Background
Lung cancer is the leading cause of cancer-related deaths in China and other countries [
1,
2]. Non-small cell lung cancer (NSCLC) accounts for 85% of lung cancers. Despite the advances in cancer research and treatment, the prognosis of NSCLC is still poor, with the 5-year survival rate being only 15% [
3]. Although new drugs such as gefitinib and erlotinib have been shown to be beneficial, especially in patients with sensitive target mutations, the survival and outcome have not changed dramatically. Therefore, it is important to understand the pathogenesis of NSCLC and identify new treatment targets.
Recently, the purinergic signaling pathway, in which extracellular ATP, ADP and adenosine are the main signaling molecules, has emerged as an important player in cancer progression [
4]. Purinergic signaling is a multistep coordinated cascade that involves stimulated release of ATP/ADP, triggering of signaling events via P2 receptors, and inactivation of the nucleotide to form adenosine, which binds to its own activated P1 receptors and influences cell survival, proliferation and cell motility [
5]. Ecto-5′-nucleotidase (CD73), a component of the purinergic signaling pathway, is a 70-kD glycosylphosphatidylinositol-anchored cell surface protein encoded by the
NT5E gene that plays a crucial role in switching on adenosinergic signaling. CD73 has both enzymatic and non-enzymatic functions in cells [
6]: as a nucleotidase, CD73 catalyzes the hydrolysis of AMP into adenosine and phosphate, and CD73-generated adenosine plays an important role in tumor immunoescape [
7]; moreover, CD73 also functions as a signal and adhesive molecule that can regulate cell interaction with extracellular matrix components, such as laminin and fibronectin, to mediate the invasive and metastatic properties of cancers [
8,
9]. Both the enzymatic and non-enzymatic functions of CD73 are involved in cancer-associated processes and are not completely independent of each other [
10]. There is ample evidence to show that CD73 is a key regulatory molecule in cancer development and is overexpressed in many cancers, including leukemia, glioblastoma, melanoma, ovarian cancer, esophageal cancer, prostate cancer and breast cancer [
10]. CD73 expression is also associated with certain clinical characteristics and the prognosis of cancer patients [
9,
11‐
15]. In particular, due to its favorable effects in tumor-bearing mouse models, which have not been investigated in the clinic, anti-CD73 therapy is now a promising approach for cancer treatment in the future [
16,
17]. However, the role of CD73 in lung cancer remains unclear. Moreover, despite its functional importance, little is known about the transcriptional regulation of CD73 [
18‐
21].
Studies have shown that the prognosis of cancer is closely related to the altered expression of miRNAs in cancer tissues and specific expression signatures or panels [
22], which can also be used to classify human cancers [
23] and distinguish between tumor subtypes [
24]. Recent research has shown that alteration in miRNA expression may be involved in the regulation of epithelial-to-mesenchymal transition in tumor progression [
25]. In particular, there is some evidence that miRNAs are closely related to the development of human lung cancer [
26,
27]. In our recent study, we used miRNA arrays to demonstrate the impact of significant miRNAs on cellular pathways and biological processes, and showed that miR-30a-5p expression was significantly downregulated in NSCLC tissues [
28]. To identify more novel targets of miR-30a-5p that may play a role in NSCLC, in the present study, we predicted its target mRNAs using computational algorithms. Interestingly, miR-30a-5p was one of only two miRNAs that could bind to the 3′-UTR of CD73 mRNA. Thus, miR-30a-5p may be involved in the regulation of CD73 in cancer progression.
Here, we aimed to evaluate the role of CD73 in the tumorigenesis of NSCLC, and to explore the possible role of miR-30a-5p in CD73 dysregulation in lung carcinogenesis.
