Background
Gastric cancer is the most common epithelial malignancy and the second leading cause of cancer-related death worldwide [
1]. Patients with advanced gastric cancer usually have a poor prognosis despite the combined therapy including gastriectomy, chemotherapy and radiotherapy. Therefore, better understanding the pathogenesis of gastric cancer and exploring novel therapeutic targets are urgent.
Hypoxia is a common feature of various cancers. It may cause the cells to acquire more aggressive phenotypes, such as cell migration, invasion, growth and metastasis, by changing genetic programs which can facilitate cellular adaptation to hypoxic stress [
2]. Hypoxia-inducible factor 1 (HIF-1), an important transcriptional regulator, is an essential mediator that plays crucial roles in the cell response to hypoxia by modulating hypoxic gene expression [
3,
4]. HIF-1 consists a hypoxia-sensitive subunit HIF-1α and a constitutively expressed subunit HIF-1β. Under normoxic condition, HIF-1α is degraded via the recruitment of an ubiquitin-protein ligase. Whereas in hypoxic condition, HIF-1α is activated via the decreased hydroxylation by PHD [
5]. Accumulating evidence shows that microRNAs (miRNAs) play important roles in the articulated molecular mechanism triggered by hypoxia [
6].
MiRNAs are a conserved family of small non-coding RNA molecules that act as important regulators of gene expression at the post-transcriptional level [
7]. They can bind to the 3′untranslated region (3′UTR) of target genes, resulting in the target mRNA degradation or translational repression [
8,
9]. MiRNAs are reported to be involved in diverse biological processes, such as cell proliferation, apoptosis and death [
10]. Previous studies demonstrate that miRNAs are important regulators of cell response to hypoxia. For instance, miR-210 is induced by HIF-1α under hypoxia and acts as an independent prognostic biomarker in breast cancer [
11]. HIF-1α-inducible miR-382 promotes angiogenesis and acts as an oncogene by directly targeting PTEN in gastric cancer under hypoxia [
12]. A previous study shows that HIF-1α and hypoxia can upregulate miR-224 in melanoma cell lines [
13] and primary human trophoblasts [
14]. MiR-224 is reported to contribute to cell invasion and metastasis in human breast cancer cells [
15] and human hepatocellular carcinoma [
16]. However, the potential roles and molecular mechanism of miR-224 remain poorly understood in human gastric cancer exposed to hypoxia.
In this study, we found that miR-224 was induced by hypoxia and HIF-1α at the transcriptional level. MiR-224 inhibition suppressed the cell growth, migration and invasion induced by hypoxia, while miR-224 overexpression had opposite effects. RASSF8 was validated to be a direct target of miR-224 and mediated the effects of miR-224. RASSF8 knockdown enhanced NF-κB transcriptional activity and p65 translocation, while the inhibition of NF-κB ameliorated the roles of RASSF8 knockdown in cell proliferation and invasion. MiR-224 and RASSF8 were inversely expressed in gastric cancer tissues.
Methods
Clinical samples, cell culture and transfections
Gastric cancer tissue samples were acquired from 29 patients with informed consent in the department of Gastroenterology of the First Hospital of Jilin University. All tissues were frozen in liquid nitrogen and stored at −80 °C. The clinicopathological information for all patients were shown in Table
1.
Table 1
The relationship between miR-224 level and the clinicopathological features of 29 gastric cancer patients
Age (year) |
< 60 | 15 | 0.37 ± 0.08 | 0.230 |
>= 60 | 14 | 0.32 ± 0.11 |
Gender |
Male | 14 | 0.36 ± 0.10 | 0.403 |
Female | 15 | 0.34 ± 0.12 |
Tumor size (cm) |
< 5 cm | 13 | 0.29 ± 0.10 | 0.015 |
>= 5 cm | 16 | 0.38 ± 0.07 |
Depth |
T1 | 12 | 0.25 ± 0.07 | 0.001 |
T2–T4 | 17 | 0.40 ± 0.07 |
Lymph node metastasis |
Absent | 13 | 0.25 ± 0.06 | 0.001 |
Present | 16 | 0.42 ± 0.05 |
TNM stage |
I, II | 11 | 0.28 ± 0.05 | 0.005 |
III, IV | 18 | 0.38 ± 0.11 |
Human gastric cancer cell lines SGC-7901 and MGC-803 were cultured in RPMI 1640 medium (Invitrogen, USA) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin (100 U/ml and 100 μg/ml, respectively). The cells were maintained in a humidified atmosphere with 5% CO2 and 20% O2 at 37 °C, which was referred as the normoxic condition. For hypoxia, the cells were maintained with 5% CO2, 1% O2 and 94% N2 in a hypoxic chamber (Invivo200, UK). The cell transfections were performed using Lipofectamine™ 2000 reagent (Invitrogen, USA) according to the manufacturer’s instructions.
