Background
Lung cancer is one of the most common cancers and the leading cause of cancer related death [
1]. Non-small cell lung cancer (NSCLC) accounts for about 80% of diagnosed patients with lung cancer [
2]. Despite the progress on targeted therapies [
3,
4], understanding the complicated pathogenesis mechanisms is still limited. Moreover, rapid progression on drug resistance affects the effectiveness of chemotherapies and targeted therapies. Thus, identification of novel biomarkers and targets are useful in developing new therapeutic options for NSCLC.
Major vault protein (MVP), also known as lung resistant protein (LRP), is ubiquitously expressed in most animal cells. It is the major component of vault that is the largest known ribonucleoprotein particle in cytoplasm [
5]. Current studies indicate that vault is involved in a broad range of cellular processes, including nuclear pore assembly, subcellular transportation, cell signaling, and interferon response [
5‐
8]. Although MVP is overexpressed in the drug-resistant cancer cells [
9‐
11], the definite role of it in NSCLC is still a disputable issue [
12,
13]. Janikova et al. reported that the MVP expression is of prognostic significance in NSCLC when examined in combination with miR-23b [
14]. MVP is required for the nuclear localization of tumor suppressor PTEN [
15,
16], which down-regulates cyclin D1, prevents the phosphorylation of MAPK, and leads to cell cycle arrest [
17]. MVP also binds to HIF1α and promotes the degradation of HIF1α. It supports the notion that MVP may function as a tumor suppressor in renal adenocarcinoma cells [
18]. Yet, MVP also promotes survival and migration of glioblastoma [
19], and suppresses apoptosis of human senescent diploid fibroblasts [
20] and human colon cancer cells [
21]. These inconsistent results suggest that insighted mechanistic researches on the role of MVP in cancer are absolutely necessary.
By examining 120 patients with NSCLC, we found that MVP expression was significantly up-regulated in cancer tissues compared with the paired normal tissues. Higher MVP expression was correlated with better clinical outcomes in patients with lung cancer, especially in patients with adenocarcinoma. When MVP was knocked down in Lewis lung carcinoma (LLC) cells and the MVP suppressed LLC cells were injected subcutaneously into mice, it promoted lung cancer growth and the tumor cell apoptosis was inhibited. This was causally linked to the activation of STAT3 signaling pathway. Our findings suggest that MVP act as a NSCLC suppressor which may be useful for discovery of novel therapy.
Methods
Patient cohort and tissue collection
This study was approved by the Institution Review Board of Bengbu Medical College and Nanjing Medical University. Patients were recruited from 2011 to 2013 in the First Affiliated Hospital of Bengbu Medical College. The cohort is composed of 44 patients with adenocarcinoma and 76 patients with squamous cell carcinomas without previous lung cancer history or preoperative chemotherapy and radiotherapy.
The lung cancer samples and matched noncancerous lung tissues (more than 5 cm from the tumoral margins) were applied for the tissue microarray (TMA) construction. The TMAs were created by contract service at Shanghai OUTDO Biotech, China. Duplicate 1.0-mm diameter cores of tissue from each sample were punched from paraffin tumor block and corresponding nontumoral tissues. As a tissue control, the biopsies of normal lung tissues were inserted in the angles of each slide.
Immunohistochemistry
The MVP antibody (1:100 dilution, Santa Cruz) was used to determine the protein expression levels. The goat IgG served as a negative control. The immunoreactivity score (IRS) were evaluated by two pathologists independently using the following semiquantitative criterion: the intensity of immunostaining was recorded as 0–3 (0, negative; 1, weak; 2, moderate; 3, strong); the percentage of immunoreactive cells was recorded as 1–4 (1, 0–25%; 2, 26–50%; 3, 51–75%; 4, 76–100%). The IRS was calculated by multiple the intensity and percentage of immunoreactive cells. Wilcoxon test (raw scores) was applied to determine the significance of MVP staining in primary lung tumors compared with the matched adjacent tumoral tissues. Mouse tumor tissues were formalin-fixed, paraffin-embedded, and sectioned. Immunohistochemistry (IHC) of mouse tissue sections was conducted with antibodies against Ki67 and CD31 (Abcam).
Animal experiments
All aspects of the animal care and experimental protocols were approved by Nanjing Medical University Committee on Animal Care. The C57BL/6 J mice were purchased from Animal Core Facility of Nanjing Medical University and kept in animal care facilities under pathogen-free conditions. Tumor xenograft assays were performed with 6 to 8-week-old mice. Briefly, 5 × 106 tumor cells per site were suspended in 0.1 ml PBS and subcutaneous injected into mice. After 3 weeks, mice were euthanized by carbon dioxide, and tumors were dissected for determining the size and weight, followed by IHC staining and flow cytometry analysis. Tumor volume was calculated using the formula: volume = 0.5236 × length × width2.
