Alpha-Actinin 4 Is Associated with Cancer Cell Motility and Is a Potential Biomarker in Non–Small Cell Lung Cancer

Original Article

Alpha-Actinin 4 Is Associated with Cancer Cell Motility and Is a Potential Biomarker in Non–Small Cell Lung Cancer Ming-Chuan Wang, PhD,* Ying-Hua Chang, PhD,* Chih-Chieh Wu, PhD, * Yu-Chang Tyan, PhD,†‡§ Hua-Chien Chang, MS,* Yih-Gang Goan, MD,║ Wu-Wei Lai, MD,¶ Pin-Nan Cheng, MD,# and Pao-Chi Liao, PhD*

Background: Differential expression and secretion of alpha-actinin 4 (ACTN4) in the lung cancer cell lines CL1-0 and CL1-5 have been reported in previous proteomic studies. The aim of this study is to investigate the functional properties of the ACTN4 protein in non–small-cell lung cancer (NSCLC) cells and evaluate its clinical importance. Methods: We used RNA interference to knock down and overexpress ACTN4 protein to evaluate the effects of this intervention on cancer cell invasion and migration, as well as on microscopic cellular morphology. Furthermore, we examined by immunohistochemistry the expression of ACTN4 protein in tissue samples at different stages of lung cancerand compared the protein levels of ACTN4 in blood plasma samples from patients with histologically confirmed lung cancer and healthy controls. Results: CL1-5 cell motility was significantly suppressed by the knockdown of ACTN4 protein. The morphology of CL1-5 cells changed from a predominantly mesenchymal-like shape into a globular shape in response to ACTN4 protein knockdown. A quantitative immunohistochemical assessment of lung cancer tissues revealed that ACTN4 protein level was considerably higher in cancerous tissues than in the adjacent normal ones, and the area under the receiver operating characteristic curve was 0.736 (p < 0.001). According to an *Department of Environmental and Occupational Health, College of Medicine, National Cheng Kung University, 138 Sheng-Li Road, Tainan 704, Taiwan.†Department of Medical Imaging and Radiological Sciences, Kaohsiung Medical University, 100 Shi-Chuan 1st Road, Kaohsiung 807, Taiwan.‡ Translational Research Center, Kaohsiung Medical University Hospital, 100 Tzyou 1st Road, Kaohsiung 807, Taiwan.§ Institute of Medical Science and Technology, National Sun Yat-Sen University, 70 Lienhai Road, Kaohsiung 804, Taiwan.║ Division of Thoracic Surgery, Kaohsing Veterans General Hospital, 386 Ta-Chung 1st RD. Kaohsiung 813, Taiwan.¶Department of Surgery, College of Medicine, National Cheng Kung University Hospital, National Cheng Kung University, 138 Sheng-Li Road, Tainan 704, Taiwan.#Department of Internal Medicine, College of Medicine, National Cheng Kung University Hospital, National Cheng Kung University, 138 Sheng-Li Road, Tainan 704, Taiwan. Disclosure: The authors have no conflicts of interest to disclose. This study was supported by grants from the National Science Council, Taiwan (NSC100-2325-B-006-004, NSC101-2325-B-006-003, and NSC102-2325-B-006-003). Address for correspondence: Pao-Chi Liao, PhD, Department of Environmental and Occupational Health, College of Medicine, National Cheng Kung University, 138 Sheng-Li Road, Tainan 704, Taiwan. E-mail: [email protected] DOI: 10.1097/JTO.0000000000000396 Copyright © 2014 by the International Association for the Study of Lung Cancer ISSN: 1556-0864/15/1002-0286

286

enzyme-linked immunosorbent assay, the plasma levels of ACTN4 protein were significantly different between cancer patients and healthy controls, and the areas under the receiver operating characteristic curves were 0.828 and 0.909, respectively, for two independent cohorts (p < 0.001). Conclusions: We demonstrate that the knockdown of ACTN4 protein inhibited cell invasion and migration. These results suggest that ACTN4 is associated with lung cancer cell motility. Thus, the level of ACTN4 in cancerous tissue and plasma is related to the presence of lung cancer. Key Words: ACTN4; non–small-cell lung cancers; CL1-0 and CL1-5 cell lines; siRNA (J Thorac Oncol. 2015;10: 286–301)

L

ung cancer is the most common cancer in the world and has been the malignancy with the highest mortality for several decades. Non–small-cell lung cancer (NSCLC) accounts for approximately 85% of all lung cancers.1 Early NSCLC is treated with surgery, with survival rates of 35% to 85% depending on tumor size and stage.2 However, most NSCLC patients are asymptomatic, and cancers are detected late. Moreover, the overall 5-year survival rates of patients with NSCLC are determined by the cancer stages, ranging from 49% for stage IA to only 1% for stage IV.3 The exceedingly low survival rates and increased mortality from cancers are believed to be due to cancer metastasis.4 Numerous reports have shown that cancer cell–secreted proteins are related to cancer metastasis; these factors aid in cell migration/invasion and the modulation of the microenvironment.5,6 From previous comparative secretomic analyses of two NSCLC cell lines, CL1-0 cells (which have a low invasive ability) and CL1-5 cells (which have a high invasive ability),7 ACTN4 protein has been identified as a relatively abundant secreted candidate protein in CL1-5 cells.8–10 ACTN4 protein is a member of the cytoskeletal protein family; it binds to actin filaments to preserve cytoskeletal structure and cell morphology.11–13 ACTN4 is expressed in nonmuscle cells and is commonly associated with focal adhesion contacts and migrating cells.14 The structure of the ACTN4 protein contains conserved functional domains, including an N-terminal actin-binding domain with two highly conserved calponin homology domains, a central domain consisting of four spectrin repeats, and a calmodulin-like domain in the C-terminus.14,15

Journal of Thoracic Oncology  ®  •  Volume 10, Number 2, February 2015

Journal of Thoracic Oncology  ®  •  Volume 10, Number 2, February 2015

Cancer cell migration in the extracellular matrix (ECM) requires that cell protrusions of lamellipodia, pseudopodia, or filopodia are formed at the leading edge of the cancer cell. The protrusion of these cells requires actin polymerization and rearrangement.16,17 Following the formation of protrusions in cancer cell membranes, the cancer cells develop new focal adhesion complexes and secrete proteins for local ECM proteolysis. Eventually, the cancer cells release their contacts at the rear and initiate migration.18 Therefore, ACTN4 may play an important role in cancer cell motility and may modulate cytoskeletal organization by cross-linking actin filaments. In the current study, we utilized small interfering RNA (siRNA)–mediated ACTN4 protein knockdown, combined with migration and invasion assays, to examine the function of ACTN4 in CL1-5 cells. To investigate whether ACTN4 has the potential to be a useful biomarker for lung cancer, a tissue microarray (TMA) was used to determine the expression of ACTN4 protein at different stages of cancer in lung tissues and in adjacent normal tissues. In addition, a sandwich enzymelinked immunosorbent assay (ELISA) was applied to measure the plasma concentration of ACTN4 protein in patients at different stages of lung cancer and in healthy controls.

