Role of small ubiquitin-like modifier proteins-1 (SUMO-1) in regulating migration and invasion of fibroblast-like synoviocytes from patients with rheumatoid arthritis

Author’s Accepted Manuscript Role of Small Ubiquitin-like Modifier Proteins-1 (SUMO-1) in Regulating Migration and Invasion of Fibroblast-like Synoviocytes from Patients with Rheumatoid Arthritis Minxi Lao, Zhongping Zhan, Nan Li, Siqi Xu, Maohua Shi, Yaoyao Zou, Mingcheng Huang, Shan Zeng, Liuqin Liang, Hanshi Xu

PII: DOI: Reference:

www.elsevier.com/locate/yexcr

S0014-4827(18)31168-6 https://doi.org/10.1016/j.yexcr.2018.12.011 YEXCR11267

To appear in: Experimental Cell Research Received date: 31 October 2018 Revised date: 11 December 2018 Accepted date: 13 December 2018 Cite this article as: Minxi Lao, Zhongping Zhan, Nan Li, Siqi Xu, Maohua Shi, Yaoyao Zou, Mingcheng Huang, Shan Zeng, Liuqin Liang and Hanshi Xu, Role of Small Ubiquitin-like Modifier Proteins-1 (SUMO-1) in Regulating Migration and Invasion of Fibroblast-like Synoviocytes from Patients with Rheumatoid A r t h r i t i s , Experimental Cell Research, https://doi.org/10.1016/j.yexcr.2018.12.011 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Role of Small Ubiquitin-like Modifier Proteins-1 (SUMO-1) in Regulating Migration and Invasion of Fibroblast-like Synoviocytes from Patients with Rheumatoid Arthritis

Minxi Laoa1, Zhongping Zhana1, Nan Lib#, Siqi Xua, Maohua Shia,c, Yaoyao Zoud, Mingcheng Huange, Shan Zengf, Liuqin Lianga, Hanshi Xua*

a

Department of Rheumatology, The First Affiliated Hospital, Sun Yat-sen University, Guangzhou,

Guangdong, China b

c

School of Traditional Chinese Medicine, Jinan University, Guangzhou, Guangdong, China

Department of Rheumatology, The First People's Hospital of Foshan, Foshan, Guangdong, China

d

Department of Rheumatology, The Second Affiliated Hospital, Sun Yat-sen University, Guangzhou,

Guangdong, China e

Department of Hematology, The First Affiliated Hospital, Sun Yat-sen University, Guangzhou,

Guangdong, China f

Department of Rheumatology, The First Affiliated Hospital of Jinan University, Guangzhou,

Guangdong, China

*

Corresponding author at: Department of Rheumatology, The First Affiliated Hospital, Sun Yat-sen

University,

No.58

Zhongshan

Road

2,

Guangzhou,

[email protected]

1

M.L., Z.Z., and N.L. contributed equally to this work.

Guangdong

510080,

China.

Abstract Rheumatoid arthritis (RA) is featured by erosive cartilage and bone destruction. The enhancing aggressive property of fibroblast-like synoviocytes (FLSs) plays a critical role in this process. Small ubiquitin-like modifier (SUMO) proteins, including SUMO-1, SUMO-2, SUMO-3 and SUMO-4, participate in regulating many cellular events such as survival, migration and signal transduction in some cell lines. However, their roles in the pathogenesis of RA are not well established. Therefore, we evaluated the role of SUMO proteins in RA FLSs migration and invasion. We found that expression of both SUMO-1 and SUMO-2 was elevated in FLSs and synovial tissues (STs) from patients with RA. SUMO-1 suppression by small interference RNA (siRNA) reduced migration and invasion as well as MMP-1 and MMP-3 expression in RA FLSs. We also demonstrated that SUMO-1 regulated lamellipodium formation during cell migration. To explore further into molecular mechanisms, we evaluated the effect of SUMO-1 knockdown on the activation of Rac1/PAK1, a critical signaling pathway that controls cell motility. Our results indicated that SUMO-1-mediated SUMOylation controlled Rac1 activation and modulated downstream PAK1 activity. Inhibition of Rac1 or PAK1 also decreased migration and invasion of RA FLSs. Our findings suggest that SUMO-1 suppression could be protective against joint destruction in RA by inhibiting aggressive behavior of RA FLSs.

Keywords: rheumatoid arthritis; small ubiquitin-like modifier; inflammation; migration; fibroblast-like synoviocytes

1.

Introduction

Rheumatoid arthritis (RA) is a chronic inflammatory joint disease characterized by progressive destruction of cartilage and bone [1]. Fibroblast-like synoviocytes (FLSs) in the synovial intimal lining are the major components that contribute to joint destruction in RA [2]. Stable activated FLSs present some tumor-like phenotype especially the aggressive properties towards cartilage and bone [3,4]. Increasing evidence suggests that suppression of FLSs migration and invasion might be potential for ameliorating joint destruction in RA [5]. Small ubiquitin-like modifier (SUMO) proteins, consisting of four members, SUMO-1, SUMO-2, SUMO-3, and SUMO-4, contribute to regulating various cellular biological processes. By attaching to the substrates, SUMO modulates cell proliferation, apoptosis, and motility in some cell lines by protein SUMOylation [ 6]. For instance, SUMOylation of microphthalmia-associated transcription factor (MITF) mediated by SUMO-1 regulates its binding to HIF1A promoter, leading to migration and invasion in melanocytic and renal cell [7]. SUMOylation was involved in the pathogenesis of RA. It was reported that SUMO-2/3 suppressed invasion of FLSs by inhibiting matrix metalloproteinase (MMP) -3, 13 secretions [8]. FLSs with elevated expression of SUMO-1 are more resistant to FasL-mediated apoptosis [9]. Suppression of the SUMO-conjugating enzyme Ubc9 attenuated cell motility of RA FLS [10]. However, the role of SUMO-1 in regulating migration and invasion of RA FLS remains to be elucidated. In the present study, we demonstrated that SUMO-1 was involved in migration and invasion of RA FLS. SUMO-1 knockdown by small interfering RNA (siRNA) suppressed migration and invasion of FLSs as well as expression of MMP-1 and MMP-3. We also showed that SUMO-1 knockdown impaired lamellipodium formation in RA FLSs and reduced the activity of Rac-1, a critical protein of Rho family that control cell motility. Therefore, our study suggests that the increased expression of synovial SUMO-1 contributes to aggressive behavior of RA FLSs. 2.