Cell culture
The mouse Lewis lung carcinoma (LLC) and human lung adenocarcinoma SPC-A1 cells (Chinese Academy of Science) were cultured in DMEM medium containing 10% FBS with the supply of 5% CO2. To generate stable knockdown cells, the lenti-viruses including hairpin (Genepharma, China) were used to infect LLC cells. After infecting for 24 h, cells were selected by 8 μg/ml puromycin for at least 2 weeks before conducting experiments. The hairpins sequences against mouse MVP were: shRNA-MVP1, CATAAGAACTCAGCACGTATTCAAGAGATACGTGCTGAGTTCTTATG; shRNA-MVP2, CCATCGAAACTGCAGATCATTCAAGAGATGATCTGCAGTTTCGATGG. The hairpin sequence against human MVP was: shRNA-hMVP, GGTGCTGTTTGATGTCACATTCAAGAGATGTGACATCAAACAGCACC.
Cell proliferation and apoptosis analyses
Cell proliferation assays were evaluated by the cell counting kit-8 assay (CCK-8) (Dojindo Laboratories, Kumamoto, Japan). Briefly, 2 × 103 cells were seeded in 96-well plates and cultured for 1 to 6 days. Then, the CCK-8 assays were conducted according to the instruction followed by 450 nm absorbance measurement using a plate reader. For colony formation assays, cells were seeded in 10 cm-plates at a density of 200 cells per plate. After culturing for 14 days, cells were fixed with 4% paraformaldehyde for 30 min and stained with 0.1% crystal violet for 20 min. Three randomly selected areas were used to count the number of colonies. For Edu assays, cells were seeded into 96-well plates at a density of 3 × 103 cells per well. Cells were treated with serum-free medium for 24 h, regular medium for 24 h, and regular medium plus the Edu for 2 h. Cells were then fixed with 4% PFA and stained with DAPI.
Flow cytometry analysis
Cells were harvested in logarithmic growth phase, washed with PBS and fixed in 70% ethanol at 4 °C for at least 12 h. Then, the cells were washed in cold PBS, stained with propidium iodide (PI) in the darkness for 30 min and resuspended in PBS at 4 °C before analyzed by BD FACS Calibur. Annexin V/PI apoptosis detection kits (BD Biosciences) were used to detect apoptotic cells. Flow cytometry data were analyzed by using BD CELLQUEST software supplied with the instrument. Flow cytometry for the LLC tumor stromal cells was conducted as previously described [
22]. Briefly, flow cytometric identification of the cells was performed through labeling with FITC-labeled CD11b antibody (AbD Serotec), APC-labeled Gr1 antibody (BD Biosciences), PE-labeled F4/80 antibody (R&D System) and APC-labeled CD11c antibody (eBioscience) or FITC-labled CD206 antibody (AbD Serotec).
Western blot
Western blot was conducted as previously described [
23]. Primary antibodies against Cyclin-D, p-Rb, Rb, Capase 3, p-STAT3, STAT3, LMNB1, p-JAK, JAK, p-Raf, Raf, p-MEK, MEK, p-ERK, ERK, p-AKT, and AKT were purchased from Cell Signaling Technology. Antibodies against MVP and GAPDH were purchased from Santa Cruz Biotechnology. Quantification was performed with Image J.
Luciferase reporter assay
LLC cells that were grown to 80% confluence in 24-well plates were co-transfected with the pGL6-STAT3-luciferase reporter plasmid and pRL-TK plasmid (Promega) at an appropriate ratio using Lipofectamine 2000 (Invitrogen). Luciferase activity was assayed after 24 h using the dual-reporter luciferase system on a GloMax-96 luminometer (Promega).
Statistical analysis
Statistical analysis was performed using Stata 15.0 software. For experimental data, continuous values were described with Mean ± standard error (SE) and tested by ANOVA followed by Bonferroni correction. For the patient cohort, the median value was regarded as cut-off value. Unpaired Student’s
t test was done for the comparison between two groups. Wilcoxon test (raw scores) was applied to determine the significance of MVP staining in primary lung tumors compared with the matched adjacent tumoral tissues. Kaplan-Meier survival analysis was performed and tested by Log-rank test. The association between the transcriptional levels of MVP and overall survival of NSCLC patients was evaluated by an online database (
http://kmplot.com/analysis/). The analysis parameters were set as following: split patients by “Auto select best cutoff”, probe set options using “only JetSet best probe set”, array quality control “exclude biased arrays”. Error bars represent SE for all figures. Statistical significance was defined as follows: *,
P < 0.05; **,
P < 0.01.
Discussion
Lung cancer, causing more deaths than any other cancer types, accounts for approximately one fourth of all cancer-related deaths [
1]. Treatment of lung cancer is of great importance clinically and new approaches on targeted therapies by molecular subtyping have been developed [
2]. For example, Gefitinib and other EGFR tyrosine kinase inhibitors (TKIs) have been developed to treat patients with EGFR mutations which may be the most successful targeted therapy [
30]. ALK inhibitor Crizotinib has been approved for treatment of patients carrying EML4-ALK fusion mutation [
4]. Therapies targeting K-RAS mutation, MET amplification, ROS1 rearrangements, and other oncogenic drivers are under clinical trials [
31]. However, only 10 to 25% NSCLC patients carry EGFR mutations [
31]. Other drivers have ever smaller incidences. Therefore, identifying novel target genes for lung cancer therapy is urgent. In the present study, MVP was found to be of anti-tumor property with potential application for treatment of NSCLC.
