Background
Lung cancer is one of the most common malignant types of tumor. Annually, over one million deaths were associated with lung cancer worldwide [
7]. Since early stages of lung cancer does not cause any signs and symptoms, the diagnosis is usually made at advanced stages of the disease, thus, the overall 5 year survival rate is pretty low with only around 20% [
3,
30]. NSCLC is a subtype of lung cancer with a poor prognosis [
12]. Approximately, 85% of lung cancer cases are NSCLC [
2,
27]. Due to the lack of sensitive early diagnosis makers and molecular targets, treatment of NSCLC patients is still not efficient. Exploring the underlying pathogenesis molecular mechanism and identifying novel diagnostic biomarkers are urgent for NSCLC treatment.
CircRNA is a group of circular RNAs derived from pre-mRNAs [
6]. In structure, they are characterized by single strand, covalently-closed loop RNAs with no Poly A tail. Due to the circular structure they are highly stable in vivo. The generation of circRNAs can be achieved through ‘exon skipping’ or ‘direct back-splicing’ [
16,
17,
32]. The primary role of circRNAs discovered so far is acting as miRNAs sponge. For example, a well-studied circular RNA CDR1as was found to have 73 seed binding sites for miR-7 [
14]. Knockdown of CDR1as in mouse decreased miR-7 and upregulated miR-7 targeted genes expression [
31]. However, despite few well studied circular RNAs, the functions of most circular RNAs remain largely unknown.
Recently more and more studies suggested that circRNAs might play important roles in tumor development [
18]. Studies on NSCLC have shown that some dysregulated circRNAs were tightly associated with NSCLC occurrence and metastasis. For example, circular RNA ATXN7 has been reported to promote NSCLC cell lines growth, metastasis, and high circATXN7 expression in NSCLC patients were associated with poorer survival rates compared with lower circATXN7 [
15]. Circular RNA microarray on 52 NSCLC patients treated with gefitinib found that 1377 circRNAs were differentially expressed in gefitinib effective and ineffective groups, further study showed that elevated hsa_circ_0109320 was associated with longer progression-free survival in gefitinib-treated NSCLC patients, indicating that hsa_circ_0109320 might be a potential biomarker for prognosis in gefitinib treated NSCLC patients [
24].
Previous genome wide ribosomal RNA‐depleted RNA sequencing with paired lung adenocarcinoma and non-tumor tissues found that over 50 circular RNAs were differentially expressed [
11]. Among them, cir_ZNF124 was found to be highly expressed in lung adenocarcinoma. However, the expression and role of circ_ZNF124 in NSCLC are still unclear. In this study we confirmed that circ_ZNF124 was highly expressed in NSCLC cells compared with normal cell, and high circ_ZNF124 expression promoted NSCLC cells proliferation. Further mechanism studies showed that the role of circ_ZNF124 in facilitating NSCLC progression was through inhibiting miR-337-3p and activating JAK2/STAT3 signaling pathway.
Materials and methods
Cell lines
Lung cancer cell lines H1975, H1299, HCC827, A549 and normal immortalized epithelial cell type BEAS-2B were used for in vitro study. H1975, H1299, HCC827 were maintained in Roswell Park Memorial Institute (RPMI-1640) medium (Gibco) added with 10% fetal bovine serum (FBS), A549 (F-12) was maintained in F-12K medium (ATCC) containing 10% FBS, BEAS-2B was cultured in Bronchial Epithelial Cell Growth Medium (Lonza). All the cells were maintained at 37 °C with 5% CO2.
Reporter assay
The wild-type circ_ZNF124 or 3′–untrans-lated region (3′–UTR) of JAK2 containing the miR-337-3p binding site was cloned into the pGL3–control vector. circ_ZNF124 and JAK2-3′UTR mutants with miR-337-3p binding sites mutation were generated by using Q5 Site-Directed Mutagenesis kit (NEB, USA). All plasmids were confirmed by sequencing. Lipofectamine 2000 (Invitrogen; Thermo Fisher Scientific, Inc.) was used to transfect plasmids into cells. In brief, 0.5 million cells were seeded into 6 well plates 1 day before transfection. 50 ng luciferase vector, 5 ng Renilla vector and different amounts of miR-337-3p mimic or circ_ZNF124 overexpression vector as indicated were transfected into cells with lipofectamine 2000 according to the manufacturer’s instructions. After 36 h, cells were harvested and washed with PBS (Phosphate-buffered saline) once. Luciferase activity was determined by using a Dual-Luciferase reporter assay kit (Promega Corporation, USA). The values of Renilla were used to normalize luciferase activity.
Oligonucleotide transfection
miR-337-3p mimic and negative control were purchased from ThermoFisher, siRNA targeting circ_ZNF114 was synthesized by Ribobio (Guangzhou, China). Cell transfection was performed with Lipofectamine RNAiMax (Life Technologies) according to manufacturer’s instructions.