MATERIALS AND METHODS Cell Lines The poorly differentiated CL1-0 and CL1-5 cells were provided by Dr. P.C. Yang (Department of Internal Medicine, National Taiwan University Hospital, Taiwan). The maintenance of CL1-0 and CL1-5 cells, including standardization of the media and growth conditions, was performed as described previously.7,9 The lung cancer cell line A549 was purchased from the American Type Culture Collection and maintained in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum (FBS) (Gibco BRL, Life Technologies, Carlsbad, CA) and 1% penicillin (Invitrogen, Life Technologies, Carlsbad, CA).

Conditioned Medium and Cell Extracts CL1-0 and CL1-5 cells were cultured at 37°C with 5% carbon dioxide in RPMI-1640 medium supplemented with 10% FBS. The cells were grown to 70% confluence and washed four times with serum-free medium before they were incubated in serum-free medium for 24 hours. The collected conditioned medium (CM) was concentrated by centrifugation in Amicon Ultra-15 tubes (3-kDa cutoff; Millipore, Billerica, MA). The cell extracts (CEs) were also collected in parallel as follows. The cultured cells were washed with phosphatebuffered saline and trypsinized (Biowest, Nuaillé, France). The collected cells were resuspended in distilled water and lysed by microsonication at 4°C. The suspension was centrifuged to remove any cell debris. The Bradford assay (Bio-Rad, Hercules, CA) was used to determine the total protein concentration in the prepared CM and CE samples.

Protein Electrophoresis and Western Blot

Aliquots (10 μg) of protein from the CM and CE samples were individually separated on NuPAGE 4%–12%

Functional Properties of the ACTN4 Protein in NSCLC

Bis–Tris gels (Invitrogen) using a Novex Mini-Cell electrophoresis system (Invitrogen). The proteins were transferred to an iBlot polyvinylidene fluoride membrane (Invitrogen) according to the manufacturer’s protocol. The transferred membrane was blocked in nonfat milk solution. The membranes were separately probed with an anti-ACTN4 antibody (1:1000; Epitomics, Burlingame, CA), an anti-PGAM1 antibody (1:1000; Epitomics), or an anti-actin antibody (1:1000; GeneTex, San Antonio, TX). The membranes were then washed with Tris-buffered saline and Tween-20 and incubated with the appropriate secondary antibodies conjugated to horseradish peroxidase (HRP) (1:100,000; Sigma-Aldrich). The membranes were washed and developed using an enhanced chemiluminescence reagent (PerkinElmer, Waltham, MA) and photographed using a UVP BioSpectrum imaging system (UVP, Upland, CA).

Small Interfering RNA-–Mediated ACTN4 Protein Knockdown The two siRNAs targeting ACTN4 protein were purchased from Santa Cruz Biotechnology (Santa Cruz, CA) and Life Technologies (Carlsbad, CA). The sequences for ACTN4 siRNA knockdown comprised two pools of RNA sequences; pool I of ACTN4 siRNA contained (1) 5′-CAGAGCUGAUUGAGUAUGA-3′, (2) 5′-CACUCUGUAUCUAUGCAAA-3′, and (3)  5′-CUGG UCUGGUAAAUAUGUA-3′ (Santa Cruz); and pool II of ACTN4 siRNAs sequences contained (1) 5′-GACCAGAGCU GAUUGAGUT-3′ and (2) 5′-CGCAAAUCAUCAACUCC AA-3′ (Life Technologies). The siRNA transfection process was performed using the different cell lines according to the manufacturer’s instructions. Briefly, the CL1-5 cells (2 × 105 cells) were seeded in a six-well culture plate. For each transfection, the final siRNA targeting ACTN4 (30 pM siRNA) or a scrambled control siRNA was added along with siRNA transfection reagent (Santa Cruz). The reactions were mixed gently, overlaid onto the cells, and incubated for 24–48 hours.

Transient Transfection for ACTN4 Protein Overexpresssion in CL1-0 Cells ACTN4 gene overexpression in CL1-0 cells was achieved through transfection with human ACTN4 DNA vector (Cat No. RC202873; OriGene Technologies, Inc., Rockville, MD) and pSV-β-Galactosidase control vector (Promega, MA) using the Fugene6 transfection reagent according to the manufacturer’s instructions (Promega). The CL1-0 cells (1 × 105 cells) were seeded in a six-well culture plate and grown to 50%–70% confluence. Fugene 6 transfection reagent (3 μl) was directly added into serum-free medium (97 μl), and then ACTN4 DNA vector (1 μg) was added into the prepared transfection medium for each ACTN4 DNA vector transfection. The transfection reagent and DNA mixture was added into cultured cells. The reactions were incubated at 37°C in the presence of 5% carbon dioxide for 48 hours. The collected cells were washed in distilled water and lysed by microsonication at 4°C. The suspension was centrifuged to remove any cell debris. The Bradford assay was used to determine the total protein concentration. Total proteins (20 μg) from the collected CEs

Copyright © 2014 by the International Association for the Study of Lung Cancer

287

Wang et al.

Journal of Thoracic Oncology  ®  •  Volume 10, Number 2, February 2015

of CL1-0 cells transfected with ACTN4 DNA vector or control vector were individually separated on NuPAGE 4%–12% Bis–Tris gels and Western blotting was performed.

3-[4,5-Dimethylthiazol-2-yl]-2,5Diphenyltetrazolium Bromide Assay The indicated cells (1 × 103 cells per well) were seeded in a 96-well plate and treated with either ACTN4 siRNA or a scrambled control for 24 hours. 3-[4,5-Dimethylthiazol-2-yl]2,5-diphenyltetrazolium bromide (MTT; 10 μl of a 500 μg/ml solution) was added to each well and incubated for 2 hours. After incubation, to each well was added 100 μl detergent reagent (10% Triton X-100 in isopropanol) and incubated at room temperature in the dark for 2 hours. The viability of each cell line was monitored every 24 hours, for 72 hours. The cells were then lysed with dimethyl sulfoxide. A microplate photometer was used to measure the absorbance at 595 nm (Multiskan EX microplate reader; Thermo Fisher Scientific Ltd., Waltham, MA). The results were obtained by culturing the cells under each condition in six different wells in three independent experiments.

Bromodeoxyuridine Assay Bromodeoxyuridine (BrdU) assays to measure cell proliferation were performed according to the manufacturer’s instructions (Roche Applied Science, Mannheim, Germany). Briefly, lung cancer cells (1 × 103 cells per well) were seeded in each well of a 96-well plate. The BrdU reagent (10 μl of 100 μM BrdU) was added to each well and incubated for 2 hours at 37°C. After prepared cells were fixed in FixDant (200 μl per well) for 30 minutes and washed, 100 μl of antiBrdU antibody conjugated with HRP (100-fold dilution) was individually added to each well. The developer substrate (100 μl of 1.25 mM tetra-methylbenzidine [TMB] per well) was then added, followed by the stop solution (25 μl of 1M H2SO4). The cell growth within each group was checked every 24 hours for 72 hours. A microplate photometer was used to measure the absorbance at 450 nm (Multiskan EX microplate reader). The results were obtained by culturing the cells under each condition in six different wells in three independent experiments.