Discussion
The findings of the present study indicate that CD73 is a critical factor in the tumorigenesis of NSCLC and that its mechanism involves the EGFR signaling pathway. This finding is consistent with the results of studies on other cancer types [
11,
32].
miR-30a has been found to be one of the miRNAs that are downregulated in various types of solid tumors, including ovarian cancer, colon cancer, prostate cancer, myeloma cells and lung cancer [
34‐
36]. Further, in an elegant study conducted by Cheng et al., miR-30a was found to be significantly downregulated in the primary tumor of patients with metastatic cancer; this implies that miR-30a has some putative function in cancer and potentially metastasis [
37]. However, the tumor suppressive role of miR-30a is poorly understood. In the present study, we have reported for the first time that miR-30a-5p plays an important role in the suppression of cell proliferation, migration and invasion in NSCLC via its effects on CD73 gene expression both in vitro and in vivo.
In the present study, we detected the expression of CD73 in NSCLC tissues and cell lines. CD73 expression was upregulated in NSCLC tissues compared with matched paracancerous tissues. Moreover, we identified the expression of CD73 in the NSCLC cell lines examined, which shows that CD73 is frequently overexpressed in NSCLC. Our findings indicate that shRNA-induced knockdown of CD73 significantly inhibits NSCLC cell proliferation by repressing the NSCLC cell cycle. The co-expression of CD73 and EGFR has been reported in other types of cancers [
11,
32]. Consistent with the previous findings, in the present study too, CD73 was found to promote EGFR expression. However, the interactions and underlying mechanisms need to be elucidated. Therefore, our findings demonstrated that CD73 may play a tumor-promoting role in NSCLC via its effects on EGFR signaling.
Although there is ample evidence for the upregulation of CD73 expression in human cancers, the underlying mechanisms are poorly understood. In this study, we focused on NSCLC cells and tissues to determine whether miRNAs can epigenetically influence CD73 expression. Intriguingly, the miR-30a-5p level was found to be reduced in NSCLC cells and tissues, which is in line with the findings of other studies [
36,
38]. The results imply that miR-30a-5p functionally contributes to the expression of CD73 in NSCLC. Although several miRNAs, including miR-30a-5p, have been implicated in prostate cancer, myeloma cells and lung cancer, the targets of miR-30a-5p have not yet been identified [
36]. Therefore, in the present study, we performed in silico prediction of microRNA targets and found that miR-30a-5p can potentially bind to two target sites of CD73 3′-UTR, and that miR-30a-5p is one of only two such molecules that can bind to the 3′-UTR of CD73 mRNA. Next, we used two methods to confirm whether CD73 is a bona fide target of miR-30a-5p. First, a luciferase reporter assay was performed, and it showed that miR-30a-5p can selectively target two putative site of the CD73 3′-UTR. Second, ectopic expression of miR-30a-5p in NSCLC cells was examined and found to be significantly reduced, and CD73 mRNA and protein expression was found to have increased. This provides a strong rationale for our findings that both low miR-30a-5p and high CD73 expression are found in NSCLC.
It is known that miR-30a-5p is frequently downregulated in human cancer tissues [
35,
38] as well as in human epithelia-derived cancer cells [
37]. Our present data showed that CD73 is required for NSCLC cell proliferation (Fig.
2). These findings together seemed to indicate the possibility that miR-30a-5p may inhibit NSCLC cell proliferation. As expected, ectopic miR-30a-5p expression was found to inhibit NSCLC cell proliferation and migration in NSCLC cells, which further confirms the tumor-promoting function of CD73. However, we cannot exclude the possibility that miR-30a-5p may have different functions, may be even tumor-inhibiting functions, in other types of cancer, because a single miRNA may have different functions in different cellular contexts [
39].
A considerable amount of evidence suggests that miRNAs play a role in fighting drug resistance. Overexpression of CD73 is also associated with resistance to antitumor agents. Further, some studies have shown the therapeutic potential of CD73 blockade for cancer therapy [
40‐
42], and the effects of CD73 on tumor cells via the EGFR signaling pathway [
11,
32]. In keeping with all these previous findings, we found that shRNA-mediated CD73 knockdown increased the sensitivity of NSCLC cells to gefitinib and lapatinib.
It was recently reported that EGFR signaling is coupled with activation of cap-dependent translation in NSCLC cells expressing wild-type EGFR [
11]. Resistance to EGFR-TKI can be mediated through multiple signaling pathways that converge upon cap-dependent translation in NSCLC cells expressing wild-type EGFR. Interestingly, our present study showed that CD73 affected the efficacy of EGFR-targeted therapies in NSCLC cells with wild-type EGFR. Thus, future investigation into the molecular mechanisms of CD73-mediated drug resistance could help develop novel CD73-based therapeutic agents to improve the treatment of NSCLC.
