Study on Attenuating Angiogenesis and Epithelial-Mesenchymal Transition (EMT) of Non-Small Cell Lung Carcinoma (NSCLC) by Regulating MAGEC2.

OBJECTIVE: To investigate the role of MAGE family member C2 in angiogenesis and epithelial-mesenchymal transition of non-small cell lung carcinoma.
METHODS: The Cancer Genome Atlas data set was analyzed to filter the highly expressed gene melanoma antigen family C2 in non-small cell lung carcinoma. Quantitative reverse transcription-polymerase chain reaction was performed to verify the overexpression of melanoma antigen family C2 in non-small cell lung carcinoma cell lines. Melanoma antigen family C2 complementary DNA and short hairpin RNA (shRNA) were transfected into SK-MES-1 cells to regulate melanoma antigen family C2 expression. Cell Counting Kit-8 assay, flow cytometry, wound healing assay, and Transwell assay were performed to investigate the effect of melanoma antigen family C2 on proliferation, apoptosis, migration, and invasion of SK-MES-1 cell line. Western blot was used to detect the expression of epithelial-mesenchymal transition markers. Enzyme-linked immunosorbent assay was performed to investigate the secretion of vascular endothelial growth factor, and tube formation assay was conducted to explore the effect of melanoma antigen family C2 on angiogenesis ability of the tumor. Tumor xenograft on nude mice and immunohistochemical/hematoxylin and eosin staining were also performed to detect the influence of melanoma antigen family C2 on growth and metastasis of non-small cell lung carcinoma cells.
RESULTS: Melanoma antigen family C2 was highly expressed in non-small cell lung carcinoma cells; melanoma antigen family C2 promoted the expression of epithelial-mesenchymal transition-related proteins as well as enhance the secretion of vascular endothelial growth factor and promote angiogenesis; melanoma antigen family C2 promoted proliferation, migration, and invasion and suppressed apoptosis of non-small cell lung carcinoma cells. It could also facilitate growth and metastasis of non-small cell lung carcinoma in vivo.
CONCLUSION: Melanoma antigen family C2 was a critical factor of angiogenesis and epithelial-mesenchymal transition in non-small cell lung carcinoma.

Introduction

Lung carcinoma is the leading cause of carcinoma-related death around the world.[1-3] According to the risk assessment given by International Agency for Research on Cancer at 2012, over 1.8 million patients are diagnosed with lung carcinoma (12.9% of the total) and about 1.6 million patients are killed by it (19.4% of the total) per year.[4] Generally, over 80% of lung carcinoma cases are classified as non-small cell lung carcinoma (NSCLC).[5-7] Treatments including chemotherapy and surgery have been widely used to cope with NSCLC, but 5-year overall survival rate is only 16% for patients with NSCLC of all stages.[8,9] The grim situation of NSCLC therapy makes it seriously necessary to further explore the mechanism of NSCLC and build new therapeutic strategies.

Initiation and development of NSCLC is partly attributed to the aberrant expression of proto-oncogenes, which lead to tumor proliferation, metastasis, and other tumor progressions.[8,10,11] With the development of bioinformatics, increasing oncogenes have been detected and explored in different kinds of carcinomas. Cancer–testis antigens (CTAs) are a group of carcinoma-associated antigens that have extremely low expression in normal tissues except adult testis but aberrant high expression in many kinds of tumors, particularly advanced carcinomas with stem cell–like characteristics.[12] Cancer–testis antigen melanoma antigen family C2 (MAGEC2), a member of type I melanoma antigen (MAGE) family, is a widely explored gene for that it is one of the most immunogenic CTAs.[13] It has been identified as the oncogene in many carcinomas, including melanoma,[14] prostate cancer,[15] breast cancer,[16] and multiple myeloma.[17]