Materials and methods

2.1 Reagents and Antibodies Recombinant human TNF-α, IL-1β and IL-17α were obtained from R&D Systems (Minneapolis, MN, USA). DMEM/F12, fetal bovine serum (FBS), antibiotics, Trypsin EDTA, PBS and other products for cell culture were obtained from Invitrogen (Carlsbad, CA, USA). Anti-SUMO-1, anti-Rac1 and anti-β-actin antibodies

were

purchased

from

Abcam (Cambridge,

UK).

Anti-PAK1

and

anti-phospho-PAK1 antibodies were purchased from Cell Signalling Technology (Beverly, MA, USA). Rac1 inhibitor NSC23766 was purchased from Selleck (Houston, USA). PAK1 inhibitor IPA-3 was purchased from Sigma (St Louis, MO, USA). 2.2 Preparation of human synovial tissues (STs) and FLSs The study was performed according to the recommendations of the Declaration of Helsinki and approved by the Medical Ethical Committee of the First Affiliated Hospital, Sun Yat-sen University, China. All patients were given informed consent to take part in the study. STs were obtained from RA (8 women and 3 men, aged 38-66 years) and OA patients (6 women and 5 men, aged 45-70 years) who were undergoing synovectomy or joint replacement. RA was diagnosed according to the 1987 revised criteria of the American College of Rheumatology [11]. OA was diagnosed according to according to the American College of Rheumatology (ACR) guidelines [12]. STs were cut into small pieces and digested with 1mg/ml collagenase for 3 h at 37°C to isolate FLSs. All cells were cultured in DMEM/F12 with 10% FBS at 37°C and 5% CO2. In our experiments, cells were used from passage 4 to 6, during which time they are a homogeneous population of cells (<1% CD11b positive, <1% phagocytic, and <1% FcgRII and FcgRIII receptor positive). 2.3 Transfection of FLSs with siRNA against SUMO-1or lentivirus overexpressing SUMO-1. For SUMO-1 knockdown, cells (5×106) were transfected with 50 nM control siRNA or SUMO-1 siRNA (The sequences of these siRNA oligonucleotides are listed as followed: SUMO-1-1: CUGGGAAUGGAGGAAGAAG; CAAGAAACUCAAAGAAUCA)

SUMO-1-2: using

CAAUGAAUUCACUCAGGUU;

Lipofectamine

reagent,

according

to

SUMO-1-3: the

protocol

recommended by the manufacturer (Invitrogen). After transfection, cells were incubated with normal culture medium for 24 h before experiments were performed. The effect of knockdown SUMO-1 protein was examined by Western blot analysis. For SUMO-1 overexpression, SUMO1 gene (NM_003352) or scramble sequence was subcloned into GV365 vector (Genechem, Shanghai, China). Lentiviruses were produced by cotransfection of HEK 293T cells with expression vectors and GFP-labeled helper plasmids using Lipofectamine 2000 (Invitrogen) according to the manufacturer’s protocol. Lentiviral particles were harvested from cell supernatants 48 and 72 h after transfection. For infection, FLSs were treated with the lentiviruses, polybrene (10mg/ml; Santa Cruz Biotechnology, Santa Cruz, CA) and RNAimax reagent (Invitrogen) for 12 h at 37˚C. Afterwards, the virus-containing medium was replaced with fresh medium. The effect

of SUMO-1 overexpression was examined using Western blot analysis 2.4 Quantitative real-time PCR (Q-RT-PCR) Total RNAs from RA FLSs were prepared by Takara PrimeScript® RT reagent kit according to the manufacturer’s protocol. Q-RT-PCR was performed using the Bio-Rad CFX96 system. The primer sequences were listed in Table 1. All experiments were carried out in triplicate. Table 1 Oligonucleotide sequences of primers used in Q-RT-PCR analysis. Gene

Forward Primer

Reverse Primer

5’-CTCTGGAGTAATGTCACACCTC

5’-TGTTGGTCCACCTTTCATCTTC-3

T-3’

’

5’-CTTTCCTGGCATCCCGAAGT-3’

5’-GGAACCGAGTCAGGTCTGTG-3’

MMP-1

MMP-3

5’5’SUMO-1

AATTCATTGGAACACCCTGTCTT-3 TGACCAGGAGGCAAAACCTTC-3’ ’

SUMO-2

5’- CTGCTTGTGTGCTCGTTTGG-3’

5’- TCCAACTGTGCAGGTGTGTC-3’

5’-GAGAGGCAGGGCTTGTCAAT-3 5’-AACCTTGCCCCCAATACCTG-3’

SUMO-3 ’ 5’-CCACGGGGATTGTCAGTGAAG-

5’-AGAATGTGGAAGGGGGAAACA

3’

A-3’

SUMO-4 5’-GCACCGTCAAGGCTGAGAAC-3 5’-TGGTGAAGACGCCAGTGGA-3’

GAPDH ’ 2.5 Western blot analysis

Protein concentrations were measured by the bicinchoninic acid (BCA) protein assay (Pierce, Rockford, IL, USA). Equal amounts of protein were solubilized in Laemmli buffer (62.5 mM TrisHCl pH 6.8, 10% glycerol, 2% SDS, 5% b-mercaptoethanol and 0.00625% bromophenol blue), boiled for 5 min and then separated by SDS-PAGE and transferred to nitrocellulose membranes. The membranes were probed with primary antibodies as indicated in Tris-buffered saline T (TBS-T) containing 5% non-fat milk at 4°C overnight. The membranes were incubated with the appropriate secondary antibodies for 1 h at room temperature. Immunoreactive bands were visualized by enhanced chemiluminescence (ECL; Amersham Pharmacia, Piscataway, NJ, USA). Each blot is a representative of at least three similar independent experiments.