RNA isolation and real-time PCR
RNAs were extracted using Trizol Reagent (Qiagen, USA) from the transfected cells exposed to hypoxia or normoxic condition according to the manufacturer’s protocols. cDNA was synthesized from 500 ng of RNAs using M-MLV reverse transcriptase (Promega) and specific miR-224 reverse transcription primer. Real-time PCR was performed using SYBR Premix EX Taq (TaKaRa) and specific miR-224 primers on a 7900HT Fast Real-Time System (Applied Biosystems) according to the manufacturer’s instructions. The PCR was carried out according to following procedures: 95 °C 5 min, followed by 40 cycles of 95 °C 1 min, 56 °C 30 s and 72 °C 30 s. U6 snRNA was used as an internal control to normalize miR-224 expression. The primers used in reverse transcription and real-time PCR were listed as follows: miR-224 reverse transcription primer: 5′ CTTGCATCACCAGAGAACGAACGGAACC 3′; U6 reverse transcription primer: 5′ AAAATATGGAACGCTTCACGAATTTG 3′. MiR-224 forward primer: 5′ GCGAGGTCAAGTCACTAGTGGT 3′; miR-224 reverse primer: 5′ CGAGAAGCTTGCATCACCAGAGAACG 3′; U6 snRNA forward primer: 5′ CTCGCTTCGGCAGCACATATACT 3′; U6 snRNA reverse primer: 5′ ACGCTTCACGAATTTGCGTGTC 3′; RASSF8 sense primer: 5′AAGTATGGGTGGATGGAGTTCAG 3′; RASSF8 antisense primer: 5′ ATGAGGTGCTAAGTGTCTTTCAG 3′; β-actin sense primer: 5′ TGGCACCCAGCACAATGAA 3′; β-actin antisense primer: 5′ TAAGTCATAGTCCGCCTAGAAGCA 3′.
Western blot assay
The cells treated under hypoxia or normoxia were harvested and lysed using RIPA lysis buffer (50 mM Tris-HCl, pH 8.8, 150 mM NaCl, 1% NP-40, 1% sodium deoxycholate, 0.1% SDS). Protein concentration was measured using a BCA protein assay kit. Fifty microgram of protein samples were resolved on 10% SDS-PAGE gels and then transferred to the PVDF membrane. The membrane was incubated with 5% milk in TBST buffer, followed by incubation with the primary antibodies. Mouse polyclonal to RASSF8 antibody and rabbit polyclonal to NF-κB p65 antibody were used as the primary antibodies purchased from Abcam Company (1:1000 dilution). HRP-conjugated goat anti-rabbit or mouse antibody was used as the secondary antibody (1:5000). The bands were visualized using enhanced chemiluminescence (ECL) detection kit according to the manufacturer’s instructions. GAPDH was employed as a loading control.
Northern blotting assay
MiRNAs were isolated from treated cells using mirVana miRNA Isolation Kit (Ambion) according to the manufacturer’s instructions. RNA concentration was measured and then subjected to Northern blotting assay according to the procedures as previously described [
17]. U6 snRNA was used as an internal control.
MTT (3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2-H-tetrazolium bromide) assay
Cell viability was measured using MTT assay according to the manufacturer’s protocols. Briefly, the transfected cells under normoxic or hypoxic conditions were incubated with MTT once the cells were adhesive to the plates. After incubation for 24, 48 and 72 h, the medium was replaced and the cells were treated with DMSO for 10 min, followed by the measurement with a spectrophotometer on 570 nm (A570 nm).
The transfected cells were seeded in 12-well plates under normoxic or hypoxic conditions. The medium was refreshed every 3 days for approximately 10 days when most of the colony contained more than 50 cells. The colony was fixed, stained with 1% crystal violet and finally counted.
In vitro scratch assay (cell migration assay)
When the transfected cells reached approximately 80% confluence in 48-well plates, a 200 μl pathogen-free tip was used to scratch the cells. The cells were then washed with PBS and cultured in RPMI 1640 medium with 1% FBS. The scratch was taken pictures at different time points under the microscope. The cell migration rate = (width at 0 h–width at different time points)/width at 0 h.
Cell invasion assay
The transfected cells (2.5 × 104 cells) were seeded in serum-free medium in modified Boyden chamber coated with Matrigel. The lower chamber contained RPMI 1640 with 10% FBS. After the cells invaded for 20 h, non-invading cells were scraped with cotton tips. The invading cells on the underside of the chambers were fixed, stained and counted under a microscope.