Epithelial–mesenchymal transition (EMT) and angiogenesis are 2 critical factors influencing tumor growth and metastasis. Epithelial–mesenchymal transition is a complex multistep event, which is characterized by the loss of E-cadherin and occludins, downregulation of epithelial maker cytokeratin, upregulation of mesenchymal marker vimentin and fibronectin, and acquisition of fibroblast-like morphology with cytoskeleton reorganization.[18] The motility, invasiveness, and metastatic ability of tumor cells are closely related to EMT.[19] Angiogenesis in tumors promotes the tumor growth and metastasis in the vascular phase,[20] which is the main reason for the failure of antitumor therapy.[21] The process of angiogenesis is actually the proliferation of vascular endothelial cells arranged by vessels,[22] and the vascular endothelial growth factor (VEGF) actively participates in the angiogenesis, lymphangiogenesis, and tube formation.[23] Many anti-VEGF medicines like endostatin have been widely used as the therapeutic strategy against the carcinomas, but they are impressionable to the tumor microenvironment.[24] For that both EMT and angiogenesis play crucial roles in the tumor progression, their correlation with oncogenes is of great research value.

Given that MAGEC2 could promote the tumor growth and metastasis, we hypothesized that MAGEC2 may function through facilitating EMT and angiogenesis in NSCLC. In our study, we regulated the expression of MAGEC2 and investigated the effects of differentially expressed MAGEC2 on the progress of EMT and angiogenesis in NSCLC. It will further improve the cognition about the function of MAGEC2 and provide a new potential target for NSCLC treatment.

Materials and Methods

Bioinformatic Analysis

The clinical data of patients with NSCLC were obtained from The Cancer Genome Atlas (TCGA) data set. R 3.4.0 (https://www.r-project.org) containing DESeq2 was used to analyze the data and draw the volcano plot and heatmap. Screening conditions were | log2(Foldchange) | >2 and adjusted P < .001. All the differentially expressed genes were showed in the volcano plot, and top 10 differentially expressed genes were showed in the heatmap.

Cell Culture

Human embryo kidney cell line HEK-293T, human NSCLC cell line SK-MES-1 (with TP53 and EGFR mutation), A549, and HCC827, human normal lung epithelial cell line BEAS-2B, and human umbilical vein endothelial cells (HUVECs) were purchased from BeNa Culture Collection (Beijing, China). HEK-293T and SK-MES-1 cells were cultured with Dulbecco modified Eagle medium (DMEM; Gibco, Grand Island, New York) containing 10% fetal bovine serum (FBS; Gibco), BEAS-2B and HCC827 cells were cultured with RPMI-1640 medium (Gibco) containing 10% FBS, A549 cells and HUVEC were cultured with Ham F12K medium (Gibco) containing 10% FBS. All the cells were incubated at 37°C in a humidified chamber containing 5% CO2.

Lentivirus Transfection

Lentivirus vectors containing MAGEC2 complementary DNA (cDNA) and shRNA were customized from GenePharma (Shanghai, China). Lentiviral packaging mixtures (containing lentivirus vectors, pMDLg/pRRE, pRSV-Rev, and pMD2.G) were cotransfected into HEK-293T cells to get lentivirus particles. SK-MES-1 cells were seeded onto 24-well plates (105/well) and incubated for 24 hours, then the culture medium was replaced with fresh DMEM containing 5μg/mL polybrene, and the virus suspension was added into the medium (multiplicity of infection [MOI] = 20). After 24-hour incubation, the culture medium was replaced with fresh DMEM in order to remove the lentivirus and polybrene. Cells with MAGEC2 upregulated or downregulated were then available after 48-hour incubation.

Quantitative Reverse Transcription-Polymerase Chain Reaction

Total RNA was extracted using TRIzol reagent (Beyotime, Shanghai, China), then DNA Reverse Transcription Kit (#4368814; Applied Biosystems, Foster City, California) was used to perform the reverse transcription, and quantitative reverse transcription-polymerase chain reaction (qRT-PCR) was performed using SYBR Select Master Mix on ABI Prism 7000 Sequence Detection system (Applied Biosystems). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was chosen as the internal reference, and the relative expressions of messenger RNAs (mRNAs) were calculated using 2−△△CT method. Experiments were repeated in triplicate for accuracy. Primer sequences are shown in Table 1.

Table 1.

Primers for qRT-PCR.