Transwell Migration and Matrigel Invasion Assays Transwell membranes (8-μm pore size; BD Biosciences, San Jose, CA) were used for the migration assay. Lung cancer cells (1 × 105) in serum-free RPMI-1640 medium were trypsinized, washed, and loaded into each upper well. The lower chambers were filled with medium containing 10% FBS to induce cell migration. Transwell membranes (8-μm pore size; BD Biosciences) coated with Matrigel basement membrane matrix (BD Biosciences) were used for the cell invasion assay. After 24 hours of incubation, the filters were fixed with methanol and then stained with Giemsa stain (Sigma-Aldrich). The migrating and invading cell numbers were evaluated in six randomly selected fields across three independent experiments under a light microscope (Zeiss Axio Imager A1, Jena, Germany) at 200× magnification.

288

In Vitro Cell Wound Healing Assay Lung cancer cells were seeded on the Ibidi CultureInsert system (Applied BioPhysics, Inc., Troy, NY) for 12 hours. Photographs were taken at the same position in the cellfree gap insert under a light microscope (Zeiss Axio Imager A1) at 100× magnification at the 0-hour and 24-hour time points. The wound closure was analyzed using ImageJ software, and the closure percentage was calculated as follows: wound closure (%) = [area of initial wound scratch − area after cell migration/area of initial wound scratch] × 100%.

Immunofluorescence and Confocal Microscopy Lung cancer cells (1 × 103 cells) cultured on glass coverslips in a six-well culture plate were separately treated with either a scrambled control siRNA or an ACTN4-targeting siRNA for 24 hours. The cultured cells were washed, fixed with 4% paraformaldehyde (prewarmed to 37°C), and permeabilized with 0.1% Triton X-100. The prepared cells were incubated with a primary rabbit anti-ACTN4 immunoglobulin G (IgG; 1:200; Epitomics). After washing, the cells were incubated with secondary goat anti-rabbit IgG antibody conjugated to Alexa Fluor594 (1:1000; Invitrogen). To detect the nuclei and actin filaments, the cells were stained with Hoechst 33258 (Sigma-Aldrich) and phalloidin-iFluor 488 conjugate (1:1000; AAT Bioquest, Sunnyvale, CA). The immunofluorescent cells were examined using a scanning confocal microscope (FV-1000; Olympus, Hamburg, Germany) or a fluorescence microscope (Zeiss Axio Imager A1).

Collection of Clinical Samples The samples of tissue (n = 84) and plasma (n = 185) of patients with lung cancer were provided by the Tissue Bank of National Cheng Kung University (NCKU) Hospital, and plasma samples from an independent cohort of lung cancer patients (n = 145) was obtained from the Kaohsing Veterans General Hospital (KVGH). The plasma samples of healthy controls (n = 147) with no history of cancer were collected from the Health Examination Center of the NCKU Hospital. The collection procedure of clinical samples from the lung cancer patients and healthy controls complied with the regulations of and was approved by Ethics Committee of the NCKU Hospital (Reference Code: ER-100-059, March, 25, 2011). All participants gave their informed consent to participate in this research study and signed the consent forms before participating in this research.

Tissue Microarray The immunohistochemistry of each tissue specimen was examined in duplicate by sampling tumor tissue cores and 4-μm-thick paraffin-embedded sections from the Tissue Bank of NCKU Hospital. The procedures were performed according to the Bond-Max automated immunohistochemistry staining protocol (Leica Biosystems Newcastle Ltd., Newcastle, UK). A rabbit monoclonal anti-ACTN4 IgG (1:250; Epitomics) was used as the primary antibody, and a mouse anti-rabbit IgG conjugated to HRP was used as the secondary antibody (Sigma-Aldrich). The signal was developed

Copyright © 2014 by the International Association for the Study of Lung Cancer

Journal of Thoracic Oncology  ®  •  Volume 10, Number 2, February 2015

with the chromogen 3, 3′-diaminobenzidene (Sigma-Aldrich). Counterstaining was carried out with hematoxylin. Image analysis of the ACTN4-positive cells was further quantified using the TissueFaxs microscopy system with the HistoQuest software module (TissueGnostics, Vienna, Austria).

Enzyme-Linked Immunosorbent Assay Microtiter plates were labeled with a mouse monoclonal anti-ACTN4 IgG (1:100; Abnova, Walnut, CA), washed, and blocked; thereafter, all of the plasma samples were loaded into the prepared microtiter plates. A rabbit polyclonal antiACTN4 IgG (1:50; Proteintech, Chicago, IL) was added into each well. The bound rabbit polyclonal anti-ACTN4 IgG was detected using diluted anti-Rabbit IgG (1:6000; SigmaAldrich) conjugated to HRP. The development was performed by SureBlue Reserve TMB Microwell HRP Substrate and Stop Solution (KPL, Gaithersburg, MD). The absorbance was read on a microplate photometer at 450 nm (Multiskan EX microplate reader).

Statistical Analysis All the data were expressed as the mean ± standard deviation (SD). The Mann–Whitney U and Kruskal–Wallis H tests were used to evaluate the differences in nonparametric groups. Analysis of variance and t tests were used to evaluate the differences in parametric groups. A value of p < 0.05 was considered to be statistically significant. All the above measures and the receiver operating characteristic (ROC) curves were determined using SPSS software (SPSS 18.0, Chicago, IL).

RESULTS Differential Expression and Secretion of ACTN4 Protein in CL1-0 and CL1-5 Cells The CL1-5 cells have a relatively higher invasive ability than the CL1-0 cells after Matrigel invasion selection.7 ACTN4 protein was reported to be one candidate factor that was associated with the invasive ability of the CL1-5 cell line; a differential expression and secretion of approximately twofold had been observed in CL1-5 cells relative to CL1-0 cells in previous label-free mass spectrometry– based proteomic quantification analyses from this laboratory.8–10 A relative intensity quantification of the Western blot data showed that the level of ACTN4 protein in the CE of CL1-5 cells was threefold greater than that in CL1-0 cells, and the level of ACTN4 protein in the CM of CL1-5 cells was sixfold greater than that in CL1-0 cells (Fig. 1A and B). These data indicate a differential expression and secretion of ACTN4 protein in CL1-0 and CL1-5 cells.