2.6 Cell migration and invasion assay Chemotaxis assay of FLS was performed by the Boyden chamber method using a filter of 6.5 mm diameter and 8.0 μm pore size (Transwell; Corning Inc., Corning, NY, USA). Briefly, DMEM containing 10% FBS as a chemoattractant was placed in the lower wells, respectively. FLSs (at a final concentration of 6×104cells/ml) were suspended in serum-free DMEM in the upper wells. The chamber was incubated at 37°C under 5% CO2 for 8 h. After incubation the non-migrating cells were removed from the upper surface of the filter using a cotton swab. The filters were fixed in methanol for 15 min and stained with 0.1% crystal violet for 15 min. Chemotaxis was quantified by counting the stained cells that migrated to the lower side of the filter using an optical microscope (magnification×100). The stained cells were counted as the mean number of cells per 3 random fields for each assay. For the in vitro invasion assay, similar experiments were performed using inserts coated with a Matrigel basement membrane matrix (BD Biosciences, Oxford, UK). 2.7 Immunohistochemistry For immunohistochemistry, ST sections obtained from patients with RA (n=5) or OA (n=5) were deparaffinized, followed by incubation with 5% serum in PBS for 2 h to block nonspecific binding and incubation with 3% H2O2 for 10 min to block endogeneous peroxidase activity. The expression of SUMO-1 was determined by staining with polyclonal rabbit anti-human SUMO-1 antibody overnight at 4°C. Irrelevant isotype-matched antibodies were used as controls. Polyclonal goat anti-rabbit antibodies labeled with horseradish peroxidase (HRP) were used as secondary antibodies for 1 h at room temperature. Results were revealed using diaminobenzidine. 2.8 IF staining RA FLSs, growing on glass coverslips at 90% confluence, were wounded with micropipette tips and then treated with 10% FBS. After 3 h of incubation, cells were fixed with 4% paraformaldehyde for 15 min and then permeated with 0.1% Triton X-100 in PBS for 10 min at room temperature. The cells were incubated with anti- SUMO-1 antibody (diluted 1:200) for 1 h at room temperature and then incubated with FITC-conjugated secondary antibody (Santa Cruz Biotechnology, Santa Cruz, CA, USA). For detection of pseudopodia organization, cells were incubated with AlexaFluor-546 rhodamin-phalloidin (Molecular Probes; Invitrogen, Eugene, OR, USA) and nuclei were visualized using 0.25 mg/ml 4,-6-diamidono-2-phenylindole, dihydrochloride (DAPI). The coverslips were mounted on glass slides with antifade mounting media and examined using a fluorescence microscopy.

2.9 FLSs proliferation assays RA FLSs were cultured for 24 h at a density of 1×10 4/well in 96-well plates in serum-free medium. After starving, the cells were incubated with medium containing 10% FBS for 72 h and then incubated with 50 μM EdU for 6 h. EdU incorporation was assessed in triplicate, using a cell proliferation ELISA kit (Ribobio, Guangzhou,

China) according to the manufacturer’s instructions.

2.10 Flow cytometry FLS were fixed in 2% paraformaldehyde and were analyzed by flow cytometry using a FACSan with CellQuest software (BD Bioscience, Oxford, United Kingdom). FLS apoptosis was assessed by labeling with annexin V (APC) and propidium iodide (PI) according to the manufacturer’s instructions (BD Bioscience) 2.11 Measurement of Rac1 activity RA FLSs were cultured for 24 h at a density of 1×10 5/well in 35-mm culture dishes in serum-free medium. After starving, the cells were incubated with TNF-α (10 ng/ml) for 10 min and then harvested for Rac1 activity detection, using a G-LISA Rac1 activation assay kit (Cytoskeleton, Denver, USA) according to the manufacturer’s instructions. 2.12 Statistical analysis Data are expressed as mean ± SEM. Student’s t-test or ANOVA test was used to evaluate differences between experimental groups. A P-value < 0.05 was considered significant. 3.

Results

3.1 Increased expression of SUMO-1 in FLSs and the STs from patients with RA Firstly, we detected mRNA expression of SUMO proteins in RA FLSs and OA FLSs. As shown in Fig.1A, when compared to OA FLSs, the gene expression of both SUMO-1 and SUMO-2 was elevated in RA FLSs, whereas SUMO-1 expression was the most prominent. However, both SUMO-3 and SUMO-4 was barely detected in RA or OA FLSs. SUMO-1 protein expression was also confirmed in STs by immunohistochemistry. As shown in Fig.1B, SUMO-1 proteins expression was prominent in STs from RA patients and mostly localized in the synovial sublining cells, whereas the expression was much less prominent in STs from OA patients. We also observed subcellular distribution of SUMO-1 in FLSs by immunofluorescence and found that RA FLSs exhibited a markedly enhanced staining for SUMO-1, which localized mainly in the nuclei (Fig.1C). We also determined the expression of SUMO-1 protein by western blot analysis. SUMO-1 expression was increased in FLSs from RA

patients compared with osteoarthritis (OA) patients (Fig.1D). In order to explore the effect of inflammation on the expression of SUMO-1, we stimulated RA FLS with pro-inflammatory cytokines. We found that SUMO-1 protein expression was up-regulated in the RA FLSs treated with TNF-α, IL-1β and IL-17α (Fig.1E). Since TNF-α was the major pro-inflammatory cytokine in RA, it was used for further study. We demonstrated that the expression of SUMO-1 stimulated by TNF-α was time-dependent. The strongest expression of SUMO-1 was found in RA FLSs when stimulated by TNF-α for 24 h (Fig.1F). 3.2 SUMO-1 inhibits in vitro migration and invasion of RA FLSs To determine the role of SUMO-1 in regulating directed migration, the chemotaxis migration of FLSs was evaluated using a Transwell chamber assay. Transfection with siRNA was used to knockdown SUMO-1 expression in RA FLS. To rule out the nonspecific interference, we constructed three different sequences of siRNA oligonucleotides for SUMO-1. As shown in Fig.2A, transfection with all of three siRNA downregulated SUMO-1 protein expression, however, the inhibitory effect of the second one (SUMO-1 siRNA-2) was the most prominent. Accordingly, SUMO-1 siRNA-2 (SUMO-1 siRNA) was used for further experiments. To evaluate the effect of SUMO-1 in regulating migration from the opposite side, we overexpressed SUMO-1 in OA FLS by lentivirus infection. The effect of SUMO-1 overexpression was shown in Figure 1B. Both the expression of conjugated- and free-SUMO-1 was enhanced compared to scramble control. Migration of RA FLSs treated with SUMO-1 siRNA was decreased compared with scramble control (Fig.2C). Conversely, overexpression of SUMO-1 enhanced in vitro migration in OA FLSs (Fig.2D). The ability to invade cartilage is a critical pathogenic behavior of RA FLSs. In vitro invasion potential of RA FLSs is well related to the rate of joint destruction in RA patients [13]. Therefore, we evaluated the effect of SUMO-1 on regulating invasive behavior of RA FLSs through Matrigel-coated transwell membranes. As shown in Fig.2E, treatment with SUMO-1 siRNA decreased Matrigel invasion compared with treatment with control siRNA in RA FLSs. Increasing invasion was observed in OA FLSs overexpressing SUMO-1 (Fig.2F). Since previous report indicated SUMO-1 participates in regulating proliferation of tumor cells [14], to rule out whether the inhibitory effect of SUMO-1 knockdown on migration and invasion is associated with proliferation or apoptosis, we addressed the role of SUMO-1 in proliferation and

apoptosis of RA FLSs. Our results showed that SUMO-1 was not involved in proliferation or apoptosis of RA FLSs (Supplement 1A, B), suggesting that the effect of SUMO-1 knockdown on migration and invasion is not related to proliferation and apoptosis of RA FLS. 3.3 SUMO-1 knockdown reduces TNF-α-induced MMP-1, MMP-3 expression The aggressive phenotype of RA FLSs is featured by the increasing expression of MMPs. To determine the role of SUMO-1 in regulating MMPs expression, RA FLSs transfected with SUMO-1 siRNA or control siRNA were stimulated with TNF-α for 24 h. As shown in Fig.3A and B, treatment with SUMO-1 siRNA markedly suppressed TNF-α-induced mRNA expression of MMP-1 and MMP-3 compared with control siRNA treatment. 3.4 SUMO-1 knockdown impairs lamellipodia formation in RA FLSs Dynamic reorganization of the actin cytoskeleton is critical for cell migration. To evaluate the role of SUMO-1 proteins in regulating actin organization in RA FLSs, fluorescent phalloidin staining was used to visualize polymerized actin in migrating cells shortly after wound healing in treatment with SUMO-1 siRNA or control siRNA. As shown in Fig.4, RA FLSs treated with control siRNA displayed flat or ruffling lamellipodia at the leading edge, however, cells transfected with SUMO-1 siRNA were significantly suppressed in lamellipodia formation. 3.5 SUMO-1 knockdown reduces TNF-α-induced activation of Rac1 and PAK1 The small guanosine triphosphatase (GTPase) Rac1 is a central regulator of cell motility. Activation of Rac1 promotes the organization of actin cytoskeleton at the leading edge, particularly, formation of lamellipodia [15,16]. Since Rac1 might be a target of SUMO-1[17], we evaluated the inhibitory effect of SUMO-1 inhibition on activation of Rac1. As shown in Fig.5A, compared with control siRNA, treatment with SUMO-1 siRNA inhibited TNF-α-induced activation of Rac1, but had no effect on the expression of total Rac1 protein. Previous study showed that phosphorylated p21-activated kinase 1 (PAK1) is involved in cell migration [18,19]. Since PAK1 is considered as a key downstream effecter of Rac1 signaling pathway, we determined the role SUMO-1 in PAK1 activation in RA FLSs. As shown in Fig.5B, treatment with SUMO-1 siRNA decreased TNF-α-induced phosphorylation of PAK1, but had no effect on the expression of PAK1. Pharmacological Rac1 inhibitor NSC23766 reduced Tnf-α-induced PAK1 phophorylation (Fig.5C). Our data suggest the role of SUMO-1 in regulating activation of Rac1/PAK1 signaling pathway.

3.6 SUMO-1 mediated RA FLS migration and invasion through Rac1 To confirm whether activation of Rac1 is required for SUMO-1-mediated migration and invasion in RA FLS, we increased the expression of SUMO-1 in RA FLSs by transfecting SUMO-1-overexpressing lentivirus (Fig.6A). Chemotaxis migration was remarkably enhanced when SUMO-1 was overexpressed. This effect, however, was partially inhibited when NSC23766 was added (Fig.6B). NSC23766 also blocked Matrigel invasion of RA FLS overexpressing SUMO-1 (Fig.6C). Our results suggested Rac1 is a mediator through which SUMO-1 regulates migration and invasion in RA FLS. 3.7 Inhibition of Rac1, PAK1 activity decreases migration and invasion of RA FLSs To confirm the role of Rac1 and PAK1 in mediating cell migration and invasion, we treated RA FLS with NSC23766 or PAK1 inhibitor IPA-3. The results showed that these specific inhibitors significantly decreased migration and invasion of RA FLSs (Supplement 2A, B). Taken together, our data suggest that Rac1/PAK1 signaling pathway mediates SUMO-1-induced aggressive behavior of RA FLSs. 4.

Discussion

We have demonstrated that among four members in SUMO protein family, mRNA expression of SUMO-1 and SUMO-2 was elevated in RA FLSs while SUMO-3 and SUMO-4 was barely detected. Considering that the role of SUMO-2 in the regulation of migration in RA FLSs has been reported, 8 in the present study, we focused on investigating whether SUMO-1 also regulates migration and invasion of RA FLSs. We found that SUMO-1 knockdown by siRNA remarkably decreased the migration and invasion of RA FLSs, as well as expression of MMP-1 and MMP-3. Furthermore, we showed that SUMO-1 knockdown downregulated activity of Rac1 and its downstream effecter PAK1. Suppression of Rac1 by specific inhibitor blocked the increasing migration as well as invasion induced by SUMO-1 overexpression. Pharmacological inhibition of Rac1 or PAK1 significantly reduced in vitro migration and invasion. Our findings suggest that SUMO-1 regulates migration and invasion through Rac1/PAK1 signaling pathway in RA FLSs, and exerts important influence on the maintenance of aggressive phenotype of RA FLSs. The aggressive property of FLSs has been considered as one of the major causes of joint destruction in RA. By producing adhesion molecules and proteolytic enzymes, RA FLSs degrade extracellular matrix and invade into the cartilage. However, although accumulating evidence suggests inhibition of migration and invasion of FLSs to be a potential target against RA, few therapies were able to prevent the damage completely. Therefore, to clarify the precise mechanisms that regulate

aggressive behavior of RA FLSs is still of great importance. SUMO proteins are the major participants during protein SUMOylation. By attaching to specific lysine residues of the substrates, SUMO can alter protein localization, stability, transcriptional activity, and eventually regulate various cellular events including survival, proliferation, and migration. Previous studies indicate that dysfunction of protein SUMOylation contributes to the survival and progression of some cancer cells. For instance, inhibition SUMO-1, -2 or -3 suppresses survival of glioblastoma cell [20]. SENP1 promotes prostate cancer metastasis and progression [21]. More interestingly, SUMO-1 is associated with tumor cell metastasis. For instance, SUMO-1, partially by promoting the production of MMP-13, is associated with the aggressiveness of bladder cancer cells [22]. Mutant p53 promotes cancer cells progression by inhibiting Rac1 SUMOylation [23]. The role of SUMO-1 in tumor cells indicates the relation between SUMO-1 and the aggressive nature of RA FLSs. SUMO-1 has been found to localize mainly at the invading edge of synovial tissue of patients with RA [24], but its role in regulating aggressive behavior of RA FLS is still unclear. In the present study, we also confirmed that the expression of SUMO-1 was elevated in both the synovial tissue and FLSs obtained from patients with RA. Moreover, we are the first to show that knockdown SUMO-1 by siRNA inhibited cell migration and invasion of RA FLSs. These data further support the notion that increasing expression of SUMO-1 plays a role in the regulation of invasive behavior of RA FLSs. Adhesion, polarization, and cytoskeleton reorganization are key steps that control directional cell movement. Lamellipodium, which presents as extended protrusion at cell membrane is the first step. Cells adhere to extracellular matrix by lamellipodia, followed by cell body contraction and move forward. We observed that while FLSs displayed lamellipodia at the edge after being wounded, however, most of the cells transfected with SUMO-1 siRNA failed to form the lamellipodia, suggesting that SUMO-1 may facilitate migration through regulating lamellipodium formation in RA FLSs. Rho GTPases function as regulators in cytoskeleton reorganization. Rac1, in particular, is in charge of lamellipodium formation by modulating actin polymerization [24,25]. When activated, Rac1 is translocated from cytoplasm to membrane and activates the downstream effecter PAK1. In this work, we demonstrated that SUMO-1 knockdown reduced activity of Rac1 and PAK1 in TNF-α-induced RA FLSs. Suppressing Rac1 activity blocked the enhancing migration and invasion induced by SUMO-1 overexpression. Pharmacological inhibition of Rac1 and PAK1 activation reduced migration and invasion of RA FLSs. Consistent with our data, suppression of Rac1 SUMOylation by inhibiting

endogenous SUMO-1 reduces Rac1 activity in MDCKII cells [17]. Taken together, these results suggest that Rac1/PAK1 signaling pathway is involved in SUMO-1-mediated motility of RA FLSs. MMPs, mainly produced by FLSs in RA, are proteinases which degrade extracellular matrix to facilitate cell migration an invasion in RA. They can be induced by pro-inflammatory cytokines, such as TNF-α and IL-1β. However, the signaling mechanisms which modulate MMPs expression in RA FLSs are not well defined. In the present study, we are the first to demonstrate that SUMO-1 knockdown decreased TNF-α-induced expression of MMP-1 and MMP-3 in RA FLSs, suggesting an important role of SUMO-1 in the maintenance of activated phenotype of RA FLSs. Consistent with our results, a previous study has shown that SUMO 2/3 is involved in regulation of MMP-3 and MMP-13 [8] SENP1, which served as a desumoylation enzyme, participated in the regulation of MMP-1 expression in RA FLSs [26]. These results suggest that one of the mechanisms of SUMO-1 involved in pathogenesis of RA is regulating MMP-1 and MMP-3 production. In conclusion, our study suggests that SUMO-1 is involved in the regulation of the aggressive behavior of RA FLSs. The elevated expression of SUMO-1 may contribute to joint destruction in RA.

Conflict of interests The authors declare no conflict of interests on the publication of this paper.

Financial support This project was supported by the grants from the National Natural Science Foundation of China (81601403, 81603435) and grants from Natural Science Foundation of Guangdong Province (2018A0303130294).

References [1] E.H. Choy, G.S.Panayi, Cytokine pathways and joint inflammation in rheumatoid arthritis, N. Engl. J. Med. 344 (2001) 907-916, https://doi.org/ 10.1056/NEJM200103223441207. [2] L.C. Huber, O. Distler, I. Tarner, R.E. Gay, S. Gay, T. Pap, Synovial fibroblasts: key players in rheumatoid arthritis, Rheumatology (Oxford). 45 (2006) 669-675, https://doi.org/10.1093/rheumatology/kel065.

[3] U. Müller-Ladner, T. Pap, R.E. Gay, M. Neidhart, S. Gay, Mechanisms of disease: the molecular and cellular basis of joint destruction in rheumatoid arthritis, Nat. Clin. Pract. Rheumatol. 1 (2005) 102-110, https://doi.org/10.1038/ncprheum0047. [4] T. Pap, I. Meinecke, U. Müller-Ladner, S. Gay, Are fibroblasts involved in joint destruction? Ann. Rheum. Dis. 64 (2005) Suppl 4:iv52-4, https://doi.org/10.1136/ard.2005.042424. [5] S. Lefevre, F.M. Meier, E. Neumann, U. Muller-Ladner. Role of synovial fibroblasts in rheumatoid arthritis, Curr. Pharm. Des. 21 (2015) 130-141. [6] J. Huang, J. Yan, J. Zhang, S. Zhu, Y. Wang, T. Shi, et al., SUMO1 modification of PTEN regulates tumorigenesis by controlling its association with the plasma membrane, Nat. Commun. 3 (2012) 911, https://doi.org/10.1038/ncomms1919. [7] Bertolotto C, Lesueur F, Giuliano S, Strub T, de Lichy M, Bille K, et al., A SUMOylation-defective MITF germline mutation predisposes to melanoma and renal carcinoma, Nature 2011;480:94-8. [8] S. Frank, M.A. Peters, C. Wehmeyer, S. Strietholt, C. Koers-Wunrau, J. Bertrand, et al., Regulation of matrixmetalloproteinase-3 and matrixmetalloproteinase-13 by SUMO-2/3 through the transcription factor NF-κB, Ann. Rheum. Dis. 72 (2013) 1874-1881, https://doi.org/10.1136/annrheumdis-2012-202080. [9] I. Meinecke, A. Cinski, A. Baier, M.A. Peters, B. Dankbar, A. Wille, et al., Modification of nuclear PML protein by SUMO-1 regulates Fas-induced apoptosis in rheumatoid arthritis synovial fibroblasts, Proc. Natl. Acad. Sci. U S A. 104 (2007) 5073-5078, https://doi.org/10.1073/pnas.0608773104. [10] F. Li, X. Li, L. Kou, Y. Li, F. Meng, F. Ma, SUMO-conjugating enzyme UBC9 promotes proliferation and migration of fibroblast-like synoviocytes in rheumatoid arthritis, Inflammation. 37 (2014) 1134-1141, https://doi.org/10.1007/s10753-014-9837-x. [11] F.C. Arnett, S.M. Edworthy, D.A. loch, D.J. McShane, J.F. Fries, N.S. Cooper, et al., The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis, Arthritis. Rheum. 31 (1998) 315-324. [12] Recommendations for the medical management of osteoarthritis of the hip and knee: 2000 update. American College of Rheumatology Subcommittee on Osteoarthritis Guidelines, Arthritis. Rheum. 43 (2000) 1905-1915, https://doi.org/10.1002/1529-0131(200009)43:9<1905::AID-ANR1>3.0.CO;2-P. [13] T.C. Tolboom, A.H. van der Helm-Van Mil, R.G. Nelissen, F.C. Breedveld, R.E. Toes, T.W. Huizinga. Invasiveness of fibroblast-like synoviocytes is an individual patient characterize associated with the rate of joint destruction in patients with rheumatoid arthritis, Arthritis. Rheum. 52 (2005) 1999-2002, https://doi.org/10.1002/art.21118. [14] E.C. Brantley, L.B. Nabors, G.Y. Gillespie, Y.H. Choi, C.A. Palmer, K. Harrison, et al., Loss of protein inhibitors of activated STAT-3 expression in glioblastoma multiforme tumors: implications for STAT-3 activation and gene expression, Clin. Cancer. Res. 14 (2008) 4694-4704, https://doi.org/10.1158/1078-0432.CCR-08-0618. [15] K. Burridge, K. Wennerberg, Rho and rac take center stage, Cell 116 (2004) 167-179. [16] A.Y. Chan, S.J. Coniglio, Y.Y. Chuang, D. Michaelson, U.G. Knaus, M.R. Philips, et al., Roles of the Rac1 and Rac3 GTPases in human tumor cell invasion, Oncogene. 24 (2005) 7821-7829,

https://doi.org/10.1038/sj.onc.1208909. [17] S. Castillo-Lluva, M.H. Tatham, R.C. Jones, E.G. Jaffray, R.D. Edmondson, R.T. Hay, et al., SUMOylation of the GTPase Rac1 is required for optimal cell migration, Nat. Cell. Biol. 12 (2010) 1078-1085, https://doi.org/10.1038/ncb2112. [18] G. Wang, Q. Zhang, Y. Song, X. Wang, Q. Guo, L. Zhang, et al., PAK1 regulates RUFY3-mediated gastric cancer cell migration and invasion, Cell. Death. Dis. 6 (2015) e1682, https://doi.org/10.1038/cddis.2015.50. [19] Y. Ma, S.K. McCarty, N.P. Kapuriya, V.J. Brendel, C. Wang, X. Zhang, et al., Development of p21 activated kinase-targeted multikinase inhibitors that inhibit thyroid cancer cell migration, J. Clin. Endocrinol. Metab. 98 (2013) E1314-1322, https://doi.org/10.1210/jc.2012-3937. [20]W. Yang, L. Wang, G. Roehn, R.D. Pearlstein, F. Ali-Osman, H. Pan, et al., Small ubiquitin-like modifier 1-3 conjugation [corrected] is activated in human astrocytic brain tumors and is required for glioblastoma cell survival, Cancer. Sci. 104 (2013) 70-77, https://doi.org/10.1111/cas.12047. [21] Q. Wang, N. Xia, T. Li, Y. Xu, Y. Zou, Y. Zuo, et al., SUMO-specific protease 1 promotes prostate cancer progression and metastasis, Oncogene. 32 (2013) 2493-2498, https://doi.org/10.1038/onc.2012.250. [22] M. Tan, H. Gong, J. Wang, L. Tao, D. Xu, E. Bao, et al., SENP2 regulates MMP13 expression in a bladder cancer cell line through SUMOylation of TBL1/TBLR1, Sci. Rep. 5 (2015) 13996, https://doi.org/10.1038/srep13996. [23] X. Yue, C. Zhang, Y. Zhao, J. Liu, A.W. Lin, V.M. Tan, et al., Gain-of-function mutant p53 activates small GTPase Rac1 through SUMOylation to promote tumor progression, Genes. Dev. 31 (2017) 1641-1654, https://doi.org/10.1101/gad.301564.117. [24] J.K. Franz, T. Pap, K.M. Hummel, M. Nawrath, W.K. Aicher, Y. Shigeyama, et al., Expression of sentrin, a novel antiapoptotic molecule, at sites of synovial invasion in rheumatoid arthritis, Arthritis. Rheum. 43 (2000) 599-607, https://doi.org/10.1002/1529-0131(200003)43:3<599::AID-ANR17>3.0.CO;2-T. [25] R.J. Petrie, A.D. Doyle, K.M. Yamada. Random versus directionally persistent cell migration, Nat. Rev. Mol. Cell. Biol. 10 (2009) 538-549, https://doi.org/10.1038/nrm2729. [26] H. Maciejewska-Rodrigues, E. Karouzakis, S. Strietholt, H. Hemmatazad, M. Neidhart, C. Ospelt, et al., Epigenetics and rheumatoid arthritis: the role of SENP1 in the regulation of MMP-1 expression, J. Autoimmun. 35 (2010) 15-22, https://doi.org/10.1016/j.jaut.2009.12.010.

Fig 1 Increased expression of SUMO-1 in FLSs and the synovial tissues (STs) from patients with RA. (A) mRNA expression of SUMO-1, -2, -3 and -4 was detected by Q-RT-PCR in FLSs from

RA

(n=10) or OA (n=10) patients. Data were normalized to GAPDH. *P<0.05 vs OA. (B) Expression of SUMO-1 (arrow) measured by immunohistochemistry in STs isolated from RA (n=5) or OA (n=5) patients. Original magnification×200 (upper panel) or ×400 (lower panel). (C) Expression of SUMO-1 measured by immunofluorescence in FLSs isolated from RA (n=5) and OA (n=5) patients. Original magnification×400. (D) Expression of SUMO-1 was detected by Western blot in FLSs from RA (n=3) and OA (n=3) patients. The average expression of the conjugated- and free-SUMO-1 was shown as mean±SEM of densitometrical quantification (lower panel) from triplicated experiments and presented as fold changes over controls after normalization by β-actin. *P<0.05 vs OA. (E) Confluent FLSs from RA patients (n=3) were treated with TNF-α (10 ng/ml), IL-1β (10 ng/ml) or IL-17α (10 ng/ml) for 24 h and expression of SUMO-1 was measured by Western blot analysis. The average expression of the conjugated- and free-SUMO-1 was shown as mean±SEM of densitometrical quantification (lower panel) from three independent experiments. *P<0.05 vs Basal. (F) Confluent FLSs from RA patients (n=3) were treated with TNF-α (10 ng/ml) for 12 h, 24 h or 48 h and expression of SUMO-1 was measured by Western blot analysis. The average expression of the conjugated- and free-SUMO-1 was shown as mean±SEM of densitometrical quantification (lower panel) from three independent experiments. *P<0.05 vs Basal. Fig 2 Targeted depletion of SUMO-1 decreases migration and invasion of RA FLSs. (A) RA FLSs were transfected with SUMO-1 siRNA-1, -2, -3 or scramble control (SC) and expression of SUMO-1 was measured by Western blot analysis. The average expression of the conjugated- and free-SUMO-1 was shown as mean±SEM of densitometrical quantification (lower panel) from three independent experiments. **P<0.01 vs SC. (B) OA FLSs were transfected with SUMO-1-overexpressing lentivirus (OE SUMO-1) and the expression of SUMO-1 was measured by Western blot analysis. The average expression of the conjugated- and free-SUMO-1 was shown as mean±SEM of densitometrical quantification (lower panel) from three independent experiments. **P<0.01 vs SC. (C) RA FLSs (n=5), transfected with SUMO-1 siRNA-2 (siSUMO-1) or SC were serum starved overnight at 24 h. Chemotaxis migration was performed in a Boyden chamber. The RA FLSs were seeded in a Boyden chamber and allowed to migrate for 8 h, fixed, and stained with Hemacolor. 10% FBS was used as chemoattractants. The migration index represents the number of migrated cells normalized to

FBS-containing media. Data are presented as mean±SEM of five independent experiments. The images are representative of migration through the membrane after staining. Original magnification×100. **P<0.01 vs SC. (D) OA FLSs (n=5), transfected with SUMO-1-overexpressing lentivirus or SC were seeded in a Boyden chamber and migrate for 8 h. The migration index represents the number of migrated cells normalized to FBS-containing media. Data are presented as mean±SEM of five independent experiments. The images are representative of migration through the membrane after staining. Original magnification×100. *P<0.05 vs SC. (E) In vitro invasion assay was performed using inserts coated with a Matrigel basement membrane matrix in Boyden chambers. RA FLSs (n=5) were allowed to invade through Matrigel toward FBS-containing media for 8 h. The number of invading cells was averaged from three×10 field-of view images, and invasion index was calculated by normalizing the mean of invaded cells to FBS-containing media. The images are representative images of stained cells that invaded through Matrigel invasion chambers. Original magnification×100. Data are presented as mean±SEM of five independent experiments. **P<0.01 vs SC. (F) OA FLSs (n=5) were allowed to invade through Matrigel toward FBS-containing media for 8 h. The number of invading cells was averaged from three×10 field-of view images, and invasion index was calculated by normalizing the mean of invaded cells to FBS-containing media. The images are representative images of stained cells that invaded through Matrigel invasion chambers. Original magnification×100. Data are presented as mean±SEM of five independent experiments. *P<0.05 vs SC. Fig 3 SUMO-1 knockdown reduces TNF-α-induced expression of MMP-1 and MMP-3 in RA FLSs. RA FLSs (n=3), transfected with scramble siRNA (SC) or SUMO-1 siRNA (siSUMO-1), were treated with TNF-α (10 ng/ml) for 24 h. The mRNA expression of MMP-1 (A) and MMP-3 (B) was determined by Q-RT-PCR. Data were normalized to GAPDH. *P<0.05 vs SC, **P<0.01 vs SC, ##P<0.01 vs SC+TNF-α. Fig 4 SUMO-1 knockdown impairs lamellipodia formation in RA FLSs. RA FLSs (n=3) were plated overnight on coverslips, and then were fixed and stained with fluorescent phalloidin 3 h after wounding to visualize polymerized actin in migrating cells. Arrows indicate lamellipodia formation. The

images

presented

are

representative

of

three

independent

experiments.

Original

magnification×400. Fig 5 SUMO-1 knockdown decreases TNF-α-induced Rac1/PAK1 activation. RA FLSs were transfected with siRNAs specific for SUMO-1 (siSUMO-1) or scramble control (SC). (A) Effect of

SUMO-1 knockdown on Rac1 activity. RA FLSs (n=5) were serum starved for 24 h and stimulated with TNF-α for 10 min. Cells were lysed, and Rac1 activity was detected by G-LISA Rac1 activation assay. Rac1 protein was measured by Western blotting. Data are presented as mean±SEM of five independent experiments. **P<0.01 vs SC. (B) Effect of SUMO-1 knockdown on phosphorylation of PAK1. Western blot analysis for p-PAK1 and total-PAK following TNF-α stimulation is shown in RA FLS (n=3) transfected with SUMO-1 siRNA or scramble control. A representative blot of three independent experiments is shown. Data are shown as the ratio of arbitrary absorption units of p-PAK1 and PAK1 (mean±SEM) (lower panel). **P<0.01 vs SC. (C) Effect of Rac1 inhibition on PAK1 activation. RA FLSs (n=3) were pre-treated with Rac1 inhibitor NSC23766 for 1 hour prior to TNF-α stimulation. The expression of p-PAK1 and PAK1 was analyzed by Western blot. A representative blot of three independent experiments is shown. Data are shown as the ratio of arbitrary absorption units of p-PAK1 and PAK1 (mean ± SEM) (lower panel). *P<0.05 vs SC, #P<0.05 vs SC+TNF-α. Fig 6 SUMO-1 promotes in vitro migration and invasion of RA FLSs through Rac1. (A) RA FLSs were transfected with SUMO-1-overexpressing lentivirus (OE SUMO-1) and the expression of SUMO-1 was measured by Western blot analysis. The average expression of the conjugated- and free-SUMO-1 was shown as mean±SEM of densitometrical quantification (lower panel) from three independent

experiments.

*P<0.05

vs

SC.

(B)

RA

FLSs

(n=5),

transfected

with

SUMO-1-overexpressing lentivirus or SC were pre-incubated with NSC23766 or control solution for 24 h. Then cells were seeded in a Boyden chamber and allowed to migrate for 8 h. 10% FBS was used as chemoattractants. The migration index represents the number of migrated cells normalized to FBS-containing media. Data are presented as mean±SEM of five independent experiments. The images are representative of migration through the membrane after staining. Original magnification×100. **P<0.01 vs SC, ## P<0.01 vs OE SUMO-1. (C) In vitro invasion assay were performed using inserts coated with a Matrigel basement membrane matrix in Boyden chambers. RA FLSs (n=5) were allowed to invade through Matrigel toward FBS-containing media for 8 h. The number of invading cells was averaged from three×10 field-of view images, and invasion index was calculated by normalizing the mean of invaded cells to FBS-containing media. The images are representative images of stained cells that invaded through Matrigel invasion chambers. Original magnification×100. Data are presented as mean±SEM of five independent experiments. **P<0.01 vs SC, ## P<0.01 vs OE SUMO-1. Supplement 1 Effect of SUMO-1 knockdown on proliferation and apoptosis of RA FLSs. (A) RA

FLSs (n=3), transfected with SUMO-1 siRNA or scramble control were cultured for 72 h. Cells proliferation was measured by Edu incorporation assay. The images are representative images of cells with Edu positivity. Data are presented as mean±SEM of three independent experiments. (B) RA FLSs (n=3), transfected with SUMO-1 siRNA or scramble control were cultured in FBS-containing median for 72 h. Cells were fixed, stained with annexin V and propidium iodide (PI) and measured by flow cytometry. Data are presented as mean±SEM of three independent experiments. Supplement 2 Rac1 or PAK1 inhibition reduces migration and invasion of RA FLSs. (A) Chemotaxis was evaluated using a Boyden chamber migration assay. RA FLSs (n=5) in serum-free condition in presence of Rac1 inhibitor NSC23766 or PAK1 inhibitor IPA-3 were placed in the upper chambers. Cells were allowed to migrate for 8 h, fixed, and stained with Hemacolor staining kit. The migration index represents the number of migrated cells normalized to FBS-containing media. *P<0.05 vs Ctrl. (B) In vitro invasion assay were performed using inserts coated with a Matrigel basement membrane matrix in Boyden chambers. RA FLSs (n=5) were allowed to invade through Matrigel toward FBS-containing media for 8 h. Invasion index was calculated by normalizing the mean of invaded cells to FBS-containing media. Data are presented as mean±SEM from five independent experiments. *P<0.05 vs Ctrl.