Target Gene		Primer Sequence
MAGEC2	Forward	5′-ACTTCGTCTATGGGGAGCCT-3′
	Reverse	5′-TGCGGTATCAATTGTGGCCT-3′
VEGF	Forward	5′-CTTGCCTTGCTGCTCTACCT-3′
	Reverse	5′-GCAGTAGCTGCGCTGATAGA-3′
GAPDH	Forward	5′-TGGAGAAAATCTGGCACCAC-3′

Abbreviations: MAGEC2, melanoma antigen family C2; qRT-PCR, quantitative reverse transcription-polymerase chain reaction; VEGF, vascular endothelial growth factor.

Western Blot

The cell proteins were extracted through The Regulation of Investigatory Powers Act (RIPA) lysis buffer (Beyotime) and quantified using bicinchoninic acid (BCA) kit (Beyotime), then 20 μg of proteins were treated by Sodium Dodecyl Sulphate - PolyAcrylamide Gel Electrophoresis (SDS-PAGE) and transferred onto polyvinylidene fluoride (PVDF) membranes (Invitrogen, Eugene, Oregon). Electrochemiluminescent detection system (Thermo Fisher Scientific, Waltham, Massachusetts) was used for blot signal detection. Specific proteins were detected by primary antibody rabbit anti-MAGEC2 (MAGEE1; #ab209667, 1:5000; Abcam, Cambridge, Massachusetts). Anti-GAPDH (#ab9485, 1:2500; Abcam) was used as the internal reference, and HRP-conjugated goat-anti-rabbit IgG (#ab6721, 1:10000; Abcam) was used as secondary antibody.

Enzyme-Linked Immunosorbent Assay

The cells were seeded onto 96-well plates and incubated at 37°C in humid air with 5% CO2 for 24 hours. Culture supernatant was collected, and enzyme-linked immunosorbent assay (ELISA) was performed using Human VEGF Quantikine ELISA Kit (#DVE00; R&D Systems, Abingdon, United Kingdom) according to the manufacturer’s protocol.

Cell Counting Kit-8 Assay

Cell Counting Kit-8 (CCK-8; #CK04; Dojindo, Kumamoto, Japan) was used to detect the cell viability. The cells were digested, then the cell suspension was planted into 96-well plates (104/well). After 24-/48-/72-/96-hour incubation, CCK-8 solution was added (10 µL/well), and the optical density at 450 nm wavelength was detected using the microplate reader (Thermo Fisher Scientific).

Tube Formation Assay

Each well of the 96-well plates was coated with 50 μL diluted Matrigel (BD Biosciences, San Jose, California), then the mixture of 2 × 104 HUVEC and transfected SK-MES-1 cells (1:1) was seeded onto the layer of Matrigel in 100 μL DMEM containing 10% FBS. Positive control group was additionally treated with 50 ng/mL VEGF (Beyotime). After 8-hour incubation, endothelial cell tube formation was observed. The ability of endothelial cells to form tubular structures in the different conditions was assessed using the optical microscope (Nikon, Japan).

Flow Cytometry

SK-MES-1 cell apoptosis was detected using Annexin V-FITC Apoptosis Detection Kit (#APOAF; Sigma-Aldrich, St Louis, Missouri), according to the manufacturer’s protocol. After digested by 0.25% trypsin for 24 hours, the cells were collected into flow tubes and washed with phosphate-buffered saline (PBS) for 3 times. Then each tube was added with 150 μL binding buffer and 1 μL Annexin V-FITC and incubated for 15 minutes in dark at room temperature. The cells were then treated by 100 μL propidium iodide, and the cell apoptosis was determined using FACSCanto II Flow Cytometer (BD Biosciences).

Wound Healing Assay

The cells were seeded in 6-well plate with 2 × 105 cells in each to grow to 80% confluence. The wounds were scratched with pipette tip gently across the center of the well. After washed 3 times with serum-free medium, the cells were incubated in humid atmosphere, 37°C and 5% CO2. The migrated cells were observed with the optical microscope at 0 and 24 hours, respectively.

Transwell Invasion Assay

Matrigel was diluted with 100 μL serum-free DMEM and mixed, then the Transwell chambers were coated with diluted Matrigel (200 mg/mL) at the density of 50 μL/well and placed into a 24-well plate. Then each chamber was filled with 250 μL cell suspension (containing 4 × 104 cells), and the 24-well plate was filled with DMEM containing 10% FBS. After 24-hour incubation, methyl alcohol and 0.1% crystal violet (Sigma-Aldrich) were respectively used for the fixation and staining of the invaded cells. The number of invaded cells was counted under the optical microscope.