Functional Role of ACTN4 Protein in CL1-5 and CL1-0 Cells Two pools of different siRNA sequences were individually transfected and investigated for ACTN4 knockdown in CL1-5 cells to generate fewer off-target effects. Figure 2A shows the effects of ACTN4 knockdown in CL1-5 cells. CL1-5 cells treated with ACTN4 siRNA exhibited only 20%

Functional Properties of the ACTN4 Protein in NSCLC

and 15% of the control levels of ACTN4 expression in the CEs and CMs, respectively. To evaluate the CL1-5 cells after treatment with ACTN4 siRNA or scrambled control for cell migration and invasion abilities, the cell viability (MTT assay) and cell proliferation (BrdU assay) experiments were performed in parallel with transwell, Matrigel, and wound-healing assays. Both MTT reduction assay and BrdU incorporation assay mimicked and reinforced the viability and proliferation ability of CL1-5 cells. The data showed the slope of absorbance increasing, and there were no significant differences between the CL1-5 cells treated with ACTN4 siRNA or scrambled control over time up to 72 hours (Fig. 2B and C). Meanwhile, CL1-5 cells treated with the ACTN4 siRNA or scrambled control were tested in transwell (Fig. 2D), Matrigel (Fig. 2E), and wound-healing assays (Fig. 2F). The data for the transwell and Matrigel assays revealed that the cell number of migrating and invasive CL1-5 cells was considerably decreased (approximately 75%) in 48 hours when the cells were treated with ACTN4 siRNA. In addition, the wound-healing assay showed a wound closure of 84.5% for the CL1-5 cells treated with scrambled control, compared with 35% for the CL1-5 cells treated with the ACTN4 siRNA (Fig. 2F). In addition, A549 lung cancer cells also showed similar data and behavior (Supplementary Fig. S1A–F). In divergent experiments, the effects of ACTN4 overexpression in CL1-0 cells were evaluated. CL1-0 cells were transfected with ACTN4 DNA vector. After 48 hours, the level of ACTN4 protein in the CL1-0 cells was significantly increased (approximately sevenfold) and was greater than that in CL1-0 cells transfected with control vector (Fig. 2G). The functional role of ACTN4 protein in CL1-0 cells was also studied, CL1-0 cells transfected with ACTN4 expression vector and control vector were tested in transwell (Fig. 2H) and Matrigel (Fig. 2I) assays. The data for the transwell and Matrigel assays revealed that the numbers of migrating and invasive CL1-0 cells transfected with ACTN4 expression vector were considerably increased, approximately 2.5- to 3-fold, relative to cells transfected with control vector, in 48 hours. These data suggested that the ACTN4 protein participates in lung cancer cell mobility.

Subcellular Distribution of ACTN4 in CL1-0 and CL1-5 Cells To define the precise localization and function of ACTN4 in CL1-0 and CL1-5 cells, the subcellular distribution of ACTN4 protein was examined by immunolocalization, combined with ACTN4 siRNA and ACTN4 overexpression in CL1-0 cells. The cellular morphology of CL1-5 cells treated with scrambled control was examined microscopically; most of these had a protruding mesenchymal- or spindle-like morphology (approximately 86%), and a small population (14%) had a globular morphology (Fig. 3A, upper panel). Notably, the morphologyof CL1-5 cells independently treated with different pools of ACTN4 siRNA sequences was altered; these CL1-5 cells predominantly had a globular shape (85%), and only a small proportion (approximately 15%) showed mesenchymal- or

Copyright © 2014 by the International Association for the Study of Lung Cancer

289

Wang et al.

Journal of Thoracic Oncology  ®  •  Volume 10, Number 2, February 2015

FIGURE 1.  Differential expression and secretion of ACTN4 protein in CL1-0 and CL1-5 cells. A, Western blots of ACTN4 protein expression in cell extracts (CE) and ACTN4 secretion into conditioned media (CM) from the CL1-0 and CL1-5 lung cancer cells. B, The quantification of relative intensity of ACTN4 protein in the CE and CM of the CL1-5 and CL1-0 lung cancer cells. The signal intensity of Western blotting was normalized to the intensity of the CL1-0 ACTN4 protein. Tubulin and phosphoglycerate mutase 1 (PGAM1) were used as loading controls for CE and CM, respectively.

spindle-like forms (Fig. 3A, middle and low panels). In addition, typical CL1-0 cells mainly showed epithelial-like morphology (Fig. 3B, upper panel). Particularly, the CL1-0 cells (approximately 68%) showed an elongated mesenchymal- or CL1-5-like morphology when they were transfected with ACTN4 DNA vector (Fig. 3B, lower panel). Under confocal immunofluorescence microscopy, CL1-0 cells showed an epithelial-like morphology and some extensions or small pseudopod extensions from cell membranes (Fig. 3C, left column). CL1-5 cells treated with scrambled control had ruffled cell membranes and elongations at the front of each cell, forming a pseudopod structure (Fig. 3C, middle column). In contrast, CL1-5 cells treated with the ACTN4 siRNA showed a globular shape, and the ACTN4 and actin proteins were retracted from the edge of the CL1-5 cell membrane and condensed in a globular shape (Fig. 3C, right column).

290

Correlation between ACTN4 Protein Expression and Lung Cancer Stages by Immunohistochemistry The NSCLC cases were included in the analyses of the T, N, and M descriptors and the subsequent analysis of TNM subsets and stage groupings.19 The T staging is determined by the size of primary tumor in long axis, the N classification defines the degree of spread to regional lymph nodes, and the M staging defines the presence of metastases beyond regional lymph nodes. The overall stage ( I, II, III, and IV) of the NSCLC is determined by the combination of T, N, and M grades. ACTN4 expression was quantified by TMA. To assess the results of immunohistochemistry for ACTN4 expression, the immunoreaction was considered to be present, showing ACTN4 immunoreactivity intensity for the surgical tumor cell specimen and the adjacent normal tissue as the standard. The representative

Copyright © 2014 by the International Association for the Study of Lung Cancer

Journal of Thoracic Oncology  ®  •  Volume 10, Number 2, February 2015

Functional Properties of the ACTN4 Protein in NSCLC

FIGURE 2.  The effects of ACTN4 protein on cell migration and invasiveness in CL1-5 cells. A, The Western blot data and relative intensity quantification of ACTN4 protein in cell extracts and conditioned media of the lung cancer CL1-5 cells after treating different pools of ACTN4 siRNA. Tubulin, actin, and PGAM1 were used as loading controls for cell extracts and conditioned media. The signal intensities of Western blotting were normalized to the intensity of the CL1-5 cells treated with scrambled control. B, MTT cell viability assay and (C) BrdU cell proliferation assay of CL1-5 cells treated with ACTN4 scrambled control or ACTN4 siRNA. Mann–Whitney U tests were carried out for each interval (p < 0.05). The results of (D) the transwell migration assay, (E) the Matrigel invasion assay, and (F) the wound healing assay of CL1-5 cells treated with scrambled control or ACTN4 siRNA. G, Western blots and relative intensity quantification of differential expression of ACTN4 protein between CL1-0 cells transfected with control vector or ACTN4 DNA vectors. The signal intensity of Western blotting was normalized to the intensity of the CL1-0 cells transfected with control vector. The results of (H) the transwell migration assay and (I) the Matrigel invasion assay of CL1-0 cells transfected with control vector or ACTN4 DNA vectors. The quantification values of the transwell migration assays and Matrigel invasion assays are expressed as the mean ± SD values from triplicate experiments, and the cell numbers were calculated from six random fields under a light microscope at 100× magnification. The wound-healing assay was quantified using ImageJ software. The gray line indicates the initial wound scratch boundary, and the black line indicates the boundary of cell migration after 24 hours. The quantification of the wound-healing assay is provided as the mean ± SD values of triplicate experiments, and a Mann–Whitney U test was performed (p < 0.05). Copyright © 2014 by the International Association for the Study of Lung Cancer

291

Wang et al.

