Background
Lung carcinoma is the predominant cause of cancer death both in China and worldwide [
1]. Lung cancer is mainly divided into non-small cell lung carcinoma (NSCLC) and small cell lung carcinoma (SCLC). NSCLC contributes to majority of 85% of lung carcinoma cases possessing its own biological characteristics and constitutes a heterogeneous population of adenocarcinoma, squamous and large cell carcinomas [
2]. Platinum-based drugs, particularly cis-diammine-dichloroplatinum (II) (cisplatin, DDP), are widely used in clinics. Cisplatin has been demonstrated to be an effective drug for lung carcinoma treatment effect, but it will develop drug-resistance later on [
3,
4].
We have previously found that ataxia telangiectasia mutated (ATM), a member of the phosphatidylinositol 3-kinase-related kinase family of Ser/Thr protein kinases, was induced by accumulated stimulation of cisplatin and was overexpressed in cisplatin-resistant NSCLC. Suppression of ATM expression could enhance the sensitivity of NSCLC to cisplatin treatment through activation of Erk, Akt, and MAPK pathways. Accumulated data also showed that chemo-resistance development is correlated with EMT process [
5]. Cisplatin resistance in gastric cancer cells is associated with HER2 upregulation-induced epithelial-mesenchymal transition [
6]. Furthermore, inhibition of EMT could overcome drug resistance in many types of cancers [
7]. The results indicate that EMT is associated with development of drug-resistance.
Recent studies showed that DNA damage promotes chemo-resistance and drives EMT in colorectal carcinoma [
8]. It is reported that Wip1 suppress ovarian cancer metastasis through inhibition of ATM/AKT/Snail pathway [
9]. Singh [
10] et al. found that ATM could mediate EMT in breast cancer. Several studies have shown that ATM expression is associated with EMT and metastatic potential of cancer cells. However, a report that loss of ATM accelerate EMT in pancreatic cancer [
11]. Thus, the relationship between ATM and EMT and their roles in cisplatin-resistant NSCLC is still unclear. Based on the information above, we suppose the key molecule of DNA damage response (DDR), ATM not only contribute to cisplatin resistance but also play an important role in epithelial-mesenchymal transition (EMT) progress and tumor metastasis.
In this study, we investigated the roles of ATM and EMT in cisplatin-resistance and cancer metastasis in lung cancer cells. We found that overexpression of ATM in cisplatin resistant NSCLC is correlated with the process EMT and further elucidated a novel mechanism of ATM in mediating EMT and metastatic increase in cisplatin-resistant NSCLC preliminarily both in vitro and in vivo. This study will explain and mimic clinical significance since increased invasive features of cisplatin-resistant NSCLC have reported [
12,
13].
Materials and methods
Cell culture
Lung cancer A549 and H157 cells were purchased from the American Type Culture Collection (ATCC, Manassas, VA) and cultured in RPMI 1640 containing 10% FBS. All cells were maintained in a humidified 5% CO2 environment at 37 °C. For inhibition studies, JAK inhibitor 1 (5 μM) (Calbiochem, CAS457081–03-7), Stattic (10 μM) (Calbiochem, CAS19983–44-9), and PD-L1 neutralizing Ab (10 μg/ml) (Invitrogen, 16–5982-82) that inhibits the ATM, JAK1,2, STAT3, and PD-L1 pathways, respectively, were added into the culture.
Induction of cisplatin-resistant lung cancer cell lines
Parental A549 and H157 cells were continuously treated with gradually increased dose of cisplatin (0.5 mg/ml saline stock solution, Sigma) for 6 months according to method described by Barr et al. 4. Briefly, cells were treated with 1 μM cisplatin for 72 h and cells were allowed to recover for the following 72 h. After repeating one more cycle at 1 μM cisplatin concentration, the cells were then treated with 2 μM cisplatin in the following two cycles. This procedure was continued with increasing cisplatin concentration up to 30 μM. During the cisplatin-resistance induction procedure, the IC50 values of every 5 passage cells were accessed in comparison with those of the parental cells until the IC50 value remained constant. The cisplatin-resistant cell lines obtained by this method were maintained in growth media containing 10 μM cisplatin.