Luciferase reporter assay
The 3′UTR of RASSF8 mRNA containing miR-224 binding sites was PCR-amplified and inserted downstream of a luciferase reporter gene in the pmirGLO vector. In addition, a mutant construct containing mutations within the binding sites was generated using the QuikChange® site-directed mutagenesis kit (Stratagene, USA) according to the manufacturer’s instructions. The cells were co-transfected with miR-224 mimics and wild-type or mutant luciferase reporter constructs, or transfected with wild-type or mutant luciferase reporter constructs under hypoxic condition. At 24 h after transfection, luciferase intensity was determined using the Dual-Luciferase Reporter Assay System (Promega) according to the manufacturer’s instructions. Renilla luciferase intensity was normalized to firefly luciferase intensity.
As for the binding of HIF-1α to miR-224 promoter, miR-224 promoter was predicted using Promoter 2.0 prediction server, cloned and inserted upstream of a luciferase open reading frame in the pGL3 promoter vector. A mutation within the binding sites between HIF-1α and miR-224 was generated using a QuikChange® site-directed mutagenesis kit (Stratagene, USA) according to the manufacturer’s recommendations. The luciferase activity was determined using the Dual-Luciferase Reporter Assay System (Promega) according to the manufacturer’s protocol. The pTK-luc (Renilla) vector was co-transfected with above constructs and served as a spiked-in control.
Chromatin immunoprecipitation (ChIP) assay
ChIP assay was performed to determine the interaction between HIF-1α and miR-224 promoter using the ChIP assay kit (Merck Millipore) according to the manufacturer’s instructions. Gastric cancer cells were incubated under hypoxic or normoxic condition before harvested. Anti-HIF-1α antibody (Abcam, USA) was used to precipitate the DNA fragment. PCR was performed to analyze the binding of HIF-1α to the promoter of miR-224. The PCR primers were ccatcacttccctcagtggt and cccttgacttttccccactt. PCR products were analyzed by gel electrophoresis on a 1.5% agarose gel.
In vivo animal study
The animal experiments were approved by the Institutional Animal Care and Use Committee of First Hospital of Jilin University and performed in accord with the institutional and NIH guidelines. Briefly, 106 gastric cancer cells transfected with miR-224 antagomir or control were injected subcutaneously into the flanks of 6 young athymic nude mice (6–8 weeks old). Tumor volumes were measured every 5 days at 5 days after injection until day 25 when the mice were sacrificed. The volume was calculated by measuring the length (L) and width (W) of the xenografts. Xenograft volume = (L2 × W)/2.
Detection of NF-κB transcription activity and p65 translocation
The cells were treated with a NF-κB luciferase reporter plasmid (Sigma, USA). The pTK-luc (Renilla) vector was transfected as a spiked-in control. The luciferase intensity was determined using the Dual-Luciferase Reporter Assay System (Promega) according to the manufacturer’s protocol.
The gastric cancer cells were subjected to subcellular fractionation using the NE-PERTM Cytoplasmic and Nuclear Extraction reagents (Thermo) according to the manufacturer’s instructions. The efficiency of fractionation was examined by Western blotting assay using antibodies against LaminB1 (the nuclear control) and GAPDH (the cytosolic control).
Statistical analysis
The data were expressed as mean ± standard deviation (SD) and obtained from three independent experiments. The difference between two groups was analyzed using two-tailed Students’ t-test. The relationship between miR-224 and RASSF8 was analyzed using Pearson correlation analysis. The analysis was performed using Graphpad Prism 5 project. The value of P < 0.05 was considered statistically significant.
Discussion
In this study, we found that miR-224 and HIF-α were upregulated under hypoxia. HIF-1α upregulated miR-224 expression at the transcriptional level. MiR-224 was involved in hypoxia-induced cell growth, migration and invasion. We validated that RASSF8 was a direct target of miR-224 and mediated the effects of miR-224 under hypoxia. In addition, RASSF8 knockdown contributed to NF-κB transcriptional activity and subcellular redistribution. MiR-224 and RASSF8 were inversely expressed in human gastric cancer tissues and related to the aggressiveness of gastric cancer.