Journal of Thoracic Oncology  ®  •  Volume 10, Number 2, February 2015

FIGURE 2.  (Continued)

292

Copyright © 2014 by the International Association for the Study of Lung Cancer

Journal of Thoracic Oncology  ®  •  Volume 10, Number 2, February 2015

Functional Properties of the ACTN4 Protein in NSCLC

FIGURE 2.  (Continued)

Copyright © 2014 by the International Association for the Study of Lung Cancer

293

Wang et al.

Journal of Thoracic Oncology  ®  •  Volume 10, Number 2, February 2015

FIGURE 2.  (Continued)

results of immunohistochemistry showed higher expression of ACTN4 in the different stages of tumor compared with the expression in adjacent normal tissues(Fig. 4A). The collected number of tissue samples were 30, 19, 27, and 8 for stages I, II, III, and IV, respectively. The distribution of age was from 21 to 79 years. Obviously, the number of sample of NSCLC stage IV is less than the numbers for the other three stages because the treatment options for stage IV of NSCLC include cytotoxic chemotherapy and radiation; in contrast, surgery is generally used in selective cases for symptom palliation. The quantitation of ACTN4 expression from the collected TMA samples (Supplementary Fig. S2) is shown by a scattergram of the immunohistochemical localization of ACTN4 protein expression in NSCLC patients (Fig. 4B) (i). The correlation between histological ACTN4 expression and clinicopathological stage is shown in Table 1. There were no significant differences in ACTN4 expression between the groups with regard to patient age and gender. In comparison, NSCLC tumors showing ACTN4 protein were strongly stained and tended toward apparent ACTN4 protein overexpression, with the highest staining at different NSCLC classifications, such as overall stages III, T4, and N2. Although the data in each group of different classifications was divergence and partial overlap for clinical samples; notably, the mean of immunohistochemistry ACTN4 expression intensity showed significant difference in the beginning between adjacent normal and overall stage I. On further analysis, the paired test for the adjacent normal versus cancer tissues also yielded significantly different results. The ACTN4 expression intensity determined by immunohistochemistry was significantly increased from overall stage II to stage III, then declining in stage IV; the physiological and significant role of ACTN4 in the NSCLC staging process needs further investigation. An ROC was plotted for the tissue ACTN4 staining

294

of cancer tissue versus normal adjacent tissue, and the area under the curve (AUC) was 0.736 (p < 0.001; Fig, 4B, (ii)).

Correlation between Plasma ACTN4 Protein Concentration and Lung Cancer Stages by ELISA Plasma was obtained from two cohorts of lung adenocarcinoma patients from NCKU (185 samples; 96, 30, 45, and 14 for stages I, II, III, and IV, respectively) and KVGH (145 samples; 40, 39, 42, and 24 for stages I, II, III, and IV, respectively) and from 147 healthy donors. The scattergrams of plasma ACTN4 protein concentration in healthy controls and lung cancer patients from NCKU and KVGH are shown in Figure 4C (i) and Figure 4D (ii). The ELISA results showed that the concentration of ACTN4 protein in healthy plasma was 3.41 μg/ml and 3.50 μg/ml (p < 0.05) for the samples from NCKU and KVGH, respectively. There were no significant differences between the ACTN4 levels with age and gender in healthy controls. The correlation between ACTN4 protein expression and the clinical plasma levels for lung adenocarcinoma from NCKU and KVGH samples is listed in Tables 2 and 3. The samples of both cohorts showed statistically significant differences in plasma ACTN4 protein concentration between healthy controls and lung cancer patients (p < 0.001). The plasma ACTN4 concentration was maintained at approximately 8 to 10 μg/ml in stage I of NSCLC and across the overall lung cancer process. It is a conspicuous increase of approximately 2.5 to 3-folds over healthy control plasma (p < 0.001). However, on further analysis, ACTN4 protein concentration did not show significant difference among the different overall stages of lung cancer during the cancer development process. The ROCs were plotted for the plasma ACTN4 protein concentration in healthy controls versus cancer patients, and the areas under the ROC curves were 0.828 and 0.909 for the samples from NCKU and KVHP, respectively (p < 0.001; Fig. 4C (ii) and Fig. 4D (ii)).

Copyright © 2014 by the International Association for the Study of Lung Cancer

Journal of Thoracic Oncology  ®  •  Volume 10, Number 2, February 2015

Functional Properties of the ACTN4 Protein in NSCLC

FIGURE 3.  Subcellular function and distribution of ACTN4 in CL1-0 and CL1-5 cells. A, Photomicrographs of CL1-5 cells treated with ACTN4 scrambled control or different pools of ACTN4 siRNA. The thin arrows indicate a typical morphology of CL1-5 cells treated with the ACTN4 scrambled control; these cells show mesenchymal-like morphology (upper panels). The thick arrows indicate a typical morphology of CL1-5 cells treated with the ACTN4 siRNA; these cells show primarily globular cells (lower panels). The scale bar represents 100 μm. Quantification of the numbers of mesenchymal-like CL1-5 cells treated with the ACTN4 scrambled control or the ACTN4 siRNA. The data are provided as the mean ± SD and were calculated from six random areas. B, Fluorescent photomicrographs of CL1-0 cells transfected with control vector and ACTN4 DNA vector and stained with actin. The typical morphology of CL1-0 cells is a typical epithelial-like shape (upper panels) and the typical CL1-0 cells transfected with ACTN4 DNA vector had elongated mesenchymal- or CL1-5-like morphology (lower panels). Quantification of the epithelial-like CL1-0 cells transfected with control vector or ACTN4 DNA vectors is provided as the mean ± SD from six random areas. C, High-magnification confocal immunofluorescence microscopy photographs. The CL1-0 cells showed an epithelial-like morphology with short extended pseudopods at the cell membranes (panels at left column). Two typical cell morphologies of CL1-5 cells treated with the ACTN4 scrambled control (panels at middle column) and the ACTN4 siRNA (panels at right column). The scale bar represents 10 μm. Actin was labeled with phalloidin-iFluor 488 (green color), the nuclei were labeled with Hoechst 33258 (blue color), and ACTN4 was detected using a goat anti-rabbit IgG conjugated to Alexa Fluor 594 (red color). Copyright © 2014 by the International Association for the Study of Lung Cancer

295

Journal of Thoracic Oncology  ®  •  Volume 10, Number 2, February 2015

Wang et al.

