Filamin A Affects Melanoma Initiation And Progression Via Regulating Epidermal Growth Factor Receptor Activition

Objective:The initiation and progression of malignant melanoma is a complexphenomenon that affected by many factors, which environmental, genetic, andethnic factors play a crucial role. Data from the clinic, epidemiology andgenetic reveal that melanoma as a heterogeneous tumor, harboring variousgenetic alterations, developing at different body sites and on sun-exposed andnon-sun-exposed regions, suggesting that melanoma arises from divergentcausal pathways. With the development of genomics, understanding themolecular mechanisms of melanoma genesis is helpful to pridect prognosisand choose the optimum therapeutics.Epidermal growth factor (EGF) and Epidermal growth factor receptor(EGFR) is the critical contributor that modulates growth, differentiation,migration and survival in normal or abnormity cells. Homo-orhetero-dimerization of ligand-activated EGFR results in receptor tyrosinekinase autophosphorylation and channeling of mitogenic signalspredominantly via the ERK, Akt and JNK pathways. Abnormal activation ofEGFR or inhibition in endocytosis of the ligand-activated receptors isimplicated in a wide range of diseases include oncogenesis. Increasing bodyof evidence revealed that overexpression of EGFR is involed in many kinds oftumors, and correlates with the poor efficacy, high exacerbation and lowsurvival.Filamin A (FLNa) is the first actin filament cross-linking protein identifiedin non-muscle cells. It forms a homo-dimer and cross-links cortical actinfilaments into a dynamic three-dimensional structure. FLNa is known toscaffold over90binding partners, including plasma membrane receptors toregulate cellular functions and processes such as cell shape, cellular motion, and intracellular transport. Oncogenesis is not only associated with persistentactivation and impaired endocytic downregulation of EGFR, but it alsoinvolves a complex network of signaling molecules that interact with FLNa, iscaused by impaired interactions between abnormal FLNa and binding partners.Despite these advances, the mechanistic link between FLNa and the signalingevents associated with malignancy remains elusive.Results from our experiment demonstrated that although the levels ofEGFR expression in FLNa-deficient M2melanoma cells were higher than thatin M2A7cells which is a subclone with stable reintroduction of human FLNa,but the phosphorylated EGFR levels in M2cells were obviously lower thanthat in M2A7cells when stimulated with EGF. These results suggested thatlevels of EGFR expression does not represent the degree of its activation,there may be other factors regulating the activation of EGFR. We can’t justaccording to the quantity of the expression of EGFR to evaluate the prognosisof the tumor in clinic. Exploring new mechanisms of the EGFR activation hasimportant theoretical and practical value.In this study, we measured the changes in proliferation, migration andinvasion of two human melanoma cell lines of distinct origins after alterationin FLNa expression. The proliferative properties of FLNa were then exploredin a xenotransplantation melanoma tumour model in nude mice. In a secondseries of experiments, the contribution of FLNa in ligand-mediated activationof EGFR and its downstream signaling pathway was investigated both in vitroand in vivo. Finally, immunohistochemical analysis was performed onsections of malignant melanoma tumours to determine whether an associationexisted between FLNa expression and overall survival of these cancer patients.It may contribute to explaining the relation between FLNa and mechanism ofcarcinogenesis and progression in malignant melanoma, provide theoreticalevidence for whether FLNa is a reference factor in pathological diagnosis anda new target for molecular therapy of malignant melanoma as well.Methods:1Effect of FLNa on proliferation, migration and invasion in human malignant melanoma cell linesThe effects of FLNa on proliferation, migration and invasion wereevaluated in two human melanoma cell lines of distinct origins:(1) theFLNa-deficient M2melanoma cells and M2A7cells, a subclone with stablereintroduction of human FLNa, and (2) UACC647melanoma cells. UACC647melanoma cells were stably transfected either with a plasmid expressingshRNA against FLNa or control shRNA and selected with puromycin,resulting in UACC647(KD) and UACC647(ctrl) cell lines. The knockdownefficiency of FLNa was measured by Western blot assay.1.1Measure effect of FLNa on proliferation in human melanoma cells in vitroMTT assay was used to measure the role of FLNa in cell viability in vitro.Cells were seeded in96-well plates and serum-starved for4h, untreated ortreated with various concentrations of EGF (4,20,100nM) for44h. MTTwas added and cells were