Tumor Xenograft

BALB/c nude mice (4-5 weeks old, weighing 18-22 g) were purchased from SLACCAS (Shanghai, China). Suspension of SK-MES-1 cells transfected with MAGEC2 cDNA or shRNA was injected into the armpit of the mice (107/0.1 mL). Size of the tumors was measured with a vernier caliper per 5 days. All the mice were decapitated after 30-day feeding, then the tumors were excised and weighed, and tissues were taken out, then formalin-fixed and paraffin-embedded for immunohistochemical (IHC) or hematoxylin and eosin (HE) staining. All the experiments involving laboratory animals have been approved by the ethic committee of Jiangxi Province Tumor Hospital.

Immunohistochemical Staining

Sections of tumor tissues were dewaxed at 67°C and then 3% hydrogen peroxide–methanol solution was used to inactivate endogenous peroxidase. Nonspecific antigens were blocked by 10% goat serum at 37°C for 1 hour. After washed with PBS, the sections were incubated with primary antibody rabbit anti-VEGFA (1:100, ab52917; Abcam) at 37°C for 1.5 hour and with secondary antibody HRP-labeled goat anti-rabbit IgG (1:2000, ab205718; Abcam) for 1 hour at room temperature. Then the sections were washed with PBS and incubated with Diaminobenzidine (DAB) (#D12384; Sigma-Aldrich) for 3 minutes in a dark room. After the coloration, the sections were washed with DDH2O for 20 minutes and then counterstained with hematoxylin for 1 minute. Finally, the sections were washed and mounted with rhamsan gum.

Hematoxylin and Eosin Staining

Lung tissues of the mice were first dewaxed using xylene and placed in 4% paraformaldehyde, immersed and fixed for 24 hours. After that, the tissues were washed using distilled water for 3 minutes, followed by staining with hematoxylin for 15 minutes. The stained tissues were disposed in 1% hydrochloric acid (acid ethanol) for color separation for 10 seconds and then soaked in warm water for 5 minutes. Eosin dye solution 0.25% was subsequently added to counterstain the tissues for 5 minutes. Hematoxylin and eosin staining results were detected with the optical microscope.

Statistical Analysis

All the data were analyzed with GraphPad Prism 6 (GraphPad Software, La Jolla, California) and presented as mean (standard deviation). Student t test was used to detect differences between groups; Pearson rank correlation coefficient was performed to assess the correlations between 2 factors. All the in vitro experiments were repeated 3 times and in vivo experiments were repeated 5 times to ensure the accuracy. The differences were considered to be statistically significant at P <.05.

Results

Melanoma antigen family C2 Was Highly Expressed in NSCLC

The Cancer Genome Atlas data set was used to screen the differentially expressed mRNAs in NSCLC. The result of bioinformatic analysis showed that MAGEC2 was one of the top 10 upregulated genes in NSCLC (log2(Foldchange) = 6.52; Figure 1A and B). Quantitative reverse transcription-polymerase chain reaction and Western blot were then used to detect the expression of MAGEC2 in 3 NSCLC cell lines SK-MES-1, A549, and HCC827, and the result conformed to the bioinformatics analysis (P < .001; Figure 1C and D), indicating that MAGEC2 was an extremely highly expressed gene in NSCLC.

Figure 1.

Melanoma antigen family C2 was highly expressed in non-small cell lung carcinoma. A, Volcano plot of differentially expressed messenger RNAs in NSCLC based on the data of The Cancer Genome Atlas (log2(Foldchange) > 2, P <.001). B, Heatmap of top 10 highly expressed mRNAs in NSCLC (P <.001). C, Quantitative reverse transcription-polymerase chain reaction: MAGEC2 mRNA was extremely highly expressed in SK-MES-1, A549, and HCC827 cell lines compared with normal lung epithelial cells BEAS-2B. D, Western blot: MAGEC2 protein was extremely highly expressed in SK-MES-1, A549, and HCC827 cell lines compared with normal lung epithelial cells BEAS-2B (**P <.01, ***P <.001). MAGEC2 indicates melanoma antigen family C2; NSCLC, non-small cell lung carcinoma; mRNAs, messenger RNAs.