FIGURE 3.  (Continued)

DISCUSSION The ability of cancer cells to migrate and invade is critical determinant in the spread of cancer from the tissue of origin and affects the subsequent implantation into other organs. 20 Factors that enhance cancer cell migration and invasion include many proteins involved in the remodeling of the actin cytoskeleton and several associated proteins, 21 such as Rac and Cdc42 proteins, which are members of the Rho guanosine triphosphatase (GTPase) superfamily, 22 and actin-binding proteins, such as fascin 23 and ACTN4. 14,24 In this study, we determined

296

that ACTN4 protein is highly expressed and secreted in the highly invasive CL1-5 cells. ACTN4 protein also participates in migration in different cancers, such as bladder, colorectal, and ovarian cancers. 25–30 However, what is the role of ACTN4 in cancer cell metastasis? In this study, we report that ACTN4 protein knockdown suppresses CL1-5 migration and invasion because the morphology of CL1-5 was altered from mesenchymal-like to globular shape. In addition, ACTN4 overexpression in CL1-0 cells caused the cell morphology to be changed from the epithelial-like to mesenchymal-like shape. It is believed

Copyright © 2014 by the International Association for the Study of Lung Cancer

Journal of Thoracic Oncology  ®  •  Volume 10, Number 2, February 2015

Functional Properties of the ACTN4 Protein in NSCLC

FIGURE 4.  The levels of ACTN4 in cancerous tissues and plasma samples of lung cancer patients. A, Representative images of the ACTN4 protein immunoreactivity of tissues at different stages of lung cancer, under a light microscope (200× magnification). The negative control samples were not incubated with primary rabbit anti-ACTN4 IgG. B, Scattergram of (i) immunohistochemical study of ACTN4 protein expression by diaminobenzidine intensity of adjacent normal and total lung cancer clinicopathological tissues, and (ii) the receiver operating characteristic curve was plotted for the ACTN4 staining of cancer tissues versus the normal adjacent tissues. Both scattergrams (i) in (C) and (D) were plotted for plasma ACTN4 protein concentration in the healthy controls and lung cancer patients from NCKU and KVGP samples, respectively. The red lines indicate the means of the ACTN4 concentration in each group. Both receiver operating characteristic curves (ii) in (C) and (D) were plotted for the plasma ACTN4 protein from the cancer patients versus the healthy controls from NCKU and KVGP samples, respectively (p < 0.001).

Copyright © 2014 by the International Association for the Study of Lung Cancer

297

Wang et al.

Journal of Thoracic Oncology  ®  •  Volume 10, Number 2, February 2015

FIGURE 4.  (Continued)

that mesenchymal-like cell is certain to form pseudopods and migrate. 21 In this study, ACTN4 participated in regulating the dynamic assembly and disassembly of the pseudopod structure. The polarity and directed migration of such cells is determined by the assembly, extension, and stabilization of a pseudopod. 31 ACTN4 knockdown in keratinocytes has shown that the keratinocytes lack polarity and have 25% increased focal contact area. 32 In the regulation of colorectal cancer metastasis, the cadherin/catenin system mediates cell–cell adhesion and acts as an invasion suppressor of epithelial malignancies. On inhibition of E-cadherin expression, ACTN4 can bind with β-catenin in the membrane protrusions of DLD-1 cells. This study shows that ACTN4 takes part in the regulation of cell–cell adhesion through dynamic binding to β-catenin from cell adhesion complex and mediates cell invasion and metastasis. 33 Therefore, these data may explain the histochemistry staining that showed the highest intensity of ACTN4 protein expression in stage III of lung cancer tissue; NSCLC cells overexpress ACTN4 protein, which regulates the loss of cell adhesion through the competition with cadherin/catenin system; NSCLC cells then develop the ability to leave the original lesion and migrate enabling distant metastasis (M stage). In MRC-5 lung fibroblast cells, subcellular ACTN4 proteins were localized in the cytoplasm and along actin stress fibers. 14 ACTN4 is an actin-binding protein that is differently localized in moving structures, such as ruffles, lamellipodia, and filopodia, 31,34 but not in the adherens junctions. 33 The expression level of ACTN4 was significantly increased in CL1-5 cells (Fig. 1 and 2), and the increased expression of ACTN4 dramatically changed the cell morphology and motility of lung cancer cells (Fig. 2A and 3A). Moreover, during cancer metastasis, abnormal tumor-derived exosomes were found to be secreted. 35 Tumor-derived exosomes are enclosed cytosolic and

298

membrane proteins derived from the parental cells, and these exosomes modulate the tumor microenvironment for cancer cell metastasis. The cytosolic molecules in exosomes include DNA, 36 RNA, 37 and protein. 9 ACTN4 was one of the candidate exosome proteins found in previous studies conducted in this laboratory and in this study. Thus, ACTN4 protein participates in migration of NSCLC cells and, after secretion, modulates the microenvironment of cancer cells for cancer development. Interestingly, from a comparison of ACTN4 expression as found in immunohistochemical data and plasma ELISA data, it can be inferred that ACTN4 protein expression varies with the clinical stage of lung cancer, with the highest staining at stage III and lower expression at stage IV; however, ACTN4 protein concentration in the plasma was maintained constant at approximately 8–10 μg/ml. It is not clear why ACTN4 protein is secreted and maintained across the overall lung cancer process. The function of ACTN4 has however been previously found; ACTN4 affects the expression of matrix metalloproteinase gene 38 and ACTN4 is upregulated by three-dimensional collagen culture conditions, leading to increased invasion and motility of ovarian cancer cells. 39 Previous research showed that matrix metalloproteinase-9 can degrade and reconstitute the ECM to facilitate tumor cell migration and invasion. 40,41 All these results suggest that ACTN4 enhances or activates other metastatsis-associated proteins and regulates cell mobility in lung cancer. In summary, this approach is based on the preliminary identification of the expressed and secreted proteins of lung cancer CL1-0 and CL1-5 cells by liquid chromatography–tandem mass spectrometry and the subsequent functional characterization of candidate ACTN4 protein. We also showed that ACTN4 protein is a mobilityrelated protein that influences the filopodia or lamellipodia structures in lung cancer CL1-5 cells. The evaluation

Copyright © 2014 by the International Association for the Study of Lung Cancer

Journal of Thoracic Oncology  ®  •  Volume 10, Number 2, February 2015

TABLE 1.  Correlation between ACTN4 Protein Expression and Clinical Pathologic Features of Lung Adenocarcinoma Using Image Analysis; Quantification of ACTN4-Positive Tissue Microarray No. of Patients

Variable

Gray Values (Mean ± SD)

p Valuesa

Total: 84 patients Age, years

Gender Overall stage

Adjacent normal vs. Stage I Stage I vs. Stage II Stage II vs. Stage III Stage III vs. Stage IV Adjacent normal vs. Cancer T stage group

N stage

M stage

7 28 32 17

74.94 ± 15.01 94.7 ± 31.7 93.9 ± 30.91 88.53 ± 22.01

0.383

0.879

Male

33

89.63 ± 20.32

Female

51

92.71 ± 33.17

I

30

81.2 ± 23.66

II

19

83.76 ± 16.46

III

27

113.58 ± 31.36

8 31

74.02 ± 16.02 71.70 ± 11.91

30 30 19 19 27 27 8 31

81.2 ± 23.66 81.2 ± 23.66 83.76 ± 16.46 83.76 ± 16.46 113.58 ± 31.36 113.58 ± 31.36 74.02 ± 16.02 71.70 ± 11.91

31 25 41 8 10 36 20 27 1 74 7

87.07 ± 30.36 91.63 ± 28.43 87.46 ± 27.97 99.01 ± 19.09 101.75 ± 37.99 82.74 ± 22.19 89.03 ± 32.36 105.45 ± 29.74 79.61 92.53 ± 28.80 73.74 ± 16.15

IV Adjacent normal Stage I Stage I Stage II Stage II Stage III Stage III Stage IV Adjacent normal Cancer T1 T2 T3 T4 N0 N1 N2 N3 M0 M1