Methods
Tissue samples
Fifty-nine paired NSCLC tissues and adjacent noncancerous lung tissues were collected, with the informed consent of the patients from the First Affiliated Hospital of Soochow University between 2009 and 2013. The patients had been diagnosed with NSCLC based on their histological and pathological characteristics according to the Revised International System for Staging Lung Cancer. They had not undergone chemotherapy or radiotherapy prior to tissue sampling. The tissue samples were snap-frozen and stored in a cryofreezer at −80 °C. This study was approved by the Academic Advisory Board of Soochow University.
Cell culture
Human NSCLC cell types A549, H1650 and SPC-A1 (lung adenocarcinoma cell lines) and H460 (giant-cell lung carcinoma cell line), 95C, 95D and H226, H520 (lung squamous carcinoma cell line) were purchased from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). The cells were grown in RPMI 1640 medium containing 10% fetal bovine serum (FBS) (Gibco, Carlsbad, CA, USA) and l-glutamine (Invitrogen, Carlsbad, CA, USA) at 37 °C in a humidified atmosphere containing 5% CO2. Genetic characteristics of the cells were determined by the Beijing Microread Genetics company using a Goldeneye™ 20A Kit and ABI 3100. All cell lines were passaged for less than 3 months and tested in January 2016.
Immunohistochemical assay
Immunohistochemical (IHC) analyses of tissues were conducted as described in our previous study [
43]. In brief, sections were incubated with EGFR (A-10) and CD73 (IE9) specific monoclonal primary antibodies (diluted 1:100; Santa Cruz Biotechnology, Santa Cruz, CA, USA) overnight at 4 °C, then incubated with biotinylated secondary antibodies. The reactions were developed using the DAB Kit (BD Bioscience, San Jose, CA, USA) and the sections were counterstained with hematoxylin.
RNA extraction and quantitative real-time PCR analysis
RNA isolation, cDNA synthesis and quantitative reverse transcription–PCR analysis were performed as previously described [
28]. The primer sequences used for CD73 mRNA detection were 5′-TCTTCTAAACAGCAGCATTCC-3′ (forward) and 5′-CATTTCATCCGT GTGTCTCAG-3′ (reverse). Ct values for CD73 mRNA and miR-30a-5p were equilibrated to
ACTB mRNA and U6, respectively, which were used as internal controls. The △△Ct method was applied to calculate the relative expression of these proteins.
Kinase and Western blot assays
5′-(N-Ethylcarboxamido) adenosine (NECA, E2387) and Adenosine 5′-(α, β-methylene) diphosphate (APCP, M3763) were purchased from Sigma. Cells were seeded into 6-well plates at a concentration of 30 × 104 cells/well. After 24 h, cells were treated with NECA (1 μM) for 24 h or APCP (10 μM) for 1 h, then were harvested and lysed in RIPA buffer (Cell Signaling Technology, Danvers, MA, USA) containing a protease and phosphatase inhibitor cocktail (Sigma-Aldrich, St. Louis, MO, USA).
For the human receptor tyrosine kinase (RTKs) assay, we used the phosphorylation antibody array-AAH-PRTK-1 (RayBiotech Inc.). Protein lysates were incubated with the array membrane and protein signal was visualized using a chemifluorescence detection system (Bio-Rad) according to the manufacturer’s protocol. Relative density of specific protein expression was determined using Quantity One software.
Western blot analysis was performed as previously described [
43]. The antibodies used in the analysis were anti-CD73 (D7F9A), anti-pEGFR (Tyr1068, 1H12), anti-pAKT (Ser473, D9E), anti-AKT, anti-pERK (Thr202/204, D13.12.4E), anti-ERK (137 F5), anti-pFak (Tyr397, D20B1), anti-FAK (D2R2E) and anti-CyclinD1 (92G2, all from Cell Signaling Technology, Danvers, MA, USA), anti-EGFR (A-10, Santa Cruz, CA, USA), anti-β-actin and anti-mouse or anti-rabbit secondary antibodies (Cell Signaling Technology).
