Title: Membrane Metalloendopeptidase (MME) Suppresses Metastasis of Esophageal Squamous Cell Carcinoma (ESCC) by Inhibiting FAK-RhoA Signaling Axis
Abstract: Esophageal squamous cell carcinoma (ESCC) is a typical neoplastic disease and a frequent cause of death in China. Although great achievements have been made in diagnostic strategies and combination therapies in recent years, the prognosis of ESCC is still poor. Metastasis/recurrence has been the major factor responsible for poor prognosis. However, the underlying mechanism of ESCC dissemination remains elusive. Membrane metalloendopeptidase (MME) is a transmembrane glycoprotein that degrades a number of substrates. This study's results indicated that the down-regulation of MME is significantly associated with advanced clinical stage (P < 0.05) and lymph node metastasis (P < 0.05). The down-regulation of MME in ESCC tumor tissues is correlated to poorer prognosis of the patients. Functional studies demonstrated that MME could significantly inhibit ESCC tumor cell metastasis in vitro and in vivo. MME overexpression could also interrupt ESCC tumor cell adhesion. Mechanistically, MME inhibits the phosphorylation of FAK thus interrupting the FAK-RhoA axis, which is important in cell movement. Taken together, these data show that MME regulates ESCC via FAK-RhoA axis. High expression of MME may indicate a beneficial outcome for patients. Esophageal squamous cell carcinoma (ESCC) is a typical neoplastic disease and a frequent cause of death in China. Although great achievements have been made in diagnostic strategies and combination therapies in recent years, the prognosis of ESCC is still poor. Metastasis/recurrence has been the major factor responsible for poor prognosis. However, the underlying mechanism of ESCC dissemination remains elusive. Membrane metalloendopeptidase (MME) is a transmembrane glycoprotein that degrades a number of substrates. This study's results indicated that the down-regulation of MME is significantly associated with advanced clinical stage (P < 0.05) and lymph node metastasis (P < 0.05). The down-regulation of MME in ESCC tumor tissues is correlated to poorer prognosis of the patients. Functional studies demonstrated that MME could significantly inhibit ESCC tumor cell metastasis in vitro and in vivo. MME overexpression could also interrupt ESCC tumor cell adhesion. Mechanistically, MME inhibits the phosphorylation of FAK thus interrupting the FAK-RhoA axis, which is important in cell movement. Taken together, these data show that MME regulates ESCC via FAK-RhoA axis. High expression of MME may indicate a beneficial outcome for patients. Esophageal squamous cell carcinoma (ESCC), along with esophageal adenocarcinoma, is the major subtype of esophageal cancer recognized, according to microanatomic categorization. In Asia, ESCC is the predominant form, which accounts for 57% of diagnoses compared with 18% in the United States.1Noone A.M. Cronin K.A. Altekruse S.F. Howlader N. Lewis D.R. Petkov V.I. Penberthy L. Cancer incidence and survival trends by subtype using data from the Surveillance Epidemiology and End Results Program, 1992-2013.Cancer Epidemiol Biomarkers Prev. 2017; 26: 632-641Crossref PubMed Scopus (247) Google Scholar It is also the fourth leading cause of cancer-related deaths in China.2Chen W. Zheng R. Baade P.D. Zhang S. Zeng H. Bray F. Jemal A. Yu X.Q. He J. Cancer statistics in China, 2015.CA Cancer J Clin. 2016; 66: 115-132Crossref PubMed Scopus (14103) Google Scholar Eastern Asia has the highest incidence of esophageal cancer, whereas it is lowest in Western Africa.3McCormack V.A. Menya D. Munishi M.O. Dzamalala C. Gasmelseed N. Leon Roux M. Assefa M. Osano O. Watts M. Mwasamwaja A.O. Mmbaga B.T. Murphy G. Abnet C.C. Dawsey S.M. Schuz J. Informing etiologic research priorities for squamous cell esophageal cancer in Africa: a review of setting-specific exposures to known and putative risk factors.Int J Cancer. 