ARTHRITIS & RHEUMATOLOGY Vol. 67, No. 1, January 2015, pp 39–50 DOI 10.1002/art.38899 © 2015, American College of Rheumatology

Transcription Factor Snail Regulates Tumor Necrosis Factor ␣–Mediated Synovial Fibroblast Activation in the Rheumatoid Joint Shih-Yao Chen,1 Ai-Li Shiau,1 Yuan-Tsung Li,1 Chi-Chen Lin,2 I-Ming Jou,1 Ming-Fei Liu,1 Chao-Liang Wu,1 and Chrong-Reen Wang1 expression of Snail, Cad-11, and ␣-smooth muscle actin (␣-SMA) in synovial fibroblasts, and anti-TNF␣ therapy down-regulated the expression of Snail, Cad-11, and ␣-SMA in the joints of rats with CIA. Although synovial fibroblast transfectants in which Snail was overexpressed showed increased expression of Cad-11 and ␣-SMA and enhanced TNF␣-mediated invasive capacity and IL-6 production, synovial fibroblast transfectants from rats with CIA in which Snail was silenced showed decreased expression and had the opposite effect on these functions. Normal joints in which Snail was overexpressed had hyperplastic synovium, with increased expression of Cad-11, ␣-SMA, and IL-6. Silencing Snail expression ameliorated arthritis, with reduced Cad-11 expression and reduced levels of extracellular matrix deposition in the joints of rats with CIA, whereas overexpression of Snail exacerbated arthritis, with increased Cad-11 expression and increased levels of extracellular matrix deposition. Conclusion. Our results demonstrate that Snail regulates TNF␣-mediated activation of synovial fibroblasts in the rheumatoid joint. These findings may contribute to the pharmacologic development of therapeutics targeting synovial fibroblasts in patients with RA.

Objective. The transcription factor Snail is involved in various biologic functions. We hypothesized that this molecule regulates tumor necrosis factor ␣ (TNF␣)–mediated synovial fibroblast activation in the rheumatoid joint. The aim of this study was to examine the role of Snail in the expression of cadherin-11 (Cad-11) and myofibroblast markers, interleukin-6 (IL-6) production, and the invasive ability of cells. Methods. Synovium samples were obtained from patients with rheumatoid arthritis (RA) and from rats with collagen-induced arthritis (CIA). Synovial fibroblasts were treated with TNF␣ or a Wnt signaling inducer, and the joints of rats with CIA were injected with a TNF␣ antagonist. Modulation of Snail expression in the synovial fibroblasts and joints was performed by lentiviral vector–mediated transfer of complementary DNA or short hairpin RNA. Results. The expression of Snail and Cad-11 was higher in synovium and synovial fibroblasts from patients with RA compared with patients with osteoarthritis and was increased in rats with CIA. TNF␣ stimulation or activation of Wnt signaling up-regulated the Supported by the Ministry of Science and Technology, Taiwan (grants NSC 98-2628-B-006-014-MY3, NSC 100-2314-B-006-031MY3, NSC 101-2320-B-006-033-MY3, and MOST 103-2314-B-006058-MY3). 1 Shih-Yao Chen, PhD, Ai-Li Shiau, PhD, Yuan-Tsung Li, PhD, I-Ming Jou, MD, Ming-Fei Liu, MD, Chao-Liang Wu, PhD, Chrong-Reen Wang, MD, PhD: National Cheng Kung University Medical College, Tainan, Taiwan; 2Chi-Chen Lin, PhD: National Chung Hsing University, Taichung, Taiwan. Drs. Chen and Shiau contributed equally to this work. Address correspondence to Chao-Liang Wu, PhD, Department of Biochemistry and Molecular Biology, National Cheng Kung University Medical College, No. 1 University Road, Tainan 70101, Taiwan (e-mail: [email protected]); or to Chrong-Reen Wang, MD, PhD, Department of Internal Medicine, National Cheng Kung University Medical College, Section of Rheumatology and Immunology, National Cheng Kung University Hospital, No. 138 ShengLi Road, Tainan 704, Taiwan (e-mail: [email protected]). Submitted for publication April 10, 2014; accepted in revised form September 25, 2014.

In rheumatoid arthritis (RA), abnormal expansion of pannus tissue containing synovial fibroblasts is a crucial feature responsible for disease progression (1). Notably, RA synovial fibroblasts (RASFs) grow in vitro in an anchorage-independent manner, and these cells maintain their invasive behavior, independent of T cells, in a mouse model. However, the molecular mechanisms underlying such phenotypic characteristics remain to be elucidated (2,3). Although the outcome and severity of RA have been significantly improved by treatment with biologic agents targeting proinflammatory cytokines or immune cell lineages, the number of patients with treatment-refractory disease is considerable, and bio39

40

logic treatment is associated with inevitable side effects, including a higher risk of infection (4). Furthermore, combination therapy with different biologic agents has been hindered because of impairment of host defense without enhancement of clinical efficacy (5). Consequently, a novel approach that modifies immune responses by immunosuppressants and simultaneously targets synovial fibroblasts has been proposed, and such a combination may offer more effective control of disease activity without interfering with the host defense against infection (1,6). In particular, cadherin 11 (Cad-11), which regulates interleukin-6 (IL-6) production and invasive behavior in synovial fibroblasts, is involved in the pathogenic mechanisms leading to inflamed synovium and cartilage erosion in the rheumatoid joint (6–8). Interestingly, Cad-11 biotherapeutics have been proven to prevent or reduce arthritis in different experimental mouse models (9). Based on these observations, identification of the critical processes that regulate the expression of Cad-11 might contribute to the development of therapeutics targeting synovial fibroblasts in patients with RA. Snail, a member of the zinc finger transcription factor family consisting of Snail, Slug, and Smuc, plays an active role in various biologic functions and is involved in many disease states (10,11). This molecule is expressed in activated fibroblasts that are localized in carcinomatous tissue or are involved in the woundhealing process and pathologic fibromatosis lesions (12). Snail-deficient fibroblasts have reduced invasiveness on 3-dimensional extracellular matrix (ECM) and fail to penetrate the surface of embryo chorioallantoic membrane (13). Snail expression is up-regulated during the transition of hepatic stellate cells into the ␣-smooth muscle actin (␣-SMA)–positive phenotype, and there is concomitant expression of Snail and ␣-SMA in adipose tissue progenitor cells during induction of the myofibroblast phenotype (14,15). Furthermore, the expression of ␣-SMA in cultured synovial fibroblasts from patients with arthritis is up-regulated by the addition of transforming growth factor ␤1 (TGF␤1), and in vivo administration of this cytokine leads to the proliferation and differentiation of synovial fibroblasts into an ␣-SMA– positive phenotype in a rat model (16,17). These data suggest that Snail is a critical regulator of synovial fibroblasts, participating in the TGF␤-induced activation process into the ␣-SMA–positive myofibroblast-like phenotype in the rheumatoid joint. Interestingly, up-regulated expression of Slug in RA synovium has been observed, and tumor necrosis factor ␣ (TNF␣)–induced conversion to a myofibroblast phenotype in dermal fibroblasts has been shown to be

CHEN ET AL

mediated via the Wnt signaling pathway and effectively reversed by TNF␣-neutralizing antibodies (18,19). Notably, this cytokine plays a fundamental role in the rheumatoid joint through pathogenic mechanisms such as induction of the proinflammatory cascade and activation of synovial fibroblasts (20,21). However, the expression of Snail and its role in TNF␣-mediated synovial fibroblast activation in the rheumatoid joint have not yet been explored. In this study, we examined the role of Snail in the expression of Cad-11 and myofibroblast markers, IL-6 production, and the invasiveness of cells, by using synovium specimens obtained from patients with RA and rats with collagen-induced arthritis (CIA). Synovial fibroblasts were treated with TNF␣ or a Wnt signaling inducer, and the joints of rats with CIA were injected with a TNF␣ antagonist. Modulation of Snail expression in synovial fibroblasts and joints was performed by lentiviral vector–mediated transfer of complementary DNA (cDNA) or short hairpin RNA (shRNA). We observed that Snail and Cad-11 were expressed at higher levels in synovium and synovial fibroblasts from patients with RA compared with patients with osteoarthritis (OA), and that expression was increased in rats with CIA. TNF␣ stimulation or activation of Wnt signaling up-regulated the expression of Snail, Cad-11, and ␣-SMA in synovial fibroblasts, and antiTNF␣ therapy down-regulated the expression of Snail, Cad-11, and ␣-SMA in the joints of rats with CIA. Although normal synovial fibroblast transfectants in which Snail was overexpressed had increased levels of Cad-11 and ␣-SMA as well as an enhanced TNF␣mediated invasive capacity and increased IL-6 production, Snail-silenced synovial fibroblast transfectants from rats with CIA showed decreased expression and had the opposite effect on these functions. Normal joints in which Snail was overexpressed had hyperplastic synovium, with increased expression of Cad-11, ␣-SMA, and IL-6. The silencing of Snail expression ameliorated arthritis, with reduced Cad-11 expression and ECM deposition in the joints of rats with CIA, whereas overexpression of Snail exacerbated arthritis, with increased expression of Cad-11 and ECM deposition. PATIENTS AND METHODS Patient samples. Synovial tissue samples from patients with RA and patients with OA were obtained during orthopedic surgery. Informed consent was obtained from all patients prior to sample collection. This study was approved by the Institutional Review Board of National Cheng Kung University Hospital. Fresh synovial fibroblasts purified from synovium became a homogeneous population after ⱖ4 passages

SNAIL REGULATES TNF␣-MEDIATED SYNOVIAL FIBROBLAST ACTIVATION

and were used for the following experiments, as described previously (22). Initiation of CIA and isolation of synovial fibroblasts. To induce CIA, male Sprague-Dawley rats (⬃8 weeks of age) were immunized with bovine type II collagen and Freund’s complete adjuvant, as described previously (23). Animal experiments were performed in accordance with the guidelines approved by the Institutional Animal Care and Use Committee of National Cheng Kung University. Fresh synovial fibroblasts were isolated from rat synovium, and different lines between the fourth and seventh passages were used, as described previously (22). Lentiviral plasmid construction. A 791-bp Snail cDNA fragment (GenBank accession no. 116490) was obtained by polymerase chain reaction (PCR) amplification of cDNA obtained from the synovial fibroblasts of rats with CIA, using the following primers: forward 5⬘-ATGCCGCGCTCCTTCCTGGTCAGGAAGCCG-3⬘ and reverse 5⬘-TCAGCGAGGGCCTCCGGAGCAGCCA-3⬘. The resulting PCR product was cloned into a yT&A cloning vector (Yeastern Biotech) to generate pyT&A-Snail and subsequently sequenced to verify its correctness. The pSin-EF2-Snail-Puro plasmid was generated from the lentiviral vector pSin-EF2-Oct4-Puro (Addgene), in which the Oct4 cDNA was replaced with the Snail cDNA. The Snail cDNA was excised from the pyT&A-Snail plasmid by digestion with Xba I and Eco RI and subcloned into pSin-EF2-Oct4-Puro by digestion with Spe I (Xba I compatible) and Eco RI, resulting in pSin-EF2-Snail-Puro. The pSinEF2-Puro control plasmid encoding no transgene was constructed from pSin-EF2-Oct4-Puro by digestion with Spe I and Eco RI to delete the Oct4 cDNA, filling in the cohesive ends with klenow enzyme, and followed by self-ligation. Lentiviral vectors and stable synovial fibroblast transfectants in which Snail is overexpressed or silenced. Snail shRNA–expressing pLKO.1-shSnail (TRCN0000218784, TRCN0000234035, and TRCN000096619) and luciferase shRNA–expressing pLKO.1-shLuc (TRCN0000072246) lentiviral plasmids were obtained from the National RNAi Core Facility (Academia Sinica, Taiwan). Recombinant lentiviral vectors, LVSnail, LVSin, LVshSnail, and LVshLuc, were produced by transient transfection of 293T cells with pSin-EF2Snail-Puro, pSin-EF2-Puro, pLKO.1-shSnail, and pLKO.1shLuc, respectively, along with the packaging plasmid psPAX2 and the envelope plasmid pMD2G, as previously described (24). The virus titers were determined by cell viability assay using A549 cells to calculate the relative infection units, according to the protocol of the National RNAi Core Facility. Based on the results of reverse transcription (RT)– PCR and immunoblot analyses, we chose 2 pLKO.1-shSnail plasmids (TRCN0000218784 and TRCN0000234035) to generate lentiviral vectors encoding Snail shRNA. To produce stable transfectants in which Snail is overexpressed or silenced, normal synovial fibroblasts and synovial fibroblasts from rats with CIA were transduced with LVSnail and LVshSnail for 48 hours, respectively, in the presence of 8 ␮g/ml Polybrene (Sigma-Aldrich), and cells were then incubated with puromycin (2 ␮g/ml) for ⬃2 weeks. Control transfectants were obtained by transduction with either LVSin or LVshLuc, followed by selection with puromycin. RT-PCR analysis. Total RNA from rat synovium was isolated with TRIzol reagents (Invitrogen), and cDNA was synthesized using a Reverse-iT First Strand cDNA synthesis kit

41

(ABgene) for RT-PCR with primer pairs specific to Snail (forward 5⬘-CTCTGAAGATGCACATCCGAAGCCAC-3⬘ and reverse 5⬘-CACGGTTGCAGTGGGAGCAGGAG-3⬘), Cad-11 (forward 5⬘-ATCCTTGCCTGCATCGTCATT-3⬘ and reverse 5⬘-CAGCCCAGGTCTAGGCATGTACTG-3⬘), and GADPH (forward 5⬘-GCCATCACTGCCACCCAG-3⬘ and reverse 5⬘-TCTTACTCCTTGGAGGCCATGT-3⬘). The PCR conditions were 30 cycles of 94°C for 30 seconds, 63°C (for Snail) or 58°C (for Cad-11) for 30 seconds, and 72°C for 30 seconds. Enzyme-linked immunosorbent assay (ELISA). Normal synovial fibroblasts in which Snail was overexpressed and Snail-silenced synovial fibroblast transfectants from rats with CIA (5 ⫻ 104 cells/well in 24-well plates) were cultured in the presence or absence of TNF␣ (10 ng/ml) for 24 hours, and the IL-6 levels in supernatants were quantified by ELISA (R&D Systems). Cell invasion assay. A cell invasion assay was performed using 24-well Transwells with inserts containing 8-␮m membrane filters (Corning Costar) coated with Matrigel (Collaborative Research). LVSnail- or LVSin-transduced normal synovial fibroblasts, LVshSnail- or LVshLuc-transduced synovial fibroblasts from rats with CIA, and parental cells (5 ⫻ 104 cells/well) were placed in the upper chamber containing TNF␣ (10 ng/ml) for incubation for 20 hours, after being serumstarved and treated with the same concentration of TNF␣ for 24 hours. Cells migrating through each membrane to the lower surface were stained with Giemsa solution and counted in 3 microscopic fields at 40⫻ magnification. TNF␣ antagonist administration into the joints of rats with CIA. The rats were immunized with collagen on days 0 and 7 and received intraarticular injections of 25 ␮g or 100 ␮g etanercept (Pfizer) into the right ankle joints and phosphate buffered saline (PBS) into the left ankle joints on day 10. Lentiviral vector–mediated gene transfer in normal joints and joints from rats with CIA. The rats received intraarticular injections of LVSnail and LVSin (control) (1 ⫻ 106 relative infection units) into the right and left ankle joints, respectively. When the rats were killed 7 days later, these joints were removed for histologic examination. Furthermore, rats receiving intraarticular injections of LVSnail and LVSin (1 ⫻ 106 relative infection units) into the right and left ankle joints, respectively, on day ⫺3 were immunized with collagen on days 0 and 7. Alternatively, rats immunized with collagen on days 0 and 7 received intraarticular injections of LVshSnail (TRCN0000218784) and LVshLuc (control) (1 ⫻ 106 relative infection units) into the right and left ankle joints, respectively, on day 10. Clinical and histopathologic examinations. Arthritis was graded clinically with an articular index. When the rats were killed (days 21 and 23), hematoxylin and eosin–stained paraffin-embedded joint sections were prepared for the overexpression and silencing experiments and evaluated, and histologic scores (0–2 scale) were assigned, as previously described (22,23). Immunohistochemical and immunofluorescence assessments. Synovial tissue samples from the patients or rats were snap-frozen and embedded in paraffin. The sections were deparaffinized in xylene, dehydrated in alcohol, treated with proteinase K, washed with H2O2 in PBS, and stained with antibodies against Snail (Cell Signaling Technology), Cad-11 (Cell Signaling Technology), ␣-SMA (Abcam), IL-6 (Santa

42

CHEN ET AL

Figure 1. Expression of Snail and cadherin 11 (Cad-11) in synovial tissue and synovial fibroblasts from patients with rheumatoid arthritis (RASFs) and patients with osteoarthritis (OASFs). A, Immunohistochemical staining of Snail and Cad-11 in synovial tissue from patients with RA and patients with OA. Bars shown on photomicrographs at ⫻200 and ⫻400 magnification correspond to 100 ␮m and 50 ␮m, respectively. Images shown at ⫻400 magnification correspond to the boxed areas in the images shown at ⫻200 magnification. B, Expression of Snail and Cad-11 in synovial tissue from patients with RA and patients with OA, as determined by immunoblotting. C, Expression of Snail and Cad-11 in synovial fibroblasts from representative patients with RA or OA, as determined by immunoblotting. D, Immunofluorescence staining of Snail (Alexa Fluor 488 stained [green]), Cad-11 (Alexa Fluor 594 stained [red]), and nucleus (DAPI stained [blue]) in RASFs. E, Immunofluorescence staining of Snail (fluorescein isothiocyanate stained [green]), Cad-11 (Texas red stained [red]), and nucleus (DAPI stained [blue]) in synovial tissue from patients with RA and patients with OA. Boxed areas are shown at higher magnification in the panels beneath them. Arrows indicate positivity for both Snail and Cad-11. Results are representative of at least 2 independent experiments.

Cruz Biotechnology), or isotype control IgG (Santa Cruz Biotechnology) in combination with the chromogen 3-amino9-ethylcarbazole (Zymed), as previously described (25). In addition, paraffin-embedded synovial tissue sections obtained from the rats were stained with sirius red (Sigma-Aldrich) for the assessment of ECM deposition, and the signal intensity was further quantitated using Image J version 1.42q (National Institutes of Health). For immunofluorescence staining, paraffin-embedded sections receiving similar pretreatment were incubated with antibodies against Snail and Cad-11, followed by fluorescein isothiocyanate– and Texas red–conjugated secondary antibodies (BD Biosciences), respectively, and observed under a fluorescence microscope. Furthermore, synovial fibroblasts purified from patients with arthritis were subjected to immunofluorescence staining with antibodies against Snail and Cad-11, followed by Alexa Fluor 488– and Alexa Fluor 594– conjugated secondary antibodies (Life Technologies), respectively, and fluorescence was detected by confocal microscopy with a FluoView FV-1000 MPE multiphoton microscope (Olympus).

Immunoblot analysis. Cell lysates of synovial fibroblasts and stable transfectants and synovium homogenates from patients or rats were subjected to immunoblot analyses with antibodies against Snail (Cell Signaling Technology), Cad-11 (Cell Signaling Technology), fibronectin (Santa Cruz Biotechnology), ␣-SMA (Abcam), IL-6 (Santa Cruz Biotechnology), phosphorylated glycogen synthase kinase 3␤ (GSK-3␤) (Cell Signaling Technology), GSK-3␤ (Cell Signaling Technology), phosphorylated Akt (Ser473; Cell Signaling Technology), or Akt (Cell Signaling Technology) in combination with a horseradish peroxidase–conjugated secondary antibody (Jackson ImmunoResearch) and quantitative control anti–␤-actin antibodies (Sigma-Aldrich), as previously described (25). Protein–protein complexes were visualized with an ECL Plus System (Amersham) and analyzed with a BioSpectrum Imaging System, UVP, for chemiluminescence detection. The signal intensity was further quantitated by densitometry. Statistical analysis. Data are expressed as the mean ⫾ SEM. Differences in Snail and Cad-11 expression levels between patients with RA and patients with OA were compared by Mann-Whitney U test. The significance of correlation was

SNAIL REGULATES TNF␣-MEDIATED SYNOVIAL FIBROBLAST ACTIVATION

43

Figure 2. Expression of Snail and Cad-11 in synovial tissue and synovial fibroblasts from normal rats and rats with collagen-induced arthritis (CIA). A, Immunohistochemical staining of Snail and Cad-11 in synovial tissue from normal rats and rats with CIA. Bars shown at ⫻40 and ⫻400 magnification correspond to 500 ␮m and 50 ␮m, respectively. B, Immunofluorescence staining of Snail (fluorescein isothiocyanate stained [green]) and Cad-11 (Texas red stained [red]) in synovial tissue from normal rats and rats with CIA (day 11). In A and B, images shown at ⫻400 magnification correspond to the boxed areas in the images shown at ⫻40 magnification. Higher-magnification views of the merged images are shown on the far right. Arrows indicate positivity for both Snail and Cad-11. C, Expression of Snail and Cad-11 in synovial tissue from normal rats and rats with CIA, as determined by reverse transcription–polymerase chain reaction. Each lane represents pooled samples (n ⫽ 3). D, Expression of Snail, Cad-11, fibronectin, ␣-smooth muscle actin (␣-SMA), and ␤-actin in synovial fibroblasts from normal rats and rats with CIA, as determined by immunoblotting. Results are representative of at least 2 independent experiments. See Figure 1 for other definitions.

calculated using Pearson’s correlation coefficient. Statistical significance between different groups was assessed with Student’s t test. Differences in articular indexes were compared by repeated-measures analysis of variance. P values less than 0.05 were considered significant.

RESULTS Increased expression of Snail and Cad-11 in synovial tissue from patients with RA and rats with CIA. The expression of Snail and Cad-11 was first examined in the synovium of patients with arthritis. Immunohistochemical staining revealed higher expression of Snail and Cad-11 in synovial tissue from patients with RA compared with that in synovial tissue from patients with OA (Figure 1A). Further quantitative analysis by immunoblotting showed that the expression of Snail and

Cad-11 was concomitantly up-regulated in RA synovium (Figure 1B), and a positive correlation between the expression levels in RA versus OA was observed (for the Snail:␤-actin ratio, P ⫽ 0.0437; for the Cad-11:␤actin ratio, P ⫽ 0.0164 [r ⫽ 0.8478, P ⫽ 0.0039]). The expression levels of Snail and Cad-11 in synovial fibroblasts were similar to those in synovial tissue (Figure 1C). We examined the expression pattern of Snail and Cad-11 in synovial fibroblasts, using confocal microscopy for immunofluorescence detection. Figure 1D shows the expression of cell membrane Cad-11 and intranuclear Snail in synovial fibroblasts from a representative patient with RA. Immunofluorescence staining of synovium demonstrated costaining of Snail and Cad-11 in fibroblastlike synoviocytes (FLS) at the lining layer (Figure 1E).

44

We further examined the expression of Snail and Cad-11 in the synovium of rats with CIA. Immunohistochemical analysis revealed increased expression of Snail and Cad-11 in synovial tissue during the progression of arthritis (Figure 2A). Immunofluorescence analysis showed costaining of Snail and Cad-11 in FLS at the synovial lining layer (Figure 2B). RT-PCR analysis showed that Snail and Cad-11 were simultaneously detected from day 11 onward, with increased expression levels during the progression of arthritis (Figure 2C). Synovial fibroblasts from rats with CIA also expressed higher levels of Snail and Cad-11 compared with normal synovial fibroblasts, as shown by immunoblotting (Figure 2D). Notably, the expression of fibronectin and ␣-SMA was up-regulated in synovial fibroblasts from rats with CIA, indicating conversion to a myofibroblastlike phenotype. Regulation of TNF␣-mediated synovial fibroblast activation by Snail. After TNF␣ (10 ng/ml) was added to the culture of normal synovial fibroblasts (105 cells/well in 6-well plates) for 24 hours, the expression of

CHEN ET AL

Snail, Cad-11, and ␣-SMA was up-regulated, as demonstrated by immunoblotting (Figure 3A). In particular, phosphorylated GSK-3␤ levels were enhanced by TNF␣ stimulation for 30 minutes, suggesting that TNF␣ may act via the Wnt signaling pathway to activate synovial fibroblasts (Figure 3A). Next, we used a selective GSK-3␤ inhibitor to induce the Wnt signaling pathway. In normal synovial fibroblasts treated with 10 ␮M SB216763 in culture for 24 hours, the expression of Snail, Cad-11, and ␣-SMA was increased (Figure 3B). Furthermore, we performed in vivo experiments to examine the effect of TNF␣ silencing on activation of synovial fibroblasts in the joints of rats with CIA. Notably, intraarticular injections of etanercept resulted in a dose-dependent reduction in articular indexes and down-regulated synovial expression of Snail, Cad-11, and ␣-SMA (Figures 3C and D). Normal synovial fibroblasts were transduced with lentiviral vectors expressing Snail, in order to detect alterations in the phenotype and function, including the expression of Cad-11 and myofibroblast markers

Figure 3. Regulation of tumor necrosis factor ␣ (TNF␣)–mediated synovial fibroblast activation by Snail. A, Left, Expression of Snail, Cad-11, ␣-smooth muscle actin (␣-SMA), and ␤-actin in normal synovial fibroblasts treated with TNF␣ for 24 hours, as determined by immunoblotting. Right, Expression of phospho–glycogen synthase kinase (phospho-GSK), GSK, and ␤-actin in normal synovial fibroblasts treated with TNF␣ for 30 minutes, as determined by immunoblotting. B, Expression of Snail, Cad-11, ␣-SMA, and ␤-actin in normal synovial fibroblasts treated with SB216763 for 24 hours, as determined by immunoblotting. C, Articular index in rats with collagen-induced arthritis (CIA) treated with intraarticular injections of etanercept (25 ␮g or 100 ␮g) or phosphate buffered saline (PBS). Values are the mean ⫾ SEM (n ⫽ 6). Arrow indicates the time at which etanercept or PBS was injected. D, Left, Expression of Snail, Cad-11, ␣-SMA, and ␤-actin in synovium extracts from the joints of rats with CIA treated with etanercept or PBS, as determined by immunoblotting. Right, Quantification of the intensity of the bands corresponding to Snail, Cad-11, and ␣-SMA compared with ␤-actin in synovium extracts. Values are the mean ⫾ SEM. Results are representative of at least 2 independent experiments. See Figure 1 for other definitions.

SNAIL REGULATES TNF␣-MEDIATED SYNOVIAL FIBROBLAST ACTIVATION

(␣-SMA and fibronectin), IL-6 production, and the invasive ability of the cells, as determined by a Matrigel invasion assay. Normal synovial fibroblast transfectants in which Snail was overexpressed displayed increased expression of Cad-11, ␣-SMA, and fibronectin in comparison with LVSin-transduced control transfectants (Figure 4A). These transfectants produced higher levels of IL-6 and exhibited greater invasive capacity in response to TNF␣ stimulation, as compared with 2 control cell lines (Figures 4B and C). In particular, the levels of IL-6 in LVSnail-transduced synovial fibroblast transfectants were higher than those in LVSin-transduced controls without the addition of TNF␣. Snail-silenced synovial fibroblast transfectants from rats with CIA demonstrated decreased expression of Cad-11, ␣-SMA, and fibronectin in comparison with LVshLuc-

45

transduced controls or parental cells; a concomitant reduction in phosphorylated Akt levels in the Snailsilenced synovial fibroblast transfectants was also observed (Figure 4D). The levels of IL-6 and the invasive ability of cells were also abrogated in Snail-silenced transfectants compared with controls (Figures 4E and F). Collectively, the results of the experiments (in vitro treatment of synovial fibroblasts with TNF␣ or a Wnt signaling inducer, modulation of Snail expression in synovial fibroblasts by generating transfectants in which Snail was overexpressed or silenced, and in vivo antiTNF␣ administration into the joints of rats with CIA) strongly suggest that Snail plays an important role in the regulation of TNF␣-mediated synovial fibroblast activation.

Figure 4. Effect of modulating Snail on Cad-11 and ␣-smooth muscle actin (␣-SMA) expression and tumor necrosis factor ␣ (TNF␣)–mediated synovial fibroblast activation. A–C, Stable transfectants of normal synovial fibroblasts transduced with lentiviral vectors expressing Snail complementary DNA. D–F, Synovial fibroblasts from rats with collagen-induced arthritis (CIA) transduced with Snail-specific short hairpin RNA (shRNA). A, Expression of Snail, Cad-11, ␣-SMA, fibronectin, and ␤-actin in LVSnail- and LVSin-transduced normal synovial fibroblasts, as determined by immunoblotting. B, Levels of interleukin-6 (IL-6) in parental cells and LVSin-transduced and LVSnail-transduced normal synovial fibroblast transfectants. C, Invasive ability of LVSin- and LVSnail-transduced normal synovial fibroblast transfectants and parental cells, as determined by Matrigel invasion assay. D, Expression of Snail, Cad-11, fibronectin, ␣-SMA, phospho-Akt (Ser473), Akt, and ␤-actin in LVShLucand LVShSnail-transduced synovial fibroblasts from rats with CIA. E, Levels of IL-6 in unstimulated cells and cells stimulated with TNF␣. F, Invasive ability of parental cells, ShLuc-transduced cells, ShSnail (218784)–transduced cells, and ShSnail (234035)–transduced cells, as determined by Matrigel invasion assay. Values in B, C, E, and F are the mean ⫾ SEM (n ⫽ 3 in B, C, and E; n ⫽ 4 in F). Results are representative of at least 2 independent experiments. See Figure 1 for other definitions.

46

Synovial hyperplasia with increased expression of Cad-11 and ␣-SMA in normal joints of rats in which Snail was overexpressed. In addition to in vitro experiments in which Snail was overexpressed in normal synovial fibroblasts, we performed in vivo experiments in which LVSnail was injected into the normal joints of rats, followed by examination of the synovium 7 days later, when the rats were killed. The joints were grossly normal, without the appearance of arthritis; however,

CHEN ET AL

histopathologic examination revealed hyperplastic synovium, and immunohistochemical and immunoblot analyses demonstrated increased expression of Cad-11, ␣-SMA, and IL-6 (Figures 5A and B). Exacerbation of arthritis by Snail overexpression and amelioration of arthritis by Snail silencing in the joints of rats with CIA. We further induced Snail overexpression in the joints of rats with CIA by administering intraarticular injections of LVSnail and ob-

Figure 5. Synovial hyperplasia in normal rats treated with LVSnail. Normal rats (n ⫽ 3) received intraarticular injections of LVSnail and LVSin into the right and left ankle joints, respectively. Expression of Snail, cadherin 11 (Cad-11), ␣-smooth muscle actin (␣-SMA), interleukin-6 (IL-6), and ␤-actin in synovial tissue was determined by immunohistochemical analysis (A) and immunoblotting (B). In A, bars shown on photomicrographs at ⫻40 and ⫻200 magnification correspond to 500 ␮m and 100 ␮m, respectively. Images shown at ⫻200 magnification correspond to the boxed areas in the images shown at ⫻40 magnification. Results are representative of 2 independent experiments. H&E ⫽ hematoxylin and eosin.

SNAIL REGULATES TNF␣-MEDIATED SYNOVIAL FIBROBLAST ACTIVATION

served higher articular indexes than those in LVsintreated control rats (Figure 6A). Histopathologic analysis of Snail-transduced joints revealed greater synovial hyperplasia, more cartilage and bone erosion, and higher histologic scores compared with control joints (Figures 6B and C). To assess whether exacerbation of arthritis in the joints of rats with CIA in which Snail was overexpressed was associated with the activation of synovial fibroblasts, ECM deposition was assessed by sirius red staining, and Cad-11 expression levels were

47

analyzed by immunohistochemistry and immunoblotting. Increased ECM accumulation and Cad-11 expression were observed in the LVSnail-injected joints of rats with CIA (Figures 6C–E), suggesting that overexpression of Snail exacerbates arthritis through enhanced synovial fibroblast activation in the joints of rats with CIA. Silencing of Snail expression in the joints of rats with CIA was performed by administering intraarticular injections of LVshSnail, and the articular indexes in these

Figure 6. Effect of Snail overexpression and Snail silencing in the joints of rats with collagen-induced arthritis (CIA). A and B, Articular indexes and histologic scores in rats treated with LVSnail or LVsin. F and G, Articular indexes and histologic scores in rats treated with short hairpin RNA (shRNA) targeting Snail (ShSnail) or ShLuc. Arrow in F indicates the time at which the lentiviral vectors were injected. C and H, Cadherin 11 (Cad-11) expression in synovial tissue, as assessed by hematoxylin and eosin (H&E), sirius red, and immunohistochemical staining on day 21 (C) and day 23 (H). Bars shown at ⫻40 and ⫻400 magnification correspond to 500 ␮m and 50 ␮m, respectively. Images shown at ⫻400 magnification correspond to the boxed areas in the images shown at ⫻40 magnification. D and I, Signal intensity of sirius red in the joints of rats with CIA treated with LVSnail (D) or LVshSnail (I). E and J, Expression of Snail and Cad-11 in the joints of rats with CIA treated with LVSnail (E) or LVshSnail (J). Values in A, B, D, F, and G are the mean ⫾ SEM (n ⫽ 6 in A, B, F, and G; n ⫽ 3 in D). Results are representative of 2 independent experiments.

48

CHEN ET AL

rats were lower than those in rats receiving LVshLuc (Figure 6F). The LVshSnail-treated joints of rats with CIA showed milder synovial hyperplasia, less cartilage and bone erosion, lower histologic scores, and decreased levels of ECM deposition and Cad-11 expression, indicating that silencing Snail expression ameliorates arthritis by reducing synovial fibroblast activation in the joints of rats with CIA (Figures 6G–J). DISCUSSION Activated tissue-resident fibroblasts can convert to ␣-SMA–positive myofibroblasts, a population of contractile cells secreting an abundant amount of ECM during the wound-healing process. Active fibrosis and the activation, proliferation, and survival of these cells are mediated by a variety of soluble factors such as TGF␤ and TNF␣. In the fibrotic condition, myofibroblasts persistently proliferate and are resistant to apoptosis, leading to assembly of ECM with subsequent dysfunction of the target organ; novel therapies targeting these cells might demonstrate an antifibrotic effect (26,27). Although increased expression of TGF␤-inducible ␣-SMA has been identified in RASFs from highly inflamed synovium, the role of TNF␣ in the conversion of synovial fibroblasts to a myofibroblast-like phenotype remains to be explored in the rheumatoid joint (28). Interestingly, TNF␣ neutralization ameliorates renal fibrosis in a rat model, with reduced expression of ␣-SMA; reduced hepatic fibrosis, with lower numbers of ␣-SMA–positive myofibroblasts, has been detected in TNF␣-knockout mice (29,30). Because experimental results increasingly support TNF␣ as a key driver in the fibrosis process, clinical trials have been initiated to examine the therapeutic effect of TNF␣ blockers on fibrotic disorders (26). Decreased progression of fibrosis has been observed in patients with idiopathic pulmonary fibrosis receiving TNF␣ antagonist injections, and antibodies against TNF␣ have been shown to reduce ␣-SMA expression and contractile activity in myofibroblasts from patients with Dupuytren’s fibroproliferative disease (19,31). Consistent with these findings, in this study, the expression of ␣-SMA in TNF␣-treated synovial fibroblasts was up-regulated, and intraarticular administration of a TNF␣ antagonist resulted in a dosedependent reduction in the expression of ␣-SMA in the joints of rats with CIA. Furthermore, the silencing of Snail expression ameliorated arthritis, with reduced deposition of ECM in the joints of rats with CIA, whereas overexpression of Snail exacerbated arthritis, with increased deposition. Taken together, these data suggest

that a link exists between a TNF␣-rich environment and enhanced conversion of synovial fibroblasts to a myofibroblast-like phenotype in the rheumatoid joint. In the current study, a significant correlation between the expression of Snail and Cad-11 in synovium was observed in patients with RA and rats with CIA. The relevant role of Snail in the regulation of synovial Cad-11 expression was demonstrated in vitro and in vivo by overexpressing Snail in normal or arthritic joints and normal synovial fibroblasts and by using a knockdown approach in joints and synovial fibroblasts from rats with CIA. In a previous study by our group, the expression of phosphorylated Akt was up-regulated in RASFs treated with TNF␣, and overexpression of PTEN in RASFs or the joints of rats with CIA reduced Akt activity, followed by enhancement of apoptosis (25). Notably, Snail has been demonstrated to be associated with and to repress the PTEN promoter, and overexpression of Snail results in decreased PTEN expression and up-regulated Akt activity (32). In tumor cells, there is a constitutive hyperactivation of the NF-␬B/Snail/PTEN/Akt pathway, which has functions such as apoptosis inhibition (33). Interestingly, in vitro stimulation with TNF␣ can increase the expression of Cad-11 in synovial fibroblasts from patients with RA and normal rats, and the expression of Cad-11 is down-regulated by treatment with a phosphatidylinositol 3-kinase (PI3K) inhibitor, implicating involvement of the PI3K/Akt pathway in the regulation of this process (34,35). Indeed, in this study, we also demonstrated that in vitro treatment of normal synovial fibroblasts with TNF␣ could enhance the expression of Cad-11, and in vivo injection of a TNF␣ antagonist into the joints of rats with CIA reduced synovial expression of Cad-11. Furthermore, the expression of phosphorylated Akt and Cad-11 in Snail-silenced synovial fibroblast transfectants from rats with CIA was concomitantly down-regulated. These observations indicate that Snail may regulate the expression of Cad-11 by suppressing PTEN-mediated repression of Akt activation. In addition, 3 zinc-finger binding motifs have been identified in the promoter region of Cad-11, raising the possibility that Snail enhances the transcriptional activation of Cad-11 expression (36). Recently, the important role of Wnt signaling in RA-related cartilage and bone erosion and in the induction of activation and proliferation of myofibroblasts was recognized (37,38). Notably, TNF␣ has been reported to activate NF-␬B, resulting in its binding to the Snail promoter, with increased transcription activity (39). This cytokine enhances activation of the Wnt pathway by inhibiting GSK-3␤ activation, and Wnt signaling can further inactivate GSK-3␤–mediated phos-

SNAIL REGULATES TNF␣-MEDIATED SYNOVIAL FIBROBLAST ACTIVATION

phorylation of Snail, with enhancement of its stabilization (19,40). In this study, quantitative RT-PCR analysis revealed that the addition of TNF␣ increased the expression of Snail messenger RNA in synovial fibroblast cultures (Chen S.-Y. et al: unpublished observations), and immunoblot analysis demonstrated that TNF␣ stimulation or Wnt activation could up-regulate the expression of Snail in synovial fibroblasts. These observations suggest that TNF␣ is a critical signal for inducing the expression of Snail through enhanced transcription and may act via the Wnt signaling pathway to increase the stabilization of Snail. In Cad-11–deficient mice, the synovial lining and pannus tissue appear architecturally aberrant and chaotic, and cartilage erosion is markedly attenuated upon induction of arthritis (9). The expression of Cad-11 and ␣-SMA is increased during differentiation of dermal fibroblasts into myofibroblasts, and targeting Cad-11 can reduce the efficiency of myofibroblast contraction during the wound-healing process (41). Podoplanin, a transmembrane glycoprotein expressed in myofibroblasts, is highly expressed in Cad-11–positive cells throughout the synovial lining layer in RA, and its expression coincides with up-regulation of ␣-SMA (42). Furthermore, in a recent study, Cad-11 was expressed in ␣-SMA–positive myofibroblasts in the fibrotic skin lesions of patients with systemic sclerosis and mice with bleomycin-induced dermal fibrosis, and attenuated fibrosis with decreased accumulation of myofibroblasts was observed in bleomycin-injected Cad-11–knockout mice (43). These observations indicate that modulation of Cad-11 expression may alter the conversion process of fibroblasts to an ␣-SMA–positive myofibroblast-like phenotype in disease states such as systemic sclerosis and RA. Our previous study identified the p53 mutation as competent for the inactivation of wild-type p53 in synovial fibroblasts from patients with RA and rats with CIA and demonstrated the effect of an E1B-55 kd– deleted adenovirus selectively targeting synovial fibroblasts on the amelioration of arthritis (22). Notably, wild-type p53 can inhibit tumor cell invasion by inducing Snail degradation via Mdm2-mediated ubiquitination, whereas cancer cells transfected with mutant p53 demonstrate greater invasiveness (44). Furthermore, overexpression of Snail has been identified in cancer patients with p53 mutations, and transfection of tumor cells harboring wild-type p53 with mutant p53 can upregulate the expression of Snail (44,45). Such observations are consistent with our findings of higher expression of Snail in synovial fibroblasts from patients with RA and rats with CIA possessing mutated p53. Interest-

49

ingly, suppression of p53 by direct binding with oncogenic K-Ras–induced Snail has been demonstrated, using small-molecule chemicals to reactivate wild-type p53 by disrupting the binding between Snail and p53 in K-Ras–mutated cancer cells (46). Taken together, these findings allow us to anticipate enhanced RASF invasiveness due to p53 mutation, with up-regulated expression of Snail. Further efforts to elucidate the critical mechanisms that regulate synovial expression of Snail by mutated p53 are under way. In conclusion, by using synovium specimens from patients with RA and an experimental model of arthritis, we demonstrated that Snail regulates TNF␣-mediated synovial fibroblast activation in the rheumatoid joint. These findings may contribute to the development of pharmacologic therapies targeting synovial fibroblasts in patients with RA. ACKNOWLEDGMENT We thank Dr. D. Trono (Ecole Polytechnique Federale de Lausanne, Lausanne, Switzerland) for providing the psPAX2 and pMD2G plasmids for the production of lentiviral vectors. 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. Drs. Wu and Wang had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study conception and design. Chen, Shiau, Wu, Wang. Acquisition of data. Chen, Li, Lin, Jou, Liu. Analysis and interpretation of data. Chen, Shiau, Wu, Wang.

REFERENCES 1. Bottini N, Firestein GS. Duality of fibroblast-like synoviocytes in RA: passive responders and imprinted aggressors. Nat Rev Rheumatol 2013;9:24–33. 2. Lafyatis RS, Remmers EF, Roberts AB, Yocum DE, Sporn MB, Wilder RL. Anchorage-independent growth of synoviocytes from arthritic and normal joints: stimulation by exogenous plateletderived growth factor and inhibition by transforming growth factor-␤ and retinoids. J Clin Invest 1989;83:1267–76. 3. Muller-Ladner U, Kriegsmann J, Franklin BN, Matsumoto S, Geiler T, Gay RE, et al. Synovial fibroblasts of patients with rheumatoid arthritis attach to and invade normal human cartilage when engrafted into SCID mice. Am J Pathol 1996;149:1607–15. 4. Woodrick RS, Ruderman EM. Safety of biologic therapy in rheumatoid arthritis. Nat Rev Rheumatol 2011;7:639–52. 5. Choy EH, Kavanaugh AF, Jones SA. The problem of choice: current biologic agents and future prospects in RA. Nat Rev Rheumatol 2013;9:154–63. 6. Chang SK, Gu Z, Brenner MB. Fibroblast-like synoviocytes in inflammatory arthritis pathology: the emerging role of cadherin11. Immunol Rev 2010;233:256–66. 7. Kiener HP, Niederreiter B, Lee DM, Jimenez-Boj E, Smolen JS, Brenner MB. Cadherin 11 promotes invasive behavior of fibroblastlike synoviocytes. Arthritis Rheum 2009;60:1305–10. 8. Chang SK, Noss EH, Chen M, Gu Z, Townsend K, Grenha R, et al. Cadherin-11 regulates fibroblast inflammation. Proc Natl Acad Sci U S A 2011;108:8402–7.

50

9. Lee DM, Kiener HP, Agarwal SK, Noss EH, Watts GF, Chisaka O, et al. Cadherin-11 in synovial lining formation and pathology in arthritis. Science 2007;315:1006–10. 10. Nieto MA. The ins and outs of the epithelial to mesenchymal transition in health and disease. Annu Rev Cell Dev Biol 2011;27: 347–76. 11. Chiang C, Ayyanathan K. Snail/Gfi-1 (SNAG) family zinc finger proteins in transcription regulation, chromatin dynamics, cell signaling, development, and disease. Cytokine Growth Factor Rev 2013;24:123–31. 12. Franci C, Takkunen M, Dave N, Alameda F, Gomez S, Rodriguez R, et al. Expression of Snail protein in tumor-stroma interface. Oncogene 2006;25:5134–44. 13. Rowe RG, Li XY, Hu Y, Saunders TL, Virtanen I, Garcia de Herreros A, et al. Mesenchymal cells reactivate Snail1 expression to drive three-dimensional invasion programs. J Cell Biol 2009; 184:399–408. 14. Scarpa M, Grillo AR, Brun P, Macchi V, Stefani A, Signori S, et al. Snail1 transcription factor is a critical mediator of hepatic stellate cell activation following hepatic injury. Am J Physiol Gastrointest Liver Physiol 2011;300:G316–26. 15. Bourlier V, Sengenes C, Zakaroff-Girard A, Decaunes P, Wdziekonski B, Galitzky J, et al. TGF␤ family members are key mediators in the induction of myofibroblast phenotype of human adipose tissue progenitor cells by macrophages. PLoS One 2012; 7:e31274. 16. Mattey DL, Dawes PT, Nixon NB, Slater H. Transforming growth factor ␤1 and interleukin 4 induced ␣ smooth muscle actin expression and myofibroblast-like differentiation in human synovial fibroblasts in vitro: modulation by basic fibroblast growth factor. Ann Rheum Dis 1997;56:426–31. 17. Watson RS, Gouze E, Levings PP, Bush ML, Kay JD, Jorgenson MS, et al. Gene delivery of TGF-␤1 induces arthrofibrosis and chondrometaplasia of synovium in vivo. Lab Invest 2010;90: 1615–27. 18. Cha HS, Bae EK, Ahn JK, Lee J, Ahn KS, Koh EM. Slug suppression induces apoptosis via Puma transactivation in rheumatoid arthritis fibroblast-like synoviocytes treated with hydrogen peroxide. Exp Mol Med 2010;42:428–36. 19. Verjee LS, Verhoekx JS, Chan JK, Krausgruber T, Nicolaidou V, Izadi D, et al. Unraveling the signaling pathways promoting fibrosis in Dupuytren’s disease reveals TNF as a therapeutic target. Proc Natl Acad Sci U S A 2013;110:E928–37. 20. Feldmann M. Development of anti-TNF therapy for rheumatoid arthritis. Nat Rev Immunol 2002;2:364–71. 21. McInnes IB, Schett G. The pathogenesis of rheumatoid arthritis. N Engl J Med 2011;365:2205–19. 22. Chen SY, Shiau AL, Shieh GS, Su CH, Lee CH, Lee HL, et al. Amelioration of experimental arthritis by a telomerase-dependent conditionally replicating adenovirus that targets synovial fibroblasts. Arthritis Rheum 2009;60:3290–302. 23. Wang CR, Chen SY, Wu CL, Liu MF, Jin YT, Chao L, et al. Prophylactic adenovirus-mediated human kallistatin gene therapy suppresses rat arthritis by inhibiting angiogenesis and inflammation. Arthritis Rheum 2005;52:1319–24. 24. Li YT, Chen SY, Wang CR, Liu MF, Lin CC, Jou IM, et al. Amelioration of collagen-induced arthritis in mice by lentivirusmediated silencing of microRNA-223. Arthritis Rheum 2012;64: 3240–5. 25. Wang CR, Shiau AL, Chen SY, Lin LL, Tai MH, Shieh GS, et al. Amelioration of collagen-induced arthritis in rats by adenovirusmediated PTEN gene transfer. Arthritis Rheum 2008;58:1650–6. 26. Wynn TA, Ramalingam TR. Mechanisms of fibrosis: therapeutic translation for fibrotic disease. Nat Med 2012;18:1028–40. 27. LeBleu VS, Taduri G, O’Connell J, Teng Y, Cooke VG, Woda C, et al. Origin and function of myofibroblasts in kidney fibrosis. Nat Med 2013;19:1047–53. 28. Kasperkovitz PV, Timmer TC, Smeets TJ, Verbeet NL, Tak PP, van Baarsen LG, et al. Fibroblast-like synoviocytes derived from

CHEN ET AL

29.

30.

31.

32.

33.

34.

35.

36.

37. 38. 39.

40.

41. 42.

43.

44. 45. 46.

patients with rheumatoid arthritis show the imprint of synovial tissue heterogeneity: evidence of a link between an increased myofibroblast-like phenotype and high-inflammation synovitis. Arthritis Rheum 2005;52:430–41. Meldrum KK, Misseri R, Metcalfe P, Dinarello CA, Hile KL, Meldrum DR. TNF-␣ neutralization ameliorates obstructioninduced renal fibrosis and dysfunction. Am J Physiol Regul Integr Comp Physiol 2007;292:R1456–64. Gabele E, Froh M, Arteel GE, Uesugi T, Hellerbrand C, Scholmerich J, et al. TNF-␣ is required for cholestasis-induced liver fibrosis in the mouse. Biochem Biophys Res Commun 2009;378:348–53. Raghu G, Brown KK, Costabel U, Cottin V, du Bois RM, Lasky JA, et al. Treatment of idiopathic pulmonary fibrosis with etanercept: an exploratory, placebo-controlled trial. Am J Respir Crit Care Med 2008;178:948–55. Escriva M, Peiro S, Herranz N, Villagrasa P, Dave N, MontserratSentis B, et al. Repression of PTEN phosphatase by Snail1 transcriptional factor during gamma radiation-induced apoptosis. Mol Cell Biol 2008;28:1528–40. Lin K, Baritaki S, Militello L, Malaponte G, Bevelacqua Y, Bonavida B, et al. The role of B-RAF mutations in melanoma and the induction of EMT via dysregulation of the NF-␬B/Snail/RKIP/ PTEN circuit. Genes Cancer 2010;1:409–20. Vandooren B, Cantaert T, ter Borg M, Noordenbos T, Kuhlman R, Gerlag D, et al. Tumor necrosis factor ␣ drives cadherin 11 expression in rheumatoid inflammation. Arthritis Rheum 2008;58: 3051–62. Wu M, Xu T, Zhou Y, Lu H, Gu Z. Pressure and inflammatory stimulation induced increase of cadherin-11 is mediated by PI3K/ Akt pathway in synovial fibroblasts from temporomandibular joint. Osteoarthritis Cartilage 2013;21:1605–12. Farre D, Roset R, Huerta M, Adsuara JE, Rosello L, Alba MM, et al. Identification of patterns in biological sequences at the ALGGEN server: PROMO and MALGEN. Nucleic Acids Res 2003;31:3651–3. Hu B, Phan SH. Myofibroblasts. Curr Opin Rheumatol 2013;25: 71–7. Lories RJ, Corr M, Lane NE. To Wnt or not to Wnt: the bone and joint health dilemma. Nat Rev Rheumatol 2013;9:328–39. Barbera MJ, Puig I, Dominguez D, Julien-Grille S, GuaitaEsteruelas S, Peiro S, et al. Regulation of Snail transcription during epithelial to mesenchymal transition of tumor cells. Oncogene 2004;23:7345–54. Zhou BP, Deng J, Xia W, Xu J, Li YM, Gunduz M, et al. Dual regulation of Snail by GSK-3␤-mediated phosphorylation in control of epithelial-mesenchymal transition. Nat Cell Biol 2004;6: 931–40. Hinz B, Pittet P, Smith-Clerc J, Chaponnier C, Meister JJ. Myofibroblast development is characterized by specific cell-cell adherens junctions. Mol Biol Cell 2004;15:4310–20. Ekwall AK, Eisler T, Anderberg C, Jin C, Karlsson N, Brisslert M, et al. The tumour-associated glycoprotein podoplanin is expressed in fibroblast-like synoviocytes of the hyperplastic synovial lining layer in rheumatoid arthritis. Arthritis Res Ther 2011;13:R40. Wu M, Pedroza M, Lafyatis R, George AT, Mayes MD, Assassi S, et al. Identification of cadherin-11 as a mediator of dermal fibrosis and possible role in systemic sclerosis. Arthritis Rheumatol 2014; 66:1010–21. Lim SO, Kim H, Jung G. p53 inhibits tumor cell invasion via the degradation of Snail protein in hepatocellular carcinoma. FEBS Lett 2010;584:2231–6. Kim NH, Kim HS, Li XY, Lee I, Choi HS, Kang SE, et al. A p53/miRNA-34 axis regulates Snail1-dependent cancer cell epithelial-mesenchymal transition. J Cell Biol 2011;195:417–33. Lee SH, Shen GN, Jung YS, Lee SJ, Chung JY, Kim HS, et al. Antitumor effect of novel small chemical inhibitors of Snail-p53 binding in K-Ras-mutated cancer cells. Oncogene 2010;29: 4576–87.