Contribution of the Interleukin-6/STAT-3 Signaling Pathway to Chondrogenic Differentiation of Human Mesenchymal Stem Cells

Transcription

1 ARTHRITIS & RHEUMATOLOGY Vol. 67, No. 5, May 2015, pp DOI /art VC 2015, American College of Rheumatology Contribution of the Interleukin-6/STAT-3 Signaling Pathway to Chondrogenic Differentiation of Human Mesenchymal Stem Cells Masahiro Kondo, 1 Kunihiro Yamaoka, 2 Kei Sakata, 1 Koshiro Sonomoto, 2 Lin Lin, 2 Kazuhisa Nakano, 2 and Yoshiya Tanaka 2 Objective. Mesenchymal stem cells (MSCs) are multipotent cells that can differentiate into chondrocytes. Articular cartilage contains MSC-like chondroprogenitor cells, which suggests their involvement in the maintenance of cartilage homeostasis by a self-repair mechanism. Interleukin-6 (IL-6) is a cytokine with a wide range of physiologic functions, which are produced by MSCs in a steady manner and in large quantities. The purpose of this study was to investigate the involvement of IL-6 signaling in MSC differentiation into chondrocytes. Methods. Human bone marrow derived MSCs were cultured using a pellet culture system in medium containing transforming growth factor b3. Chondrogenic Supported in part by Research Grants-In-Aid for Scientific Research from the Ministry of Health, Labor, and Welfare of Japan and the Ministry of Education, Culture, Sports, Science, and Technology of Japan and by a Grant for Advanced Research from the University of Occupational and Environmental Health, Japan. 1 Masahiro Kondo, MSc, Kei Sakata, MSc: University of Occupational and Environmental Health, Kitakyushu, Japan, and Mitsubishi Tanabe Pharma, Osaka, Japan; 2 Kunihiro Yamaoka, MD, PhD, Koshiro Sonomoto, MD, PhD, Lin Lin, MD, Kazuhisa Nakano, MD, PhD, Yoshiya Tanaka, MD, PhD: University of Occupational and Environmental Health, Japan, Kitakyushu, Japan. Dr. Yamaoka has received consulting fees and/or speaking fees from Pfizer, Chugai, Mitsubishi Tanabe Pharma, Takeda Industrial Pharma, GlaxoSmithKline, Nippon Shinyaku, Eli Lilly, Janssen, Eisai, Astellas, and Actelion (less than $10,000 each). Dr. Tanaka has received consulting fees, speaking fees, and/or honoraria from Mitsubishi Tanabe Pharma, Eisai, Pfizer, Abbott Immunology, Janssen, Takeda Industrial Pharma, Santen, AstraZeneca, Astellas, Asahi Kasei, UCB, and GlaxoSmithKline (less than $10,000 each) and from AbbVie and Chugai (more than $10,000 each) and has received research grant support from Bristol-Myers Squibb, Mitsubishi Tanabe, MSD, Takeda Industrial Pharma, Astellas, Eisai, Chugai, Pfizer, and Daiichi-Sankyo. Address correspondence to Yoshiya Tanaka, MD, PhD, First Department of Internal Medicine, University of Occupational and Environmental Health, Japan, 1-1 Iseigaoka, Yahatanishi-ku, Kitakyushu, Fukuoka , Japan. tanaka@med.uoeh-u.ac.jp. Submitted for publication February 7, 2014; accepted in revised form January 13, differentiation was detected by cartilage matrix accumulation and chondrogenic marker gene expression. Results. IL-6 was detected at a high concentration in culture supernatants during chondrogenic differentiation. The expression of the IL-6 receptor (IL-6R) was significantly increased, accompanied by markedly increased phosphorylation and expression of STAT-3. Addition of IL-6 and soluble IL-6R (sil-6r) to the chondrogenic culture resulted in concentration-dependent increases in cartilage matrix accumulation and cartilage marker gene expression (type II collagen/aggrecan/type X collagen). Phosphorylation of the master transcription factor SOX9 was enhanced upon addition of IL-6 and sil-6r. STAT-3 knockdown suppressed chondrogenic differentiation. IL-6 and the MSC markers CD166 and nestin were colocalized in macroscopically normal human cartilage taken from the lateral femoral compartment of knees with medial tibiofemoral osteoarthritis. Conclusion. During differentiation of human MSCs into chondrocytes, the activation of IL-6/STAT-3 signaling positively regulated chondrogenic differentiation. The presence of IL-6 around MSC-like cells in the cartilage tissue was identified, suggesting that IL-6 contributes to homeostasis and cartilage self-repair by promoting chondrogenic differentiation. Articular cartilage is a structurally unique tissue, lacking blood, lymph vessels, and nerves, and it is considered to be in a low-nutrient, low-oxygen environment because of its dependency on nutrient and oxygen supplies primarily from the synovial fluid. Chondrocytes scattered in the cartilage matrix show virtually no proliferative capacity, and it has not been fully elucidated how homeostasis of cartilage tissue is maintained (1). Cartilage was once considered to consist of 1250

2 CONTRIBUTION OF THE IL-6/STAT-3 SIGNALING PATHWAY TO CHONDROGENESIS 1251 only chondrocytes. However, several studies in recent years have revealed that cartilage contains chondrogenic progenitor cells with mesenchymal stem cell (MSC) like characteristics (2 5), suggesting that the homeostasis of cartilage is maintained through differentiation from chondroprogenitor cells into chondrocytes, which allows the turnover of cartilage cells. MSCs are multipotent cells that can be readily collected from the bone marrow, adipose tissue, and other mesodermal tissues. Furthermore, MSCs are capable of differentiating into mesenchymal lineages, including osteoblasts, chondrocytes, and adipocytes (6). Besides their differentiation potential, MSCs exhibit immunosuppressive activity (7) and inhibitory effects on osteoclast differentiation (8) via trophic effects, by releasing various humoral factors (9). Previous studies have shown that interleukin-6 (IL-6) is one of the humoral factors produced in the largest quantities by MSCs (10). The proinflammatory functions of IL-6 have been widely recognized (11). However, IL-6 is a pleiotropic cytokine that not only acts on the immune system, but also is involved in various physiologic events, including hematopoietic stem cell proliferation and differentiation, tumorigenesis, proliferation, and differentiation of neural cells (12); thus, IL-6 can be beneficial or harmful depending on target cells, organs, and in vivo environment. For example, anti IL-6 therapy is effective in some diseases, such as rheumatoid arthritis (RA) (11), yet IL-6 exhibits antiinflammatory activity in certain pathologic states, such as acute lung injury and sepsis (13). Interestingly, MSCs