2017; 140: 259-271Crossref PubMed Scopus (83) Google Scholar Although the incidence rates from 2000 to 2011 decreased for ESCC, and scientists have performed numerous studies to identify prognostic markers, the outcome for ESCC patients remains grim. Thus, understanding the detailed molecular mechanisms in ESCC progression and developing novel strategies for treatment are urgently needed to improve the survival rates for these sufferers. The membrane metalloendopeptidase (MME) gene is located at human chromosome 3q21-27. It encodes a 100-kD type II transmembrane glycoprotein,4Barker P.E. Shipp M.A. D'Adamio L. Masteller E.L. Reinherz E.L. The common acute lymphoblastic leukemia antigen gene maps to chromosomal region 3 (q21-q27).J Immunol. 1989; 142: 283-287PubMed Google Scholar a widely expressed membrane metalloendopeptidase that degrades a number of substrates. The active site of the enzyme faces the extracellular space.5Depondt C. Donatello S. Rai M. Wang F.C. Manto M. Simonis N. Pandolfo M. MME mutation in dominant spinocerebellar ataxia with neuropathy (SCA43).Neurol Genet. 2016; 2: e94Crossref PubMed Scopus (37) Google Scholar MME is widely expressed in a broad range of tissues, and in kidney and lung tissues, it has particularly abundant expression.6Sahli S. Stump B. Welti T. Schweizer W.B. Diederich F. Blum-Kaelin D. Aebi J.D. Böhm H.-J. A new class of inhibitors for the metalloprotease neprilysin based on a central imidazole scaffold.Helv Chim Acta. 2005; 88: 707-730Crossref Scopus (32) Google Scholar The molecule was first identified as a tumor-specific antigen (common acute lymphoblastic leukemia antigen) in leukemia and has been used for diagnosing B-lineage acute lymphoblastic leukemia in combination with other B-lineage markers.7Wood B.L. Arroz M. Barnett D. DiGiuseppe J. Greig B. Kussick S.J. Oldaker T. Shenkin M. Stone E. Wallace P. 2006 Bethesda International Consensus recommendations on the immunophenotypic analysis of hematolymphoid neoplasia by flow cytometry: optimal reagents and reporting for the flow cytometric diagnosis of hematopoietic neoplasia.Cytometry B Clin Cytom. 2007; 72 Suppl 1: S14-S22Crossref PubMed Scopus (178) Google Scholar Its expression can indicate prognosis in acute lymphoblastic leukemia.8Choi W.W. Weisenburger D.D. Greiner T.C. Piris M.A. Banham A.H. Delabie J. Braziel R.M. Geng H. Iqbal J. Lenz G. Vose J.M. Hans C.P. Fu K. Smith L.M. Li M. Liu Z. Gascoyne R.D. Rosenwald A. Ott G. Rimsza L.M. Campo E. Jaffe E.S. Jaye D.L. Staudt L.M. Chan W.C. A new immunostain algorithm classifies diffuse large B-cell lymphoma into molecular subtypes with high accuracy.Clin Cancer Res. 2009; 15: 5494-5502Crossref PubMed Scopus (534) Google Scholar Although researchers have observed associations between MME expression and a few types of cancer, the specific association remains obscure. Interestingly, in some cancers, such as colorectal carcinoma, pancreatic endocrine tumor, and advanced melanomas, MME is overexpressed.9Mizerska-Kowalska M. Bojarska-Junak A. Jakubowicz-Gil J. Kandefer-Szerszen M. Neutral endopeptidase (NEP) is differentially involved in biological activities and cell signaling of colon cancer cell lines derived from various stages of tumor development.Tumour Biol. 2016; 37: 13355-13368Crossref PubMed Scopus (10) Google Scholar In some types of tumors, most notably lung cancers and ovarian cancer, MME is down-regulated. In prostatic carcinoma, MME is expressed in androgen-sensitive LNCaP cells and possesses tumor suppressive ability.10Dai J. Shen R. Sumitomo M. Goldberg J.S. Geng Y. Navarro D. Xu S. Koutcher J.A. Garzotto M. Powell C.T. Nanus D.M. Tumor-suppressive effects of neutral endopeptidase in androgen-independent prostate cancer cells.Clin Cancer Res. 2001; 7: 1370-1377PubMed Google Scholar MME not only is a cell surface marker, but also has a critical role in regulating many biological processes.11Makretsov N.A. Hayes M. Carter B.A. Dabiri S. Gilks C.B. Huntsman D.G. Stromal CD10 expression in invasive breast carcinoma correlates with poor prognosis, estrogen receptor negativity, and high grade.Mod Pathol. 2007; 20: 84-89Crossref PubMed Scopus (71) Google Scholar However, the role of MME has not been investigated in the ESCC oncogenesis and progression. In this study, the clinical relevance of MME expression was investigated in human ESCC samples. MME down-regulation was correlated with poorer prognosis of ESCC patients. Also, MME introduction inhibited tumor cell metastasis, while not affecting tumor cell proliferation. The underlying mechanism of how MME inhibits ESCC tumor cell was also explored. The HKESC1, EC18, and EC109 cell lines were kindly provided by Prof. G. Srivastava (Department of Pathology, The University of Hong Kong, Hong Kong, China). An immortalized nonmalignant esophageal cell line was provided by Prof. G.S. Tsao (Department of Anatomy, The University of Hong Kong). ESCC cell lines (KYSE520, KYSE510, KYSE410, KYSE180, KYSE140, and KYSE30) were acquired from the German Biological Resource Center (DSMZ, Braunschweig, Germany). 293FT cell were obtained from Invitrogen (Thermo Fisher Scientific, Waltham, MA). ESCC cell lines were cultured in Dulbecco's modified Eagle's medium with 10% supplements of fetal bovine serum Gibco, Thermo Fisher Scientific. The cells were authenticated by short tandem repeat profiling. The study was performed under the review of the Committees for Ethical Review of Research Involving Human Subjects, both in the Sun Yat-sen University Cancer Center and Zhengzhou University. The ESCC tumor tissues along with the adjacent nontumor parts (at least 5 cm away from the tumor site) were obtained from Linzhou Cancer Hospital, and the tissue microarray containing 300 pairs of tumor specimens and the corresponding non-tumor tissues were constructed.12Li Y. Chen L. Nie C.J. Zeng T.T. Liu H. Mao X. Qin Y. Zhu Y.H. Fu L. Guan X.Y. Downregulation of RBMS3 is associated with poor prognosis in esophageal squamous cell carcinoma.Cancer Res. 2011; 71: 6106-6115Crossref PubMed Scopus (40) Google Scholar The paraffin wax of the specimens was removed by xylene, then the slides were washed in a series of graded alcohol: 100%, 95%, 75%, and 50% for rehydration and then incubated with 3% hydrogen peroxide. For antigen retrieval, the slides were boiled in EDTA buffer for 30 minutes. Five percent bovine serum albumin in phosphate-buffered saline (PBS) was used to block the nonspecific binding. After incubation with primary antibody against MME (PB0026, Boster Biological Technology, Wuhan, China) at 4°C overnight, the slides were washed and incubated with horseradish peroxidase–conjugated secondary antibody before stained with the DAB substrate (Dako, Glostrup, Denmark). An immunoreactivity pathological score system which is widely used was applied as previously described.12Li Y. Chen L. Nie C.J. Zeng T.T. Liu H. Mao X. Qin Y. Zhu Y.H. Fu L. Guan X.Y. Downregulation of RBMS3 is associated with poor prognosis in esophageal squamous cell carcinoma.Cancer Res. 2011; 71: 6106-6115Crossref PubMed Scopus (40) Google Scholar In brief, the proportion of MME stained cells from 0% to 100% (<5%, ≥5% to <25%, ≥25% to <50%, ≥50% to <75%, ≥75% to 100%) got a proportion score (PS) of 0 to 4. The intensity of MME staining was also scored as 0 to 4 for different degrees of staining, from negative to strong. The staining index (SI) was calculated using a product formula: positive percentage × intensity to evaluate the score. Because the average SI of nontumor tissues (NSI) was equal to 5.2, the down-regulation of MME was determined as TSI < 5. The recombinant construct pLV105-MME and vector control were bought from GeneCopia (Rockville, MD). Lentivirus was obtained by the cotransfection of recombinant constructs with three lentivirus packaging mix (pLp1, pLp2, and pLp-VSVG) (Invitrogen) into 293FT (Invitrogen). ESCC cells were infected with lentivirus and stably overexpressed cells were established by selecting with 5 μg/mL of puromycin for about 14 days. The shRNA products were purchased from Genechem (Shanghai, China). The knockdown efficiency was tested by quantitative RT-PCR and Western blot. The primers and shRNAs are listed in Table 1.Table 1Primers and shRNAsPrimer nameSequenceMMEF: 5′-AGTCGGAAACTGGCAGATAGC-3′MMER: 5′-GGTAGTGTTGTACTGGGCCAAT-3′β-ActinF: 5′-CTCCATCCTGGCCTCGCTGT-3′β-ActinR: 5′-GCTGTCACCTTCACCGTTCC-3′β-TubulinF: 5′-GAGCTGTTCAAGCGCATCTC-3′β-TubulinR: 5′-TCCTCCTCGTCGTCTTCGTA-3′shRNAs targeting MME Sh15′-TCCAGGAGTATGTCACCTT-3′ Sh25′-CAAGGCTGATGGAAGACTT-3′ Sh35′-GGTGAACTTCCAGGAGTAT-3′F, forward; R, reverse. Open table in a new tab F, forward; R, reverse. Cell fixation with 4% paraformaldehyde was followed by permeabilization with 0.1% Triton X-100. After incubating with primary antibodies (rabbit anti-FAK, Cell Signaling Technology, Danvers, MA) at 4°C overnight, cells were washed and incubated with fluorescent secondary antibody for 1 hour. The cells were counterstained with DAPI (Thermo Fisher Scientific) for 3 minutes at room temperature. For F-actin staining, direct immunofluorescence of fluorescein isothiocyanate–conjugated phalloidin (Sigma-Aldrich, St. Louis, MO) was applied at 1% in PBS to stain the fixed cells for 30 minutes. The slides were then mounted in VECTASHIELD with DAPI (Vector Laboratories, Burlingame, CA).13Watkins S.J. Borthwick G.M. Arthur H.M. The H9C2 cell line and primary neonatal cardiomyocyte cells show similar hypertrophic responses in vitro.In Vitro Cell Dev Biol Anim. 2010; 47: 125-131Crossref PubMed Scopus (278) Google Scholar All images were visualized under Olympus FLUOVIEW FV1000 confocal laser scanning microscope or OLYMPUS BX53 fluorescent microscope (Olympus Corporation, Tokyo, Japan). A 10-μL pipette tip was used to scratch a straight wound in the experimental group cells or the control cells in 35-mm dishes. The migration photos were captured at 0, 24, or 48 hours after scratching to compare the gap closure speed. Two kinds of 24-well chambers were used in the assays (BD Biosciences, Franklin, NJ), both of them have 8-μm microporous membranes to separate the chambers. In the invasion assay, the membranes were coated with matrix gel. Tumor cells were seeded in the top compartment with serum-free medium. The bottom compartments were filled with Dulbecco's modified Eagle's medium with 10% fetal bovine serum as the chemoattractant. After 18 or 28 hours, the cells that migrated through the pores were fixed. The cells were stained with crystal violet, observed, and then quantified in over 10 fields under the microscope. Cells, 1.0 × 103 per well, were seeded in 6-well plates. After 10-day culture, colonies that formed were stained using crystal violet (Sigma-Aldrich) and counted. Three independent experiments were repeated. The animal experiments were approved by the Institutional Animal care and Use Committee of Sun Yat-sen University Cancer Center. Cell proliferation in vivo was evaluated by the subcutaneous tumor models. Briefly, MME overexpressed cells and vector control cells were suspended in PBS, then subcutaneously implanted in the lateral flank of 4-week–old BALB/c nude mice (n = 5). Tumor size of the mice was measured every 7 days. Tumor volume was calculated as: V = L × W2 × 0.5 (mm3). At the end of the experiment, the animals were sacrificed, and tumors were weighed. To establish metastasis models, a total of 5 × 105 EC109 or K510 derivative cells were injected via the tail vein of Nu/Nu BALB/c mice (4-week–old, female, n = 5 for each group). After 2 months, the animals were sacrificed. The animals' lungs were isolated and fixed and the number of nodules was counted. The lung sections were examined by hematoxylin and eosin stain. For the lymph node metastasis model, cells in PBS solution (1.5 × 105/20 μL) were injected into one hind foot per animal. The mice were euthanized 2 months later. The popliteal lymph nodes were excised and fixed. Each experimental group consisted of five Nu/Nu BALB/c mice. Cells were lyzed with radioimmunoprecipitation assay buffer (Cell Signaling Technology) containing protease and phosphatase inhibitor (Roche, Indianapolis, IN). The antibodies were: β-tubulin, β-actin, FAK, P-FAK (Y397), ERK, P-ERK (Thr202/204), MEK, and P-MEK (Ser217/221) (Cell Signaling Technology), and Pan-CK (Zhongshan Golden Bridge Bio-technology, Beijing, China). The 24-well plate was first coated with fibronectin and washed twice with PBS before plating cells. MME- or vector-transfected cells were digested to prepare suspensions containing 2 × 105 cells/mL in Dulbecco's modified Eagle's medium + 3% fetal bovine serum. The cell suspension (500 μL) was added and allowed to attach for 2, 4, or 8 hours at 37°C. Unattached cells were then gently removed. Ultimately, the cells were stained by crystal violet and observed under the microscope. All of the experiments were repeated three times in duplicate wells.14Ponda M.P. Breslow J.L. Serum stimulation of CCR7 chemotaxis due to coagulation factor XIIa-dependent production of high-molecular-weight kininogen domain 5.Proc Natl Acad Sci U S A. 2016; 113: E7059-E7068Crossref PubMed Scopus (10) Google Scholar To assess the potential contribution of GTPases to MME-induced inhibition of ESCC cell motility, the levels of active RhoA in cells were measured by GST-rhotekin pulldown assays according to the manufacturer's directions (Millipore, Denver, CO).15Liebl J. Weitensteiner S.B. Vereb G. Takács L. Fürst R. Vollmar A.M. Zahler S. Cyclin-dependent kinase 5 regulates endothelial cell migration and angiogenesis.J Biol Chem. 2010; 285: 35932-35943Crossref PubMed Scopus (81) Google Scholar In brief, MME- or vector-transfected cells were selected by puromycin and then starved in medium without fetal bovine serum overnight before treatment. Cells were harvested in Mg2+ lysis buffer, clarified by centrifugation. Fifteen to 23 μL (20 to 30 μg) of the Rho Assay Reagent slurry were added to the cell extract. After washing 3 times, the agarose beads were resuspended in 25 μL of 2 × Laemmli reducing sample buffer, then boiled for 5 minutes. Activated RhoA of equal amounts of total protein was detected by Western blotting in each sample. All steps were performed at 4°C or on ice to reduce hydrolysis. Samples were also analyzed to verify that MME knockdown did not have secondary effects on RhoA expression. After normalization of the results to the control samples, the results of three independent experiments were combined and analyzed. All results were presented in forms of means ± SD. One-way analysis of variance tests were applied for comparisons of the statistical differences among more than two groups. Other statistical differences were evaluated via the unpaired t-tests. The χ2 test was performed to evaluate the association between the clinicopathological parameters and MME levels. Survival curves were plotted using Kaplan-Meier analysis and log rank test was adopted. Using the Cox proportional hazards model, the significance of various clinicopathological variables for survival was analyzed. P-value of less than 0.05 was considered significant. The TMA was stained with MME and the correlation analysis of MME expression with clinicopathological features was performed with 212 informative ESCC tumor cases. Noninformative samples were not taken into account: few tumor cells in the sample, unrepresentative samples or lost samples. The average staining index of MME in nontumor tissues was 5.2 and the down-regulation of MME was determined as SI < 5. In 212 valid ESCC tumor tissues, as the results showed, 86 cases (40.5%) showed down-regulation of MME (Figure 1A ). The correlation between MME level and clinicopathological features of patients with ESCC was studied next. MME down-regulation was significantly associated with lymph node metastasis (Pearson χ2 test, P < 0.05), advanced clinical stage (Pearson χ2 test, P < 0.05) (Table 2). Kaplan-Meier survival analysis revealed that patients with MME down-expression underwent poorer overall survival (P < 0.01) (Figure 1B). Univariate and multivariate survival analyses were also achieved using the Cox regression model. Univariate analysis demonstrated that down-regulation of MME (P < 0.01), along with poor differentiation (P = 0.013) and lymph node metastasis, was considered a significant prognostic factor for overall survival (Table 3). Multivariate analysis taking into account the effects of all variables possible also proved that for ESCC patient overall survival, MME down-regulation was an independent prognostic factor.Table 2Correlation of MME Expression with Clinicopathological Features in ESCC PatientsFeaturesTotalMME down-regulation∗MME down-regulation was defined as TSI < 5.P valueSex, n (%) Male11350 (44.2%)0.152 Female9936 (36.4%)Age in years, n (%) ≤ 609536 (37.9%)0.284 > 6011750 (42.7%)Differentiation, n (%) Well/moderate18971 (40.7%)0.535 Poor2315 (39.1%)Tumor stage, n (%) I–II11941 (34.4%)0.040 III–IV9345 (48.4%)Lymph node metastasis, n (%) N011639 (33.6%)0.024 N19647 (49.0%)∗ MME down-regulation was defined as TSI < 5. Open table in a new tab Table 3Univariate and Multivariate Analysis of Different Prognostic Variables in Patients with ESCCFeaturesUnivariate analysis∗Cox regression model.Multivariate analysis∗Cox regression model.HR95% CIP valueHR95% CIP valueSex1.1400.809–1.6060.454NANANAAge1.3010.922–1.8370.134NANANADifferentiation0.4030.197–0.8240.0130.3650.178–0.7500.006LN metastasis2.1891.547–3.097<0.0012.0651.456–2.927<0.001MME down-regulation0.3480.246–0.493<0.0010.3650.256–0.518<0.001HR, hazard ratio; LN, lymph node; NA, not applicable.∗ Cox regression model. Open table in a new tab HR, hazard ratio; LN, lymph node; NA, not applicable. MME protein was detected in nine ESCC cell lines and NE1 (an immortalized normal esophageal epithelial cell line) using Western blotting. MME showed frequent down-regulation in the majority of ESCC cell lines in comparison with NE1 (Figure 2A). EC109 and KYSE510 cells with MME stable expression were established by lentiviral transduction (Figure 2B). Foci formation results showed that MME overexpression in EC109 could mildly decrease the tumorgenicity, whereas in KYSE510, it has no significant influence on cell growth in vitro and in vivo (Supplemental Figure S1). Wound-healing assay showed that MME-transfected cells obtained slower migration rate compared with the control cells (Figure 2C). Cells' ability to migrate and invade significantly decreased with MME overexpression (P < 0.01) (Figure 2D). MME was also stably knocked down in KYSE180 and KYSE410 using shRNAs. Gene silencing was confirmed at both the RNA and protein level (Figure 3, A and B ). In vitro functional assays also found silencing MME could enhance wound healing closure (Figure 3C), cell migration, and invasion ability (Figure 3D). It was next investigated whether MME had an effect on cell adhesion to the matrix as it is an important stage for cancer cell metastasis. Cell–cell adhesiveness is generally reduced in human cancers, but cell–matrix junctions that form firm adhesions increased.16Parsons J.T. Horwitz A.R. Schwartz M.A. Cell adhesion: integrating cytoskeletal dynamics and cellular tension.Nat Rev Mol Cell Biol. 2010; 11: 633Crossref PubMed Scopus (1354) Google Scholar, 17Corey S.J. Yu J. Cancer cell migration: when red light switched to green.Asian J Androl. 2011; 13: 177-178Crossref PubMed Scopus (1) Google Scholar The results demonstrated that MME overexpression significantly decreased cell adhesion to the plate (Figure 4, A and B ), which is not good for tumor cell metastasis. Because previous studies reported that MME could inhibit FAK phosphorylation, and FAK happens to be an essential nonreceptor tyrosine kinase regulating cell adhesive signaling and migrating,18Sumitomo M. Shen R. Walburg M. Dai J. Geng Y. Navarro D. Boileau G. Papandreou C.N. Giancotti F.G. Knudsen B. Nanus D.M. Neutral endopeptidase inhibits prostate cancer cell migration by blocking focal adhesion kinase signaling.J Clin Invest. 2000; 106: 1399-1407Crossref PubMed Scopus (124) Google Scholar, 19Sulzmaier F.J. Jean C. Schlaepfer D.D. FAK in cancer: mechanistic findings and clinical applications.Nat Rev Cancer. 2014; 14: 598-610Crossref PubMed Scopus (852) Google Scholar phosphorylated FAK protein level decreased in the MME overexpressing cells (Figure 4C). The results were further confirmed using immunofluorescence, which showed that the immunoreactivity against phosphorylated FAK was markedly weakened compared with the vector control, both in EC109 and KYSE510 (Figure 4D). The protein levels of MME and phosphorylated FAK were also detected in ESCC tumor tissues and paired nontumor tissues. Reciprocal correlation between MME and phosphorylated FAK was observed in 6 of 10 ESCC cases (Figure 4E). Recent studies proved that FAK is responsible for RhoA activity.20Schmidt T.T. Tauseef M. Yue L. Bonini M.G. Gothert J. Shen T.-L. Guan J.-L. Predescu S. Sadikot R. Mehta D. Conditional deletion of FAK in mice endothelium disrupts lung vascular barrier function due to destabilization of RhoA and Rac1 activities.Am J Physiol Lung Cell Mol Physiol. 2013; 305: L291-L300Crossref PubMed Scopus (41) Google Scholar It is indicated that RhoA activity has close associations with the actin reorganization, namely stress fiber formation, in epithelial cells, which also plays a vital part in cell metastasis. The RhoA activity assay indicated that the MME overexpression dramatically suppressed the activity of RhoA in EC109 and KYSE510 cells, whereas silencing MME increased GTP-RhoA in KYSE410 and KYSE180 cells (Figure 5A). As is well known, the arrangement of the actin cytoskeleton can significantly influence cell morphology. This characteristic offers us insight into further discovering the signaling pathway of migrating or attached cells. With fluorescein isothiocyanate–phalloidin staining, the stress fiber formation was found to be decreased in MME overexpressed cells compared with the control cells (Figure 5B). The downstream pathway signaling of FAK-RhoA was next studied, and it was found that phosphorylated Erk1/2 (Thr202/204) and MEK1/2 (Ser217/221) decreased in MME overexpressed EC109 and KYSE510 cells (Figure 5C). The reverse validation was also performed in MME silenced cells (Figure 5D). Because in vitro assays showed that MME could suppress ESCC cell motility, animal pulmonary metastasis experiments were performed by injecting MME-transfected EC109 cells into Nu/Nu BALB/c mice via the tail vein. Tumor nodules that formed in the lungs were counted, and the number of nodules deceased significantly in the MME-EC109 group (P < 0.05) (Figure 6, A and B ). Similar results were observed in the KYSE510-MME cell group (Figure 6C). EC109 derivative cells (MME and vector control) were inoculated via footpads of the mice to establish the lymph node metastasis animal models. Two months later, the popliteal lymph nodes were isolated and fixed for further examination (Figure 6D). Hematoxylin and eosin, and human CK staining results demonstrated that tumor cells were observed in 5 of 5 in the vector group versus only 2 of 5 in the MME group (Figure 6E). Tumor metastasis is an important cause of human cancer-related deaths.21Rankin E.B. Giaccia A.J. Hypoxic control of metastasis.Science. 2016; 352: 175-180Crossref PubMed Scopus (755) Google Scholar, 22Massague J. Obenauf A.C. Metastatic colonization by circulating tumour cells.Nature. 2016; 529: 298-306Crossref PubMed Scopus (1115) Google Scholar The poor clinical outcome of ESCC is mainly attributed to its nature to metastasize and invade easily. It has been reported that skipped metastasis was observed in 60% of early esophageal cancer patients,23Cho J.W. Choi S.C. Jang J.Y. Shin S.K. Choi K.D. Lee J.H. Kim S.G. Sung J.K. Jeon S.W. Choi I.J. Kim G.H. Jee S.R. Lee W.S. Jung H.-Y. Korean E.S.D.S.G. Lymph node metastases in esophageal carcinoma: an endoscopist's view.Clin Endosc. 2014; 47: 523-529Crossref PubMed Scopus (62) Google Scholar making it urgent to uncover the molecular mechanisms of ESCC metastasis,24Chiang A.C. Massagué J. Molecular basis of metastasis.N Engl J Med. 2008; 359: 2814-2823Crossref PubMed Scopus (800) Google Scholar and identify and characterize molecules that are responsible for ESCC metastasis. MME is a cell surface zinc metalloprotease. It is widely expressed in a variety of normal cell types. First identified as an oncogene in leukemia,4Barker P.E. Shipp M.A. D'Adamio L. Masteller E.L. Reinherz E.L. The common acute lymphoblastic leukemia antigen gene maps to chromosomal region 3 (q21-q27).J Immunol. 1989; 142: 283-287PubMed Google Scholar it has been studied in several different types of cancer.25Maguer-Satta V. Besancon R. Bachelard-Cascales E. Concise review: neutral endopeptidase (CD10): a multifaceted environment actor in stem cells, physiological mechanisms, and cancer.Stem Cells. 2011; 29: 389-396Crossref PubMed Scopus (134) Google Scholar It seems that the function of MME in cancer is cell-type–dependent: In colorectal cancer patients, tumoral MME expression has been substantially associated with liver metastasis.26Fujimoto Y. Nakanishi Y. Sekine S. Yoshimura K. Akasu T. Moriya Y. Shimoda T. CD10 expression in colorectal carcinoma correlates with liver metastasis.Dis Colon Rectum. 2005; 48: 1883-1889Crossref PubMed Scopus (58) Google Scholar MME-positive expression correlates with tumor progression in pancreatic endocrine tumor27Deschamps L. Handra-Luca A. O'Toole D. Sauvanet A. Ruszniewski P. Belghiti J. Bedossa P. Couvelard A. CD10 expression in pancreatic endocrine tumors: correlation with prognostic factors and survival.Hum Pathol. 2006; 37: 802-808Crossref PubMed Scopus (44) Google Scholar and malignant melanoma.28Oba J. Nakahara T. Hayashida S. Kido M. Xie L. Takahara M. Uchi H. Miyazaki S. Abe T. Hagihara A. Moroi Y. Furue M. Expression of CD10 predicts tumor progression