TABLE 2.  Correlation between ACTN4 Protein Expression and the Clinical Plasma Levels in Lung Adenocarcinoma Using an ELISA Experiment. The Cancer Plasma Samples Were from NCKU

Healthy: Total 147

Gender

<0.001

0.545 <0.003 <0.00 <0.001

0.326

0.007

0.094

a The Kolmogorov–Smirnov test was examined for the normality of collected data distribution. (1) Nonparametric statistics Age: Kruskal–Wallis H Sex: Mann–Whitney U Adjacent normal vs. stage I: Mann–Whitney U Each stage vs. each stage: Mann–Whitney U Adjacent normal vs. cancer: Wilcoxon matched–pairs test Tumor size (T): Kruskal–Wallis H Tumor size group: Kruskal–Wallis H Lymph node metastasis (N): Kruskal–Wallis H Lymph node metastasis group: Mann–Whitney U (2) Parametric statistics Stage: One-way analysis of variance Distant metastasis (M): t test

results of clinical samples suggest that ACTN4 protein is a potential indicator to discriminate between plasma samples of healthy controls and NSCLC patients.

21–49 50–59 60–69 70–79 ≥80 Male Female

25 38 41 33 10 71 76

4.09 ± 2.55 3.42 ± 1.98 2.84 ± 1.70 3.26 ± 1.83 4.49 ± 2.57 3.30 ± 2.15 3.51 ± 2.44

0.083

8.50 ± 6.58 8.57 ± 5.23 8.54 ± 6.80 13.09 ± 18.07 15.89 ± 14.91 10.58 ± 9.62 8.97 ± 10.91 3.52 ± 2.35 9.71 ± 10.34 10.51 ± 13.35 8.24 ± 5.12 9.11 ± 5.78 9.28 ± 5.17 3.41 ± 2.30 10.51 ± 13.35 10.51 ± 13.35 8.24 ± 5.12 8.24 ± 5.12 9.11 ± 5.78 9.11 ± 5.78 9.28 ± 5.17 9.17 ± 8.93 10.77 ± 13.44 10.02 ± 6.55 8.40 ± 3.83 10.64 ± 12.71 6.26 ± 3.84 9.44 ± 5.86 9.24 ± 3.44 9.65 ± 11.00 9.37 ± 8.06

0.583

0.801

Patients: Total 185 Age, years

<0.002

No. of ACTN4 Level p Patients (μg/ml) (Mean ± SD) Valuesa

Variable

Age, years

21–49 50–59 60–69 70–79

Functional Properties of the ACTN4 Protein in NSCLC

21–49 50–59 60–69 70–79 ≥80 Gender Male Female Healthy vs. patients Healthy Patients Overall stage I II III IV Healthy vs. Stage I Healthy Stage I Stage I vs. Stage II Stage I Stage II Stage II vs. Stage III Stage II Stage III Stage III vs. Stage III Stage IV Stage IV T stage group T1 T2 T3 T4 N stage N0 N1 N2 N3 M stage M0 M1

16 63 61 41 4 85 100 147 185 96 30 45 14 147 96 96 30 30 45 45 14 90 66 12 16 108 26 46 4 142 13

0.028c <0.001c 0.679

<0.001 0.652 0.344 0.347 0.527

0.330

0.667

a The Kolmogorov–Smirnov test was examined for the normality of collected data distribution. Nonparametric statistics Age: Kruskal–Wallis H Sex: Mann–Whitney U Healthy controls vs. patients: Mann–Whitney U Stage: Kruskal–Wallis H Each stage vs. each stage: Mann–Whitney U Tumor size (T): Kruskal–Wallis H Lymph node metastasis (N): Kruskal–Wallis H Distant metastasis (M): Mann–Whitney U c p < 0.05 is considered to be a significant difference.

Copyright © 2014 by the International Association for the Study of Lung Cancer

299

Journal of Thoracic Oncology  ®  •  Volume 10, Number 2, February 2015

Wang et al.

TABLE 3.  Correlation between ACTN4 Protein Expression and the Clinical Plasma Levels in Lung Adenocarcinoma Using an ELISA Experiment. The Cancer Plasma Samples Were from KVGH No. of ACTN4 level (μg/ml) p Patients (Mean ± SD) Values a

Variable

Patients: Total 145 Healthy vs. patients Overall stage

Healthy vs. Stage I

Healthy Patients I

147 145 40

3.50 ± 2.38 8.22 ± 7.77 8.01 ± 5.24

II

39

8.11 ± 4.57

III

42

9.28 ± 4.35

IV

24

8.97 ± 5.17

Healthy

147

3.50 ± 2.38

40

8.01 ± 5.24

Stage I Stage I vs. Stage II

Stage I

40

8.01 ± 524

Stage II

39

8.11 ± 4.57

Stage I

39

8.11 ± 4.57

Stage III

42

9.28 ± 3.35

Stage III vs. Stage IV Stage III

42

9.28 ± 4.35

Stage IV

24

8.97 ± 5.17

Stage II vs. Stage III

T stage group

N stage

M stage

T1

49

8.33 ± 3.13

T2

77

8.47 ± 4.25

T3

14

9.66 ± 4.99

T4 N0

5 71

8.85 ± 5.69 7.93 ± 5.03

N1

29

8.82 ± 4.32

N2

36

8.97 ± 3.16

N3

9

9.32 ± 3.72

M0 M1

121 24

8.01 ± 3.84 8.98 ± 4.96

<0.001 0.606

<0.001 0.817 0.680 0.298 0.892

0.156

0.11

a The Kolmogorov–Smirnov test was examined for the normality of collected data distribution. Nonparametric statistics Age: Kruskal–Wallis H Sex: Mann–Whitney U Healthy controls vs. patients: Mann–Whitney U Stage: Kruskal–Wallis H Each stage vs. each stage: Mann–Whitney U Tumor size (T): Kruskal–Wallis H Lymph node metastasis (N): Kruskal–Wallis H Distant metastasis (M): Mann–Whitney U

Acknowledgments The authors thank the National Cheng Kung University Research Core Laboratory staff for their assistance with mass spectrometry analysis and consultations. The authors also thank the National Cheng Kung University Hospital and the Kaohsing Veterans General Hospital for providing the lung cancer tissue and plasma samples used in this study. REFERENCES 1. D’Addario G, Fruh M, Reck M, et al. Metastatic non-small-cell lung cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol 2010;21 Suppl 5:v116–v119.