continue to incubate for4h. At the end ofexperiment, DMSO was added and the absorbance was measured at570nm.Colony formation assay was used to evaluate the effect of FLNa inanchorage-dependent cell growth potential in vitro. Cells were seeded in35mm dishes and incubated in medium containing1%or10%fetal bovineserum (FBS) with or without addition of EGF for two weeks. The cellcolonies were stained with hematoxylin and counted under a microscope.Soft agar assay was used to evaluate the effect of FLNa inanchorage-independent cell growth in vitro. Cells were suspended in culturemedium containing0.35%low melting agarose, and plated onto solidified0.6%agar containing culture medium in six-well plates at a density of300cells per well. Cells were incubated in medium containing1%or10%FBSwith or without addition of EGF for two weeks. The number of colonies weremanually counted under a microscope.1.2Wound-healing assay was used to measure the role of FLNa in humanmelanoma cell migration in vitroMelanoma cells were seeded in24-well plates and serum-starved for4h.The monolayer was scratched with a sterile pipette tip and washed with serum-free medium (SFM) to remove floating cells. Following addition ofvehicle or20nM epidermal growth factor, cells were photographed at thesame field every2h until closure of the scratch.1.3Transwell assay was used to measure the role of FLNa in humanmelanoma cell invasion in vitroSerum-starved cells were placed in the top chamber of transwell invasionchambers. The lower chamber was filled with MEM or1640supplementedwith10%FBS. After a24h incubation period, noninvasive cells wereremoved from the upper surface of the transwell membrane with a cottonswab, and invasive cells on the lower membrane surface were fixed withmethanol, stained with H&E, photographed, and counted under a microscope.2Effect of FLNa on EGF-EGFR signal pathway in human malignantmelanoma cell linesSerum-starved cells were left untreated or treated with EGF (20nM) for5,10, and30min, after which cells were lysed with cold lysis buffer. Afterdetermining the protein concentration, equal amount of proteins wereseparated on SDS-PAGE and transferred onto polyvinylidene difluoridemembranes. The membranes were blocked in5%nonfat milk or3%bovineserum albumin, incubated with various primary antibodies that includingFLNa, EGFR, phosphotyrosine (clone4G10), ERK1/2, phospho-ERK1/2and-actin, followed by incubation with HRP-conjugated secondary antibodies,and visualization with enhanced chemiluminescence (ECL) detection reagents.Signals were quantitated by densitometry using the ImageJ software.Cells were lysed in immune precipitation buffer. The clarified lysates wereincubated with mouse anti-EGFR antibodies and non-immune IgG followedby the addition of protein G-agarose. Total cell extracts andimmunoprecipitated complexes were subjected to Western blotting.3Effect of FLNa in vivo xenograft model for tumor growthUACC647(ctrl) and UACC647(KD) cells were inoculated subcutaneouslywithin the right thigh region to generate tumors in five weeks old femaleBALB/c nude mice. Tumor size was measured with calipers twice weekly. Mice were sacrificed4weeks after tumor inoculation. Tumor tissue wasweighted and photographed. Lysates from frozen tissues were separated onSDS-PAGE and immunoblot analyses were carried out. Immunostaining wasperformed on5-m formalin-fixed, paraffin-embedded tissue sections usingan immunoperoxidase method with mouse anti-FLNa monoclonal antibodyand rabbit anti-Ki-67monoclonal antibody. Two pathologists blinded to theexperiments assessed the extent and intensity of immunostaining.Semi-quantitative scoring for FLNa was performed using the H-score method.The percentage of nuclear Ki-67immunoreactivity was calculated bydetermining the average expression of positive Ki-67signal in500nuclei.4Immunohistochemical (IHC) staining of FLNa and Ki-67in tissueThirty cases of malignant melanoma tissues were obtained from the surgicalpathology files at Bethune International Peace Hospital with the approvalfrom the Institutional Review Board. FLNa and Ki-67immunostaining wasperformed using a procedure similar to that indicated in the experiments innude mice.Results:1FLNa Promotes the Proliferation, Migration and invasion in HumanMelanoma Cell LinesStable knockdown of FLNa gene was carried out in UACC647cells tocreate the UACC647(KD) cells. These cells were found to have a70%reduction in FLNa protein levels as compared either with UACC647cellstransfected with control shRNA (ctrl) or mock-transfected cells.In MTT assay, M2A7cells exhibited a higher cell viability than M2cellswhen challenged with4-100nM EGF, but not in basal conditions.Knockdown of FLNa in UACC647(KD) cells was accompanied by significantreduction in basal and EGF-stimulated cell growth as compared toUACC647(ctrl) cells. In