Actin-Binding Protein, Espin: a Novel Metastatic Regulator for Melanoma

Published OnlineFirst December 13, 2013; DOI: 10.1158/1541-7786.MCR-13-0468-T

Molecular Cancer Oncogenes and Tumor Suppressors Research

Actin-Binding Protein, Espin: A Novel Metastatic Regulator for Melanoma

Takeshi Yanagishita1,3, Ichiro Yajima1,5, Mayuko Kumasaka1,5, Yoshiyuki Kawamoto2, Toyonori Tsuzuki4, Yoshinari Matsumoto3, Daisuke Watanabe3, and Masashi Kato1,5

Abstract Espin is a multifunctional actin-bundling protein with multiple isoforms, and has special connections to hair cell stereocilia and microvillar specializations of sensory cells in the inner ear. However, there have been no reports showing the expression and function of Espin in cancers, including melanoma. Here, it is demonstrated that Espin expression is significantly increased in melanomas that spontaneously developed in RET-transgenic mice (RET- mice). Importantly, the invasion capacity of Espin-depleted Mel-ret melanoma cells derived from a tumor of the RET-mouse was dramatically less than that of control melanoma cells with reductions of lamellipodia, focal adhesion kinase (FAK), and GTP-Rac1 activities. Correspondingly, the ratio of metastatic foci in Espin-depleted Mel-ret melanoma cells was significantly less than that of control melanoma cells in an in vivo melanoma metastasis model. Moreover, Espin could be a novel biomarker of melanoma in humans, because our immunohistochemical analysis data reveal that percentages of Espin-positive cells in human primary and metastatic melanomas were significantly higher than that of cells in melanocytic nevi. Together, these results indicate that Espin is not only a metastatic regulator for melanoma but also a potential biomarker of disease progression. Implications: Actin-binding protein Espin is expressed in melanoma, affects metastasis, and is a potential target for melanoma therapy. Mol Cancer Res; 12(3); 440–6. 2013 AACR.

Introduction The ESPIN gene has various isoforms and encodes an fi – The incidence of melanoma has increased by 3.1% every actin- lament binding protein (Supplementary Fig. S1; refs. 4, 5). All isoforms of ESPIN encode the same region year and its incidence rate has doubled over a 10-year period – (1). Melanoma is the most serious skin cancer and is highly of a 116 amino acid actin-bundling module (ABM), which invasive and resistant to conventional therapy (2). Identifi- is necessary for potent actin-bundling and microvillar par- cation of a molecule associated with melanoma growth, allel actin bundle (PAB) elongating activities (4, 5). Muta- tion in Wiskott–Aldrich syndrome protein homology 2 progression, and metastasis will provide new insights into ESPIN the design of a biomarker and a novel therapeutic strategy for (WH2) and ABM domains of gene causes hereditary melanoma. Preventing or eliminating metastasis is one of the deafness and vestibular dysfunction accompanied by stereo- most important challenges for therapeutic intervention. ciliary shortening in jerker mice and humans (5, 6). Espins Several regulatory molecules for cancer invasion and metas- are also capable of affecting the actin cytoskeleton, resulting tasis have recently been reported as candidates for clinical in a special connection to microvillar specializations of applications (3). However, an effective therapy for melano- sensory cells such as taste receptor cells, solitary chemore- ma has not yet been established (2). ceptor cells, and Merkel cells (7). On the other hand, dynamic remodeling of the actin cytoskeleton is required for cancer invasion. To activate cancer invasion, remodeling Authors' Affiliations: Units of 1Environmental Health Sciences and of the actin cytoskeleton results in the formation of lamelli- 2Immunology, Department of Biomedical Sciences, College of Life and Health Sciences, Chubu University, Kasugai; 3Department of Dermatology, podia, sheets of F-actin, regulated by Rho family GTPases, Aichi Medical University School of Medicine, Nagakute; 4Department of including Rac1, whose activity is regulated by focal adhesion Pathology, Nagoya Daini Red Cross Hospital; and 5Department of Occu- kinase (FAK; ref. 8). These previous findings indicate the pational and Environmental Health, Nagoya University Graduate School of Medicine, Nagoya, Aichi, Japan possibility that Espin is functionally correlated with the invasion of cancer. To our knowledge, however, there is no Note: Supplementary data for this article are available at Molecular Cancer Research Online (http://mcr.aacrjournals.org/). information about the expression levels of Espin in cancer cells or its role in cancers including melanoma. Corresponding Author: Masashi Kato, Department of Occupational and RET Environmental Health, Nagoya University Graduate School of Medicine, 65 We previously established -transgenic mice of line Tsurumai-cho, Showa-ku, Nagoya, Aichi 466-8550, Japan. Phone: 81-52- 304/B6 (RET-mice) carrying oncogenic RET (RFP/RET) 744-2122; Fax: 81-52-744-2122; E-mail: [email protected] under the control of the metallothionein-I promoter (9). doi: 10.1158/1541-7786.MCR-13-0468-T Hyperpigmented skin, benign melanocytic tumors, and 2013 American Association for Cancer Research. melanomas with metastases develop stepwise in RET-mice