300

2. Johnson DH. Management of small cell lung cancer: current state of the art. Chest 1999;116:525S–530S. 3. Ettinger DS. Overview and state of the art in the management of lung cancer. Oncology (Williston Park) 2004;18:3–9. 4. Mehlen P, Puisieux A. Metastasis: a question of life or death. Nat Rev Cancer 2006;6:449–458. 5. Leber MF, Efferth T. Molecular principles of cancer invasion and metastasis (review). Int J Oncol 2009;34:881–895. 6. Volmer MW, Stühler K, Zapatka M, et al. Differential proteome analysis of conditioned media to detect Smad4 regulated secreted biomarkers in colon cancer. Proteomics 2005;5:2587–2601. 7. Chu YW, Yang PC, Yang SC, et al. Selection of invasive and metastatic subpopulations from a human lung adenocarcinoma cell line. Am J Respir Cell Mol Biol 1997;17:353–360. 8. Chiu KH, Chang YH, Wu YS, Lee SH, Liao PC. Quantitative secretome analysis reveals that COL6A1 is a metastasis-associated protein using stacking gel-aided purification combined with iTRAQ labeling. J Proteome Res 2011;10:1110–1125. 9. Chang YH, Lee SH, Liao IC, Huang SH, Cheng HC, Liao PC. Secretomic analysis identifies alpha-1 antitrypsin (A1AT) as a required protein in cancer cell migration, invasion, and pericellular fibronectin assembly for facilitating lung colonization of lung adenocarcinoma cells. Mol Cell Proteomics 2012;11:1320–1339. 10. Chang YH, Lee SH, Chang HC, et al. Comparative secretome analyses using a hollow fiber culture system with label-free quantitative proteomics indicates the influence of PARK7 on cell proliferation and migration/ invasion in lung adenocarcinoma. J Proteome Res 2012;11:5167–5185. 11. Ebashi S, Ebashi F. A new protein factor promoting contraction of actomyosin. Nature 1964;203:645–646. 12. Goto H, Wakui H, Komatsuda A, et al. Renal alpha-actinin-4: purification and puromycin aminonucleoside-binding property. Nephron Exp Nephrol 2003;93:e27–e35. 13. Condeelis J, Segall JE. Intravital imaging of cell movement in tumours. Nat Rev Cancer 2003;3:921–930. 14. Honda K, Yamada T, Endo R, et al. Actinin-4, a novel actin-bundling protein associated with cell motility and cancer invasion. J Cell Biol 1998;140:1383–1393. 15. Virel A, Backman L. Molecular evolution and structure of alpha-actinin. Mol Biol Evol 2004;21:1024–1031. 16. Mejillano MR, Kojima S, Applewhite DA, Gertler FB, Svitkina TM, Borisy GG. Lamellipodial versus filopodial mode of the actin nanomachinery: pivotal role of the filament barbed end. Cell 2004;118:363–373. 17. Khurana S, George SP. The role of actin bundling proteins in the assembly of filopodia in epithelial cells. Cell Adh Migr 2011;5:409–420. 18. Lauffenburger DA, Horwitz AF. Cell migration: a physically integrated molecular process. Cell 1996;84:359–369. 19. Mirsadraee S, Oswal D, Alizadeh Y, Caulo A, van Beek E Jr. The 7th lung cancer TNM classification and staging system: review of the changes and implications. World J Radiol 2012;4:128–134. 20. Yilmaz M, Christofori G. Mechanisms of motility in metastasizing cells. Mol Cancer Res 2010;8:629–642. 21. Tojkander S, Gateva G, Lappalainen P. Actin stress fibers–assembly, dynamics and biological roles. J Cell Sci 2012;125(Pt 8):1855–1864. 22. Raftopoulou M, Hall A. Cell migration: Rho GTPases lead the way. Dev Biol 2004;265:23–32. 23. Adams JC. Roles of fascin in cell adhesion and motility. Curr Opin Cell Biol 2004;16:590–596. 24. Shao H, Wang JH, Pollak MR, Wells A. α-Actinin-4 is essential for maintaining the spreading, motility and contractility of fibroblasts. PLoS One 2010;5:e13921. 25. Yamamoto S, Tsuda H, Honda K, et al. ACTN4 gene amplification and actinin-4 protein overexpression drive tumour development and histological progression in a high-grade subset of ovarian clear-cell adenocarcinomas. Histopathology 2012;60:1073–1083. 26. Welton JL, Khanna S, Giles PJ, et al. Proteomics analysis of bladder cancer exosomes. Mol Cell Proteomics 2010;9:1324–1338. 27. Choi DS, Lee JM, Park GW, et al. Proteomic analysis of microvesicles derived from human colorectal cancer cells. J Proteome Res 2007;6:4646–4655. 28. Hegmans JP, Bard MP, Hemmes A, et al. Proteomic analysis of exosomes secreted by human mesothelioma cells. Am J Pathol 2004;164:1807–1815.

Copyright © 2014 by the International Association for the Study of Lung Cancer

Journal of Thoracic Oncology  ®  •  Volume 10, Number 2, February 2015

29. Gonzalez-Begne M, Lu B, Han X, et al. Proteomic analysis of human parotid gland exosomes by multidimensional protein identification technology (MudPIT). J Proteome Res 2009;8:1304–1314. 30. Yamada S, Yanamoto S, Yoshida H, et al. RNAi-mediated down-regulation of alpha-actinin-4 decreases invasion potential in oral squamous cell carcinoma. Int J Oral Maxillofac Surg 2010;39:61–67. 31. Pollard TD, Cooper JA. Actin, a central player in cell shape and movement. Science 2009;326:1208–1212. 32. Hamill KJ, Hopkinson SB, Skalli O, Jones JC. Actinin-4 in keratinocytes regulates motility via an effect on lamellipodia stability and matrix adhesions. FASEB J 2013;27:546–556. 33. Hayashida Y, Honda K, Idogawa M, et al. E-cadherin regulates the association between beta-catenin and actinin-4. Cancer Res 2005;65:8836–8845. 34. Araki N, Hatae T, Yamada T, Hirohashi S. Actinin-4 is preferentially involved in circular ruffling and macropinocytosis in mouse macrophages: analysis by fluorescence ratio imaging. J Cell Sci 2000;113(Pt 18):3329–3340. 35. Yang C, Robbins PD. The roles of tumor-derived exosomes in cancer pathogenesis. Clin Dev Immunol 2011;2011:842849.

Functional Properties of the ACTN4 Protein in NSCLC

36. Paci M, Maramotti S, Bellesia E, et al. Circulating plasma DNA as diagnostic biomarker in non-small cell lung cancer. Lung Cancer 2009;64:92–97. 37. Valadi H, Ekstrom K, Bossios A, Sjöstrand M, Lee JJ, Lötvall JO. Exosome-mediated transfer of mRNAs and microRNAs is a novel mechanism of genetic exchange between cells. Nat Cell Biol 2007;9:654–659. 38. Aksenova V, Turoverova L, Khotin M, et al. Actin-binding protein alphaactinin 4 (ACTN4) is a transcriptional co-activator of RelA/p65 sub-unit of NF-kB. Oncotarget 2013;4:362–372. 39. Barbolina MV, Adley BP, Kelly DL, et al. Motility-related actinin alpha-4 is associated with advanced and metastatic ovarian carcinoma. Lab Invest 2008;88:602–614. 40. Kim YH, Kwon HJ, Kim DS. Matrix metalloproteinase 9 (MMP-9)dependent processing of βig-h3 protein regulates cell migration, invasion, and adhesion. J Biol Chem 2012;287:38957–38969. 41. Cox G, Jones JL, O’Byrne KJ. Matrix metalloproteinase 9 and the epidermal growth factor signal pathway in operable non-small cell lung cancer. Clin Cancer Res 2000;6:2349–2355.

Copyright © 2014 by the International Association for the Study of Lung Cancer

301