steadily produce large quantities of IL-6 even under nonstimulatory conditions. It has been reported that the IL-6 produced by MSCs contributes to the maintenance of the stemness of MSCs and exerts proliferative and antiapoptotic effects by acting as an autocrine/paracrine factor (14). The detailed physiologic significance is yet to be explored, however. The effect of IL-6 signaling on the chondrogenic differentiation of MSCs is also not well understood. In the present study, we investigated the involvement of IL-6 signaling in the chondrogenic differentiation of human MSCs using an in vitro pellet culture system based on a chondrogenic induction medium containing transforming growth factor b3 (TGFb3). Activation of the IL-6/STAT-3 pathway was shown to be important for the chondrogenic differentiation of MSCs and to positively regulate chondrogenic differentiation. In addition, MSC-like cells and IL-6 colocalized in human articular cartilage, suggesting that besides its known functions, IL-6 also aids in the maintenance of homeostasis in cartilage. MATERIALS AND METHODS Cell culture. Human MSCs were purchased from Lonza. Multipotency was confirmed by the differentiation of MSCs into osteoblasts, chondrocytes, and adipocytes. Cell surface markers were positive for CD29, CD44, CD105, and CD166 and were negative for CD14, CD34, and CD45, as determined by the manufacturer. Following the instructions recommended by the manufacturer, the cells were cultured at 37 C in an atmosphere containing 5% CO 2 in MSC growth medium (Lonza) and maintained at subconfluence to prevent spontaneous differentiation. Cells from passages 2 4 were used in this study. Chondrogenesis and cell treatment. Human MSCs at subconfluent conditions were trypsinized, and aliquots of cells/well were added to a 96-well, roundbottomed, ultra low attachment surface plate (Corning), and the plate was spun at 400g for 5 minutes. For differentiation into chondrocytes, cells were cultured in a commercial chondrogenic induction medium (hmsc Differentiation BulletKitchondrogenic; Lonza) in the absence or presence of 10 ng/ml of recombinant human TGFb3 (Lonza). The cell pellets formed free-floating aggregates within the first 24 hours. Recombinant human IL-6 (Miltenyi Biotec) and human soluble IL-6 receptor (sil-6r; R&D Systems) or anti IL-6R antibody (tocilizumab; Chugai Pharmaceutical) were added to the culture medium at concentrations indicated below during induction of chondrogenesis. The medium was replaced every 2 3 days, and aggregates or culture supernatants were collected at the indicated time points for analysis. Measurement of IL-6 in culture supernatants. Culture supernatants were collected 2 days after the last medium change at the indicated time points and stored at 280 C until analyzed. IL-6 concentrations were measured by cytometric bead array (Becton Dickinson) using a FACSVerse flow cytometer (Becton Dickinson), and data were analyzed using the FCAP Array software (Becton Dickinson). Western blotting. Chondrogenic aggregates were homogenized in a lysis buffer containing 50 mm Tris (ph 8.0), 150 mm NaCl, and 1% Nonidet P40, supplemented with a protease and phosphatase inhibitor cocktail tablet (Complete Mini and PhosStop, respectively; Roche) using a microhomogenizer (Sarstedt). To analyze the effect of IL-6 and sil-6r on Smad2 expression, cells were seeded at a density of /cm 2 on a 24-well plastic plate in Dulbecco s modified Eagle s medium (DMEM) containing 5% fetal calf serum (FCS) and antibiotics. Cells were incubated at 37 C for 24 hours, and then placed in DMEM without FCS. After 16 hours of starvation in an FCS-free medium, TGFb3 (10 ng/ ml) was administered for the indicated times. Cells were then washed with ice-cold phosphate buffered saline (PBS). IL-6 and sil-6r were added throughout the culture period. For preparation of whole cell lysates, the cells were lysed with lysis buffer. To analyze the gene knockdown efficiency of STAT- 3, MSCs were seeded at a density of cells/cm 2 on a 24-well plastic plate in MSC growth medium and then transfected with STAT-3 small interfering RNA (sirna). Two days after transfection, the cells were lysed with lysis buffer. Protein concentrations were determined using a bicinchoninic acid protein assay kit (Pierce), and an equal amount of protein was loaded in each experiment. Immunoblotting was

3 1252 KONDO ET AL performed with antibodies against IL-6Ra (MCA822; Bio- Rad), phospho STAT-3 (catalog no. 9131; Cell Signaling Technology), STAT-3 (catalog no. 9132; Cell Signaling Technology), STAT-1 (catalog no. 9172; Cell Signaling Technology), phospho-smad2 (138D4; Cell Signaling Technology), Smad2 (D43B4; Cell Signaling Technology), phospho-sox9 (ab59252; Abcam), or SOX9 (ab26414; Abcam), followed by the appropriate secondary antibodies (GE Healthcare). As a loading control, b-actin (catalog no. A-1978; Sigma) was used. To quantify the band intensities, densitometric analyses were performed using CS Analyzer version 3.0 software (Atto). The relative value of each band was calculated as the intensity of the target band divided by the intensity of the loading control. Histologic analysis and immunohistochemical staining. Aggregates were harvested 21 days after chondrogenic induction, fixed for 3 hours at room temperature in 10% buffered formalin, and prepared for paraffin embedding. To detect matrix proteoglycans, sections (4 mm thick) were stained with 0.1% Safranin O solution (Muto Pure Chemicals) for 2 minutes and counterstained with hematoxylin. For immunohistochemistry, the sections were deparaffinized, hydrated, and incubated for 5 minutes in 0.4 mg/ml of proteinase K (Dako). Endogenous peroxidases were quenched for 30 minutes with 3% hydrogen peroxide solution (Wako Pure Chemicals). After washing with PBS and incubation for 1 hour in blocking solution (ProteinBlock; Dako), the slides were incubated with polyclonal rabbit anti-human type II collagen antibody (ab34712; Abcam) at a 1:200 dilution for 1 hour at room temperature. The sections were then washed with PBS and incubated for 30 minutes with horseradish peroxidase conjugated goat antirabbit secondary antibody (Nichirei). Antigens were visualized using a 3,39-diaminobenzidine tetrahydrochloride substrate (Dako) and counterstained with hematoxylin. Slides were coverslipped and examined using BIOREVO BZ-9000 fluorescence microscope (Keyence) equipped with a Plan Apo 103/0.45 objective (Nikon). For immunofluorescence microscopy of human cartilage, the sections were deparaffinized, hydrated, and incubated for 10 minutes in 0.4 mg/ml of proteinase K. After washing with PBS and incubation for 1 hour in blocking solution, the slides were incubated for 2 hours at room temperature with rabbit anti-human IL-6 antibody (C42362; LS-Bio) and mouse anti-human CD166 (ab175422; Abcam) or mouse anti-human nestin antibody (ab22035; Abcam) at a 1:200 dilution with Can Get Signal immunostain solution A (Toyobo). Sections were then washed with PBS and incubated for 1 hour with fluorescein isothiocyanate conjugated anti-rabbit secondary antibody (catalog no. F9887; Sigma) and rhodamine-conjugated anti-mouse secondary antibody (catalog no ; Thermo) at a 1:200 dilution. After washing with PBS, slides were mounted with Vectashield mounting medium (Vector). Slides were examined with a BIOREVO BZ-9000 fluorescence microscope equipped with a Plan Apo 43/0.2 and A Plan Fluor ELWD 203/0.45 objective (Nikon). BZ-II Viewer software (Keyence) was used for image acquisition and processing. Gene expression analysis by real-time polymerase chain reaction (PCR). Total RNA was extracted from each aggregate using a microhomogenizer (Sarstedt) in RLT buffer (Qiagen). Total RNA was purified using an RNeasy Mini kit (Qiagen), and first-strand complementary DNA (cdna) was prepared using a high-capacity RNA-to-cDNA kit (Applied Biosystems) according to the specifications provided by the manufacturer. Real-time PCR was performed in a StepOne Plus system (Applied Biosystems). TaqMan Gene Expression Assay (Applied Biosystems) primer/probe pairs were used to analyze the expression of b-actin (Hs _m1), Figure 1. Interleukin-6 (IL-6)/STAT-3 pathway activation during chondrogenesis. A, Human mesenchymal stem cells (MSCs) were cultured in monolayer on 24-well plates at a density of 1,000 cells/ well with MSC growth medium. After 2 days of culture, IL-6 levels in the supernatants were measured. Values are the mean 6 SD of 3 samples. B, MSCs were cultured as aggregates in chondrogenic induction medium supplemented with transforming growth factor b3. IL-6 levels in the supernatants were evaluated at the indicated time points. Values are the mean 6 SD of 5 samples. C, IL-6 receptor a (IL-6Ra) levels in chondrogenically induced aggregates were determined at the indicated time points by Western blotting. D, Total and phosphorylated STAT-3 in aggregate lysates were determined at the indicated time points by Western blotting. Asterisk indicates unknown cross-reactive bands. In C and D, b-actin was used as a loading control. Results are representative of 2 independent experiments with similar findings.

4 CONTRIBUTION OF THE IL-6/STAT-3 SIGNALING PATHWAY TO CHONDROGENESIS 1253 type II collagen (Hs _m1), aggrecan (Hs _ m1), type X collagen (Hs _m1), and STAT-3 (Hs _m1) genes. The relative expression of each gene was normalized to that of b-actin, and the relative quantities of transcript were compared with an MSC control (cultured under conventional 2-dimensional conditions) and analyzed using the 2 2DDCt method (15). Transfection with sirna. The following sirnas were purchased from Invitrogen: STAT-3 sirna-1 (59-GCCAAUU GUGAUGCUUCCCUGAUUG-39), STAT-3 sirna-2 (59- GA UAACGUCAUUAGCAGAAUCCUCAA-39), and negative control sirna (high GC content control sirna; ). Transfection was performed using Lipofectamine RNAiMAX (Invitrogen). Briefly, human MSCs were cultured in a culture flask (Easy Flask 75; Nunc) in 8 ml of MSC growth medium free of antibiotics. At subconfluence, transfection reagents containing 12 pmoles of sirna and 16 ml oflipofectamine RNAiMAX in a final volume of 1.6 ml with Opti-MEM I (Invitrogen) was added to each flask, and cells were incubated at 37 C for 48 hours prior to chondrogenic differentiation assay. Preparation of human cartilage sections. Human cartilage was obtained from the knee joints of patients with medial knee osteoarthritis (OA) who underwent total knee arthroplasty at the University of Occupational and Environmental Health Hospital. Patients with systemic inflammatory diseases such as RA were excluded. Macroscopically normal cartilage taken from the lateral femoral compartment of 3 female patients (ages 70 years, 83 years, and 79 years) were used in this experiment. The study was approved by the ethics committees of the University of Occupational and Environmental Health, and all patients gave their informed consent. The samples were prepared using a handsaw and were directly fixed in 4% paraformaldehyde for 1 day. Next, fixed samples were decalcified by soaking in 80% ethanol for 1 day, and then the solution was changed to 10% EDTA. The EDTA solution was replaced every 3 4 days for 2 weeks. Decalcified samples were cut into small pieces, fixed for 2 hours in 10% buffered formalin, and then prepared for paraffin embedding and sliced into 4-mm thick sections. Figure 2. Interleukin-6 (IL-6) and soluble IL-6 receptor a (sil-6ra) enhancement of transforming growth factor b3 (TGFb3) induced chondrogenic differentiation. A, Mesenchymal stem cells (MSCs) were cultured in pellets in the presence or absence of TGFb3. Shown are photographs of the aggregates cultured for 14 days with IL-6 and/or sil-6r. Scale bar 5 1 mm. B, Aggregates were cultured for 21 days with IL-6 and/or sil-6r and were stained with Safranin O or anti type II collagen antibody. Scale bar mm. Results in A and B are representative of 2 independent experiments with similar findings. C, The wet weight of aggregates treated for 14 days with IL-6 and/or sil-6r was determined. Values are the mean 6 SD of 6 aggregates from 3 independent experiments. * 5 P, 0.05 by Dunnett s multiple comparison test; ## 5 P, 0.01 by Student s t-test. D, COL2A1, ACAN, and COL10A1 mrna levels were determined on day 21 by real-time polymerase chain reaction (PCR). Results are expressed relative to b-actin expression. E, Aggregates were cultured for 21 days in the presence of 100 ng/ml of sil-6r without additional IL-6 and were stained with anti type II collagen. Scale bar mm. F, Chondrogenic marker gene expression by aggregates induced in the presence of 100 ng/ml of sil-6r for 21 days was determined by real-time PCR. Values in D and F are the mean 6 SD of 3 aggregates from 1 of 3 independent experiments, each with similar findings.