Melanoma antigen family C2 Promoted Proliferation, Migration, and Invasion and Suppressed Apoptosis of NSCLC

To further explore the influence of MAGEC2 on NSCLC, we used lentiviral vectors to transfect MAGEC2 cDNA or shRNA into SK-MES-1 cells, and qRT-PCR was performed to verify the regulation of MAGEC2. Three designed shRNAs were respectively transfected into SK-MES-1 cells, and shRNA2 showed the best silencing efficiency (the lowest MAGEC2 mRNA expression compared with NC; P < .01; Figure 2A). Therefore, shRNA2 was selected for performing following experiments and labeled as MAGEC2 shRNA. The MAGEC2 expression was significantly enhanced after transfecting MAGEC2 cDNA, while MAGEC2 expression was significantly declined after transfecting MAGEC2 shRNA (P < .01; Figure 2B).

Figure 2.

Melanoma antigen family C2 promoted proliferation, migration, invasion, and EMT, while suppressed apoptosis of NSCLC cells. A, Quantitative reverse transcription-polymerase chain reaction: Silencing efficiency of 3 designed shRNAs on MAGEC2 mRNA. shRNA2 was selected for following experiments and labeled as MAGEC2 shRNA. B, Quantitative reverse transcription-polymerase chain reaction: MAGEC2 was significantly upregulated after MAGEC2 cDNA transfection, while MAGEC2 was significantly downregulated after MAGEC2 shRNA transfection. C, Cell Counting Kit-8 assay: Cell proliferation of MAGEC2 cDNA group was significantly enhanced while MAGEC2 shRNA group was significantly suppressed, compared to negative control group. D, Flow cytometry: Basal cell apoptosis of MAGEC2 cDNA group was significantly reduced while MAGEC2 shRNA group was significantly ascended, compared with NC group. E, Wound healing assay: Closed wound area of MAGEC2 cDNA group at 24 hours was significantly bigger, while MAGEC2 shRNA group was significantly smaller, compared with NC group. F, Transwell assay: Invasive cell number of MAGEC2 cDNA group was significantly bigger while MAGEC2 shRNA group was significantly smaller, compared with NC group. G, Western blot: Expressions of E-cadherin and cytokeratin were significantly declined after MAGEC2 cDNA transfection or induced after MAGEC2 shRNA transfection; expressions of MAGEC2, vimentin, and fibronectin were significantly induced after MAGEC2 cDNA transfection or declined after MAGEC2 shRNA transfection (*P < .05, **P < .01). cDNA indicates complementary DNA; EMT, epithelial–mesenchymal transition; MAGEC2, melanoma antigen family C2; mRNA, messenger RNA; NSCLC, non-small cell lung carcinoma; NC, negative control; qRT-PCR, quantitative reverse transcription-polymerase chain reaction.

Cell function experiments were then conducted to investigate the effect of MAGEC2 on viability of NSCLC cells. Cell Counting Kit-8 assay showed that MAGEC2 could positively regulate proliferation of SK-MES-1 cells (P < .01; Figure 2C); flow cytometry revealed that MAGEC2 could significantly suppress basal apoptosis of SK-MES-1 cells (P < .01; Figure 2D); wound healing assay and Transwell assay indicated that MAGEC2 could positively regulate migration and invasion ability of SK-MES-1 cells (P < .01; Figure 2E and F). All the results revealed that MAGEC2 might be a key factor of NSCLC progress.

Melanoma antigen family C2 Promoted the Progression of EMT

Given that MAGEC2 could positively affect migration and invasion of NSCLC cells, which was strongly associated with EMT, we conducted Western blot to detect the differential expression of EMT marker proteins after regulating MAGEC2. The result showed that MAGEC2 could negatively regulate the expression of E-cadherin and cytokeratin and positively regulate the expression of vimentin and fibronectin (P < .01; Figure 2G). These results confirmed that MAGEC2 could promote the progression of EMT in NSCLC cells.

Melanoma antigen family C2 Facilitated Angiogenesis Ability of NSCLC

Since that angiogenesis was one of the critical factor of tumor growth, we performed ELISA and tube formation assay to explore the potential relationship between MAGEC2 and angiogenesis. Quantitative reverse transcription-polymerase chain reaction indicated that VEGF mRNA expression was positively regulated by the expression of MAGEC2 (P < .01; Figure 3A); similarly, ELISA showed that the VEGF concentration in supernatant of culture medium was positively regulated by the expression of MAGEC2 (P < .01; Figure 3B). Tube formation assay revealed that HUVEC cells mixed with MAGEC2-overexpressed SK-MES-1 cells had better angiogenesis, while HUVEC cells mixed with MAGEC2-knockdown SK-MES-1 cells had poorer angiogenesis, compared to HUVEC cells alone. Human umbilical vein endothelial cells with exogenous VEGF had similar angiogenesis with those mixed with MAGEC2-overexpressed SK-MES-1 cells and were diagnosed as the positive control (P < .01; Figure 3C). These results indicated that MAGEC2 could promote the expression of VEGF and facilitate angiogenesis in NSCLC.

Figure 3.

Melanoma antigen family C2 promoted angiogenesis of non-small cell lung carcinoma. A, Quantitative reverse transcription-polymerase chain reaction: VEGF mRNA expression was significantly upregulated after MAGEC2 cDNA transfection or downregulated after MAGEC2 shRNA transfection. B, Enzyme-linked immunosorbent assay: VEGF secretion was significantly induced after MAGEC2 cDNA transfection or reduced after MAGEC2 shRNA transfection. C, Tube formation assay: Tube formation of VEGF (positive control) or MAGEC2 cDNA group was significantly more intensive while MAGEC2 shRNA group was less intensive, compared with NC group. (*P <.05, **P <.01). cDNA indicates complementary DNA; ELISA, enzyme-linked immunosorbent assay; MAGEC2, melanoma antigen family C2; mRNA, messenger RNA; NC, negative control; VEGF, vascular endothelial growth factor.

Melanoma antigen family C2 Facilitated the Growth, Angiogenesis, and Metastasis of NSCLC In Vivo

Tumor xenograft on nude mice was performed to explore the influence of MAGEC2 on the growth and metastasis of NSCLC in vivo. Tumor volume of MAGEC2 cDNA group was significantly higher compared to the control group among the 30-day feeding, while which of MAGEC2 shRNA group was significantly lower (P < .01; Figure 4A and B). Similar to tumor volume, tumor weight of MAGEC2 cDNA group was also significantly higher, while which of MAGEC2 shRNA group was notably lower, compared to control group (P < .01; Figure 4C). Serum of all the nude mice was extracted for ELISA, and the result showed that the concentration of serum VEGF of MAGEC2 cDNA group was significantly higher, while MAGEC2 shRNA group was significantly lower (P < .01; Figure 4D). Vascular endothelial growth factor IHC staining of tumor tissues (Figure 4E) provided the similar result that VEGF-positive rate of MAGEC2 cDNA group was significantly higher while MAGEC2 shRNA group was lower compared to NC group. Hematoxylin and eosin staining of mice lung tissues revealed that the tumor of MAGEC2 cDNA group had a severer pulmonary metastasis, while MAGEC2 shRNA group had a slighter metastasis (Figure 4F). All these results indicated that MAGEC2 could facilitate the growth, angiogenesis, and metastasis of NSCLC in vivo.

Figure 4.

Melanoma antigen family C2 promoted tumor growth, angiogenesis, and metastasis of NSCLC in vivo. A, Photograph of tumors taken from the mice. B, Tumor volume of MAGEC2 cDNA group was significantly bigger while MAGEC2 shRNA group was significantly smaller, compared with NC group. C, Tumor weight of MAGEC2 cDNA group was significantly bigger while MAGEC2 shRNA group was significantly smaller, compared with NC group. D, Enzyme-linked immunosorbent assay: Mice serum VEGF concentration of MAGEC2 cDNA group was significantly higher while MAGEC2 shRNA group was significantly lower, compared with NC group. E, Vascular endothelial growth factor immunohistochemical staining of the tumor tissues (100× magnified): MAGEC2 cDNA group had a higher VEGF positive rate while MAGEC2 shRNA group had a lower positive rate, compared with NC group. F, Hematoxylin and eosin staining of lung tissues (200× magnified) of MAGEC2 cDNA group had more pulmonary nodules while MAGEC2 shRNA group had less, compared with NC group (**P < .01). cDNA indicates complementary DNA; MAGEC2, melanoma antigen family C2; NC, negative control; NSCLC, non-small cell lung carcinoma; NC, negative control; VEGF, vascular endothelial growth factor.

Discussion

In the present study, we demonstrated that MAGEC2 was a positive regulator of the growth and metastasis of NSCLC. According to the analysis result of TCGA data set, MAGEC2 was an extremely highly expressed gene in NSCLC tissues and it was barely expressed in adjacent tissues. Although the aberrant expression of MAGEC2 and its promotion to the tumor progression have been confirmed in many kinds of carcinomas, its role in NSCLC has not been clearly explored before. Firstly, we preformed both in vitro and in vivo experiments to confirm that MAGEC2 worked as an oncogene in NSCLC, which was corresponded with previous studies[14-17]; moreover, our study paid attention on the relationship between MAGEC2 and critical physiological process of NSCLC. The results substantiated our hypothesis that MAGEC2 could promote the EMT and angiogenesis of NSCLC.

Epithelial–mesenchymal transition and angiogenesis are 2 crucial tumor progressions. For that most carcinomas are originated from the mutation of epithelial cells, EMT is a necessary progression for tumor cells. It is not a simple process to acquire migration and invasion ability, but a complicated reprogramming involved in metabolism, epigenetics, and differentiation, through which differentiated epithelial cancer cells express semblable characteristics, such as expressing stem cell markers and then acquiring stem cell-like functions.[19] Numerous researches have revealed the relationship between EMT and its upstream genes. For example, an oncogene ZEB1 has been widely recognized as an EMT inducer[25]; the effect of transforming growth factor-β on regulating EMT was also investigated by many researches[26]; SNAI2 and TWIST1 were reported to regulate EMT progression in thyroid carcinomas.[27] The gene involved in our study, MAGEC2, has been considered as an EMT inducer in breast cancer (PMID: 24687377). Our study confirmed that MAGEC2 could also promote EMT in NSCLC, which would improve the understanding about NSCLC progression.

Angiogenesis is an important part of the vascular phase in tumor growth and metastasis. Extensive researches have demonstrated the effectiveness of tumor therapy targeting vascular endothelial cells.[28-30] For that there is a natural synergy between targeting vascular endothelial cells and tumor cells, it is urgent to determine gene regulation of tumor angiogenesis and develop new angiogenic inhibitors. In tumors, VEGF is a regulator of endothelial cell proliferation, migration, invasion, and survival by mediating angiogenesis and vascular tube formation.[31,32] Vascular endothelial growth factor overexpression is related to intratumoral microvessel density and a poor prognosis in varieties of cancers.[33] Downregulation of VEGF could inhibit cell proliferation and lead to tumor cell apoptosis.[34] In our study, we found that MAGEC2 could enhance the expression and secretion of VEGF, and silencing MAGEC2 could significantly suppress the expression of VEGF and the progression of angiogenesis, indicating that MAGEC2 could positively regulate VEGF activation and it might be a potential target for antiangiogenesis therapy.

In summary, activation of MAGEC2 in NSCLC functions as an important part in EMT and angiogenesis. Melanoma antigen family C2 promotes EMT progression and VEGF-induced vascular endothelial cell growth, migration, and tube formation, then facilitates angiogenesis and tumor development. Our findings also showed that silencing MAGEC2 could suppress NSCLC cell proliferation, migration, and tube formation as well as accelerate NSCLC cell apoptosis. Taken together, targeting MAGEC2 might be a potential gene therapy for NSCLC treatment. However, the specific relationship among MAGEC2, EMT, and angiogenesis was not intensively studied in our research for some restrictions. We believe that there are several signaling pathways participating in these progressions and that will be the target of our further study.