Western blot
Cells were harvested and washed once with PBS. Protein was extracted by using RIPA (Radioimmunoprecipitation assay buffer) lysis buffer (10 mM Tris, pH 7.4. 100 mM NaCl. 1 mM EDTA. 1 mM EGTA. 1% Triton X-100. 10% glycerol. 0.1% SDS. 0.5% deoxycholate.) containing protein inhibitors followed by ultracentrifuge at 13,000 rpm for 5 min. Supernatants were collected, and protein was quantitated by BSA assay. Proteins with different molecular weight were then separated by SDS-PAGE (sodium dodecyl sulphate-polyacrylamide gel electrophoresis) electrophoresis and transferred to NC (Nitroellulose) membrane. Membrane was blotted with 5% non-fat milk. Primary antibodies targeting proteins of interests were incubated with membrane with shaking at 4 °C overnight. The next day, primary antibodies were collected, and membrane was continue incubated with HRP tagged secondary antibodies at room temperature (RT) for 1 h. Membrane was imaged with Odyssey CLx.
RNA immunoprecipitation (RIP)
RNA immunoprecipitation used to verify circ_ZNF124 and miR-337-3p interaction in vivo was performed by using Magna RIP RNA-Binding Protein Immunoprecipitation Kit (Millipore, Bedford, MA). In brief, 20 million A549 cells were harvested and lysed with RIP lysis buffer containing protease and RNase inhibitors. Cell lysates were then split, and RNA was pulled down by incubating with anti-Argonaute 2 (AGO2) antibody (Millipore) or control rabbit IgG (Millipore), followed by rotating at 4 °C overnight. After proteinase K treatment, the immunoprecipitated RNAs were extracted. The abundance of circ_ZNF124 and miR-337-3p were determined by qRT-PCR.
Cell growth and cell cycle
After transfection for 24 h, cells were harvested and seeded into 10 cm dish at 0.1 million cells/ml. At days 1, 2, 3 and 5 cells were collected and counted with Hemacytometers. At least 3 replicates were performed. CellTiter-Glo (CTG) luminescent cell viability assay used to measure cell viability was performed at day 5 after cell outgrowth. In brief, cells were trypsin digested and washed once with PBS. Cells was re-suspended with 500 µl PBS. 30 µl cell suspension was aspirated and transferred into 96 well white plate, CTG reagent was added to cells followed by incubation at RT for 10 min, cell viability was recorded with Luminometer.
Cell cycle was measured with propidium iodide (PI) staining. In brief, harvested cells were washed with PBS and re-suspended with 100 µl cold PBS. Cells were fixed with 900 µl cold ethanol by adding dropwise with vortex. Cell fixation was performed at 4 °C for at least 2 h. Fixed cells were stained with 500 µl PI staining solution (50 µg/ml; 1 mg/ml of RNase A, 0.1% Triton X-100 in PBS) at RT for 30 min. FACS was then applied to detect cell cycle.
A549 and H1299 cells were transfected with Circ_ZNF124 siRNA or control siRNA. After 24 h, cells were collected and counted. Cells were seeded into 6-well plates and continue culture in complete medium at 37 °C with 5% CO2 for 2 weeks until the cell colonies were clearly observed. Colonies were first fixed with methanol for 10 min followed by staining with crystal violet (0.1% concentration; comWin Biotech, Beijing) for 5 min at RT. Stained colonies were recorded with scanner and colony numbers were counted for statistical analysis.
Cell migration
After transfection for 24 h, cells were harvested and seeded into 6 well plates. The next day, sterile 200 µl sharp tips were used to draw lines on the cell. Cell migration was determined 24 h after cell scratch. At least 10 views per well were recorded for statistical analysis.
RNA extraction and qRT-PCR
Total RNA was extracted by using TRIzol (Thermo Fisher, USA). 1 µg RNA was used to reverse transcript to cDNA by using PrimeScript RT Master Mix (Takara, Japan). The expression of circ_ZNF124, JAK2, HIF1a, BCL2, c-FOS and miR-337-3p were determined by real-time PCR analyses by using SYBR master mix (Bio-rad). U6 small nuclear RNA levels served as an internal control for miRNA expression detection. GAPDH (Glyceraldehyde 3-phosphate dehydrogenase) was used as an internal control for gene expression. The relative RNA expression level was calculated with 2
−∆∆ct. Primers used were listed in Table
1.