colony formation assay and soft agar assay, M2A7and UACC647(ctrl) cells with sufficient FLNa expression exhibited a higherrate of colony formation than the FLNa-deficient M2and UACC647(KD)cells in EGF stimulated. Using wound-healing assays, it was determined that the ability of migrationin FLNa-expressing M2A7and UACC647(ctrl) cells was significantlystronger than FLNa-deficient M2and UACC647(ctrl) cells treated oruntreated with EGF. The relative width index was0in M2A7after12h inEGF stimulated, while the relative width index in M2cell was0.58±0.03. Therelative width index was0in UACC647(ctrl) cells after15h in EGFstimulated, while the relative width index in UACC647(KD) cell was0.51±0.01.In Transwell assays, M2A7cells had a1.9-fold higher invasive potential inMatrigel invasion assays than M2cells (P <0.01). Similarly, there was a1.8-fold increase in invasive potential of UACC647(ctrl) cells as compared toUACC647(KD) cells (P <0.01).2Role of FLNa in ligand-induced EGFR signal pathway in human melanomacellsAs anticipated, the lack of FLNa rendered M2cells somewhat unresponsiveto EGF-induced phosphorylation of EGFR, even though these cells expressednoticeably higher levels of total EGFR as compared with M2A7cells. As aresult, the pY-EGFR to total EGFR ratio was significantly higher inEGF-stimulated M2A7cells. Similar results were observed when anti-EGFRimmunoprecipitates were subjected to Western blot analysis using4G10. Asanticipated, pERK1/2levels were found to be significantly higher in M2A7cells as compared to M2cells upon EGF stimulation. These experiments wererepeated using UACC647(ctrl) and UACC647(KD) cells and the resultsindicated that knockdown of FLNa was accompanied by sharp reduction inEGF responsiveness when looking at pY-EGFR and pERK1/2levels. Thephosphorylated to total EGFR protein ratio was70%lower in UACC647(KD)cells as compared with UACC647(ctrl) cells after a10min stimulation withEGF. Immunoprecipitation experiments were carried out to further confirmthe differential tyrosine phosphorylation of EGFR in UACC647(ctrl) cells vs.UACC647(KD) cells. The depletion of FLNa in UACC647(KD) cells wasaccompanied by a significant1.4-fold reduction in EGF-dependent increase in pERK1/2levels when compared to UACC647(ctrl) cells. It would appear,therefore, that FLNa enhances EGF-induced phosphorylation of EGFR andactivation of the Raf-MEK-ERK cascade.3FLNa increases human melanoma cell proliferation in nude mice xenograftmodelThe in vivo proliferative properties of FLNa were tested after subcutaneousimplantation of UACC647(ctrl) and UACC647(KD) cells in nude mice. Asignificant reduction in melanoma tumor growth was observed inFLNa-deficient UACC647(KD) cells as compared to UACC647(ctrl) cells.Immunohistochemical analysis illustrated that the diffuse and strong nuclearstaining of Ki-67correlated with strong FLNa cytoplasmic signal in tumoursfrom UACC647(ctrl) cells. In contrast, tumor tissues from UACC647(KD)cells exhibited weak staining of Ki-67and low levels of FLNa staining. TheH-score for FLNa was64.33±10.59, and a significant positive correlationwas detected between the immunoreactivity for FLNa and Ki-67(r=0.88; P <0.01). Western blot analysis revealed a significant attenuation in pY-EGFRand pERK1/2levels in UACC647(KD) cells by80%and46%, respectively.These data support the notion that knockdown of FLNa decreases proliferationand melanoma tumor growth both in vitro and in vivo.4Reduction of FLNa predicts better survival in malignant melanomaBased on the scoring by two independent pathologists, the H-score value(mean±SEM) for FLNa in30cases was69.77±5.17. There were significantpositive correlations between the immunoreactivity for FLNa and Ki-67(r=0.60; P <0.01). There was an inverse correlation between FLNaimmunoreactivity and over survival (r=-0.45; P=0.013). Gather the casesbased on the median value of H-score, patients with high FLNa expression intheir tumours had a markedly lower survival (median survival was594days)than those with low FLNa content (median survival was1949days; P=0.0011). However, in this study there was no significant relationship betweenFLNa immunoreactivity and patient age, sex and recurrence or metastases. Conclusions:1Overexpression or knockdown of FLNa in human malignant melanomacells promote or inhibit proliferation, migration and invasion in melanomacells in vitro.2Knockdown of FLNa in human malignant melanoma cell inhibit tumorgrowth in vivo.3FLNa may promot the growth of malignant melanoma cells via enhancingEGF-induced EGFR activation (phosphorylation) and downstream signaltransduction.4Expression of FLNa in malignant melanoma tissues was positivecorrelated with the expression of Ki-67, while correlated inversely with oversurvival in malignant melanoma. Low level of FLNa predicts better survival inmalignant melanoma.