440 Mol Cancer Res; 12(3) March 2014

Role of Espin in Melanoma

(9). Because histopathologic characteristics of melanoma in into the lateral veins of 6-week-old female mice (n ¼ 5in RET-mice are similar to those of melanoma in humans, each group), and the mice were sacrificed at 42 days after RET-mice have been used as a standard model of cutaneous inoculation. Green fluorescence protein (GFP)-positive melanoma in recent studies (9, 10). In this study, we metastatic melanomas in lungs were analyzed by MZ16F examined the effect of Espin on the pathogenesis of mela- light microscopy (Leica Microsystems). noma and found for the first time a correlation between Espin and cancer. Ethics statement This study was performed in Chubu University, Japan. The study was approved by the Animal Care and Use Materials and Methods Committee (approval no. 2410030), Recombination DNA Cells Advisory Committee (approval no. 10-08), and Ethical A murine Mel-ret melanoma cell line derived from a Committee (approval no. 20008-10) at the Chubu tumor in the RET-mouse (11) and human melanoma cell University. lines (G361, SK-Mel28, Colo679, HMVII, A375P, and A375M) (10) were cultured in RPMI 1640 supplemented with 10% FBS. G361, SK-Mel28, Colo679, and HMVII Results were provided by the Riken Bioresource Center Cell Bank Expression and localization of Espin in murine and A375P and A375M were kindly provided by Dr. melanomas Dorothy C Bennett (St. George's, University of London, We first showed that Espin transcript expression levels in UK). Primary culture of normal human epithelial melano- melanomas were more than 200-fold higher than those in cytes (NHEM) was performed according to the manufac- benign melanocytic tumors in RET-mice by quantitative turer's protocol (Cell Applications). real-time PCR (qRT-PCR) analysis using primers (primer set total Espin shown in Supplementary Fig. S1), which Immunoblotting, immunohistochemistry, and recognize all isoforms of the Espin transcript (Supplementary immunofluorescence analysis Fig. S2A). These results indicated that the Espin transcript is An anti-Espin-ru rabbit polyclonal antibody was devel- highly expressed in melanomas in RET-mice. We then oped using peptides (SSSSTGSTKSFN) in this study examined the isoform of the Espin transcript that is highly (Supplementary Fig. S1). Rabbit polyclonal antibodies expressed in melanoma using various primers (primer sets 1, against phosphorylated tyrosine 397 in FAK (Invitrogen), 2A-A0, 2B-B0,1–3, and 4) shown in Supplementary Fig. S1. phosphorylated tyrosine 181 in paxillin (Epitomics), Transcript expression levels of Espin-1, -2, and -3 and Espin- mouse monoclonal antibodies against glyceraldehyde-3- 4 isoforms detected by primer sets 1–3 and 4 (Supplemen- phosphate dehydrogenase (GAPDH; Cell Signaling Tech- tary Fig. S1) in melanomas were 286- and 28-fold higher nology), FAK (BD Transducton Laboratories), vinculin than those in benign tumors, respectively (Supplementary (Sigma), and paxillin (BD) were used as primary anti- Fig. S2B). However, transcript expression levels of Espin-1, bodies. A pull-down assay was performed using RhoA and Espin-2A, and -2A0 and Espin-1, 2B, and -2B0 isoforms Rac1 Activation Assay Kits (Cell Biolabs) according to the detected by primer sets 1, 2A-2A0, and 2B-B0 (Supplemen- manufacturer's protocol. Immunohistochemical estima- tary Fig. S1) in melanomas were comparable with those in tion of ESPIN (positive/negative) was performed using benign tumors (Supplementary Fig. S2C). These results thesoftwareprogramWinROOF(MitaniCorporation) showed higher expression of Espin-3, lower expression of according to the method previously reported (12). Meth- Espin-4, and no expression of Espin-1 and Espin-2 transcripts ods for immunoblotting, immunofluorescence analysis, in melanomas. Because ESPIN-1 and -3, but not ESPIN-2 and immunohistochemistry using human tissue array and -4, were detected in humans in previous studies (5, 7), slides (ME103; US Biomax, Inc.) are described in Sup- we newly developed a rabbit polyclonal anti-Espin-ru anti- plementary Materials and Methods. body targeting the common sequence for Espin-3 protein in mice and humans (Supplementary Figs. S1 and S3). Wound-healing assay and in vitro invasion assay Corresponding to the results of qRT-PCR analysis, Wound-healing assays using stably expressed short hairpin immunoblot analysis using anti-Espin-ru-antibody revealed RNA (shRNA) and control Mel-ret cells seeded on type I a more than 44-fold increased Espin protein expression level collagen (10 mg/mL)–coated 6-well plastic dishes at a density in melanomas compared with that in benign tumors in RET- of 2 105 cells per well were performed according to the mice (Fig. 1A). Immunofluorescence analysis showed that method previously described (13). In vitro invasion assays Espin protein was localized in the cytoplasm and lamelli- using 2 105 cells with stably expressed shRNA and 2 105 podia (Fig. 1B) and was colocalized with F-actin protein in control Mel-ret cells were performed according to the lamellipodia (Fig. 1B, arrowheads). Disruption of lamelli- method previously described (14). podia formation and rugged shape were observed in stable clones of Espin-depleted Mel-ret cells (clone E4 in Fig. 1C Melanoma xenograft studies and D), whereas lamellipodia were organized in the mobile Mel-ret melanoma cells (1 106 in PBS) stably trans- edge of cells in the control clones (clone C6 in Fig. 1C and fected with Espin shRNA or control shRNA were injected D). Cortical thick stress fibers were also identified in the

www.aacrjournals.org Mol Cancer Res; 12(3) March 2014 441

Yanagishita et al.

Figure 1. Morphologic effects of Espin expression on melanoma cells. A, Espin (Espin-3; 35 kDa) protein expression levels in the skin of wild-type-mice (WT), RET-mice (RET), benign melanocytic tumors, and melanoma from RET-mice are shown (left). Intensities of bands are compared between benign melanocytic tumors (mean SD; n ¼ 4) and melanoma (mean SD; n ¼ 3; right). B, confocal immunofluorescence images of merged Espin (green) and F-actin (magenta) in Mel-ret cells are presented. Arrowheads indicate lamellipodia. C, Espin expression levels in stable clones established from Mel-ret cells transfected with control (clones C3 and C6) and Espin-shRNA (clones E2 and E4) vectors are shown (top). Intensities of bands are presented as percentages (mean SD; n ¼ 3) relative to control clone C3 (bottom). D, morphologic change of Mel-ret cells by stable transfection of shRNA vectors. Cells were stained with Alexa568-phalloidin to visualize the structures of actin filaments. Arrowheads indicate lamellipodia in Mel-ret cells. Arrows indicate cortical thick stress fibers in Espin-depleted Mel-ret cells (clone E4; left). Percentages of lamellipodia formation (n ¼ 100) are compared between control cells (clones C3 and C6) and Espin-depleted cells (clones E2 and E4; right). Significantly different ( , P < 0.01) from the control (clone C3) by the Student t test. Scale bar, 20 mm.

Espin-depleted clone (arrows of clone E4 in Fig. 1D). depleted Mel-ret cells. Activities of cell migration in the Statistical analysis further showed about 55% (clones C3 Espin-depleted clones (clones E2 and E4 in Fig. 2A) were and C6) and about 12% (clones E2 and E4) of lamellipodia about 77% reduced compared with those in the control formed in control (clones C3 and C6 in Fig. 1D) and Espin- cells (clones C3 and C6 in Fig. 2A) in wound-healing depleted Mel-ret cells (clones E2 and E4 in Fig. 1D), assays. Activities of cell invasion in Espin-depleted cells respectively. (clones E2 and E4 in Fig. 2B) were also about 80% reduced compared with those in control cells (clones Effects of Espin on migration and invasion activities of C3 and C6 in Fig. 2B) in invasion assays. Levels of FAK, melanoma cells paxillin and Rac1 activities, and vinculin expression in We next performed wound-healing assays and in vitro Espin-depleted cells (clone E4) were decreased compared invasion assays in stable clones of control and Espin- with those in control cells (clone C6; Fig. 2C). However,

442 Mol Cancer Res; 12(3) March 2014 Molecular Cancer Research

Role of Espin in Melanoma

Figure 2. Decreased cell migration and invasion of Espin-depleted melanoma cells. A, wound-healing assay was performed with control (clones C3 and C6) and Espin- depleted (clones E2 and E4) Mel-ret cells (left). The graph shows the wound-filled area evaluated as percentage (mean SD; n ¼ 3) at 16 hours after scratching (right). B, Matrigel-invasion assay was performed with control (clones C3 and C6) and Espin-depleted (clones E2 and E4) Mel-ret cells. Photographs of cells invading the membrane stained with HE are shown (left). After invading cells had been counted in five random microscopic fields in each Matrigel-invasion assay, the results of three independent assays were normalized and are presented as an invasion index (right). C, phosphorylation and expression levels of FAK and paxillin and expression levels of vinculin and GAPDH in control (clone C6) and Espin-depleted (clone E4) Mel-ret cells are shown. Amounts of GTP-bound forms and total Rac1 and RhoA proteins in control (clone C6) and Espin- depleted (clone E4) Mel-ret cells are also shown. Intensities of bands are presented as percentages (mean SD; n ¼ 3) relative to control clone C3. Significantly different ( , P < 0.01; , P < 0.05) from the control (C6) by the Student t test. Scale bar, 200 mm.

activity levels of RhoA were comparable in control cells (clone E4) was smaller than that in control cells (clone andEspin-depletedcells(Fig.2C). C6; Fig. 3A). Statistical analysis also showed that the number of GFP-positive metastatic foci on the surface of Effects of Espin on metastasis of melanoma cells in vivo the lung in Espin-depleted cells (clone E4) was about 25% Because many studies have shown that migration and of that in control cells (clone C6; Fig. 3B). In addition, invasion involve metastasis (15), we then examined the lung metastasis in Espin-depleted Mel-ret cells (clone E4) effect of Espin on melanoma metastasis in vivo.GFP- and control cells (clone C6) was microscopically con- tagged Espin-depleted Mel-ret cells (clone E4; n ¼ 5) and firmed by hematoxylin-eosin (HE) staining (Fig. 3A). control cells (clone C6; n ¼ 5) were injected into the tail veins of nude mice (Fig. 3). Macroscopic analysis for ESPIN expression in human melanomas fluorescence intensity on the surface of the lung showed We finally examined ESPIN protein expression levels that the number of metastatic foci in Espin-depleted cells in melanoma cell lines and tissues in humans.

www.aacrjournals.org Mol Cancer Res; 12(3) March 2014 443

Yanagishita et al.

Figure 3. Decreased lung metastasis in Espin-depleted melanoma cells in vivo. A, results of morphologic analysis of lung metastasis of control (clone C6) and Espin-depleted (clone E4) Mel- ret cells injected into tail veins of nude mice (n ¼ 5) are presented. Animals were dissected to observe lung metastases at 42 days after inoculation. Lung metastases were macroscopically visualized by GFP fluorescence images. Metastatic foci (arrow) derived from control (clone C6) and Espin-depleted (clone E4) cells were microscopically confirmed by HE staining. B, number of GFP- positive metastatic foci per lung surface (mean SD; n ¼ 5) is presented. Significantly different ( , P < 0.01) from the control (C6) by the Student t test.

Immunoblot analysis showed that ESPIN was expressed matrix (16, 17). The structure is induced in many biologic in all of the six melanoma cell lines (G361, SK-Mel28, processes including cell migration and drives progression of Colo679, HMVII, A375P, and A375M) but not in cancer invasion and metastasis via dynamic remodeling of NHEM (Fig. 4A). Immunohistochemical analysis also the actin cytoskeleton (18). Because various actin-binding showed that ESPIN protein was expressed in more than proteins constructing lamellipodia were previously reported 80% and 90% of primary melanomas (n ¼ 77) and to play an important role in cancer invasion (14), we metastatic melanomas for lymph nodes (n ¼ 23), respec- hypothesized that Espin regulates the migration and inva- tively, whereas no expression of ESPIN was detected in sion of melanoma cells. We found localization of Espin at more than 60% of melanocytic nevi (n ¼ 36;Fig.4Band lamellipodia in migrating melanoma cells on the type I C). Statistical analysis showed that percentages of collagen matrix as an ECM in addition to decreased ESPIN-positive primary and metastatic melanomas were migration activity in Espin-depleted melanoma cells with significantly higher (P < 0.01) than the percentage of cortical thick stress fibers, which inhibit cell motility (18), ESPIN-positive melanocytic nevi (Fig. 4C). in the wound-healing assay. Together with the results showing decreased invasion activity in Espin-depleted melanoma cells, these results suggest that Espin regulates Discussion the dynamics of actin polymerization at lamellipodia and We first demonstrated that the expression of Espin in is one of the drivers of melanoma cell motility. skin melanomas is definitely higher than that in benign Cells in the process of migration must acquire a spatial melanocytic tumors in RET-mice. Because our previous asymmetry enabling them to turn intracellularly generated study showed that melanomas developed from benign forces into net cell body translocation (19). One mani- melanocytictumorsinRET-mice(9),theresultssuggest festation of this asymmetry is a polarized morphology, that increased Espin expression level in melanoma is which is clearly distinct between the cell front and cell rear associated with malignant transformation. We also dem- (19). Because central regulators of the cell front are Rho onstrated that levels of ESPIN expression in melanoma family small GTP-binding proteins (Rho GTPases), cell lines and primary and metastatic melanomas are including Rac and Rho, they are pivotal regulators of higher than those in NHEM and melanocytic nevi, actin and adhesion organization and control the formation respectively, in humans. These results suggest that ESPIN of lamellipodia (20). Rac1 is a direct regulator of the could be a potential biomarker for primary and metastatic organization of lamellipodial structure (8, 16). A previous melanomas. study showed that Rac1 is downstream of signaling from We then examined the biologic significance of increased the complex of paxillin, FAK, and vinculin in focal Espin expression in melanoma cells. We showed that Espin adhesions and cell–cell contacts (8). FAK is not only a was colocalized with F-actin in lamellipodia. Lamellipodia central regulator in focal adhesions but also a critical player are formed when cancer cells migrate on an extracellular in melanoma cell motility via interaction with the ECM matrix (ECM) such as the type I collagen or fibronectin and subsequent activation of downstream signals toward a

444 Mol Cancer Res; 12(3) March 2014 Molecular Cancer Research

Role of Espin in Melanoma

Figure 4. Espin expression in human melanoma cell lines and melanoma tissues. A, expression levels of Espin protein (Espin-3; 34 kDa) in NHEM and six human melanoma cell lines are shown. Intensities of bands are presented as ratio (mean SD; n ¼ 3) relative to NHEM. Significantly different ( , P < 0.01) from NHEM by the Student t test. B, HE-stained nevus (top), primary melanoma (middle), and lymph node metastasis of melanoma (bottom) in humans are presented (scale bar: 100 mm). Representative Espin protein expressions (purple) in melanocytic nevus (top), primary melanoma (middle), and lymph node metastasis of melanoma (bottom) evaluated by immunohistochemical analysis counterstained by methyl green with anti-Espin-ru antibody are presented in lower (scale bar: 100 mm) and higher (scale bar: 20 mm) magnifications. C, percentages of Espin-positive (black) and -negative (white) cells in melanocytic nevus, primary melanomas, and lymph node metastases of melanomas are presented. Number of Espin- positive cells was divided by number of total cells in five fields with 200-fold magnification in each tissue. Significantly different ( , P < 0.01) from nevi by the Fisher exact test.

malignant phenotype (8). Because our results showed tasis of melanoma by regulation of migration and invasion decreased levels of FAK, paxillin, and Rac1 activities and with modulation of FAK/paxillin/vinculin/Rac1 signaling. vinculin expression in Espin-depleted cells with disruption Our results indicate that ESPIN is potentially useful for of lamellipodia, Espin may be associated with lamellipodia therapy for metastatic melanoma as well as a biomarker for formation via regulation of FAK/paxillin/vinculin/Rac1 melanoma. signaling in melanoma cells. Together with our results showing an approximately 75% decrease of lung metas- Disclosure of Potential Conflicts of Interest tasis in Espin-depleted cells inoculated in nude mice, the No potential conflicts of interest were disclosed. results indicated that Espin is a molecular target for therapy for metastatic melanoma. fi Authors' Contributions In summary, we demonstrated for the rst time that Conception and design: T. Yanagishita, I. Yajima, Y. Matsumoto, M. Kato Espin/ESPIN is expressed in melanoma and affects metas- Development of methodology: T. Yanagishita, I. Yajima, M. Kumasaka

www.aacrjournals.org Mol Cancer Res; 12(3) March 2014 445

Yanagishita et al.

Acquisition of data (provided animals, acquired and managed patients, provided Grant Support facilities, etc.): T. Yanagishita, M. Kumasaka, T. Tsuzuki, M. Kato This study was supported in part by Grants-in-Aid for Scientiﬁc Research Analysis and interpretation of data (e.g., statistical analysis, biostatistics, compu- (B: 24390157 and 24406002; C: 25340052), Research Fellow of Japan Society for tational analysis): T. Yanagishita, I. Yajima, Y. Kawamoto, T. Tsuzuki, D. Watanabe, the Promotion of Science (25-40080), Grant-in-Aid for Challenging Exploratory M. Kato Research (23650241), Grant-in-Aid for Scientiﬁc Research on Innovative Areas Writing, review, and/or revision of the manuscript: T. Yanagishita, I. Yajima, (24108001) from the Ministry of Education, Culture, Sports, Science and Technology M. Kato (MEXT), the Uehara Memorial Foundation, the Naito Foundation, and the Cos- Administrative, technical, or material support (i.e., reporting or organizing data, metology Research Foundation. constructing databases): T. Yanagishita, I. Yajima, M. Kumasaka, T. Tsuzuki The costs of publication of this article were defrayed in part by the payment of page Study supervision: T. Yanagishita, Y. Matsumoto, D. Watanabe, M. Kato charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Acknowledgments The authors would like to thank Kazumi Shigeta and Yoko Kato for technical Received September 4, 2013; revised November 12, 2013; accepted December 3, assistance. 2013; published OnlineFirst December 13, 2013.

References 1. Linos E, Swetter SM, Cockburn MG, Colditz GA, Clarke CA. Increasing 11. KatoM,LiuW,AkhandAA,DaiY,Ohbayashi M, Tuzuki T, et al. burden of melanoma in the United States. J Invest Dermatol 2009; Linkage between melanocytic tumor development and early 129:1666–74. burst of Ret protein expression for tolerance induction in metal- 2. Chin L, Garraway LA, Fisher DE. Malignant melanoma: genetics and lothionein-I/ret transgenic mouse lines. Oncogene 1999;21: therapeutics in the genomic era. Genes Dev 2006;20:2149–82. 837–42. 3. Gray-Schopfer V, Wellbrock C, Marais R. Melanoma biology and new 12. Ohgami N, Ida-Eto M, Shimotake T, Sakashita N, Sone M, Naka- targeted therapy. Nature 2007;22:851–57. shima T, et al. c-Ret-mediated hearing loss in mice with 4. Bartles JR, Wierda A, Zheng L. Identification and characterization of Hirschsprung disease. Proc Natl Acad Sci U S A 2010;107: espin, an actin-binding protein localized to the F-actin-rich junctional 13051–6. plaques of Sertoli cell ectoplasmic specializations. J Cell Sci 1996;109: 13. Jiang P, Enomoto A, Jijiwa M, Kato T, Hasegawa T, Ishida M, et al. An 1229–39. actin-binding protein Girdin regulates the motility of breast cancer 5. Sekerkova G, Zheng L, Loomis PA, Mugnaini E, Bartles JR. Espins and cells. Cancer Res 2008;68:1310–8. actin cytoskeleton of hair cell stereosilia and sensory cell microvilli. Cell 14. Thang ND, Yajima I, Kumasaka MY, Ohnuma S, Yanagishita T, Hayashi Mol Life Sci 2006;63:2329–41. R, et al. Barium promotes anchorage-independent growth and inva- 6. Zheng L, Sekerkova G, Vranich K, Tilney LG, Mugnaini E, Bartles JR. sion of human HaCaT keratinocytes via activation of c-SRC kinase. The deaf jerker mouse has a mutation in the gene encoding the espin PLoS ONE 2011;6:e25636. actin-bundling proteins of hair cell stereocilia and lacks espins. Cell 15. Yajima I, Kumasaka MY, Thang ND, Goto Y, Takeda K, Yamanoshita O, 2006;102:377–85. et al. RAS/RAF/MEK/ERK and PI3K/PTEN/AKT signaling in malignant 7. Sekerkova G, Zheng L, Loomis PA, Changyaleket B, Whitlon DS, melanoma progression and therapy. Dermatol Res Pract 2012;2012: Mugnaini E, et al. Espins are multifunctional actin cytoskeletal regu- 354191. latory proteins in the microvilli of chemosensory and mechanosensory 16. Machesky LM. Lamellipodia and filopodia in metastasis and invasion. cells. J Neurosci 2004;9:5445–56. FEBS Lett 2008;582:2102–11. 8. Besson A, Assoian RK, Roberts JM. Regulation of the cytoskeleton: an 17. Even-Ram S, Yamada KM. Cell migration in 3D matrix. Curr Opin Cell oncogenic function for CDK inhibitors? Nat Rev Cancer 2004;4: Biol 2005;17:524–32. 948–55. 18. Yamaguchi H, Wyckoff J, Condeelis J. Cell migration in tumors. Curr 9. Kato M, Takahashi M, Akhand AA, Liu W, Dai Y, Shimizu S, et al. Opin Cell Biol 2005;17:559–64. Transgenic mouse model for skin malignant melanoma. Oncogene 19. Ridley AJ, Schwartz MA, Burridge K, Firtel RA, Ginsberg MH, Borisy G, 1998;8:1885–88. et al. Cell migration: integrating signals from front to back. Science 10. Yajima I, Kumasaka M, Thang ND, Yanagishita T, Ohgami N, Kallen- 2003;5:1704–9. berg D, et al. Zinc finger protein 28 as a novel melanoma-related 20. Lauffenburger DA, Horwitz AF. Cell migration: a physically integrated molecule. J Dermatol Sci 2009;55:68–70. molecular process. Cell 1996;84:359–69.

446 Mol Cancer Res; 12(3) March 2014 Molecular Cancer Research

Actin-Binding Protein, Espin: A Novel Metastatic Regulator for Melanoma

Takeshi Yanagishita, Ichiro Yajima, Mayuko Kumasaka, et al.

Mol Cancer Res 2014;12:440-446. Published OnlineFirst December 13, 2013.

Updated version Access the most recent version of this article at: doi:10.1158/1541-7786.MCR-13-0468-T

Supplementary Access the most recent supplemental material at: Material http://mcr.aacrjournals.org/content/suppl/2013/12/13/1541-7786.MCR-13-0468-T.DC1

Cited articles This article cites 20 articles, 4 of which you can access for free at: http://mcr.aacrjournals.org/content/12/3/440.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://mcr.aacrjournals.org/content/12/3/440. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.