It is intriguing that MVP expression was increased in surgical removed NSCLC compared with paired normal-adjacent tissues. Similar reaction pattern of MVP is also found in pancreatic tumor [
32]. The up-regulation of MVP is likely triggered by the increased transcription activity in tumor cells. The core promoter sequence of MVP contains several putative transcription factor binding sites for p53 and YB-1, which are increased in tumorigenesis [
32‐
34]. The up-regulation of tumor suppressor like p53 may constitute a feedback response in which transformed cells manage to restore normal growth controls. Malignant lesions are probably the consequences of escaping or overcoming the MVP-associated anti-tumor regulations. Indeed, our in vivo and in vitro studies reveal that suppression of MVP in lung cancer cells resulted in robust tumor growth and suppressive tumor cell apoptosis. Interestingly, the up-regulation of MVP has also been found in other situations including chemotherapy resistance, malignant transformation, as well as exposure to diverse antineoplastic drugs [
5], suggesting the complexity in regulation of MVP expression in tumor cells. Yet, our population investigation reveals a better clinical outcome for the high expression of MVP in lung cancer, especially in adenocarcinoma. Therefore, MVP may be associated with the suppression of lung cancer.
NSCLC is thoughted as a group of distinct diseases with genetic and cellular heterogeneity [
35]. There are many genetic and epigenetic differences between the main NSCLC histologic subtypes, squamous cell carcinoma and adenocarcinoma. Following the identification of
KRAS and
BRAF mutations, epidermal growth factor receptor (
EGFR) mutations were discovered in patients with lung adenocarcinoma and were associated with response to EGFR inhibitors. Instead, for lung squamous cell carcinoma, genes such as
DDR2,
FGFR and genes in the PI3K pathway seem to be more commonly mutated. [
35,
36] Difference in the prognosis of MVP expression for adenocarcinoma and squamous cell carcinoma may reflect the complexity of their pathogenesis. For example, STAT3 mediates the oncogenic effects of EGFR kinase domain mutations in human lung adenocarcinoma [
37,
38]. The efficacy of EGFR tyrosine kinase inhibitors (TKIs) in EGFR-mutant NSCLC is limited by adaptive activation of STAT3 [
39]. The association of high expression of MVP with better clinical outcome in adenocarcinoma may be attributed partly to the suppressive effect of MVP on STAT3 signaling pathway.
However, the anti-tumorigenesis effect of MVP has been challenged in glioblastoma [
19] and colon cancer cells [
21], in which MVP promotes tumor cell survival and clonogenicity and inhibits cell apoptosis. Furthermore, MVP-mediated selective sorting of tumor suppressor miRNA into exosomes promotes tumor cell growth and colon cancer progression [
40]. As an ubiquitously expressed protein, MVP may exert impacts on tumorigenesis in a context-dependent manner. The genetic background, cell linage, and alterations in other pathways may shape the role of MVP in the niche. Indeed, molecules including TGF-β [
41], Notch [
42], Runx [
43], and KLF4 [
44], exhibit opposite functions in different cancer settings. The requirement of MVP for the nuclear localization of tumor suppressor PTEN would lead to tumor cell cycle arrest [
17]. MVP can also promote the degradation of HIF1α, a known tumor promoting factor, and function as a tumor suppressor in renal adenocarcinoma cells [
18]. We demonstrate that MVP may directly inhibit lung cell proliferation and stimulate cell apoptosis. As such, MVP may act as a suppressor for tumorigenesis in lung cancer.
Effect of MVP on signal transduction is also inconsistent as its characteristics in tumorigenesis. MVP has been reported to activate the EGFR/PI3K/AKT signaling pathway in glioblastoma [
19] and colon cancer cells [
21]. However, MVP also functions as a negative regulator for the growth signaling. For examples, MVP binds with YPEL4 and inhibits YPEL4 catalyzed activation of Elk-1 in the MAPK signaling pathway [
45]. Our results reveal that MVP inhibited lung cancer growth by suppression of STAT3 signal pathway, which is regulated by JAK2 and RAF-MEK-ERK pathways. This is consistent with the observation that MVP binds with and inhibits Src kinase activity and, thus, suppresses ERK activation in stomach cancer cells [
46]. However, in human airway smooth muscle cells knockdown of MVP induces cell death by inhibiting STAT3 and Akt signaling [
47], which is contradictory to our observation in lung tumor cells. Indeed, cancer cells like NSCLC cells are usually endowed with enhanced progrowth signals to satisfy their uncontrolled proliferation [
48]. The property of being phosphorylated suggests that MVP be an intrinsic signal transducer [
46,
49]. MVP can bind with some signal molecules, such as Src, SHP2, ERK [
46,
49], and transcription factors C-FOS and C/EBPβ [
50]. Therefore, the mechanisms underlying suppression of STAT3 signaling in lung cancer cells by MVP is warranted further exploration.