5 1254 KONDO ET AL Statistical analysis. All quantitative data were expressed as the mean 6 SD. Differences between 2 groups were tested for statistical significance by Student s unpaired 2-tailed t-test. For comparison of.3 groups, analysis of variance (ANOVA) was used. If the ANOVA result was significant, Dunnett s multiple comparison test was used as a post hoc test. Statistical analyses were performed using GraphPad Prism version 4.00 software. P values less than 0.05 were considered significant. effect of more potent and earlier activation of STAT-3 on the chondrogenic differentiation of MSCs. Human MSCs were led to differentiate into chondrocytes by pellet-culturing in a TGFb3-containing chondrogenic RESULTS MSC expression of IL-6 and IL-6R and activation of STAT-3 during chondrogenic differentiation. Because it has been reported that MSCs produce IL-6 in large quantities (10,11,16), we initially confirmed that our human MSC samples also produced IL-6 in nonstimulatory monolayer cultures (Figure 1A). We also analyzed IL-6 production by MSCs during chondrogenic differentiation induced by pellet culturing in a chondrogenic differentiation induction medium containing TGFb3. IL-6 levels were measured in culture supernatants collected periodically (2 days after each medium replacement). MSCs also produced high levels of IL-6 during chondrogenic differentiation, with peak production on day 3 in the initial phase of induction (10,000 pg/ml), followed by a time-dependent decrease (Figure 1B). To confirm the effect of self-produced IL-6 on MSCs, we examined the expression of IL-6Ra during chondrogenic differentiation. While IL-6Ra was not appreciably expressed in undifferentiated MSCs, its expression level increased markedly after day 7, and although it decreased on day 21, it remained higher thereafter compared with the expression of undifferentiated MSCs (Figure 1C). The phosphorylation of STAT-3, a downstream signaling molecule of IL-6, was hardly detectable in undifferentiated MSCs; however, during differentiation, phosphorylation was increased after day 4, peaking on days 7 14, with a subsequent decrease on day 21. The expression of STAT-3 also followed a pattern similar to that of STAT-3 phosphorylation (Figure 1D). We consider that anti phospho STAT-3 antibody detected a shorter form of phosphorylated STAT-3, because the expression of these low molecular weight bands was similar to that of phospho STAT-3. These results suggest that MSCs become capable of receiving stimulation with self-produced IL-6 upon expression of IL-6Ra and activate STAT-3 during chondrogenic differentiation. IL-6/sIL-6R promotion of the chondrogenic differentiation of MSCs. By adding IL-6 and sil-6r concomitantly to the cell cultures, we evaluated the Figure 3. Lack of effect of the interleukin-6 (IL-6)/STAT-3 pathway on the transforming growth factor b3 (TGFB3)/Smad2 pathway, but enhancement of SOX9 phosphorylation by chondrogenic culture of human mesenchymal stem cells (MSCs). A, MSCs were cultured in monolayer, placed in serum-starved medium (without fetal bovine serum), and after 16 hours, were stimulated with TGFb3 (10 ng/ml) for the indicated time periods. Whole cell lysates were analyzed for Smad2 and phospho-smad2 expression by Western blotting. b-actin was used as a loading control. B, MSCs were cultured for 30 minutes with the indicated concentrations of IL-6 and soluble IL-6 receptor (sil-6r), TGFb3 was added, and after 15 minutes, whole cell lysates were analyzed for Smad2 and phospho-smad2 by Western blotting. Results in A and B are representative of 2 independent experiments with similar findings. C, MSCs were cultured as aggregates in chondrogenic induction medium with the indicated concentrations of IL-6 and sil-6r, and on day 7, aggregate lysates were analyzed for total and phospho-sox9 by Western blotting. D, Densitometric analysis of the Western blots in C was performed, and the data were normalized to b-actin. Results are expressed as the intensity of the band relative to that of the control (TGFb3 without cytokine; broken line). Values are the mean 6 SD of 8 independent experiments. * 5 P, 0.05; ** 5 P, 0.01 by Dunnett s multiple comparison test.