Generation of ATM knocked-down and overexpression cell lines by lentiviral transduction
Lentivirus constructs carrying either ATMsiRNA or ATMRNA, or scramble sequence (pLenti-II vector, Addgene) were transfected into 293 T cells with a mixture of pLent-II-ATMsiRNA, psPAX2 (virus-packaging plasmid), and pMD2G (envelope plasmid) (4:3:2 ratio) using PolyFect Transfection reagent (Qiagen, Valencia, CA). These virus were harvested and storage for use. ATMsiRNA and ATMRNA, or sc virus infected After A549 and H157 cells overnight, the culture media containing the virus was replaced with normal culture media 48 h after infection, and maintained under normal cell culture conditions. After sub-culturing cells, the ATM knocked-down cells, overexpressing cells and sc cells were selected in the presence of puromycin (2 μg/ml) (Sigma) and then maintained in media containing 0.1 μg/ml puromycin. Stable cells were identified by determining the expression of ATM using Western blot.
MTT assay
Cisplatin-cytotoxicity was analyzed by MTT (3-[4,5-dimethylthiazol-2-yl]-2,5-dipheny ltetrazolium bromide, 5 mg/ml, Sigma, USA) assay. Cells were seeded on 96-well plates (7 × 103 cells/well) and treated with various concentrations of cisplatin for 48 h. MTT test was then performed and absorbance at 490 nm was measured. Cell viability was calculated using the formula: OD sample/OD blank control × 100. Triplicate experiments were performed and average values with mean ± SEM were represented.
Wound healing assay
Tumor cells (A549P/cisR, H157P/cisR cells and their control cells) were subjected to an in vitro wound assay with images captured at 12 h after using a microscope. Cells were seeded onto 6-well plates. When the cells reached 90–100% density, cells were scratched with 10□L sterile pipette tips. Cells were washed with PBS three times to remove detached cells and medium was replaced by fresh serum-free medium. The rate of migration was measured by quantifying the distance that cells moved from the edge of scratch toward the center of the scratch.
Transwell assay
Tumor cells (A549P/cisR, H157P/cisR cells, either A549sc/siATM and H157sc/siATM cells, or inhibitor treated cells, 1 × 104, in serum-free media) were plated in upper chamber of transwell plates and 10% FBS-containing media (as chemotactic factor) was added in bottom wells. Before the assay, the membranes were pre-coated with 8% Matrigel. The Cells were cultured for 24 h. Invaded cells at the end of 24 h of incubation were visualized by staining with a crystal violet solution and counted under a microscope. Three independent experiments (with triplicates each experiment) were done and average numbers of positively stained cells in three randomly picked areas were presented in quantitation.
In vivo mice studies
The luciferase tagged A549P, A549CisR cells and control cells (1 × 106) that were obtained by transfection of luciferase reporter gene and selection procedure were orthotopically injected through pleural (1 × 106 cells in media with Matrigel, 1:1 ratio in volume) into 8 weeks old female nude (National Cancer Institution, NCI). Tumor growth was monitored once a week by in vivo Image System (IVIS) with luciferin injection. When luminescence reached to 5 × 105 to 1 × 106 radiance(p/sec/cm2/sr), which corresponding to tumor size of 300-400 mm3, the A549cisR cells-inoculated mice were intraperitoneally (i.p.) injected with CP466722 (10 mg/kg) or vehicle (10%DMSO) every day. Tumor metastasis was monitored by IVIS once a week. After 8 weeks monitoring the metastasis, mice were sacrificed by euthanasia, the tumors were taken and fixed by formalin.
