As shown in Figure 2D,E, the induction of RunX2 in Ncam −/− MSCs was higher than that in WT cells. The immunofluorescence assay was also employed to examine the expression of another hypertrophic marker Col X. Like the expression pattern of RunX2, Col X expression was upregulated in Ncam −/− MSCs ( Figure 2F,G). These findings further support the conclusion that loss of NCAM function promotes hypertrophic differentiation during chondrogenic differentiation of MSCs. Together, these data further support our hypothesis that NCAM plays an important role in chondrocyte hypertrophy during chondrogenic differentiation.

| ERK signaling contributes to NCAM deficiencyinduced chondrocyte hypertrophy
It has been shown that the ERK signaling pathway plays an essential role in hypertrophic and terminal differentiation events of growth plate chondrocytes. 28 The activation of ERK signaling was examined to explore the mechanism underlying NCAM deficiency induced hypertrophic chondrocyte differentiation. As shown in Figure 4A, a stronger phosphorylation of ERK was observed in chondrocyte-differentiated Ncam −/− MSCs as compared to the WT cells. To determine the significance of ERK activation under these conditions, we applied the ERK inhibitor U0126 during chondrogenic induction. As expected, phosphorylation of ERK was inhibited by U0126 ( Figure 4B). As a result of ERK inhibition, the mRNA expression and protein production of Col X and RunX2 in Ncam −/− MSCs were markedly downregulated ( Figure 4C,D). Accordingly, matrix mineralization and accumulation of calcium were also inhibited by U0126 ( Figure 4E,F) in Ncam −/− MSCs. These data indicate that ERK signaling is involved in NCAM deficiency-induced chondrocyte hypertrophy.
To further illustrate the contribution of ERK signaling in hypertrophic differentiation, we upregulated ERK activation after chondrogenic induction for 30 minutes in mouse MSCs by transfecting the constitutively active form of MEK (EMEK) ( Figure S4A). As shown in Figure S4B, the introduction of EMEK could increase the expression of RunX2 at protein levels. Alizarin red staining and quantitative analysis also revealed that ERK activation leads to the increase in mineral accumulation ( Figure S4C,D).

| NCAM inhibits chondrocyte hypertrophy in cellular OA model
It has been reported that the level of interleukin (IL)-1β is elevated in the synovial fluid, subchondral bone, and cartilage of joints in OA patients. 29,30 Research has shown that IL-1β induces calcification and increases the expression of hypertrophic gene MMP-13 and Col X. 31 F I G U R E 5 Neural cell adhesion molecule (NCAM) expression is downregulated in IL-1β-stimulated mesenchymal stem cells (MSCs) and ATDC5 cells. A, B, MSCs were induced with or without IL-1β (0.01, 0.1, and 1 ng/mL) for 1 hour and then underwent chondrogenic induction for 3 days. A, The mRNA levels of NCAM were determined by qPCR, the results are expressed as the mean ± SEM of three independent experiments. **P < .01, ***P < .001, compared with cells without IL-1β treatment. B, The protein levels of NCAM were determined by Western blotting. C, Level of NCAM was quantified by densitometry and normalized to β-actin (n = 3; mean ± SEM; **P < .01, compared with cells without IL-1β treatment). D, E, ATDC5 cells were stimulated with or without IL-1β (0.01 and 0.1 ng/mL) for 1 hour and then induced with differentiation medium for 3 days. D, The mRNA levels of NCAM were determined by qPCR (n = 3; mean ± SEM; ***P < .001, compared with cells without IL-1β treatment). E, The protein levels of NCAM were determined by Western blotting. F, Level of NCAM was quantified by densitometry and normalized to β-actin. Data are representative of three independent experiments and values are means ± SEM. **P < .01, ***P < .001, compared with cells without IL-1β treatment We showed here that IL-1β upregulates the expression of hypertrophic makers RunX2 and Col X in mouse MSCs ( Figure S5A). These data suggested that IL-1β stimulation was a useful cellular OA model in vitro.
In such model of OA, we found that the expression of NCAM was signif-  The expression of NCAM was analyzed by immunoblotting with antibodies against NCAM or c-Myc tag, respectively. B, Cells were treated with chondrocyte-differentiation medium for 3 days in the presence of IL-1β (0.01 ng/mL) and the expressions of RunX2 and Col X were detected by immunoblotting. C, D, cells were differentiated for 3 days with 0.01 and 0.1 ng/mL IL-1β, and qPCR was performed to examine the mRNA levels of RunX2 (C) and Col X (D). (n = 3; mean ± SEM; *P < .05, **P < .01, compared with Veh group). E, The expression of Col X in differentiated Veh and NCAM-overexpressing cells with IL-1β (0.01 ng/mL) was detected by immunofluorescence microscopy (original magnification ×200, scale bars 100 μm) of the growth plate. 32,33 In the present study, we provide the first evi- Multiple signaling molecules were shown to regulate the differentiation of chondrocytes from the initial induction of mesenchymal progenitor cells to the terminal maturation of hypertrophic chondrocytes, 9,11 including mitogen-activated protein kinase (MAPK) pathways. 37,38 The role of ERK/MAPK pathway remains puzzling that some studies report a positive, and others a negative, action on chondrocyte proliferation and hypertrophy. 39 In our study, we demonstrated that the activation of ERK signaling was increased in Ncam −/− MSCs. We also showed that hypertrophic markers and mineral accumulation were blocked by ERK inhibitor  43 These results are consistent with our data that NCAM deficiency elevated chondrocyte hypertrophy by ERK activation, supporting a positive role of ERK signaling in OA cartilage hypertrophic changes. In our previous work, ERK is found to be activated by NCAM and the activation of ERK is partially responsible for the migration of MSCs. 36 However, the present study showed a negative regulation of NCAM on ERK in the chondrocyte hypertrophy in chondrogenic differentiation. The results are very interesting and we are curious why NCAM behaves so differently in cell migration and chondrogenic differentiation of MSCs, which needs to be further investigated.
OA is characterized by inflammation and catabolism in joint environment, leading to progressive degeneration of cartilage. 5 In the present study, we found that expression of NCAM was decreased in both MSCs and chondrogenic cells stimulated with IL-1β, an inflammatory cytokine frequently used in developing an OA cellular model. Chondrocyte hypertrophy-like changes, such as hypertrophy genes expression and cartilage calcification, are reported in (experimental) OA. 44 In our study, the features of excessively hypertrophic chondrocyte differentiation induced by NCAM deficiency are similar to those of hypertrophy-like chondrocytes in OA. Adhesion of adhesion molecules such as N-cadherin and NCAM are essential for differentiation of mesenchymal cells into chondrocytes. 32 Here, the role of NCAM was determined in WT and Ncam −/− OA mice. Our data imply that NCAM can also be considered as a potential regulator of chondrocyte hypertrophy in the pathogenesis of OA. In recent years, a great deal of attention has been focused on cell-based therapeutic strategy for cartilage degeneration using MSCs, 45 however, chondrogenic differentiation of MSCs is usually inefficient that excessive chondrocyte hypertrophy is observed under inflammatory intra-articular conditions caused by OA. How to inhibit chondrocyte hypertrophy-like changes in treatment of cartilage damage using MSCs remains a challenge. 46,47 In this study, we demonstrated that overexpression of NCAM pre-

CONFLICT OF INTEREST
The authors indicated no potential conflicts of interest.