6 CONTRIBUTION OF THE IL-6/STAT-3 SIGNALING PATHWAY TO CHONDROGENESIS 1255 induction medium, and MSCs cultured in the absence of TGFb3 were used as a nondifferentiation induction control. Upon induction of differentiation in the TGFb3- containing chondrogenic induction medium, the sizes of the cell aggregates were found to increase as compared with the noninduction control (Figure 2A). Furthermore, an enhanced accumulation of the cartilage matrix was detected by Safranin O staining and immunostaining with anti type II collagen antibody (Figure 2B). Staining specificity was ensured by confirming the absence of detectable signal with isotype-matched control antibodies (data not shown). The concomitant addition of IL-6/sIL-6R to the culture medium promoted enlargement of the aggregates and accumulation of cartilage matrix in a concentrationdependent manner (Figures 2A and B). The wet weight of the aggregates also increased according to the IL-6/ sil-6r concentration; the increase observed at 100 ng/ml was significant (Figure 2C). Furthermore, evaluation of the expression of the chondrocyte marker genes COL2A1, ACAN, and COL10A1 revealed that ACAN and COL2A1 increased depending on the concentration of IL-6/sIL-6R, whereas no significant change in the expression of COL10A1 was found (Figure 2D). These results indicate that IL-6/sIL-6R promotes chondrogenic differentiation of MSCs. Moreover, the addition of sil-6r alone showed a tendency to enhance the accumulation of cartilage matrix (Figure 2E) and the expression of chondrogenic marker genes (Figure 2F). Figure 4. Role of the STAT-3 signaling pathway in chondrogenic differentiation of mesenchymal stem cells (MSCs). MSCs were transfected with control small interfering RNA (sirna) or 2 different STAT-3 sirnas (STAT-3 sirna-1 [STAT-3 #1] and STAT-3 sirna-2). A, The gene-knockdown effects of the STAT-3 sirnas and influence on STAT-1 expression were evaluated by Western blotting 2 days after transfection. b-actin was used as a loading control. B, STAT-3 mrna levels were determined on day 21 by real-time polymerase chain reaction (PCR). MSCs were incubated alone or in chondrogenic culture after no transfection or after transfection with the indicated sirnas and harvested 2 days after transfection. Aggregate culture was initiated in chondrogenic induction medium supplemented with transforming growth factor b3. C, Paraffin sections of nontransfected, control sirna treated, or STAT-3 sirna treated aggregates obtained on day 21 were stained with Safranin O or anti type II collagen antibody. Scale bar mm. D, ACAN, COL2A1, and COL10A1 mrna levels in STAT-3 sirna treated aggregates were determined on day 21 by real-time PCR. E, Aggregate lysates were evaluated on day 7 for total and phospho-sox9 by Western blotting. Results are representative of 3 independent experiments with similar findings. F, Paraffin sections of control IgG treated or anti interleukin-6 receptor (anti IL-6R) antibody (Ab) treated aggregates obtained on day 21 were stained with anti type II collagen antibody. Scale bar mm. G, ACAN, COL2A1, and COL10A1 mrna levels in untreated, control IgG treated, or anti IL-6R antibody treated aggregates were determined on day 21 by real-time PCR. Results in A, C, and F are representative of 2 independent experiments with similar findings. Values in B, D, and G are the mean 6 SD of 3 aggregates from 1 of 2 independent experiments with similar findings.

7 1256 KONDO ET AL Figure 5. Colocalization of mesenchymal stem cell (MSC) like cells and interleukin-6 (IL-6) in human articular cartilage. Cartilage was obtained from the knee joint of a 70-year-old woman with medial tibiofemoral osteoarthritis who underwent total knee arthroplasty. A, Photographs show the cartilage/bone specimen that was evaluated. The medial femoral compartment had a cartilage defect (demarcated by the broken line). The lateral femoral compartment had normal cartilage thickness and no superficial fissures or cracks, and this was used for these experiments. B, Paraffin sections of cartilage were stained with anti IL-6 antibody followed by fluorescein isothiocyanate (FITC) conjugated anti-rabbit IgG antibody (green) or with anti-cd166 antibody followed by rhodamine-conjugated anti-mouse IgG antibody (red). The distribution of IL-6 and CD166 was visualized by fluorescence microscopy at the indicated magnifications. Brightfield (BF) images are shown at the left. Merged images are shown at the right. Boxed areas are shown at higher magnification in the bottom row. C, Paraffin sections of cartilage were stained with anti IL-6 antibody followed by FITC-conjugated anti-rabbit IgG antibody (green) or with anti-nestin antibody followed by rhodamine-conjugated anti-mouse IgG antibody (red). Brightfield and merged images are shown at the left and right, respectively. Images in A C are representative of 3 different patients with similar findings. Involvement of SOX9 phosphorylation in the chondrogenic differentiation promoting effect of IL-6/ sil-6. TGFb3 activates Smad2-mediated intracellular signaling and acts as an essential factor in chondrogenic differentiation. To investigate the direct effect of IL-6/sIL-6R on the TGFb3 signaling pathway, Smad2 phosphorylation was examined. Phosphorylation of Smad2 increased 15 minutes after stimulation with TGFb3 and gradually decreased thereafter (Figure 3A). The addition of IL-6/sIL-6R did not alter Smad2 phosphorylation (Figure 3B). Next, we examined their effect on SOX9, which is a master transcription factor in chondrogenic differentiation (17,18), and its transcriptional activity is regulated by phosphorylation (19,20). As we reported previously (21), the phosphorylation and expression levels of SOX9 increased on day 7 in chondrogenic differentiation by pellet culture with TGFb3, and addition of IL-6/sIL-6R significantly increased SOX9 phosphorylation (by ;30%) as compared with SOX9 phosphorylation without the addition of IL-6/sIL-6R (Figures 3C and D). These results suggest that the phosphorylation of SOX9 plays a partial role in the chondrogenic differentiation enhancing effects of IL-6/sIL- 6R. Importance of the IL-6/STAT-3 signaling pathway for the chondrogenic differentiation of MSCs. To analyze the role of the IL-6 signaling pathway during chondrogenic differentiation, gene knockdown of STAT-3 was performed in MSCs. Knockdown efficiency at 48 hours after transfection of 2 STAT-3 sirnas with different sequences was evaluated by Western blotting. The results indicate that STAT-3 sirna-1 almost completely suppressed the expression of STAT-3a, but not STAT-3b, whereas STAT-3 sirna-2 suppressed the expression of both STAT-3a and STAT-3b. Neither