RNA extraction and quantitative real-time PCR (qPCR) analysis
Total RNA (1 μg) was subjected to reverse transcription using Superscript III transcriptase (Invitrogen). qPCR was conducted using the appropriate primers and a Bio-Rad CFX96 system with SYBR green to determine the mRNA expression levels of genes of interest using the following protocol: 95 °C for 30 s, followed by 40 cycles of 95 °C for 5 s, 55 °C for 30 s, and 72 °C for 30 s. Each sample was detected in triplicate. Expression levels were normalized to (glyceraldehyde-3-phosphate dehydrogenase, GAPDH) level. The primer sequences for ATM, E-cad, N-cad, Snail, Vimentin, JAK1, JAK2, STAT3 and GAPDH were designed as follows: ATM sense primer 5′-CAGGGTAGTTTAGTTGAGGTTGACAG-3′,antisense primer 5′-CTATACTGGTGGTCAGTGCCAAAGT-3′.E-cad sense primer 5′-CAGAAAGTTTTCCACCAAAG-3′,antisense primer 5′-AAATGTGAGCAATTVTGCTT-3′.N-cad sense primer 5′-AGCCTGACACTGTGGAGCCT-3′,antisense primer 5′-TCAGCGTGGATGGGTCTTTC-3′. Snail sense primer 5′-GAGGCGGTGGCAGACTAGAGT-3′,antisense primer 5′-CGGGCCCCCAGAATAGTTC-3′.Vimentin sense primer 5′-GGCTCAGATTCAGGAACAGC-3′,antisense primer 5′-GCTTCAACGGCAAAGTTCTC-3′.JAK1 sense primer 5′-ACCGAGGACGGAGGAAAC-3′,antisense primer 5′-ACTGCCGAGAACCCAAAT-3′.JAK2 sense primer 5′-CAGCAGCTTGGCAAAGGTAACTTC-3′,antisense primer 5′-TCAGTGCTGTGCTGGAGTTTCTTC-3′.STAT3 sense primer 5′-CAGAAAGTGTCCTACAAGGGCG-3′,antisense primer 5′-CGTTGTTAGACTCCTCCATGTTC-3′.PD-L1 sense primer 5′-TATGGTGGTGCCGACTACAA-3′,antisense primer 5′-TGGCTCCCAGAATTACCAAG-3′.GAPDH sense primer 5′-CTCCTCCACATTTGACGCTG-3′,antisense primer 5′-TCCTCTTGTGCTCTTGCTGG-3′.A melting curve analysis was performed to monitor PCR product purity and the 2 − ΔΔCt method was used to quantify the expression of these indicated genes.
Western blot analysis
Cells were lysed in RIPA buffer (50 mM Tris-Cl at pH 7.5, 150 mM NaCl, 1% NP-40, 0.5% sodium deoxycholate, 1 mM EDTA, 1 μg/mL leupeptin, 1 μg/mL aprotinin, 0.2 mM PMSF) and proteins (20–40 μg) were separated on 8–10% SDS/PAGE gel and then transferred onto PVDF membranes (Millipore, Billerica, MA). After blocking procedure, membranes were incubated with primary antibodies (1:1000), HRP-conjugated secondary antibodies (1:5000), and visualized in Imager (Bio-Rad) using ECL system (Thermo Fisher Scientific, Rochester, NY). Antibodies of ATM, p-ATM, JAK1, p-JAK1, JAK2, p-JAK2, STAT3, and p-STAT3, were from Gene Tex (Irvine, CA), Antibodies of E-cadherin, N-cadherin, Vimentin, Snail, Zeb1, Twist and VEGF were obtained from Abgent (San Diego, CA) and antibodies of PD-L1, GAPDH, were from Cell Signaling (Danvers, MA).
Immunofluorescent analysis
Cells were cultured on coverslips in a 24-well plate, fixed with 4% paraformaldehyde for 15 min and permeabilized with 0.5% Triton X-100 solution for 5 min at room temperature. Then the coverslips were blocked with 5% bovine serum albumin in Tris buffered saline with Triton X-100 for 1 h and incubated with rabbit anti-E-cad, rabbit anti-N-cad (Cell Signaling Technology) and mouse anti-Snail (Abcam Technology) primary antibodies overnight at 4 °C. After being washed with phosphatebuffered saline three times, the coverslips were incubated with FITC conjugated anti-rabbit IgG and Cy3-conjugated anti-mouse IgG secondary antibodies (Biosharp, Shanghai, China) for 1 h at room temperature. Finally, cells were labeled with 4′-6-diamidino-2-phenylindole (Beyotime) and examined using a confocal laser scanning microscopy (LSM700, Zeiss, Oberkochen, Germany).