Table 1
The sequences of specific primers
GAPDH-F | GTAGAGCGGCCGCCATG |
GAPDH-R | GATTTCCATTGATGACAAGC |
qRT-GAPDH-F | ACTTCAACAGCGACACCCACTC |
qRT-GAPDH-R | TCTCTTCCTCTTGTGCTCTTGCT |
qRT-HIF1A-F | GAACGTCGAAAAGAAAAGTCTCG |
qRT-HIF1A-R | CCTTATCAAGATGCGAACTCACA |
qRT-cFOS-F | CCGGGGATAGCCTCTCTTACT |
qRT-cFOS-R | CCAGGTCCGTGCAGAAGTC |
qRT-BCL2-F | GGTGGGGTCATGTGTGTGG |
qRT-BCL2-R | CGGTTCAGGTACTCAGTCATCC |
qRT-circZNF_124-F | GACCAGAGCATTGAAGA |
qRT-circZNF_124-R | AGGATCCAACAAAGCC |
qRT-JAK2-F | ATCCACCCAACCATGTCTTCC |
qRT-JAK2-R | ATTCCATGCCGATAGGCTCTG |
JAK2sg1-F | CACCGAGAAAACGATCAAACCCCAC |
JAK2sg1-R | AAACGTGGGGTTTGATCGTTTTCTC |
JAK2-3′UTR-F | AACAGATCTGTTTTCTAAT TTTTCC |
JAK2-3′UTR-R | GTTAGATCTCAACAACG AACAACCCCT |
circ_ZNF124-pGL3-F | AGACGTGATGCAGG |
circ_ZNF124-pGL3-R | GGATCCAACAAAGCC |
qRT-U6-F | TTGGTCTGATCTGGCACATATAC |
qRT-U6-R | AAAAATATGGAGCGCTTCACG |
qRT-miR-377-3p-F | CGCAGCTTCTTTCCGTAGT |
qRT-miR-377-3p-R | GGTCCAGTTTTTTTTTTTTTTTGAG |
JAK2-3′UTR-mirmutant-F | GTTTTATTTGGTCGACAAATTTCCCTGACCCTAAATAATAC |
JAK2-3′UTR-mirmutant-R | TAGACATTTGTTCCCTTTATG |
circ_ZNF124-mirmutant-F | TCTGGCTTCCGACAAAAACAAAGGGGAAGAC |
circ_ZNF124-mirmutant-R | TTCCTGAAGGTTTCCTGC |
Biotin pull down
Wild type circ_ZNF124 probe and miR-337-3p binding site mutated circ_ZNF124 probe were labeled with biotin and incubate with Streptavidin magnetic beads to generate probe-coated beads. 10 million A549 cells were lysed and incubated with probe-coated beads with shaking at 4 °C overnight. RNA bound to the beads was eluted. The enrichment of miR-337-3p was evaluated by qRT-PCR.
CRISPR (clustered regularly interspaced short palindromic repeats)
Lentiviruses expressing Cas9 were made by transfecting 293T cells with VSVG, PsPAX2 and lentiCas9-Blast. Lentiviruses were harvested at day 1 and day 2 after transfection, harvested viruses were used to infect A549 and H1975. Cells stably expressing Cas9 were selected by blasticidin for 5 days. The expression of Cas9 was verified by western blot. To delete JAK2, sgRNA targeting JAK2 was cloned into pLentiGuide-puro vector and packaged into lentiviruses with VSVG, PsPAX2 in 293T cells. Harvested lentiviruses were then used to infect A549 and H1945 cells in which Cas9 were stably expressed. After transfection, puromycin was used to select lentiviruses infected cells. The efficiency of JAK2 deletion was tested by western blot after puromycin selection for 3 days.
Statistical analysis
Statistical analysis was performed using SPSS 20.0 software (Chicago, IL, USA). Student’s t test was applied to compare the differences between experimental group and control group. All data were shown as mean ± SEM. *P < 0.05 was considered statistically significant.
Discussion
Despite great advances in the treatment of NSCLC through the use of radiotherapy [
4], surgery [
26], chemotherapy, and other interventions, the 5 year survival rate of NSCLC patients is still low [
36]. The lack of efficient NSCLC biomarkers limited the development of targeted chemical therapy. Nowadays, adjuvant chemotherapy remains the standard of care for patients with resected NSCLC. Among them, cisplatin-based postoperative chemotherapy cisplatin is a widely used chemotherapeutic drug in the clinical treatment of NSCLC [
21,
33]. However, due to the genetic background diversity in NSCLC patients, endogenous and acquired drug resistance limits its clinical efficacy [
23]. Therefore, identify early NSCLC diagnostic biomarkers and novel molecular targets is of clinical important for NSCLC treatment.