8 CONTRIBUTION OF THE IL-6/STAT-3 SIGNALING PATHWAY TO CHONDROGENESIS 1257 sirna-1 nor sirna-2 affected the expression of STAT- 1 (Figure 4A). MSCs recovered at 48 hours after transfection of STAT-3 sirnas were subjected to pellet culture, and evaluations were conducted after 21 days. Realtime PCR confirmed that both STAT-3 sirnas completely suppressed STAT3 gene expression (Figure 4B). Furthermore, compared with the control sirna, the accumulation of cartilage matrix was suppressed in cell aggregates of MSCs subjected to STAT-3 knockdown regardless of the sirna used (Figure 4C). In addition, while the increase in COL2A1 and ACAN expression upon induction of differentiation was equally suppressed by both sirnas, COL10A1 was suppressed only by sirna-2, and not by sirna-1 (Figure 4D). On day 7, the SOX9 phosphorylation level was reduced in aggregates of STAT-3 knockdown MSCs (Figure 4E). Furthermore, the addition of anti IL-6R antibody showed inhibitory effects on cartilage matrix accumulation (Figure 4F) and chondrogenic marker gene expression (Figure 4G). These results suggest that IL- 6/STAT-3 plays an important role in the chondrogenic differentiation of MSCs. Colocalization of IL-6 and either the MSC marker CD166 or nestin in human cartilage tissue. We also analyzed cartilage tissues excised during knee replacement arthroplasty in patients with medial knee OA. While the articular cartilage in the medial compartment showed macroscopic OA-like changes, the cartilage in the lateral tibiofemoral compartment showed normal thickness and no macroscopic fissures or cracks (Figure 5A). In the lateral femoral cartilage, double immunofluorescence staining was performed using antibodies against IL-6 and CD166 (one of the MSC surface markers). IL-6 positive cells were observed at high density in the superficial zone of the cartilage and were also detected in the calcified and deep zones. CD166-positive cells were detected at high density in the superficial zone. The IL-6 positive cells and the CD166-positive cells detected in the superficial zone were completely superimposable (Figure 5B). Staining with nestin, a marker of multilineage progenitor cells (22 24), showed a pattern similar to that of CD166 (Figure 5C). Staining specificity was ensured by confirming the absence of detectable signal with isotype-matched control antibodies (data not shown). The results shown in Figure 5 are representative of a single patient (patient 1). Similar results were obtained with tissues from an additional 2 patients (patients 2 and 3); however, the density of the distribution of IL-6 and CD166/nestin-positive cells varied between donors (data not shown). These results revealed that IL-6 was present in the vicinity of MSClike cells in the cartilage. DISCUSSION IL-6 is a cytokine with pleiotropic functions that is involved in a wide range of biologic functions, including immunoregulation, hematopoiesis, inflammation, and tumorigenesis (12). Its effect on the chondrogenic differentiation of MSCs, however, is not well understood. In the present study, activation of the IL-6/ STAT-3 pathway was shown to positively regulate chondrogenic differentiation. Moreover, MSC-like cells and IL-6 were demonstrated to colocalize in human articular cartilage. As reported previously (10), the human bone marrow derived MSCs that we used also showed IL-6 producing capability at steady state (Figure 1A). We also examined the humoral factors produced by MSCs at steady state with an antibody-based protein array (80 factors) and found that IL-6 was the factor produced in the largest quantity (data not shown). During chondrogenic differentiation by pellet culture, MSCs had the ability to produce high levels of IL-6, particularly during the early stage of differentiation (Figure 1B), and this was reduced after day 14. This suggests that the ability to produce high levels of IL-6 is a property of undifferentiated MSCs, which is lost as chondrogenic differentiation advances. In contrast, the activation of STAT-3 was not detected in undifferentiated MSCs (Figure 1D), despite the availability of the IL-6 produced by MSCs. This is presumably because undifferentiated MSCs do not express IL-6R (Figure 1C) and, thus, are unable to undergo stimulation by IL-6 despite its availability. Therefore, the level of IL-6R, rather than IL-6, may be the rate-limiting factor. Involvement of IL-6/STAT-3 signaling in chondrogenic differentiation was studied by the concomitant addition of IL-6/sIL-6R, which promoted chondrogenic differentiation. As shown in Figure 1C, IL-6R expression was low at the early differentiation stage (before day 4), and the costimulation was intended to provide an earlier and stronger stimulation. However, because the IL-6 produced by MSCs was available, the addition of sil-6r alone showed a tendency to promote chondrogenic differentiation (Figures 2E and F). The promotion of chondrogenic differentiation by IL-6/sIL-6R also correlated with the increased phosphorylation of SOX9, the master transcription factor for chondrogenic

9 1258 KONDO ET AL differentiation, suggesting enhanced transcriptional activity of downstream cartilage-associated genes (Figures 3C, 3D, and 4E). However, only an ;30% increase in SOX9 phosphorylation was induced by IL- 6/sIL-6R, suggesting the involvement of additional unknown mechanisms, which will be the subject of future studies. We focused on the effect of IL-6 on chondrogenic differentiation; however, it is also known that IL-6/IL-6R treatment enhances osteoblastic differentiation (25) and inhibits adipogenic differentiation (26) of human MSCs in vitro. In the STAT-3 knockdown experiments, we used two sirnas. There are two STAT-3 isoforms, STAT-3a and STAT-3b (full-length STAT-3a and truncated STAT-3b) (27), and sirna-1 failed to suppress STAT-3b (Figure 4A). STAT-3b knockdown incompetency may underlie the inability of sirna-1 to suppress the expression of COL10A1 (Figure 4D). CD166 and nestin were used to confirm the presence and distribution of MSC-like cells in human articular cartilage. CD166 is an adhesion molecule (also known as the activated leukocyte cell adhesion molecule [ALCAM]) and nestin is an intermediate filament that has been originally identified as a neural stem cell marker (also known as a common marker for multilineage progenitor cells including MSCs) (22 