Histology and immunohistochemistry
Tissues obtained were fixed in 10% (v/v) formaldehyde in PBS, embedded in paraffin, and cut into 5-μm sections. Tumor tissue sections were deparaffinized in xylene solution, rehydrated, and immunostaining was performed using the IHC kit (Santa Cruz, Santa Cruz). Antibodies of E-cadherin, N-cadherin, Vimentin, Snail (Cell Signaling, Danvers, MA), and ATM (Bethyl Laboratories, Montgomery, TX) (all antibodies at 1:250 dilution) were applied in staining. For ATM staining, the antigen retrieval process was performed in 10 mM Citric buffer, pH 6.0 for 20 min using a pressure cooker prior to staining. After staining, tissues were counterstained by Hematoxylin. Three microscopic visions were picked by random, and positively stained cells were determined.
Statistics
The data values were presented as the mean ± SEM. Differences in mean values between two groups were analyzed by two-tailed Student’s t test. P ≤ 0.05 was considered statistically significant.
Discussion
In our previous study, we found that increased ATM expression is related to cisplatin resistance formation, and knockdown ATM can enhance the sensitivity of cisplatin treatment in lung cancer cells. In this study, we further investigated the role of ATM in EMT and metastasis of lung cancer cells. Some studies have indicated that ATM expression is related to EMT and tumor metastasis in other tumor cells. Chen S [
22] et al. reported that ATM was involved in EMT in pancreatic cancer by regulation of long non-coding RNA ANRIL. Liu R [
23] et al. reported that depletion of ATM inhibited colon cancer proliferation and migration via Chk1/P53/CD44 cascades. In this study, we for the first time demonstrated that increased ATM expression contribute to increased EMT and metastatic potential in cisplatin-resistant NSCLC cells.
Studies have shown that ATM mainly functions as a central regulator of DNA damage response (DDR) [
24], and help responding to heat stress, hypoxia, peroxide, inflammation and radiation [
25]. So far, the mechanism by which DDR participates in drug resistance (such as cisplatin resistance) is not completely understood. While it is well known that cisplatin may inhibit DNA duplication and cause double strands breaking and crosslinking of DNA, resulting in impeding cell proliferation and inhibiting tumor growing. We further wonder whether there is any relationship between drug resistance and EMT as well as tumor metastasis. It has reported that a number of signaling pathways is involved both in chemo-resistance and EMT, including HER2-snail axis, MEK/ERK, Stat3 and AKT [
6,
7,
9,
26] In addition, Kim HP [
26] et al. showed that EMT signaling confers to acquired resistance to gastric cancer. Oliveras-Ferraros C [
27] et al. reported that EMT confers to primary resistance to trastuzumab (Herceptin). EMT process functions as a driving factor contributing to drug resistance. It could be easy to understand that cisplatin-resistant NSCLC cells had more malignant phonotype not only showing drug resistance but also exhibiting potential metastatic ability. The clinic studies have shown that the patients with cisplatin-resistance had a poor curative effect and poor prognosis. In this study, our results showed that cisplatin-resistant lung cancer cells expressed high level ATM, and over-expressing lung cancer cells exhibited cisplatin-resistance and EMT. Thus, ATM not only contributes to drug resistance, but also is an important regulator of EMT and metastasis. But the molecular mechanism by which ATM mediates chemo-resistance, EMT and tumor metastasis is not fully understood.