With the advancement of RNA-seq technologies and the development of bioinformatic analysis, more and more circRNAs have been identified [
8,
13,
40]. CircRNA is a type of non-polyadenylated circular RNA which is produced during transcription [
6]. Recently, more and more research has focused on the functional studies of circular RNA. For example, Ashwal-Fluss showed that some circRNA co-transcription with mRNA from the same locus could function as an RNA trap through competing with their corresponding mRNA expression [
1]. Hansen proved that circRNA ciRS-7 can inhibit miRNA miR-7 activity by serving as miR-7 sponge [
14]. Furthermore, in contrast to our previous concept, some researchers have demonstrated that like mRNA, some circRNA can also encode proteins, and these circRNA encoded proteins are also functional in cells [
19,
29,
38]. Despite molecular mechanism studies, the roles of circRNA in cancers have also been extensively investigated. By using circRNA microarray, Li identified that hsa_circ_0004277 was downregulated in acute myeloid leukemia, and chemotherapy treatment could significantly restore hsa_circ_0004277 expression, indicating that hsa_circ_0004277 could be a new biomarker for Acute Myeloid Leukemia [
22]. In the research of bladder carcinoma, Zhong discovered that circTCF25 was highly expressed in bladder carcinoma, further mechanism studies demonstrated that circTCF25 could promote bladder carcinoma cell line proliferation and migration through suppressing miR-103a-3p and miR-107 expression, and increasing CDK6 expression [
41].
Previous studies showed that circZNF_124 was upregulated in lung adenocarcinoma, however, the function of circZNF_124 in NSCLC is still unknown. In this study, we provided insights into the clinical significance, function and molecular mechanism of circZNF_124 in NSCLC. The potential association of circZNF_124 with lung cancer cells proliferation was first investigated. The findings indicated that circZNF_124 expression was highly upregulated in NSCLC cells compared with normal epithelial cells. these results indicated that NSCLC could be a potential novel biomarker for NSCLC diagnosis and prognosis.
To interrogate the function of circZNF_124 on NSCLC, siRNA was used to silence circZNF_124 expression. The results showed that abolishing circZNF_124 expression promoted cell cycle arrested in sub-G1 phase and significantly decreased NSCLC cell line A549 and H1975 growth rate, suggesting the oncogenic role of circZNF_124 in NSCLC. Recent studies on circRNA indicated that many circRNA may function as a sponge of miRNA. For example Chi showed that Circular RNA circPIP5K1A promotes NSCLC proliferation and metastasis through targeting miR-600 [
9]. Wan found that by acting as a ceRNA, circ_0020123 released miR-488-3p mediated ADAM9 downregulation, which further promoted NSCLC progression [
37]. To understand the oncogenic mechanism of circZNF_124 in NSCLC, we used circinteractome to predict potential circZNF_124 targets. Results showed that miR-337-3p was scored the highest among all the potential miRNA targets. RNA immunoprecipitation confirmed their directly interaction in vivo. Luciferase assay showed that miR-337-3p could greatly affect circZNF_124 activity through binding to circZNF_124. These results implicated that miR-337-3p is the target of circZNF_124 in NSCLC.
We next asked whether miR-337-3p is also associated with NSCLC development. The expression and the effect of miR-337-3p on NSCLC were investigated. Contrary to circZNF_124, miR-337-3p expression was downregulated in NSCLC cells compared with normal cell. However, whether circZNF_124 can directly regulate miR-337-3p expression needs further investigation. The roles of miR-337-3p in NSCLC were also investigated. As indicated, miR-337-3p mimic significantly suppressed NSCLC cell lines A549 and H1975 growth rate and colony formation ability. While in the presence of circZNF_124 the function of miR-337-3p in NSCLC is heavily impaired.
Dysregulation of JAK2/STAT3 signaling pathway is associated with many cancer progression and metastasis [
5,
20,
39]. miRDB and TargetScan prediction indicated that JAK2 is the target of miR-337-3p. Indeed, when miR-337-3p mimic was transfected into A549 and H1975, JAK2 protein level was greatly reduced, luciferase assay confirmed the regulatory role of miR-337-3p on JAK2 is through directly binding to JAK2 3′UTR, indicating JAK2 is the target of miR-337-3p. Previous studies showed that STAT3 is also a target of miR-337-3p, in consistent with this study, we found that overexpression of miR-337-3p downregulated both JAK2 and STAT3, implicating multi-targets of miR-337-3p. BCL2, c-FOS and HIF1a are transcription factors, whose dysregulation were broadly reported to be associated with tumor inflammation, angiogenesis, and suppression of apoptosis among others [
10,
25,
28,
34,
35]. As the downstream genes of JAK2/STAT3 signaling pathway, we found that impaired JAK2/STAT3 signaling pathway mediated by miR-337-3p also downregulated BCL2, c-FOS and HIF1a expression, while circZNF_124 rescued JAK2 activity and these genes expression caused by miR-337-3p. These results strongly support the hypothesis that circZNF_124 facilitates JAK2/STAT3 signaling pathway activation by acting as a competing endogenous RNA (ceRNA) of miR-337-3p.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.