24). Although the expression of these molecules may not be limited to MSCs, given the fact that CD166 expression is limited to cell populations that have proliferative and migratory activity and that nestin is not expressed in mature chondrocytes, the CD166/nestin-positive cells shown in Figure 5 are likely to be MSC-like cells, which presumably function as a chondroprogenitor cell population. In addition, the dense distribution of MSC-like cells in the superficial zone of cartilage is also consistent with previous reports (4,28). Considering that MSCs produce IL-6 in large amounts even in the steady state (Figure 1A), MSClike cells represent the likely source of IL-6 detected in chondral superficial zone. This suggests that the IL-6 produced by MSC-like cells in cartilage promotes chondrogenic differentiation by acting as an autocrine/paracrine factor and helps to maintain cartilage homeostasis. However, IL-6 and CD166/nestin were detected in cartilage lacunae, where chondrocytes are observed. Chondrocytes are also capable of producing IL-6 upon stimulation, although under physiologic conditions (i.e., embedded in the cartilage matrix), chondrocytes are reported to have little ability to produce IL-6 (29). Therefore, chondrocytes cannot be completely ruled out as a possible source of IL-6. Nevertheless, the present study demonstrated the presence of IL-6 in proximity to MSC-like cells and the involvement of IL-6 in chondrogenic differentiation. The limitation of our study is that the cartilage used for immunohistochemistry was obtained from OA patients, and its integrity was determined only by macroscopic observation. Further evaluation using cartilage from healthy donors will be necessary to reinforce the contribution of IL-6 and MSC-like cells to cartilage homeostasis. In this study, we showed that IL-6/STAT-3 signaling positively regulates the chondrogenic differentiation of MSCs. Previous studies in aged male IL-6 knockout mice have shown that the animals spontaneously develop OA-like cartilage erosions (30) and that the production of proteoglycans is reduced (31). In addition, it has been shown that the addition of IL-6 results in enhanced cartilage matrix production in an in vitro cartilage regeneration system using cells isolated from human cartilage (32). These reported findings suggest anabolic functions of IL-6 in cartilage metabolism and are consistent with our hypothesis that IL-6 serves as a beneficial factor in the maintenance of homeostasis in cartilage tissue. In contrast, some studies have shown that IL-6 is cartilage-destructive in a mouse model of OA induced by mechanical stress or hypoxia inducible factor 2a (33) and that IL-6 stimulated the chondrocytes to exert catabolic actions by producing cartilage matrix degrading enzymes, such as aggrecanase (34). These contradictions suggest that IL-6 produces different effects on the balance between generation and decomposition, depending on its concentration and surrounding environment. The IL-6 concentration in synovial fluid from patients with OA, the typical cartilage-destructive disease, is higher than that in synovial fluid from healthy individuals (32,35) and correlates with radiographic knee OA findings (36). However, the IL-6 concentration in synovial fluid from OA patients was substantially lower than the concentration at which the differentiation-promoting effect was observed in our in vitro chondrogenic differentiation system (Figure 2), and thus, we could not address the relevance of an IL-6 increase in the synovial fluid promoting the chondrogenic differentiation of MSC-like cells. We believe that the key factor for the IL-6 promoting effects on chondrogenic differentiation is local activity in proximity to IL-6 producing cells, and therefore, further investigations will compare healthy subjects and OA patients in terms of the IL-6 levels detected in cartilage tissues. Tocilizumab (TCZ), a biologic agent that targets IL-6R, is currently used for the treatment of RA and has demonstrated clinical efficacy with improvement in joint space narrowing, which suggests an improvement

10 CONTRIBUTION OF THE IL-6/STAT-3 SIGNALING PATHWAY TO CHONDROGENESIS 1259 in cartilage destruction (11). Although our present results suggest that blockade of IL-6 signaling could lead to negative effects on cartilage homeostasis, TCZ treatment is also known to reduce the progression of cartilage degeneration (37). It is conceivable that TCZ is unable to act on MSC-like cells embedded in dense cartilage matrix, because cartilage has not been shown to be permeable to large molecules, including antibodies (38 40). The results of the present study suggest that the activation of IL-6/STAT-3 signaling positively regulates the chondrogenic differentiation of human MSCs. Furthermore, IL-6 was shown to be present in the proximity of MSC-like cells in normal cartilage tissues, suggesting that IL-6 contributes to homeostasis maintenance and self-repair of cartilage tissue by promoting chondrogenic differentiation. ACKNOWLEDGMENTS The authors thank Ms. K. Noda for technical support with the histologic analysis. We thank all medical staff members at the associated institutions for providing the data. AUTHOR CONTRIBUTIONS All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Tanaka had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study conception and design. Kondo, Yamaoka, Sonomoto, Nakano, Tanaka. Acquisition of data. Kondo, Yamaoka, Sakata, Lin. Analysis and interpretation of data. Kondo, Yamaoka, Sakata, Sonomoto, Lin, Nakano, Tanaka. ADDITIONAL DISCLOSURES Authors Kondo and Sakata are employees of Mitsubishi Tanabe Pharma. REFERENCES 1. Goldring MB, Marcu KB. Cartilage homeostasis in health and rheumatic diseases. Arthritis Res Ther 2009;11: Alsalameh S, Amin R, Gemba T, Lotz M. Identification of mesenchymal progenitor cells in normal and osteoarthritic human articular cartilage. Arthritis Rheum 2004;50: Dowthwaite GP, Bishop JC, Redman SN, Khan IM, Rooney P, Evans DJR, et al. The surface of articular cartilage contains a progenitor cell population. J Cell Sci 2004;117: Pretzel D, Linss S, Rochler S, Endres M, Kaps C, Alsalameh S, et al. Relative percentage and zonal distribution of mesenchymal progenitor cells in human osteoarthritic and normal cartilage. Arthritis Res Ther 2011;13:R Grogan SP, Miyaki S, Asahara H, D Lima DD, Lotz MK. Mesenchymal progenitor cell markers in human articular cartilage: normal distribution and changes in osteoarthritis. Arthritis Res Ther 