Metastasis is the primary cause of death in cancers, especially in NSCLC and breast cancers [
28]. Tumor metastasis is a compatible process starting with primary tumor cell invasion in which EMT plays a critical role. EMT is characterized by endothelial cell exhibiting mesenchymal phonotype by loss of cell-to-cell adhesion and increase of cell motility [
28,
29]. In this study, first, we observed that cisplatin-resistant A549cisR and H157cisR cells showed a reduced cell-to-cell contact and decreased number of cell colony compared to parental A549P and H157P cells. We observed one interesting phenomena in which A549cisR and H157cisR cells spread more directions compared to the parental cells when these cells were cultured at low concentration of serum medium during the cell migration. This phenomenon indicates that cisplatin-resistant cell subclone own more migration character, especially in environment of infertile nutrition. This special phenotype is depending on the conditioned medium containing 10 mM cisplatin. The cell phenotype and EMT ability could be reversed if cisplatin-resistant cells have been cultured in cisplatin free medium for more than 4–6 weeks (Additional file
2: Figure S2A&B). Secondly, we found A549cisR and H157cisR cells possessed an increased ability of EMT accompanied by decreasing expression of E-cadherin and increasing expressions of N-cadherin, Vimentin, Twist and Snail. Previous studies showed activated Snail can bind to the E-boxes of E-cadherin inhibiting its expression at plasma membrane and ATM can stabilize Snail via phosphorylation at serine 100 [
30,
31]. Our results showed that ATM inhibitor treatment in A549cisR and H157cisR cells or siRNA knock-down of ATM in A549cisR and H157cisR cells resumed epithelial cell phonotype, showing an increased cell-to-cell contact, round shape, and colony formation. In addition, increased E-cadherin and decreased N-cadherin, Vimentin, Twist, snail, Zeb were also observed in ATM inhibitor treated cells or A549cisR-siATM and H157cisR-siATM cells, suggesting that ATM inhibition can reverse EMT and metastasis. In addition, ATM inhibitor-treated A549cisR and H157cisR cells or A549cisR-siATM and H157cisR-siATM cells abrogated the increased ability of cell invasion. In summary, all above findings showed an essential role of ATM in EMT and metastasis in cisplatin-resistant cells.
It has been reported that ATM effects on EMT and metastatic potential through activation of JAK/STAT3 pathway [
32]. Our previous results showed that continuous cisplatin treatment induce ATM expression and EMT [
33]. Recent studies have shown that PD-L1 its function on promoting cell migration and invasion are gradually emphasized. Kim [
17] et al. proved PD-L1 expression is associated with EMT in adenocarcinoma of lung. Wang Y [
34] et al. reported PD-L1 induced EMT via activating SREBP-1c in renal cell carcinoma. PD-L1 (CD274) is regarded as an important mediator of T cell proliferation and PD-L1/PD1 signaling inhibits immune response and results in immune escape of tumor cell [
35‐
37]. Studies have shown that PD-L1 expression can be induced by extrinsic stimulation such as interferon-gamma produced by surrounding tumor cells [
38], and by the activation of intrinsic oncogenic pathways, such as STAT
3, an activating EGFR mutation or ALK translocation [
39,
40]. In this study, we found that ATM regulated PD-L1 expression via activation of JAK
1,2/STAT
3 pathway, consisting with previous reports [
41]. JAK/STAT signaling activation, which is required for diverse process during embryogenesis and now, is thought to be associated with cancers. JAK(
S) could be activated after an extracellular ligand binds to its receptor. The phosphorylated JAK/STAT proteins then dimerize and shuttle into the nucleus where they function as a transcription factor. STAT
3 is regarded as one of the master regulators of EMT programs including twist, snail, slug, foxc2, zeb1 and zeb2. In this study we found that STAT
3 signaling activation regulated PD-L1 expression in cisplatin-resistant lung cancer cells.
So far, no studies have revealed the detail mechanisms how ATM regulates PD-L1 expression through JAK
1,2/STAT
3 pathway. Zhang [
42] et al. found that ATM was involved in phosphorylation of STAT3. Similarly, SUN [
31] et al. reported that ATM stabilized Snail via phosphorylation at serine 100, and allowed Snail to down-regulate E-cadherin expression resulting in EMT finally. Since ATM protein is not a transcription factor, we wonder whether it may regulate EMT through downstream signaling molecules via mythelation, phosphorylation or ubiquitination. We found that ATM doesn’t regulate PD-L1 directly, but activate JAK/STAT3 signaling.