Plasmids construction, transient transfection and luciferase assay
To construct a plasmid containing the CD73 3′-untranslated region (3′-UTR) fused to the 3′-end of a luciferase reporter, 1746-bp sequences containing the predicted miR-30a-5p target sites were synthesized and ligated into the pGL3-control vector (Promega, Madison, WI, USA). CD73 3′-UTR was amplified with the primers 5′-GGCTAGTCTAGACTGCCTTTTAGGACCTGGCT-3′ (forward) and 5′-GGCTAGTCTAGA ACCGAGGCTATTATTTTGCTGC-3′ (reverse) and were subcloned into a pGL3 control vector with the restriction endonuclease XbaI site (italic font) to generate pGL3-CD73-3′-UTR. The 3′-UTR of CD73 containing two putative miR-30a-5p-binding sites was amplified and cloned into a pGL3-control vector separately. In the mutated fragment, four mutational bases were introduced into the predicted miR-30a-5p target sites. The wild-type and mutated fragments were directly synthesized (Genewiz, Suzhou, China), digested with XbaI and subcloned into the pGL3-control vector. Subsequently, cells were seeded into 24-well plates and co-transfected with the constructed plasmid, the pRL-TK plasmid and with either miR-30a-5p mimics or miR-NC; miR-30a-5p inhibitor and matched NC were purchased from RiboBio Co., Ltd (Guangzhou, China). After 48 h, cells were collected and their luciferase activity was measured using the Dual-Luciferase Reporter Assay Kit (Promega). The results are expressed as the relative firefly luciferase activity, which is obtained after normalization to Renilla luciferase activity. All the transient transfections, including the miR-30a-5p mimic, inhibitor and miR-NC transfections, were performed using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA).
Establishment of CD73-silenced stable cell lines
To establish stable cell lines with silenced CD73 expression, two DNA fragments (CD73 shRNA-1, 5′-GATCCGGAATCGTTGGATACACTTCCTTCAAGAGAGGAAGTG TATCCAACGATTCCTTTTTTG-3′; CD73 shRNA-2, 5′-GATCCGCCGCTTTAGAGAATG CAACATTCAAGAGATGTTGCATTCTCTAAAGCGGCTTTTTTG-3′) were subcloned into the lentiviral vector pGLV2-U6-Puro (GenePharma, Shanghai, China) containing the endonucleases AgeI and EcoRI. A scrambled sequence (underscored) of CD73 shRNA, which served as the negative control, was used: 5′-GATCCGTTCTCCGAACGTGTCACGTT TCAAGAGAACGTGACACGTTCGGAGAACTTTTTTG-3′. Then, the CD73-silenced construct or negative control was co-transfected with packaging plasmids into human embryonic kidney 293 T cells using Lipofectamine 2000 (Invitrogen). After 48 h, cells were infected with the packaged lentiviruses and cultured for 2 days before being selected with 0.4 μg/ml and 2 μg/ml of puromycin, respectively (Sigma-Aldrich, St Louis, MO, USA).
Cell proliferation analysis and drug treatment
Cell proliferation was examined using Cell Counting Kit-8 (Boster, Wuhan, China). Cells or the corresponding negative control cells were seeded in 96-well plates at a density of 2 × 103 cells per well and further grown under normal culture conditions for 24, 48 and 72 h. Cell viability was determined according to the manufacturer’s instructions. The experiment was performed in triplicate. We also detected cell proliferation using a clonogenic assay. In brief, cells transfected with miR-30a-5p mimics and sh-CD73 or sh-NC were suspended in complete culture medium and 200 cells were reseeded into a 60 mm plate. After incubation for 14 – 20 days, foci of least 50 cells were stained with Giemsa and counted. Cell viability was measured according to manufacturer’s instructions at several time points (24, 48 and 72 h). Each experiment was performed in triplicate. For drug treatment, stable CD73-knockdown cells were plated into 96-well plates, and gefitinib and laptinib (gefitinib C#s1025, laptinib C#s2111, Selleck Chemicals, Houston, TX, USA) were added to the cultures. Cell viability was assessed 72 h after the drug treatment.
Wound healing, migration and invasion assays
The motility of cells was analyzed as previously described [
43]. For the wound healing assay, cells were seeded into 6-well plates and cultured for 48 h to form a monolayer. The monolayer was then scratched with a new 10-μl pipette tip across the center of the well, washed gently twice with PBS and replenished with fresh medium. The cells were grown for an additional 24 h and were visualized on a microscope (CKX41, Olympus).
For the migration assay, 5 × 104 cells in medium containing 1% FBS were seeded into the upper chamber of a Transwell insert, 800 μl medium containing 10% FBS was into the lower chamber and then incubated at 37 °C for 24 h, according to the manufacturer’s instructions. For the invasion assay, the inserts were coated with the Matrigel matrix (BD Science, Sparks, MD, USA) and cells were diluted in serum-free medium before plating and incubated at 37 °C for 2 h. In both assays, the cells were then photographed and counted.
Cell cycle analysis
According to the instructions of the Cell Cycle Analysis kit (Beyotime, Shanghai, China), cells were cultured in 6-well plates. The cells were collected, washed with cold PBS, fixed in 70% ethanol at 4 °C for 24 h, washed with cold PBS again, and stained with a propidium iodide (PI)/RNase A mixture. The cells were then kept in the dark at 37 °C for 30 min and analyzed with a fluorescence-activated cell sorting (FACS) caliber system (Beckman Coulter, Brea, California).
Animal experiments and immunocytochemistry staining
Female BALB/c athymic nude mice (4–6 weeks old and weighing 16–20 g) were purchased from the Experimental Animal Center of Soochow University and bred under pathogen-free conditions. All the animal experiments were carried out in accordance with the Guide for the Care and Use of Experimental Animals Center of Soochow University. To establish the lung carcinoma xenograft model, A549 cells and stable CD73-knockdown A549 cells were suspended in 100 ml PBS and inoculated subcutaneously into the flanks of nude mice. After 8–10 days, the nude mice with transplanted A549 cells were randomly divided into two groups (8 mice in each group). The mice with transplanted CD73-knockdown A549 cells were also split into two groups. An miR-30a-5p agomir and NC agomir (RiboBio Co. Ltd., Guangzhou, China) were directly injected into the A549 implanted tumors at a dose of 2 nmol (in 20 μl PBS) per mouse every 4 days seven times (28 days total). Chemically stabilized miRNAs may have markedly improved pharmacological properties, as described before [
44]. Tumor volume (
V) was determined by measuring the length (
L) and width (
W) with a vernier caliper and applying the formula
V = (
L × W
2) × 0.5.
Statistical analysis
Differences in CD73 and miR-30a-5p expression between NSCLC tissues (T) and adjacent noncancerous lung tissues (N) were analyzed using a paired t-test (two-tailed). For cell lines, differences between two groups were assessed using an unpaired t-test (two-tailed). The clinicopathologic characteristics and expression levels of mRNA and miRNA in the NSCLC samples were compared using nonparametric tests (Mann–Whitney U-test for comparison between two groups, and the Kruskall-Wallis test for comparison between three or more groups). Two-way ANOVA was used to determine the difference in cell growth between two groups. Differences were considered to be significant at P < 0.05. All statistical analyses were performed using GraphPad Prism 5.02 (GraphPad, San Diego, CA, USA) and the SPSS 16.0 software (SPSS, Chicago, IL, USA).
Acknowledgements
We thank all patients who participated in this study for their cooperation. This work was supported by grants from the National Natural Science Foundation of China (No. 81201575 to Z-Y. Liu, No. 31270940 to J-A. Huang.), the Jiangsu Province Colleges and Universities Natural Science Research Foundation (No. 12KJB310016 to Z-Y. Liu, No. 14KJB0017 to Z. Lei), the Science and Technology Plan Projects of Suzhou (No. SYS201612), the Clinical Medical Center of Suzhou (No. Szzx201502), and the Clinical Key Specialty Project of China.