Accumulating evidence demonstrates that miR-224 plays important roles in the pathogenesis of diverse cancers. MiR-224 is upregulated and acts as an oncogene in non-small cell lung cancer [
20,
21]. MiR-224 is also upregulated in esophageal squamous cell carcinoma and promotes cell proliferation, migration and invasion, and suppresses cell apoptosis, functioning as an oncogenic miRNA [
22]. It has been reported that miR-224 enhances cell proliferation and suppresses cell apoptosis in meningioma cells by targeting ERG2 [
23]. Previous studies have shown that miR-224 is induced under hypoxic condition in melanoma [
13] and primary human trophoblasts [
14]. Hypoxia has been validated to modulate miRNAs’ expression and plays important roles in miRNA functions [
24]. In line with the findings in the above studies, we found that miR-224 expression was increased by hypoxia. In vitro and in vivo studies demonstrated that miR-224 promoted cell growth, while miR-224 inhibition suppressed cell growth induced by hypoxia. MiR-224 also promoted cell migration and invasion, and inhibition of miR-224 resulted in a decrease in cell migration and invasion induced by hypoxia.
HIF-1α, a crucial transcription factor, is known to regulate several hypoxia-related miRNAs [
25]. Huang et. al has shown that HIF-1α regulates miR-210 expression through binding to the hypoxia-responsive element (HRE) in the region of miR-210 promoter in tumor initiation [
26,
27]. HIF-1α can also regulate miR-155 expression via the HRE in miR-155 promoter during the prolonged hypoxia [
28]. In agree with the mechanism of the regulation of miR-210 and miR-155 by HIF-1α, our study validated that HIF-1α upregulated miR-224 expression by binding to the HRE in the region of miR-224 promoter. We found that hypoxia increased the promoter activity of miR-224 and silencing of HIF-1α reduced the miR-224 promoter activity by luciferase assay. Moreover, when the mutations within the HRE were generated, the effects of hypoxia or HIF-1α on miR-224 promoter activity disappeared. Furthermore, ChIP assay showed that DNA fragment precipitated with anti-HIF-1α antibody generated a specific PCR product flanking the HRE.
RASSF8 is one of the members of RAS association domain family and is ubiquitously expressed in all major organs and tissues. The RASS family contains the classical RASSF proteins (RASSF1-6) and the four recently added N-terminal proteins (FASSF7-10) that are linked to biological processes, including cell proliferation, death, and response to hypoxia [
29]. In our study, we validated that RASSF8 was a target of miR-224. RASSF8 was downregulated under hypoxia, opposite to that of miR-224. MiR-224 suppressed RASSF8 protein levels and the luciferase intensity controlled by RASSF8 3′UTR, while mutations within the binding sites between miR-224 and RASSF8 3′UTR abrogated the inhibitory roles of miR-224 in RASSF8 3′UTR. MiR-224 and RASSF8 expression levels were inversely expressed in gastric cancer. Several studies indicate that RASSF8 functions as a tumor suppressor in diverse cancers. In lung cancer, RASSF8 knockdown contributes to cell migration and invasion, acting as a tumor suppressor [
19,
30]. RASSF8 downregulation enhances lymphangiogenesis and metastasis in esophageal squamous cell carcinoma (ESCC) and inversely correlates with patients’ survival. RASSF8 knockdown also enhances the expression of nuclear NF-κB p65, and NF-κB transcriptional activity in ESCC [
31]. In cutaneous melanoma, RASSF8 knockdown promotes the cell growth, migration and invasion by increasing the expression of NF-κB p65 [
32]. In agreement with the findings in the previous studies, our study found that RASSF8 overexpression suppressed the cell growth and invasion, while RASSF8 knockdown enhanced the cell growth and invasion that were inhibited by miR-224 inhibitor under hypoxia. In addition, we also found that RASSF8 overexpression inhibited the NF-κB transcriptional activity and nuclear p65 expression, while RASSF8 downregulation contributed to NF-κB transcriptional activity and p65 expression in nucleus. Moreover, inhibition of NF-κB by a specific inhibitor PDTC suppressed the cell growth and invasion that were induced by RASSF8 knockdown. A previous study also shows that hypoxia can activate NF-κB activation that is required for organism survival under hypoxia [
33]. In hypoxic hepatocarcinoma cells, hypoxia can activate NF-κB, and NF-κB p65 and p50 could bind to HIF-1α promoter to increase its transcription [
34]. In prostate cancer, inhibition of estrogen receptor β or hypoxia can stabilize HIF-1α which can contribute to the transcription of IKKβ, resulting in the activation of NF-κB [
35]. In in endometrial carcinoma cells, hypoxia can activate NF-κB pathway, resulting in the transactivation of HIF-1α gene, while HIF-1α can enhance NF-κB transcriptional activity [
36]. The above positive feedback loop between NF-κB and HIF-1α facilitates the tumor adaptation to microenvironmental hypoxia in cancer cells. Therefore, we propose that NF-κB may enhance HIF-1α expression in gastric cancer cells under hypoxia, which needs further studies.
Acknowledgements
This study was supported by the Health Research Program of Jilin Province (No.3D5125543428). All authors declared no conflict of interest.