2009;11:R Pittenger MF, Mackay AM, Beck SC, Jaiswal RK, Douglas R, Mosca JD, et al. Multilineage potential of adult human mesenchymal stem cells. Science 1999;284: Ren G, Chen X, Dong F, Li W, Ren X, Zhang Y, et al. Mesenchymal stem cells and translational medicine: emerging issues. Stem Cells Transl Med 2012;1: Oshita K, Yamaoka K, Udagawa N, Fukuyo S, Sonomoto K, Maeshima K, et al. Human mesenchymal stem cells inhibit osteoclastogenesis through osteoprotegerin production. Arthritis Rheum 2011;63: Caplan AI, Dennis JE. Mesenchymal stem cells as trophic mediators. J Cell Biochem 2006;98: Chen L, Tredget EE, Wu PY, Wu Y. Paracrine factors of mesenchymal stem cells recruit macrophages and endothelial lineage cells and enhance wound healing. PLoS One 2008;3:e Smolen JS, Schoels MM, Nishimoto N, Breedveld FC, Burmester GR, Dougados M, et al. Consensus statement on blocking the effects of interleukin-6 and in particular by interleukin-6 receptor inhibition in rheumatoid arthritis and other inflammatory conditions. Ann Rheum Dis 2013;72: Naka T, Nishimoto N, Kishimoto T. The paradigm of IL-6: from basic science to medicine. Arthritis Res 2002;4 Suppl 3:S Xing Z, Gauldie J, Cox G, Baumann H, Jordana M, Lei XF, et al. IL-6 is an antiinflammatory cytokine required for controlling local or systemic acute inflammatory responses. J Clin Invest 1998;101: Pricola KL, Kuhn NZ, Haleem-Smith H, Song Y, Tuan RS. Interleukin-6 maintains bone marrow-derived mesenchymal stem cell stemness by an ERK1/2-dependent mechanism. J Cell Biochem 2009;108: Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2 2DDCt Method. Methods 2001;25: Aggarwal S, Pittenger MF. Human mesenchymal stem cells modulate allogeneic immune cell responses. Blood 2005;105: Akiyama H, Chaboissier MC, Martin JF, 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 2002;16: Bi W, Deng JM, Zhang Z, Behringer RR, de Crombrugghe B. Sox9 is required for cartilage formation. Nat Genet 1999;22: Huang W, Zhou X, Lefebvre V, de Crombrugghe B. Phosphorylation of SOX9 by cyclic AMP-dependent protein kinase A enhances SOX9 s ability to transactivate acol2a1 chondrocytespecific enhancer. Mol Cell Biol 2000;20: Huang W, Chung UI, Kronenberg HM, de Crombrugghe B. The chondrogenic transcription factor Sox9 is a target of signaling by the parathyroid hormone-related peptide in the growth plate of endochondral bones. Proc Natl Acad Sci 2001;98: Kondo M, Yamaoka K, Sonomoto K, Fukuyo S, Oshita K, Okada Y, et al. IL-17 inhibits chondrogenic differentiation of human mesenchymal stem cells. PLoS One 2013;8:e Wiese C, Rolletschek A, Kania G, Blyszczuk P, Tarasov KV, Tarasova Y, et al. Nestin expression: a property of multi-lineage progenitor cells? Cell Mol Life Sci 2004;61: Mendez-Ferrer S, Michurina TV, Ferraro F, Mazloom AR, Macarthur BD, Lira SA, et al. Mesenchymal and haematopoietic stem cells form a unique bone marrow niche. Nature 2010;466: Tondreau T, Lagneaux L, Dejeneffe M, Massy M, Mortier C, Delforge A, et al. Bone marrow-derived mesenchymal stem cells already express specific neural proteins before any differentiation. Differentiation 2004;72: Sonomoto K, Yamaoka K, Oshita K, Fukuyo S, Zhang X, Nakano K, et al. Interleukin-1b induces differentiation of human mesenchymal stem cells into osteoblasts via the Wnt-5a/receptor tyrosine kinase like orphan receptor 2 pathway. Arthritis Rheum 2012;64:

11 1260 KONDO ET AL 26. Okada A, Yamasaki S, Koga T, Kawashiri SY, Tamai M, Origuchi T, et al. Adipogenesis of the mesenchymal stromal cells and bone oedema in rheumatoid arthritis. Clin Exp Rheumatol 2012;30: Schaefer TS, Sanders LK, Park OK, Nathans D. Functional differences between Stat3a and Stat3b. Mol Cell Biol 1997;17: Ozbey O, Sahin Z, Acar N, Ustunel I. Distribution of CD105 and CD166 positive cells in the proximal epiphysis of developing rat humerus. Histol Histopathol 2010;25: Beekhuizen M, Bastiaansen-Jenniskens YM, Koevoet W, Saris DB, Dhert WJ, Creemers LB, et al. Osteoarthritic synovial tissue inhibition of proteoglycan production in human osteoarthritic knee cartilage: establishment and characterization of a long-term cartilage synovium coculture. Arthritis Rheum 2011;63: De Hooge AS, van de Loo FA, Bennink MB, Arntz OJ, de Hooge P, van den Berg WB. Male IL-6 gene knock out mice developed more advanced osteoarthritis upon aging. Osteoarthritis Cartilage 2005;13: Van de Loo FA, Kuiper S, van Enckevort FH, Arntz OJ, van den Berg WB. Interleukin-6 reduces cartilage destruction during experimental arthritis: a study in interleukin-6-deficient mice. Am J Pathol 1997;151: Tsuchida AI, Beekhuizen M, Rutgers M, van Osch GJ, Bekkers JE, Bot AG, et al. Interleukin-6 is elevated in synovial fluid of patients with focal cartilage defects and stimulates cartilage matrix production in an in vitro regeneration model. Arthritis Res Ther 2012;14:R Ryu JH, Yang S, Shin Y, Rhee J, Chun CH, Chun JS. Interleukin-6 plays an essential role in hypoxia-inducible factor 2a induced experimental osteoarthritic cartilage destruction in mice. Arthritis Rheum 2011;63: Flannery CR, Little CB, Hughes CE, Curtis CL, Caterson B, Jones SA. IL-6 and its soluble receptor augment aggrecanasemediated proteoglycan catabolism in articular cartilage. Matrix Biol 2000;19: Doss F, Menard J, Hauschild M, Kreutzer HJ, Mittlmeier T, Muller-Steinhardt M, et al. Elevated IL-6 levels in the synovial fluid of osteoarthritis patients stem from plasma cells. Scand J Rheumatol 2007;36: Livshits G, Zhai G, Hart DJ, Kato BS, Wang H, Williams FM, et al. Interleukin-6 is a significant predictor of radiographic knee osteoarthritis: the Chingford Study. Arthritis Rheum 2009;60: Hashimoto J, Garnero P, van der Heijde D, Miyasaka N, Yamamoto K, Kawai S, et al. Humanized anti-interleukin-6- receptor antibody (tocilizumab) monotherapy is more effective in slowing radiographic progression in patients with rheumatoid arthritis at high baseline risk for structural damage evaluated with levels of biomarkers, radiography, and BMI: data from the SAMURAI study. Mod Rheumatol 2011;21: Maroudas A. Transport of solutes through cartilage: permeability to large molecules. J Anat 1976;122: Van Lent PL, van den Berg WB, Schalkwijk J, van de Putte LB, van den Bersselaar L. The impact of protein size and charge on its retention in articular cartilage. J Rheumatol 1987; 14: Van Lent PL, van den Bersselaar L, Grutters GJ, van den Berg WB. Fate of antigen after intravenous and intraarticular injection into mice. Role of molecular weight and charge. J Rheumatol 1989;16: