P38 Stabilizes Snail by Suppressing DYRK2- Mediated Phosphorylation

Published OnlineFirst June 17, 2019; DOI: 10.1158/0008-5472.CAN-19-0049

Cancer Molecular Cell Biology Research

p38 Stabilizes Snail by Suppressing DYRK2- Mediated Phosphorylation That Is Required for GSK3b-bTrCP–Induced Snail Degradation Ki-Jun Ryu1, Sun-Mi Park1, Seung-Ho Park2, In-Kyu Kim1, Hyeontak Han1, Hyo-Jin Kim1, Seon-Hee Kim1, Keun-Seok Hong1, Hyemin Kim1, Minju Kim1, Sung-Jin Yoon2, Killivalavan Asaithambi3,4, Kon Ho Lee3,4, Jae-Yong Park5, Young-Sool Hah6,7, Hee Jun Cho8, Jong In Yook9, Jung Wook Yang10, Gyung-Hyuck Ko10,11, Gyemin Lee12, Yang Jae Kang1,13, Cheol Hwangbo1,13, Kwang Dong Kim1,13, Young-Jun Park2, and Jiyun Yoo1,13

Abstract

Snail is a key regulator of epithelial–mesenchymal tran- tively suppressed DYRK2-mediated Ser104 phosphoryla- sition (EMT), which is a major step in tumor metastasis. tion, which is critical for GSK3b-dependent Snail phosphor- Although the induction of Snail transcription precedes EMT, ylation and bTrCP-mediated Snail ubiquitination and deg- posttranslational regulation, especially phosphorylation of radation. Importantly, functional studies and analysis of Snail, is critical for determining Snail protein levels or clinical samples established a crucial role for the p38–Snail stability, subcellular localization, and the ability to induce axis in regulating ovarian cancer EMT and metastasis. These EMT. To date, several kinases are known that enhance the results indicate the potential therapeutic value of targeting stability of Snail by preventing its ubiquitination; however, the p38–Snail axis in ovarian cancer. the molecular mechanism(s) underlying this are still unclear. Here, we identified p38 MAPK as a crucial post- Significance: These findings identify p38 MAPK as a novel translational regulator that enhances the stability of Snail. regulator of Snail protein stability and potential therapeutic p38 directly phosphorylated Snail at Ser107, and this effec- target in ovarian cancer.

Introduction epithelial cells initially lose apical–basal polarity and cell–cell Ovarian cancer is the leading cause of death from gyneco- contact while shifting to a mesenchymal phenotype (4). This logical malignancy (1). Despite introduction of targeted ther- loss of epithelial features is often accompanied by increased apies, survival has not signiﬁcantly improved in the last cell motility and expression of mesenchymal genes, a process decade (2). The 5-year survival for ovarian cancer is remarkably that is collectively referred to as the epithelial–mesenchymal low, as this disease frequently recurs and quickly metastasizes transition (EMT) and considered a key step during the throughout the peritoneal cavity. Therefore, a better under- progression of tumors toward metastasis (5, 6). Thus, EMT standing of the molecular events that contribute to tumor regulators, which likely play important roles in cancer invasion and metastasis is crucial for developing novel progression, have been extensively studied. One such regulator treatment strategies for ovarian cancer. is the zinc ﬁnger protein Snail, which induces EMT by directly Most cancer-related deaths are attributable to local invasion repressing E-cadherin transcription during development or and distant metastasis of tumor cells (3). During metastasis, tumor progression (7, 8).

1Division of Applied Life Science (BK21 Plus), Research Institute of Life Sciences, Gyeongsang National University, Jinju, Korea. 13Division of Life Science, College Gyeongsang National University, Jinju, Korea. 2Environmental Disease Research of Natural Sciences, Gyeongsang National University, Jinju, Korea. Center, Korea Research Institute of Bioscience and Biotechnology, Daejeon, Note: Supplementary data for this article are available at Cancer Research 3 Korea. Department of Convergence Medical Science (BK21 Plus), Graduate Online (http://cancerres.aacrjournals.org/). School Gyeongsang National University, Jinju, Korea. 4Department of Microbi- ology, Gyeongsang National University School of medicine, Jinju, Korea. 5School K.J. Ryu, S.M. Park, S.H. Park, and I.K. Kim contributed equally to this article. of Biosystem and Biomedical Science, College of Health Science, Korea Uni- Corresponding Authors: Jiyun Yoo, Gyeongsang National University, Jinjudaero versity, Seoul, Korea. 6Biomedical Research Institute, Gyeongsang National 501, Jinju 52828, Korea. Phone: 82-55-772-1327; Fax: 82-55-772-2553; E-mail: University Hospital, Jinju, Korea. 7Institute of Health Sciences, Gyeongsang [email protected]; and Young-Jun Park, Environmental Disease Research Center, National University School of Medicine, Jinju, Korea. 8Immunotherapy Conver- Korea Research Institute of Bioscience and Biotechnology, Daejeon 34141, gence Research Center, Korea Research Institute of Bioscience and Biotech- Korea. Phone: 82-42-879-4219; Fax: 82-42-879-8595; E-mail: nology, Daejeon, Korea. 9Department of Oral Pathology, Oral Cancer Research [email protected] 10 Institute, College of Dentistry, Yonsei University, Seoul, Korea. Department of Cancer Res 2019;79:4135–48 Pathology, Gyeongsang National University Hospital, Jinju, Korea. 11Department of Pathology, Gyeongsang National University School of Medicine, Jinju, Korea. doi: 10.1158/0008-5472.CAN-19-0049 12Department of Information and Statistics, College of Natural Sciences, 2019 American Association for Cancer Research.

www.aacrjournals.org 4135

Ryu et al.

Snail expression is controlled at the transcriptional level by to express the markers, such as ADE2, HIS3, and LacZ. The many growth factors and cytokines including HGF, TNFa,or pDEST–GADT7 plasmids were isolated from the positive colo- TGFb (9–11); however, Snail mRNA is constitutively present nies, and transformed into Escherichia coli to amplify, and then in many cell types, even in the absence of activation of these cDNA inserts were sequenced. signaling pathway (12). Snail is an extremely unstable protein, and its subcellular level or protein stability is mainly regulated Plasmid construction and transfection by the many different kinases. Snail phosphorylation by The full-length human Snail gene was subcloned into a GSK3b, for example, negatively regulates Snail function by induc- pDONR207 vector (Entry vector) using the Gateway Cloning ing nuclear export and ubiquitination-dependent cytosolic System (Invitrogen) following the manufacturer's instructions. degradation (12–14). In contrast, other kinases positively regu- The entry clones were converted into several destination vectors, late Snail function by inducing nuclear import, nuclear retention, pDEST-GFP-C, pDEST-FLAG-C, and pDEST-HA-C. Site-directed and enhancing its stability (15–19). mutagenesis was performed with a QuikChange Mutagenesis To date, three kinases (ATM, Erk2, and DNA-PKcs) have been Kit, according to the manufacturer's instructions. For transient known to enhance the stability of Snail preventing its ubiquitina- transfection, HEK293T cells were seeded in 6-well or 100-mm- tion (17–19); however, the molecular mechanism whereby these diameter dish for 24 h and transfected with the indicated plasmid kinases prevent Snail ubiquitination is unknown. Here, we show by using Fugene 6 transfection reagent (Roche) following man- that p38 MAPK suppresses DYRK2-mediated prime phosphory- ufacturer's instruction. After 48 hours, the cells were harvested lation required for GSK3b-bTrCP–mediated Snail ubiquitination and used for Western blot analysis. Two different siRNA oligo and degradation, and the resulting stabilized Snail protein can duplexes for targeting human p38 or murine p38, respectively, promote EMT and tumor metastasis. were purchased from Bioneer. Transient transfection of siRNA oligo duplex was accomplished using siLentFect Reagent (Bio-Rad) following manufacturer's instruction. For stable Materials and Methods transfection, SKOV3 cells were transfected with Flag-tar (Con), Cell culture Flag-WT-Snail, or Flag-S107A-Snail expressing plasmid by using All cell lines used in this study, except 4T1 cells (ATCC), were the Fugene 6 transfection reagent (Roche). After 48 hours incu- obtained from the Korean Cell Line Bank, where they were bation, 500 mg/mL of G418 was added to the cultures to select characterized by DNA-fingerprinting and isozyme detection, and for G418-resistant clones. Three to four weeks later, independent cultured according to ATCC instructions. All cell lines were used colonies were picked using cloning cylinder (Sigma), subcultured, within 3 to 20 passages of thawing the original stocks and were and tested for Snail expression by Western blot analysis. tested every 3 months for mycoplasma contamination. The cell lines were maintained for no more than 3 passages between Immunoprecipitation experiments. Human HEK293T, HeLa, A549, human colon cancer Cells were lysed in lysis buffer: 20 mmol/L Tris pH 7.4, cell lines (HCT116, SW480, SW620, DLD-1, KM-12), and human 2 mmol/L EDTA, 150 mmol/L sodium chloride, 1 mmol/L breast cancer cell lines (MCF-7, MDA-MB-231, T47D) were cul- sodium deoxycholate, 1% Triton X-100, 10% glycerol, 2 pills tured in DMEM (Invitrogen) supplemented with 10% FBS and 1% protease inhibitor cocktail (Roche), mixed by vortexing and penicillin and streptomycin. Human ovarian cancer cell lines incubated 30 minutes on ice. Lysates were precleared using (SKOV3, SNU-8, OVCAR-3), human prostate cancer cell lines protein A/G beads (Santa Cruz Biotechnology), incubated with (LNCaP, DU145, PC-3), and 4T1 murine mammary carcinoma the specific antibodies for 2 hours at 4C and then incubated cells were cultured in RPMI (Invitrogen) supplemented with 10% with beads for overnight at 4C with gentle mixing. Beads were FBS and 1% penicillin and streptomycin. then washed five times with lysis buffer and eluted with 30 mLof 2 SDS sample buffer. Western blot analysis was then performed. Kinase library construction and yeast two-hybrid screening Primary antibodies used for immunoprecipitation (IP) were as The full-length human kinase genes (approximately 300 kinase follows: Flag (Cell Signaling Technology, No. 14793; 5 mg/mL), clones; Supplementary Table S1) were subcloned into a HA (Abm G036; 5 mg/mL), Snail (Cell Signaling Technology, pDONR207 vector (Entry vector) using the Gateway Cloning No. 3879; 5 mg/mL), and p38 (Cell Signaling Technology, System (Invitrogen) following the manufacturer's instructions. No. 9212; 5 mg/mL). The entry clones were converted into a destination vector, pDEST-GADT7, which expresses fusion protein with the GAL4 Western blot analysis activation domain. The full-length Snail protein was produced as Protein samples were subjected to SDS-PAGE and transferred a fusion protein with the GAL4 DNA-binding domain in plasmid to polyvinylidene fluoride membranes. Membranes were incu- pGBKT7 and used as a bait for yeast two-hybrid screening. Yeast bated with indicated primary antibodies for overnight at 4C. two-hybrid screening was performed with the MATCHMARKER After washing with TBS-T (TBS containing 0.1% Tween-20), two-hybrid system (Clontech) following the manufacturer's membranes were incubated with corresponding horseradish instructions. The pGBKT7-Snail plasmid was transformed into peroxidase-conjugated secondary antibodies (1:5,000) for 1 hour the MATa strain Y187 and human kinase library-containing at room temperature. Blots were developed with enhanced pDEST-GADT7 plasmids were transformed into MATa strain chemiluminescence (ECL; GE Healthcare) reaction according AH109. When two transformants were mated to each other, to manufacturer's instructions. Primary antibodies used diploid cells were formed that contained reporter gene. If the were: anti-Snail (Cell Signaling Technology, No. 3895; fusion proteins interact each other, mated cells were grown in 1:1,000), anti-phospho-p38 (Cell Signaling Technology, No. the selection medium. A total of 1.5 107 transformants were 4511; 1:1,000), anti-p38 (Cell Signaling Technology, No. 9212; screened. Positive colonies were selected based on their capacity 1:1,000), anti-DYRK2 (Cell Signaling Technology, No. 8143;

4136 Cancer Res; 79(16) August 15, 2019 Cancer Research

p38 Regulates Snail Protein Stability

1:1,000), anti-MKK6 (Cell Signaling Technology, No. 8550; degradation of the Snail protein before lysed with Triton X-100 1:1,000), anti-GSK3b (Cell Signaling Technology, No. 9315; lysis buffer. Cell lysates were then collected and immunopreci- 1:1,000), anti-phospho-GSK3b(S9) (Cell Signaling Technolo- pitated with anti-GFP antibody (Abcam, ab6556; 5 mg/mL) to gy, No. 5558; 1:1,000), anti-phospho-b-catenin (S33/37/T41) specifically pull down GFP-Snail protein. Pulled down samples (Cell Signaling Technology, No. 9561; 1:1,000), anti-PTEN were subject to immunoblotting with anti-HA (ubiquitin) to (Cell Signaling Technology, No. 9552; 1:1,000), anti-Flag visualize polyubiquitinated Snail protein bands. (Abm G191; 1:1,000), anti-HA (Abm G036; 1:1,000), anti- vimentin (Santa Cruz Biotechnology, sc-6260, 1;1,000), anti- Docking calculations Slug (Cell Signaling Technology, No. 9585; 1:1,000), anti-GFP Docking simulations were done using AUTODOCK 4.2 (20). (Santa Cruz Biotechnology, sc-9996; 1:1,000), anti-phosphoserine- The structure of DYRK2 (PDB ID: 3K2L) was used for docking as a proline/phospho-threonine-proline (Abcam, ab9344; 1:1,000), target with a peptide (residues 100-SQPPSPPSP-108 of Snail) anti-E-cadherin (BD Biosciences, No. 610181; 1:1,000), carrying either unphosphorylated or phosphorylated Ser107. The anti-N-cadherin (BD Biosciences, No. 610921; 1:1,000), anti- peptide was generated by using Coot (21). The AutoDockTools fibronectin (BD Biosciences, No. 610078; 1:1,000), anti-b-catenin program (20) was used to generate the docking input files (BD Biosciences, No. 610154; 1:1,000), anti-occludin (Thermo using the implemented empirical free energy function and Fisher Scientific, No. 71-1500; 1:1,000), anti-phospho-GSK3b the Lamarckian genetic algorithm. In all docking, a grid box size (S389) (Proteintech, No. 14850-1-AP; 1:1,000), and anti-a-tubulin of 60 60 60 points in x, y, and z directions was built and (Sigma, T6199; 1:1,000). the maps were centered in the catalytic site of the protein. All of the other docking parameters were used with the default values Total RNA extraction and RT-PCR according to the program manual. A hundred independent dock- Total RNA was extracted from the cultured cells using RNeasy ing runs were performed. The best docked conformations with the Mini Kit (Qiagen) following the manufacturer's instructions. lowest binding and highest scores were found. Figures were RT-PCR was performed using AccuPower RT-PCR PreMix prepared using PyMOL (22). Kit (Bioneer) according to the manufacturer's instructions. Amplification was performed using Thermo Electron PCR ther- Migration assay mal cycler. Primers used were: Snail (F: 50-ATGACTGAAAAA- Cells were seeded into Culture-Insert (Ibidi) at 5.0 105 cells/ GCCCCA-30;R:50-TCATTCTGTCCACTCCTT-30), E-cadherin (F: insert. After the cells were confluent, the Culture-Insert was 50-GAGTACCCTGATGAGATCGAG-30;R:50-TCACCGCCTCGG- removed and washed with PBS for 3 times to rinse off the detached CTTGTCACA-30), vimentin (F: 50-CCTGACGATGGCCTGGA- cells. Cells were then cultured with appropriate fresh media for GTGT-30;R:50-GCCATGTGTCACCTTCGCAG-30), b-actin (F: further 24 hours. The wound closure was observed and photo- 50-GTGGGGCGCCCCAGGCACCA-30;R:50-CTCCTTAATGT- graphed at 0 and 16 hours, using a phase-contrast microscope CACGCACGAT-30). with digital camera.

In vitro kinase assay Invasion assay In vitro kinase assays were performed by incubating recombi- Invasion assays were assessed using QCM 24-Well Cell Inva- nant active p38 kinase (Millipore) or DYRK2 (CARNA Bio- sion Assay (Fluorormetric) Kit (Millipore) following the manu- sciences) kinase proteins with purified MBP-fused WT-Snail or facturer's instruction. Cells were serum-starved for 24 hours and mutant Snail proteins (S82A-, S104A-, and S107A-Snail), which 2.5 105 cells in 250 mL of serum-free medium were seeded into were expressed in pDEST-MBP vector in E. coli, in kinase buffer upper chambers in the absence or presence of 50 ng/mL of EGF. [25 mmol/L Tris-HCl pH 7.5, 5 mmol/L b-glycerophosphate, The lower chambers were filled with 500 mL of appropriate media 2 mmol/L dithiothreitol (DTT), 0.1 mmol/L Na3VO4, containing 20% FBS. Twenty hours after incubation, noninvaded 32 10 mmol/L MgCl2] in the presence or absence of [g- P]-ATP cells/medium remaining on the upper chambers were removed by (BMS). The reaction was carried out for 30 minutes at 30C, and pipetting. The upper chambers were transferred into a clean well was terminated with SDS sample buffer. Each sample was then containing 225 mL of prewarmed Cell Detachment Solution, and boiled for 10 minutes at 100C, followed by SDS-PAGE, and incubated for 30 minutes at 37C. The upper chambers were autoradiography or immunoblot analysis using indicated removed from the well. Seventy-five microliters of lysis buffer/dye antibodies. solution (CyQuant GR Dye 1:75 with 4 lysis buffer) was added into each well and incubated for 15 minutes at room temperature. Cycloheximide pulse-chase assay Two hundred microliters of the mixture were transferred into a HEK293T cells were seeded on 12-well plate at a density of 96-well plate and assessed with a fluorescence plate reader using 2.5 105 cells per well. After culturing overnight, the cells a 480/520 nm filter set. were transfected with plasmids as desired. Two days after transfection, the cells were treated with 50 mg/mL of cycloheximide Proliferation assay (CHX). Total protein lysate were collected at different time points Cells were seeded in 6-well plates at 2 104 cells/well. After and subjected to immunoblotting for HA or Flag. incubation for 1 to 4 days, cells were trypsinized and resuspended in 1 mL of appropriate medium. The viable cells were stained with Ubiquitination assay trypan blue and counted with a hemocytometer. Ubiquitination assay was done following an IP protocol. HEK293T cells were transfected with HA-Ub, GFP-Snail (WT or Mice and animal housing S107A), p38, and MKK6. Two days after transfection, cells were Female BALB/c nude mice at 6 to 8 weeks of age were purchased treated with 10 mmol/L MG132 for 12 hours to block proteasomal from Charles River Laboratories and housed in a pathogen-free

www.aacrjournals.org Cancer Res; 79(16) August 15, 2019 4137

Ryu et al.

barrier room in Animal Care Facility at Korea Research Institute of 21 software were used in this study. All experiments were Bioscience and Biotechnology (KRIBB). All experiments using repeated at least 3 independent times. Animal studies were animals were conducted under the Institutional Animal Care and performed with adequate n numbers to ensure statistical Use Committee (IACUC)-approved protocols at KRIBB in accor- evaluation. No statistical method was used to predetermine dance with institutional guidelines. sample size. Sample size was chosen on the basis of literature in the field. Xenografts studies For peritoneal dissemination analysis of SKOV3 human ovarian cancer models, 2 106 of SKOV3 (Con), WT-Snail- Results expressing SKOV3 (WT-Snail), or S107A-Snail-expressing p38 interacts with Snail and phosphorylates residue Ser107 SKOV3 (S107A-Snail) cells were injected intraperitoneally into To find novel Snail-interacting kinases that could influence BALB/c female nude mice (n ¼ 6 for each group). Five weeks its function, we performed yeast two-hybrid screening with after the injection, the mice were euthanized, and primary home-made human kinase cDNA library (composed of approx- tumor masses in the peritoneum for ovarian cancer were imately 300 kinase clones; Supplementary Table S1) by using excised and weighed. human Snail as the bait. Through this screening, we identified 13 independent kinases that might interact with Snail. Four of Ovarian cancer tissue specimens these, namely GSK3b (12), PAK (15), PKA (23), and PKD1 (24), De-identified and paraffin-embedded human ovarian cancer have already been shown to influence Snail function by tissue specimens (39 cases) were collected from 2001 to 2012 at direct phosphorylation. In this study, we focused on p38 Gyeongsang National University Hospital, Jinju, Korea. These MAPK, which is known to induce EMT in many cancer cell clinical ovarian cancer tissue specimens were examined and types (25–27). We first confirmed the interaction between p38 diagnosed by pathologists at Gyeongsang National University and Snail by reciprocal immunoprecipitation in HEK293T cells Hospital. Tumor collections with written informed consents and cotransfected with vectors encoding HA-Snail and/or Flag-p38 analyses were approved by the Institutional Review Board at (Fig.1A).Next,weconfirmed the interaction between endog- Gyeongsang National University Hospital, Jinju, Korea. enous p38 and Snail in SW620 and HCT116 colon cancer cells, both of which express high levels of Snail (Fig. 1B). Finally, to Immunohistochemistry verify the direct binding of p38 with Snail in vitro, a commer- Three-millimeter-diameter core tissues were obtained from cially available, purified recombinant p38 protein was tested individual formalin-fixed and paraffin-embedded tissue, and for its ability to bind to purified MBP–Snail fusion proteins. arranged in new recipient paraffin blocks. Two tissue cores from The results clearly showed that p38 directly interacts with Snail the most representative tumor areas were analyzed. IHC was without the need for any other proteins (Fig. 1C). performed on 4-mm-thick paraffin sections using a BenchMark Given the interaction between p38 and Snail, we speculated ULTRA (Ventana Medical Systems Inc.) and Optiview DAB IHC that Snail might be a direct substrate of the p38 kinase. A search Detection Kit (Ventana Medical Systems Inc.). Polyclonal anti- for the consensus p38 phosphorylation motif (SP/TP) revealed bodies specific to Snail21 (1:750) and phospho-p38 (Santa Cruz that Snail contains 3 potential p38-phosphorylation motifs Biotechnology, sc-166182; 1:50), and a polyclonal antibody for (Ser 82, 104, and 107), and these serine residues are evolu- DYRK2 (LifeSpan, LS-B7095-50; 1:200) were used for IHC. tionarily conserved in other species (Fig. 1D; Supplementary ULTRA Cell Conditioning 1 (Ventana Medical Systems Inc.) was Fig. S1A). To determine which serine residue could be phos- used (56 minutes, 37C) for antigen retrieval for Snail and phorylated by p38, we performed in vitro kinase assays using phospho-p38. Antigen retrieval was not done for DYRK2. Incu- commercially available recombinant active p38 kinase and bation time for primary antibodies was 32 minutes. purified MBP-fused wild-type (WT) or mutant Snail proteins. Expression of each proteins was examined with blindness to We found that WT-, S82A-, and S104A-Snail (but not the each IHC results. Snail, phospho-p38, and DYRK2 were expressed S107A-Snail mutant) were phosphorylated by active p38 in nucleus, nucleus and cytoplasm, and cytoplasm, respectively. in vitro (Fig. 1E; Supplementary Fig. S1B). We independently Staining intensity was scored as 0 (negative), 1 (mild), confirmed these results in experiments performed with a mono- 2 (moderate), or 3 (marked). Proportional score of stained tumor clonal antibody that recognizes phospho-serine or phospho- cells was classified into 0 (<5%), 1 (5%–25%), 2 (26%–50%), threonine, followed by a proline (pS/TP; Supplementary Fig. 3 (51%–75%), and 4 (76%–100%). Expression score was S1C). We also performed in vitro kinase assays using recombi- calculated by multiplying the intensity score by the proportional nant active p38 with a synthetic peptide (residues 96- score (0–12; Supplementary Table S2). An expression score higher SGKGSQPPSPPSPAPSSFSS-115 of Snail) in the absence of than 4 was considered high expression, if not low expression. radioactive isotope. Analysis by mass spectrometry revealed clearly phosphorylation at Ser107 of Snail by p38 kinase Statistical analysis (Supplementary Fig. S1D). Finally, to recapitulate p38-medi- Quantitative data in this study are presented as means SD ated Snail phosphorylation in vivo, we cotransfected WT-Snail and were analyzed by a 2-tailed unpaired Student t test to or S107A-Snail with p38 and MKK6 (an upstream kinase compare the difference between groups. P < 0.05 was consid- required for p38 activation) and found that WT-Snail, but not ered statistically significant. For quantification of protein sta- S107A-Snail, was phosphorylated in HEK293T cells with acti- bility following treatment of CHX, Snail and a-tubulin pro- vated p38 (Fig. 1F). The p38-binding ability of S107A-Snail was teins detected by immunoblotting were quantified using Ima- not different from that of WT-Snail (Supplementary Fig. S1E). geJ software. For normalization, a-tubulin expression was Collectively, these results suggest that p38 directly interacts used as a control. GraphPad Prism version 7 and SPSS version with Snail and phosphorylates residue Ser107.

4138 Cancer Res; 79(16) August 15, 2019 Cancer Research

p38 Regulates Snail Protein Stability

Figure 1. p38 interacts with Snail and phosphorylates Ser107 residue. A, Coimmunoprecipitation assay in HEK293T cells cotransfected with HA-Snail and Flag- p38 plasmids. B, IP assay in SW620 and HCT116 cells with antibodies against p38 or Snail. C, Purified recombinant p38 protein was incubated with purified MBP–Snail fusion protein, after which, an MBP pull-down assay was performed. D, Conservation of Snail serine residues at amino acids 82, 104, and 107 in diverse species. E, In vitro kinase assays were performed by incubating purified recombinant active p38 protein with purified MBP-fused WT-Snail or Snail serine mutants (S82A, S104A, or S107A) in the presence of [g-32P]-ATP. The resultant products were subjected to SDS- PAGE and autoradiography. F, Flag-WT-Snail or Flag-S107A-Snail was cotransfected with plasmids expressing p38 and MKK6 into HEK293T cells, and then the cells were treated with 10 mmol/L MG132 for 12 hours. Cell lysates were immunoprecipitated using an anti-pS/TP antibody and then analyzed by immunoblotting using an anti-Flag antibody.

p38 enhances Snail protein stability by suppressing expressed Snail was also increased by cotransfection of the p38 ubiquitination-dependent Snail degradation expression vector and enhanced further still by transiently Because the phosphorylation status of Snail is important for overexpressing both p38 and MKK6 in HEK293T cells regulating its stability (17–19), we next investigated whether (Fig. 2D). Importantly, inhibiting proteasome function with p38 could affect Snail protein stability. In several cancer cell MG132 markedly enhanced exogenous Snail expression to the lines that we tested, the activation level of p38 (phosphoryla- same level irrespective of p38 activation (Fig. 2D), suggesting tion of p38 kinase at Thr180/Tyr182) correlated positively with that p38 activation could enhanceSnailproteinexpressionby the Snail expression level (Supplementary Fig. S2A). This pos- inhibiting proteasome-dependent Snail degradation. Perform- itive relationship between p38 activity and Snail protein expres- ing a CHX pulse-chase analysis demonstrated that p38 activa- sion prompted us to study the ability of p38 to upregulate Snail tion markedly extended the half-life of the WT-Snail protein stability. We noticed a significant increase in HA-Snail expres- (from 60 to 120 minutes, Fig. 2E). sion when cotransfected with Flag-p38 in HEK293T cells To explore the effects of Snail phosphorylation at Ser107 more (Fig. 1A, right). Activation of p38 markedly enhanced endog- precisely, we transiently overexpressed WT- and S107A-Snail in enous Snail protein levels in HEK293T, SKOV3, and HeLa cells HEK293T, SKOV3, and HeLa cells to compare their expression (Fig. 2A) without affecting their mRNA-expression levels (Sup- levels. As shown in Supplementary Fig. S3A, S107A-Snail showed plementary Fig. S2B). This effect depends on p38 kinase activ- much lower expression when compared with WT-Snail. However, ity, as cotransfection of a kinase-dead (KD) p38 form with the S107A-Snail and WT-Snail expression levels were increased to MKK6 did not affect the Snail level (Fig. 2A). RNAi-mediated the same level in response to MG132, indicating that S107A-Snail depletion of p38 or inhibition of p38 activity using a p38- is more susceptible to proteasome-dependent degradation com- specific inhibitor (SB203580) resulted in decreased endoge- pared with WT-Snail (Supplementary Fig. S3A). In addition, nous Snail protein level in SW620, HCT116, and 4T1 cells, S107A-Snail degraded significantly faster than WT-Snail in the which have high expression of phospho-p38 and Snail proteins presence of CHX (protein half-lives in HEK293T cells: WT-Snail, (Fig.2BandC),withoutaffecting the level of Snail mRNA 60 minutes; S107A-Snail, 30 minutes; Supplementary Fig. S3B), (Supplementary Fig. S2C and S2D). The level of exogenously and p38 activation could not extend the half-life of the

www.aacrjournals.org Cancer Res; 79(16) August 15, 2019 4139

Ryu et al.

Figure 2. p38 enhances Snail protein stability by suppressing ubiquitination-dependent Snail degradation. A, The WT or KD form of p38 was cotransfected with MKK6 into HEK293T, SKOV3, or HeLa cells. Cell lysates were immunoblotted with the indicated antibodies. B, Immunoblot analysis in p38-depleted SW620, HCT116, or 4T1 cells. C, Immunoblot analysis in SB203580 (SB)-treated SW620, HCT116, or 4T1 cells. D, HA-Snail was cotransfected with plasmids expressing p38 or p38 and MKK6 into HEK293T cells (top left), and then treated with 10 mmol/L MG132 for 12 hours (bottom left). Cell lysates were immunoblotted with the indicated antibodies. The data are representative of three independent experiments and relative Snail levels were quantified using ImageJ software (right). For normalization, a-tubulin expression was used as a control. E, HA-WT-Snail or HA-S107A-Snail was cotransfected with plasmids expressing p38 and MKK6 into HEK293T cells in the presence of CHX (100 mg/mL) for the indicated times. Cell lysates were immunoblotted by antibodies as indicated (top). The data were quantified using ImageJ software (bottom). For normalization, a-tubulin expression was used as a control. F, HA-WT-Snail or HA-S107A-Snail was cotransfected with plasmids expressing the WT or KD form of p38 and MKK6 into HEK293T cells, and then the cells were treated with 10 mmol/L MG132 for 12 hours. Cell lysates were immunoblotted with the indicated antibodies (top). The data are representative of three independent experiments and relative Snail levels were quantified using ImageJ software (bottom). For normalization, a-tubulin expression was used as a control. G, GFP-WT-Snail or GFP-S107A-Snail was cotransfected with plasmids expressing HA-ubiquitin, p38, and MKK6 as indicated into HEK293T cells, and then the cells were treated with 10 mmol/L MG132 for 12 hours. Cell lysates were immunoprecipitated using an anti-GFP antibody and then analyzed by immunoblotting using an anti-HA tag antibody.

4140 Cancer Res; 79(16) August 15, 2019 Cancer Research

p38 Regulates Snail Protein Stability

S107A-Snail (Fig. 2E). We also found that the degree of ubiqui- that DYRK2 markedly reduced WT-Snail expression, but this tination of S107A-Snail was significantly increased compared phenomenon was not observed in the presence of active p38 with that of WT-Snail (Supplementary Fig. S3C). Furthermore, (Fig. 3J). In contrast, the expression of S107A-Snail was p38 activation significantly increased the WT-Snail protein levels evidently reduced by DYRK2 but could not be recovered by and decreased ubiquitination, but the expression and ubiquitina- active p38 expression (Fig. 3J). All of these results suggest that tion of S107A-Snail did not change significantly (Fig. 2F and G). Ser107 phosphorylation of Snail by p38 directly inhibits Taken together, these results suggest that p38-mediated Ser107 DYRK2-mediated Ser104 phosphorylation, which is critical phosphorylation leads to Snail stabilization by suppressing for GSK3b-bTrCP–induced Snail degradation, resulting in ubiquitination-dependent Snail degradation. enhanced Snail expression. It has been known that GSK3b could be inactivated by phos- Phosphorylation of Snail at Ser107 by p38 inhibits DYRK2- phorylation at Thr390 (human)/Ser389 (mouse) by p38 mediated Snail phosphorylation at Ser104, which acts as a kinase (31). To verify whether p38 could directly phosphorylate prime phosphorylation for GSK3b and inactivate GSK3b, we checked the phosphorylation level of Snail can undergo bTrCP-mediated ubiquitination and pro- GSK3b and the expression level of the target protein for GSK3b teasomal degradation after GSK3b-dependent phosphorylation when p38 is activated in HEK293T cells, but could not find any at Ser96/100 (12–14). Considering that p38-mediated phos- difference (Supplementary Fig. S4). These results suggest that p38 phorylation of Snail at Ser107 decreased Snail ubiquitination does not affect the activity of GSK3b under our experimental (Fig. 2G) and that S107A-Snail showed increased ubiquitina- condition. tion compared with WT-Snail (Supplementary Fig. S3C), we next investigated whether p38 could impair the interaction Ser107 phosphorylation is required for Snail-promoted EMT between GSK3b and Snail, which is prerequisite for bTrCP- and metastasis in the SKOV3 ovarian cancer cell line mediated Snail ubiquitination. We found that WT-Snail bound To understand the role of Ser107 phosphorylation in Snail less to GSK3b than S107A-Snail (Fig. 3A) and that p38 activa- function, we next examined whether this phosphorylation tion markedly reduced GSK3b binding to WT-Snail, although might contribute to Snail-promoted EMT. For this purpose, S107A-Snail showed stronger binding regardless of p38 acti- we used SKOV3 human ovarian epithelial cancer cells because vation (Fig. 3B). In many cases, the association between GSK3b exogenously expressed WT-Snail is more stable in this cell line and its substrates requires the substrate to be phosphorylated at and has a longer protein half-life, compared with S107A-Snail a priming site (28, 29), and in case of Snail, GSK3b can (Supplementary Fig. S3A and S3B). Interestingly, among 3 phosphorylate Snail at Ser96/100 only if Ser104 has already SKOV3 cell lines expressing Flag-tag (Con), WT-Snail, or been phosphorylated by DYRK2 (30). We also found that the S107A-Snail, the morphology of WT-Snail-expressing cells was degree of binding of WT-Snail to GSK3b was strikingly distinct from that of the control and S107A-Snail–expressing increased by DYRK2, but decreased when p38 was activated cells. Confocal microscopy analysis of phalloidin-stained cells (Fig.3C).However,theinteractionbetweenSnailandDYRK2 confirmedthepresenceoffilopodia, lamellipodia, and micro- was maintained even if the Snailproteinwasphosphorylated spikes in WT-Snail–expressing SKOV3 cells, whereas control by p38 kinase (Fig. 3D). These results suggest that p38- cells and S107A-Snail–expressing cells exhibited less staining mediated Ser107 phosphorylation of Snail likely inhibits the and no cellular protrusions (Fig. 4A). These results prompted us GSK3b–Snail interaction by preventing DYRK2-mediated to measure the expression levels of EMT marker genes in these prime phosphorylation of Snail at Ser104. cells. WT-Snail protein was readily detected, whereas the Because no site-specific phospho-antibody for Ser104 is avail- S107A-Snail protein was barely detected because of its insta- able, to examine whether Ser107 phosphorylation of Snail by bility, even though both cell lines expressed comparable p38 could directly inhibit DYRK2-mediated Ser104 phosphor- amounts of Snail protein in the presence of MG132 ylation, we first performed an in vitro kinase assay with active (Fig. 4B). We also found that depletion of p38 expression p38 and WT-Snail in the presence of unlabeled cold ATP and significantly reduced WT-Snail but did not alter S107A-Snail then added a p38-specific inhibitor and DYRK2 with [g-32P]-ATP expression (Supplementary Fig. S5A), in contrast depletion of (Fig. 3E). DYRK2 phosphorylated Snail at residue Ser104 DYRK2 expression could only increase S107A-Snail expression (Fig. 3F), as also shown previously (30). The p38 inhibitor (Supplementary Fig. S5B). These results suggest that the expres- effectively suppressed its auto-phosphorylation, as well as Snail sion level of S107A-Snail protein is reduced by DYRK2- phosphorylation by p38 kinase (Fig. 3G). Although DYRK2- mediated phosphorylation of Snail and p38 can reverse this mediated Snail phosphorylation was not inhibited by the p38 effect on WT-Snail. As expected, WT-Snail expression inhibited inhibitor, pre-incubation with p38 completely reduced DYRK2- the expression of epithelial markers (including E-cadherin mediated Snail phosphorylation (Fig. 3H). These results, along and Occludin), but promoted the expression of mesenchymal with the results of Fig. 3D, suggest that the Snail protein markers (including N-cadherin, fibronectin, and vimentin) in phosphorylated by p38 kinase can interact with DYRK2 but not SKOV3 cells (Fig. 4B). However, S107A-Snail expression failed phosphorylated by DYRK2. Furthermore, the results of docking to regulate these markers and only inhibited E-cadherin expres- simulations using DYRK2 (PDB ID: 3K2L) with a peptide sion (Fig. 4B). These results suggest that Ser107 phosphoryla- (residues 100-SQPPSPPSP-108 of Snail) carrying either unpho- tion is required for Snail-dependent EMT promotion. Next, we sphorylated or phosphorylated Ser107 predicted that the investigated whether Snail expression could alter the migratory unphosphorylated peptide binds closely to the active site of properties of SKOV3 cells by conducting a wound-healing DYRK2, but the phospho-Ser107–containing peptide cannot assay. After 16 hours, we found that when compared with access the active site of DYRK2 due to steric hindrance by control and S107A-Snail–expressing cells, an increased number phosphorylated Ser107 residue (Fig. 3I). Finally, we found of WT-Snail–expressing SKOV3 cells had migrated into the

www.aacrjournals.org Cancer Res; 79(16) August 15, 2019 4141

Ryu et al.

Figure 3. Phosphorylation of Snail at Ser107 by p38 inhibits DYRK2-mediated Snail phosphorylation at Ser104, which acts as a prime phosphorylation for GSK3b. A, Flag-WT-Snail or Flag-S107A-Snail was cotransfected with plasmids expressing HA-GSK3b into HEK293T cells, and then the cells were treated with 10 mmol/L MG132 for 12 hours. Cell lysates were immunoprecipitated using an anti-HA antibody and then analyzed by immunoblotting using an anti-Flag antibody. B, Flag-WT-Snail or Flag-S107A-Snail was cotransfected with plasmids expressing HA-GSK3b, p38, and MKK6 (as indicated) into HEK293T cells, and then the cells were treated with 10 mmol/L MG132 for 12 hours. Cell lysates were immunoprecipitated using an anti-HA antibody and then analyzed by immunoblotting using an anti-Flag antibody. C, GFP-WT-Snail was cotransfected with plasmids expressing HA-GSK3b, Flag-DYRK2, p38, and MKK6 (as indicated) into HEK293T cells, and then the cells were treated with 10 mmol/L MG132 for 12 hours. Cell lysates were immunoprecipitated using an anti-HA antibody and then analyzed by immunoblotting with an anti-GFP antibody. (Continued on the following page.)

4142 Cancer Res; 79(16) August 15, 2019 Cancer Research

p38 Regulates Snail Protein Stability

scratch wound (Fig. 4C). Similarly, in an in vitro invasion assay, p38 activity suppressed this effect, whereas S107A-Snail protein we detected a meaningful increase in the number of invasive levels were not changed regardless of EGF treatment (Fig. 5A). WT-Snail–expressing SKOV3 cells, relative to control and These results suggest that activation of EGF signaling pathways S107A-Snail–expressing cells (Fig. 4D). In addition, WT- stabilizes Snail protein expression by inducing p38-mediated Snail–expressing SKOV3 cells showed a slightly decreased Snail phosphorylation at Ser107. We next investigated the growth rate under the same growth conditions (Fig. 4E), indi- role of Snail Ser107 phosphorylation in EGF-induced cell cating that the increased migration and invasion observed with migration and invasion. The migration and invasion abilities WT-Snail expression is independent of the growth rate. Fur- of EGF-treated S107A-Snail-expressing cells were not signifi- thermore, depletion of p38 expression significantly inhibited cantly different when compared with control cells, whereas the migration ability of WT-Snail–expressing cells, but not that WT-Snail–expressing cells showed robustly increased migration of S107A-Snail–expressing cells, in contrast depletion of and invasion abilities (Fig. 5B and C). Together, these results DYRK2 expression could only increase the migration ability suggest that Ser107 phosphorylation play an important role in of S107A-Snail–expressing cells (Supplementary Fig. S5C). mediating EGF-induced and Snail-promoted cancer cell migra- These results suggest that the migration ability of S107A- tion and invasion. Snail–expressing SKOV3 cells is inhibited by DYRK2-mediated phosphorylation of Snail and p38 can reverse this effect on Activation of p38 is the main cause of increased Snail WT-Snail. expression in patients with human ovarian cancer We also investigated whether Ser107 phosphorylation in Snail expression has been shown to be highly correlated with Snail is critical for the metastasis of ovarian cancer. Ovarian both the tumor stage and metastatic potential in human ovar- cancer primarily metastasizes to the peritoneum and rarely ian cancer (36–38). To verify whether p38 activation plays an migrates to distant sites. The resulting peritoneal implants are important role in increasing the Snail expression in patients characterized by the adhesion and invasion of tumor cells into with ovarian cancer, we performed tissue microarray analysis of the peritoneum, leading to miliary dissemination (32). There- 39 ovarian cancer tissue specimens and observed a strong fore, to analyze the in vivo roles of Ser107 phosphorylation of positive correlation between p38 activity (i.e., the phospho- Snail in the peritoneal dissemination of ovarian cancer cells, p38 level) and the Snail protein expression. We found that 86% female BALB/c nude mice were injected intraperitoneally of the patient samples with high Snail expression also showed with control SKOV3 (Con), WT-Snail–expressing SKOV3 high p-p38 levels and that all patient samples with low Snail (WT-Snail), or S107A-Snail–expressing SKOV3 (S107A-Snail) expression showed low p-p38 levels (Fig. 6A). However, we did cells, after which, the tumors were allowed to grow for 5 weeks. not find any inverse correlation between DYRK2 and Snail After 5 weeks, the peritoneal surface tumor burden was expression.DYRK2expressionwashighin9of11patients significantly higher (P < 0.05) in mice that had received (82%) with low Snail expression and in 26 of 28 patients SKOV3/WT-Snail cells (1.22 0.17 g) versus control cells (93%) with high Snail expression (Fig. 6A). Interestingly, we (0.53 0.12 g), but not in mice administered SKOV3/ observed that even though DYRK2 was highly expressed, all S107A-Snail cells (0.31 0.08 g; Fig. 4F and G). These results patient samples with high p-p38 levels showed high Snail indicate that Ser107 phosphorylation of Snail is critical for expression and 69% of patient samples with low p-p38 levels enhancing the metastatic potential of ovarian cancer cells. showed low Snail expression (Fig. 6B). We also found that in EGF can activate MAPK pathways (such as p38 MAPK) and patients with high DYRK2 expression, the level of Snail expres- regulate gene expression to promote EMT and cancer sion was higher in patients with high p-p38 levels than in metastasis (33–35). To examine whether EGF can enhance patients with low p-p38 levels (Fig. 6C; Supplementary Table Snail stability and whether p38-mediated Ser107 phosphory- S2). All of these results indicate that activated p38 can enhance lation is important for EGF-induced increased Snail stability, Snail protein levels in patients with ovarian cancer by suppres- we treated SKOV3 cell lines stably expressing either WT-Snail or sing DYRK2-mediated negative regulation of Snail, which is S107A-Snail with EGF. Under the serum-starved condition, we consistent with the results we have shown earlier. observed similar, low-level basal expression levels of WT-Snail In conclusion, our study reveals a functionally important and S107A-Snail in each cell line; however, EGF treatment posttranslational control mechanism for the EMT master regula- significantly increased WT-Snail protein levels and inhibiting tor Snail. Our results suggest a model in which DYRK2 and GSK3b

(Continued.) D, Flag-DYRK2 was cotransfected with plasmids expressing HA-Snail, p38, and MKK6 (as indicated) into HEK293T cells, and then the cells were treated with 10 mmol/L MG132 for 12 hours. Cell lysates were immunoprecipitated using an anti-HA antibody and then analyzed by immunoblotting with an anti-Flag antibody. E, Schematic diagram for in vitro kinase assays in F to H. F, In vitro kinase assays were performed by incubating purified recombinant active DYRK2 protein with purified MBP-fused WT-Snail or S107A-Snail in the presence of [g-32P]-ATP. The resultant products were subjected to SDS-PAGE and autoradiography. G, In vitro kinase assays were performed by incubating purified recombinant active p38 protein with purified MBP-fused WT-Snail in the presence or absence of SB203580. The resultant products were subjected to SDS-PAGE and autoradiography. H, In vitro kinase assays were performed by incubating purified recombinant active p38 protein with purified MBP-fused WT-Snail in the presence of unlabeled ATP, after which, SB203580 and DYRK2 were added, and the resulting mixtures were incubated for 30 minutes in the presence of [g-32P]- ATP. The resultant products were subjected to SDS-PAGE and autoradiography. I, Docking simulations were done using AUTODOCK 4.2. Red and blue shading correspond to negatively and positively charged regions, respectively, on the surface of DYRK2. The peptides (residues 100-SQPPSPPSP-108 of Snail) bound to the DYRK2 are represented in stick format. The unphosphorylated peptide is represented with carbons in white, nitrogens in blue, and oxygens in red. The phosphorylated peptide is represented with carbons in orange, nitrogens in blue, and oxygens in red. The active site is marked with a dashed white circle. J, HA-WT-Snail or HA-S107A-Snail was cotransfected with plasmids expressing DYRK2 or DYRK2, p38, and MKK6 into HEK293T cells, and then the cells were treated with 10 mmol/L MG132 for 12 hours. Cell lysates were immunoblotted with the indicated antibodies (left). The data are representative of three independent experiments and relative Snail levels were quantified using ImageJ software (right). For normalization, a-tubulin expression was used as a control.

www.aacrjournals.org Cancer Res; 79(16) August 15, 2019 4143

Ryu et al.

Figure 4. Ser107 phosphorylation is required for Snail-promoted EMT and metastasis in the SKOV3 ovarian cancer cell line. A, Morphologic changes of SKOV3 cell lines expressing Flag-tag (Con), WT-Snail, or S107A-Snail were visualized by confocal microscopy after staining with TRITC-conjugated phalloidin. B, Expression levels of EMT marker proteins or mRNAs in SKOV3 cell lines expressing Flag-tag (Con), WT-Snail, or S107A-Snail were analyzed by immunoblotting or RT-PCR analysis. C, Flag-tag- (Con), WT-Snail-, or S107A-Snail-expressing SKOV3 cells were analyzed in wound-healing assays by visualizing wound closure via phase- contrast microscopy (top). Wound areas were measured using WimScratch software (Wimasis). The data shown represent the percentage of the wound area and are expressed as the means SD of three individual experiments (bottom). , P < 0.05 as determined by t test. ns, not signiﬁcant. D, Flag-tag- (Con), WT- Snail-, or S107A-Snail-expressing SKOV3 cells were seeded onto Matrigel matrix-coated top chambers, and the fold-changes of invading cells were measured after 24 hours. The data shown are expressed as the means SD of three individual experiments, each performed in triplicate. , P < 0.01 as determined by t test. E, The indicated cells were seeded in a 6-well plate at a concentration of 2 104 cells per well. After incubation for 1 to 4 days, the viable cells were counted with a hemocytometer after trypan blue staining. F, In vivo roles of Snail phosphorylation at residue Ser107. Flag-tag- (Con), WT-Snail-, or S107A-Snail-expressing SKOV3 cells (2 106) were suspended in 200 mL PBS and intraperitoneally injected into BALB/c nude mice (6 mice/group). After 5 weeks, tumor burdens and ascites formation were estimated. Arrows, disseminated tumors. G, At autopsy, tumors were excised and weighed. The data are shown as the means SD of 6 mice/group. Statistical signiﬁcances were determined by t test.

are strongly activated in normal or early-stage tumor cells, where activated in malignant tumor cells and activated p38 can suppress they effectively degrade Snail through bTrCP-mediated ubiquiti- DYRK2-mediated prime phosphorylation required for GSK3b- nation, thus suppressing EMT. In contrast, p38 MAPK is highly bTrCP–mediated Snail ubiquitination and degradation, where

4144 Cancer Res; 79(16) August 15, 2019 Cancer Research

p38 Regulates Snail Protein Stability

Figure 5. Ser107 phosphorylation is required for EGF-induced ovarian cancer cell invasion. A, WT-Snail or S107A-Snail-expressing SKOV3 cells were treated with EGF in the presence or absence of SB203580. Cell lysates were immunoblotted with antibody against Snail (top). Relative Snail levels were quantiﬁed using ImageJ software (bottom). For normalization, a-tubulin expression was used as a control. B, Flag-tag- (Con), WT-Snail-, or S107A- Snail-expressing SKOV3 cells were treated with EGF and analyzed in a wound-healing assays to visualize wound closure by phase-contrast microscopy (top). Wound areas were measured by using the WimScratch software (Wimasis). The data shown represent the percentage of the wound area and are expressed as the means SD of three individual experiments (bottom). C, Flag-tag- (Con), WT-Snail-, or S107A-Snail-expressing SKOV3 cells were seeded onto Matrigel matrix-coated top chambers and treated with EGF, and fold-changes in the numbers of invading cells were measured after 24 hours. The data are shown as the mean SD of three individual experiments performed in triplicate. , P < 0.01; , P < 0.001, as determined by t test. ns, not signiﬁcant.

the resulting high levels of Snail expression can promote EMT and tion (15). Although the protein machinery used for nuclear tumor metastasis (Fig. 6D). Snail import is not precisely known, Snail phosphorylation by PAK1 probably increases the interaction of Snail with the protein that is important for its nuclear trafﬁcking. A second Discussion way of increasing Snail stability might be to allow it to stay in Similar to other signaling proteins, the activities of transcrip- the nucleus longer. Lats2 phosphorylates Snail at Thr203 in the tion factors are commonly regulated by phosphorylation in nucleus and prevents its nuclear export, thereby leading to its response to various cellular signals. In the case of Snail, phos- stabilization (16). Although the precise molecular mechanism phorylation can control the protein stability and function in explaining how Snail phosphorylation at Thr203 causes nuclear various ways. Because Snail is known to be negatively regulated retention is unclear, Lats2-mediated Thr203 phosphorylation by ubiquitination-dependent proteasomal degradation in the might alter Snail binding with the nuclear export machinery or cytoplasm, one way to positively increase Snail stability might with some resident nuclear proteins. A third way might be to be to promote its delivery into the nucleus. PAK1 phosphor- prevent Snail ubiquitination. Phosphorylation of Snail at ylates Snail at Ser246, which is located in the zinc ﬁnger Ser100 by ATM and DNA-PKcs decreases Snail ubiquitination domain that is critical for its nuclear localization (39), which or reduces Snail interaction with GSK3b, which is critical for was found to enhance its protein-expression level and func- Snail degradation, thus enhances its protein stability (17, 19).

www.aacrjournals.org Cancer Res; 79(16) August 15, 2019 4145

Ryu et al.

Figure 6. The Snail protein expression level correlates positively with the phospho-p38 protein level in patients with ovarian cancer. A, Left, representative IHC images of phospho-p38 and Snail in ovarian tumors. Scale bars, 50 mm. Right, correlation of the Snail expression level with the phospho-p38 or DYRK2 levels, as determined using an ovarian tumor tissue microarray. Statistical significances were determined by Pearson x2 test and Fisher exact test. B, Top, representative IHC images of DYRK2, phospho-p38, and Snail in ovarian tumors with high DYRK2 expression. Scale bars, 50 mm. Bottom, correlation of the Snail expression level with phospho-p38 levels in patients with ovarian cancer who had high DYRK2 expression. Statistical significances were determined by Pearson x2 test and Fisher exact test. C, Scatter dot plots comparing Snail and phospho-p38 levels in patients with ovarian cancer who had high DYRK2 expression. Statistical significances were determined by t test. D, Proposed model to illustrate how stabilization of Snail by p38 may lead to EMT.

4146 Cancer Res; 79(16) August 15, 2019 Cancer Research

p38 Regulates Snail Protein Stability

Erk2-mediated Ser82/Ser104 phosphorylation of Snail also In this study, we showed the precise molecular mechanism by protected it from ubiquitination and subsequent proteasomal which p38 promotes the EMT and metastasis of tumor cells by degradation (18). However, none of these reports explained regulating the stability of the Snail. Our study not only reveals a exactly how phosphorylation of Snail inhibits ubiquitination. critical mechanism underlying p38-induced cancer metastasis, In this study, we showed that the precise molecular mechanism but also has important implications in the development of by which phosphorylation can inhibit Snail ubiquitination and treatment strategies for metastatic cancers. thereby enhance its stability: p38 MAPK enhances Snail stability by suppressing the DYRK2-mediated prime phosphorylation for Disclosure of Potential Conflicts of Interest GSK3b, which is critical for bTrCP-mediated Snail ubiquitination No potential conflicts of interest were disclosed. and subsequent degradation. Recent studies have suggested that GSK3b-dependent phosphorylation of Snail is also crucial for SPSB3- and FBW7-mediated Snail ubiquitination and degrada- Authors' Contributions tion (40, 41). These results suggest that p38-mediated Snail Conception and design: K.-J. Ryu, S.-M. Park, I.-K. Kim, J.I. Yook, J. Yoo phosphorylation may also inhibit ubiquitination of Snail by Development of methodology: K.-J. Ryu, S.-H. Park, K.D. Kim Acquisition of data (provided animals, acquired and managed patients, these E3 ubiquitin ligases. provided facilities, etc.): K.-J. Ryu, S.-M. Park, S.-H. Park, I.-K. Kim, fi Besides phosphorylation, other posttranslational modi ca- H. Han, H.-J. Kim, S.-H. Kim, K.-S. Hong, H. Kim, M. Kim, S.-J. Yoon, tions also can control Snail stability. For example, lysyl- J.-Y. Park, H.J. Cho, J.W. Yang, G.-H. Ko, Y.-J. Park oxidase-like 2 (LOXL2) modifies Snail at Lys98/Lys137 and leads Analysis and interpretation of data (e.g., statistical analysis, biostatistics, to increased protein stability by preventing Snail ubiquitination computational analysis): S.-M. Park, S.-H. Park, I.-K. Kim, K. Asaithambi, and its interaction with GSK3b (42). In addition, O-GlcNAc K.H. Lee, G. Lee, Y.J. Kang, J. Yoo fi Writing, review, and/or revision of the manuscript: K.-J. Ryu, J. Yoo modi cation of Snail at Ser112 increases Snail protein stability Administrative, technical, or material support (i.e., reporting or organizing by blocking Snail ubiquitination and its interaction with data, constructing databases): Y.-S. Hah, C. Hwangbo GSK3b (43). Recently, it was shown that A20-mediated multi- Study supervision: J. Yoo monoubiquitination of Snail at Lys206/Lys234/Lys235 can fi increase Snail protein stability by reducing the binding af nity Acknowledgments of Snail for GSK3b (44). Interestingly, most posttranslational This work was supported by the National Research Foundation of Korea fi modi cations that can block Snail ubiquitination (phosphoryla- (NRF) grants funded by the Ministry of Education (NRF-2013R1A1A2057753 tion at Ser100 by DNA-PKcs and ATM, phosphorylation at Ser104 to J. Yoo), the Korea government (MSIP; NRF-2017R1A2B4003185 to J. Yoo by Erk2, lysyl-oxidation at Lys98, and O-GlcNAc modification at and NRF-2018M3A9H3023077 to Y.-J. Park), and grants from the KRIBB Ser112) occur near the phosphorylation sites for DYRK2 and Research Initiative Program, Republic of Korea. GSK3b. These findings, combined with the results of this study, suggest that additional studies should be performed to determine The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked whether these various modifications could inhibit DYRK2-medi- advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate ated prime phosphorylation for GSK3b. this fact. p38 MAPK has been reported to contribute to the EMT of cells in the primary tumor and to the acquisition of invasion and Received January 4, 2019; revised May 7, 2019; accepted June 11, 2019; migration capabilities in several types of cancers (25–27, 45, 46). published first June 17, 2019.

References 1. Coleman RL, Monk BJ, Sood AK, Herzog TJ. Latest research and treatment of 9. Grotegut S, von Schweinitz D, Christofori G, Lehembre F. Hepatocyte advanced-stage epithelial ovarian cancer. Nat Rev Clin Oncol 2013;10: growth factor induces cell scattering through MAPK/Egr-1-mediated upre- 211–24. gulation of Snail. EMBO J 2006;25:3534–45. 2. Burger RA, Brady MF, Bookman MA, Fleming GF, Monk BJ, Huang H, et al. 10. Dong R, Wang Q, He XL, Chu YK, Lu JG, Ma QJ. Role of nuclear factor kappa Incorporation of bevacizumab in the primary treatment of ovarian cancer. B and reactive oxygen species in the tumor necrosis factor-a-induced N Engl J Med 2011;365:2473–83. epithelial-mesenchymal transition of MCF-7 cells. Braz J Med Biol Res 3. Wan L, Pantel K, Kang Y. Tumor metastasis: moving new biological insights 2007;40:1071–8. into the clinic. Nat Med 2013;19:1450–64. 11. Brandl M, Seidler B, Haller F, Adamski J, Schmid RM, Saur D, et al. 4. Ikenouchi J, Matsuda M, Furuse M, Tsukia S. Regulation of tight junctions IKK(a) controls canonical TGF(ß)-SMAD signaling to regulate genes during the epithelium-mesenchyme transition: direct repression of the expressing SNAIL and SLUG during EMT in panc1 cells. J Cell Sci 2010; gene expression of claudins/occludin by Snail. J Cell Sci 2003;116: 123:4231–9. 1959–67. 12. Zhou BP, Deng J, Xia W, Xu J, Li YM, Gunduz M, et al. Dual regulation of 5. Nieto MA, Huang RY, Jackson RA, Thiery JP. EMT: 2016. Cell 2016;166: Snail by GSK-3beta-mediated phosphorylation in control of epithelial- 21–45. mesenchymal transition. Nat Cell Biol 2004;6:931–40. 6. Chaffer CL, San Juan BP, Lim E, Weinberg RA. EMT, cell plasticity and 13. Yook JI, Li XY, Ota I, Fearon ER, Weiss SJ. Wnt dependent regulation of the metastasis. Cancer Metastasis Rev 2016;35:645–54. E-cadherin repressor Snail. J Biol Chem 2005;280:11740–8. 7. Cano A, Perez-Moreno MA, Rodrigo I, Locascio A, Blanco MJ, del Barrio 14. Yook JI, Li XY, Ota I, Hu C, Kim HS, Kim NH, et al. A Wnt-Axin2-GSK3beta MG, et al. The transcription factor snail controls epithelial-mesenchymal cascade regulates Snail1 activity in breast cancer cells. Nat Cell Biol 2006;8: transitions by repressing E-cadherin expression. Nat Cell Biol 2000;2: 1398–406. 76–83. 15. Yang Z, Rayala S, Nguyen D, Vadlamudi RK, Chen S, Kumar R. Pak1 8. Batlle E, Sancho E, Franci C, Dominguez D, Monfar M, Baulida J, et al. The phosphorylation of snail, a master regulator of epithelial-to-mesenchyme transcription factor snail is a repressor of E-cadherin gene expression in transition, modulates snail's subcellular localization and functions. epithelial tumour cells. Nat Cell Biol 2000;2:84–9. Cancer Res 2005;65:3179–84.

www.aacrjournals.org Cancer Res; 79(16) August 15, 2019 4147

Ryu et al.

16. Zhang K, Rodriguez-Aznar E, Yabuta N, Owen RJ, Mingot JM, Nojima H, 32. Lengyel E. Ovarian cancer development and metastasis. Am J Pathol 2010; et al. Lats2 kinase potentiates Snail1 activity by promoting nuclear reten- 177:1053–64. tion upon phosphorylation. EMBO J 2012;31:29–43. 33. Cheng JC, Klausen C, Leung PC. Hydrogen peroxide mediates EGF-induced 17. Sun M, Guo X, Qian X, Wang H, Yang C, Brinkman KL, et al. Activation of down-regulation of E-cadherin expression via p38 MAPK and snail in the ATM-Snail pathway promotes breast cancer metastasis. J Mol Cell Biol human ovarian cancer cells. Mol Endocrinol 2010;24:1569–80. 2012;4:304–15. 34. Xu Z, Jiang Y, Steed H, Davidge S, Fu Y. TGFb and EGF synergistically induce 18. Zhang K, Corsa CA, Ponik SM, Prior JL, Piwnica-Worms D, Eliceiri KW, et al. a more invasive phenotype of epithelial ovarian cancer cells. The collagen receptor discoidin domain receptor 2 stabilizes SNAIL1 to Biochem Biophys Res Commun 2010;401:376–81. facilitate breast cancer metastasis. Nat Cell Biol 2013;15:677–87. 35. Bae GY, Choi SJ, Lee JS, Jo J, Lee J, Kim J, et al. Loss of E-cadherin activates 19. Pyun BJ, Seo HR, Lee HJ, Jin YB, Kim EJ, Kim NH, et al. Mutual regulation EGFR-MEK/ERK signaling, which promotes invasion via the ZEB1/MMP2 between DNA-PKcs and snail1 leads to increased genomic instability and axis in non-small cell lung cancer. Oncotarget 2013;4:2512–22. aggressive tumor characteristics. Cell Death Dis 2013;4:e517. 36. Blechschmidt K, Sassen S, Schmalfeldt B, Schuster T, Hofler H, Becker KF. 20. Morris GM, Huey R, Lindstrom W, Sanner MF, Belew RK, Goodsell DS, et al. The E-cadherin repressor Snail is associated with lower overall survival of AutoDock4 and AutoDockTools4: automated docking with selective recep- ovarian cancer patients. Br J Cancer 2008;98:489–95. tor flexibility. J Comput Chem 2009;30:2785–91. 37. Jin H, Yu Y, Zhang T, Zhou X, Zhou J, Jia L, et al. Snail is critical for tumor 21. Emsley P, Cowtan K. Coot: model-building tools for molecular graphics. growth and metastasis of ovarian carcinoma. Int J Cancer 2010;126: Acta Crystallogr D Biol Crystallogr 2004;66:486–501. 2102–11. 22. Mooers BH. Simplifying and enhancing the use of PyMOL with horizontal 38. Taki M, Abiko K, Baba T, Hamanishi J, Yamaguchi K, Murakami R, et al. scripts. Protein Sci 2016;25:1873–82. Snail promotes ovarian cancer progression by recruiting myeloid-derived 23. MacPherson MR, Molina P, Souchelnytskyi S, Wernstedt C, Martin-Perez J, suppressor cells via CXCR2 ligand upregulation. Nat Commun 2018;9: Portillo F, et al. Phosphorylation of serine 11 and serine 92 as new positive 1685. regulators of human Snail1 function: potential involvement of casein 39. Yamasaki H, Sekimoto T, Ohkubo T, Douchi T, Nagata Y, Ozawa M, et al. kinase-2 and the cAMP-activated kinase protein kinase A. Mol Biol Cell Zinc finger domain of Snail functions as a nuclear localization signal for 2010;21:244–53. importin beta-mediated nuclear import pathway. Genes Cells 2005;10: 24. Du C, Zhang C, Hassan S, Biswas MH, Balaji KC. Protein kinase D1 455–64. suppresses epithelial-to-mesenchymal transition through phosphoryla- 40. Liu Y, Zhou H, Zhu R, Ding F, Li Y, Cao X, et al. SPSB3 targets SNAIL for tion of snail. Cancer Res 2010;70:7810–9. degradation in GSK-3b phosphorylation-dependent manner and regulates 25. Hipp S, Berg D, Ergin B, Schuster T, Hapfelmeier A, Walch A, et al. metastasis. Oncogene 2018;37:768–76. Interaction of Snail and p38 mitogen-activated protein kinase results in 41.ZhangY,ZhangZ,YeM,JingP,XiongJ,HanZ,etal.FBW7loss shorter overall survival of ovarian cancer patients. Virchows Arch 2010; promotes epithelial-to-mesenchymal transition in non-small cell lung 457:705–13. cancer through the stabilization of Snail protein. Cancer Lett 2018;419: 26. Kolosova I, Nethery D, Kern JA. Role of Smad2/3 and p38 MAP kinase in 75–83. TGF-beta1-induced epithelial-mesenchymal transition of pulmonary epi- 42. Peinado H, Del Carmen Iglesias-de la Cruz M, Olmeda D, Csiszar K, Fong thelial cells. J Cell Physiol 2011;226:1248–54. KS, Vega S, et al. A molecular role for lysyl oxidase-like 2 enzyme in snail 27. Hong J, Zhou J, Fu J, He T, Qin J, Wang L, et al. Phosphorylation of serine 68 regulation and tumor progression. EMBO J 2005;24:3446–58. of Twist1 by MAPKs stabilizes Twist1 protein and promotes breast cancer 43. Park SY, Kim HS, Kim NH, Ji S, Cha SY, Kang JG, et al. Snail1 is stabilized by cell invasiveness. Cancer Res 2011;71:3980–90. O-GlcNAc modification in hyperglycaemic condition. EMBO J 2010;29: 28. Cohen P, Frame S. The renaissance of GSK3. Nat Rev Mol Cell Biol 2001;2: 3787–96. 769–76. 44. Lee JH, Jung SM, Yang KM, Bae E, Ahn SG, Park JS, et al. A20 promotes 29. Doble BW, Woodgett JR. GSK-3: tricks of the trade for a multi-tasking metastasis of aggressive basal-like breast cancers through multi- kinase. J Cell Sci 2003;116:1175–86. monoubiquitylation of Snail1. Nat Cell Biol 2017;19:1260–73. 30. Mimoto R, Taira N, Takahashi H, Yamaguchi T, Okabe M, Uchida K, et al. 45. Greenberg AK, Basu S, Hu J, Yie TA, Tchou-Wong KM, Rom WN, et al. DYRK2 controls the epithelial-mesenchymal transition in breast cancer by Selective p38 activation in human non-small cell lung cancer. Am J Respir degrading Snail. Cancer Lett 2013;339:214–25. Cell Mol Biol 2002;26:558–64. 31. Thornton TM, Pedraza-Alva G, Deng B, Wood CD, Aronshtam A, Clements 46. Iyoda K, Sasaki Y, Horimoto M, Toyama T, Yakushijin T, Sakakibara M, JL, et al. Phosphorylation by p38 MAPK as an alternative pathway for et al. Involvement of the p38 mitogen-activated protein kinase cascade in GSK3beta inactivation. Science 2008;320:667–70. hepatocellular carcinoma. Cancer 2003;97:3017–26.

4148 Cancer Res; 79(16) August 15, 2019 Cancer Research

p38 Stabilizes Snail by Suppressing DYRK2-Mediated Phosphorylation That Is Required for GSK3 β-βTrCP−Induced Snail Degradation

Ki-Jun Ryu, Sun-Mi Park, Seung-Ho Park, et al.

Cancer Res 2019;79:4135-4148. Published OnlineFirst June 17, 2019.

Updated version Access the most recent version of this article at: doi:10.1158/0008-5472.CAN-19-0049

Supplementary Access the most recent supplemental material at: Material http://cancerres.aacrjournals.org/content/suppl/2021/03/11/0008-5472.CAN-19-0049.DC1

Cited articles This article cites 46 articles, 13 of which you can access for free at: http://cancerres.aacrjournals.org/content/79/16/4135.full#ref-list-1

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/79/16/4135. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.