- Email: [email protected]
Negative regulation of lncRNA GAS5 by miR-196a inhibits esophageal squamous cell carcinoma growth
Biochemical and Biophysical Research Communications 495 (2018) 1151e1157
Contents lists available at ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
Negative regulation of lncRNA GAS5 by miR-196a inhibits esophageal squamous cell carcinoma growth Kai Wang a, Juan Li a, Gang Xiong b, Gang He a, Xingying Guan a, Kang Yang b, **, Yun Bai a, * a b
Department of Medical Genetics, Department of Basic Medicine, Third Military Medical University, Chongqing 400038, PR China Department of Thoracic and Cardiac Surgery, Southwest Hospital, Third Military Medical University, Chongqing 400038, PR China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 16 November 2017 Accepted 19 November 2017 Available online 21 November 2017
MiR-196a could play important roles in carcinogenesis by targeting many protein coding genes. However, little is known about whether miR-196a can target any long non-coding RNAs (lncRNAs). In the present study, we screen lncRNAs which are regulated by miRNA-196a in human esophageal squamous cell carcinoma (ESCC). We found that miR-196a could suppress the expression of lncRNA growth arrestspecific 5(GAS5). GAS5 is frequently down-regulated in 86 paired human ESCC tissues. Importantly, there was lower GAS5 expression in the late stage of ESCC patients. The reduced expression of GAS5 in ESCC may not be related to DNA methylation but related to the high expression of miR-196a. In vitro and in vivo studies indicated that GAS5 could inhibit the growth of ESCC cells. Using Chromatin Isolation by RNA PurificationeqPCR, we found that miR-196a could bind to GAS5. The Luciferase Reporter Assay indicated that miR-196a could bind to the seventh exon of GAS5. Additionally, both GAS5 and miR-196a could bind to Ago2 which is a key component of the RNA-induced silencing complex (RISC). Together, these results suggest that GAS5 functions as a tumor suppressor gene in ESCC and is regulated by miR196a involved in RISC. © 2017 Elsevier Inc. All rights reserved.
Keywords: ncRNA miR-196a GAS5 Esophageal squamous cell carcinoma
1. Introduction MicroRNAs are small non-protein-coding RNAs. Increasing evidence has indicated that miRNAs play important roles in carcinogenesis through post-transcriptional gene silencing. Many miRNAs are dysregulated in cancers and function as tumor suppressors or oncogenes. Among them, miR-196a has been found to function as an oncogene or tumor suppressor gene in many types of cancers. For example, miR-196a is upregulated in many cancers including gastrointestinal stromal tumor [1], pancreatic cancer [2] and head and neck squamous cell carcinoma [3]. In our previous miRNA microarray study (GSE61047), miR-196a was one of the upregulated miRNAs in ESCC tissues (fold change ¼ 3.5, p < 0.01). As a cancer related gene, miR-196a regulates many genes involved in cell proliferation, cell apoptosis, migration and invasion [2,4,5]. MiR-196a functions as a cancer related gene by inhibiting target genes. Some studies have found that HOXC8 and HOXB7 are targets
* Corresponding author. ** Corresponding author. E-mail addresses: [email protected] (K. Yang), [email protected] (Y. Bai). https://doi.org/10.1016/j.bbrc.2017.11.119 0006-291X/© 2017 Elsevier Inc. All rights reserved.
of miR-196a [6,7]. Many studies indicated that the HOX genes were abnormally expressed in a number of cancers [8]. Our previous study suggested that miR-196a could participate in ESCC migration and invasion by targeting RAP1A [9]. Currently, there are at least 10 experimentally validated targets of miR-196a and the number of miR-196a targets continues to grow. However, the targets of miR196a are all protein coding genes. It has been estimated that only 2% of the genome is transcribed into protein-coding RNAs. The majority of the transcribed RNAs are non protein coding RNAs. LncRNAs are non protein coding RNAs larger than 200bp. Therefore, more and more studies suggested that miRNAs could regulate the expression of lncRNAs [10,11]. However, whether there are any lncRNAs regulated by miR-196a is unknown. Currently, several protein coding genes have been identified as miR-196a targets. While, many lncRNAs have been characterized and shown to be associated with cancer development. In the present study, we asked whether miR-196a can regulate cancer related lncRNAs. Our results indicated that lncRNA GAS5 is a novel target of miR196a. Moreover, GAS5 is frequently down-regulated in ESCC tissues and is associated with clinical stages. The reduced expression of GAS5 in ESCC may not be related to DNA methylation of the GAS5 promoter but instead to the high expression of miR-196a. GAS5 can
1152
K. Wang et al. / Biochemical and Biophysical Research Communications 495 (2018) 1151e1157
suppress the tumor growth of ESCC in vitro and in vivo. Finally, miR196a negatively regulates the expression of GAS5 possibly through the RNA-induced silencing complex (RISC).
2. Materials and methods 2.1. Tumor tissues and cell lines The ESCC patients were histopathologically diagnosed and conformed at the Southwest Hospital, the Third Military Medical University. Informed consent was obtained from the subjects, and the study was performed with the approval of the ethical committee of the Third Military Medical University. Tumors and adjacent non-cancerous tissues were collected from patients who underwent surgery at the Thoracic and Cardiac Surgery of Southwest Hospital. Three esophageal squamous cancer cell lines (EC109, KYSE150 and KYSE450) were purchased from the Cell Bank of the Chinese Academy of Science. A human esophageal epithelial cell (HET-1A) was purchased from ATCC. All the cells were genotyped for identity by STR method at Key Laboratory of Birth Defects and Reproductive Health (Fig. S1).
2.2. Plasmid construction The entire GAS5 sequence was amplified with cDNA and then cloned into the expression vector pcDNA3.1. Two step PCR was performed to clone of GAS5 with mutations at the putative miR196a binding site. We used two sets of primers for the wild-type GAS5 gene as a template to generate two overlapped PCR products. Then we mixed the two PCR products as a template to generate a mutated product and then cloned it into the pMIRREPORT Firefly Luciferase reporter vector. All constructed vectors were verified by DNA sequencing. All the primer sequences for vector construction are presented in Table S1.
2.5. LncRNA quantitative reverse transcription-polymerase chain reaction (qRT-PCR) Total RNA was isolated using the RNAiso Plus Kit. Then, 100 ng of RNA from each sample was reverse-transcribed into cDNA and subjected to conventional polymerase chain reaction (PCR). Primer sequences for PCR were presented in supplementary table S1. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used for normalization. The analysis was performed in triplicate. 2.6. RNA immunoprecipitaion (RIP) RIP was performed by using the Magna RIP RNA-Binding Protein Immunoprecipitation Kit and the Ago2 antibody according to the manufacturer's protocol. After the antibody was recovered by protein beads, qRT-PCR was performed to detect GAS5 and miR196a in the precipitates. 2.7. Cell viability assay The EC109 cells were planted in 96 well plates one day prior to transfection. When the cells reached 50% to 70% confluency, plasmids or siRNA were transfected. After incubation for 24 h, 48 h, 72 h, 96 h and 130 h, 100 ml fresh medium and 10 ul Cell Counting Kit-8 (Dojindo, Japan) were incubated together for 2 h. Absorbance readings were detected at 450 nm. The cell viability assay was performed for three times. 2.8. Cell proliferation assay Cell-Light Edu Apollo DNA in vitro Kit was used to detect the ability of cell proliferation. The detailed procedure has been described previously [12]. 2.9. In vivo tumorigenicity
2.3. Luciferase reporter assay For luciferase activity analysis, EC109 cells were co-transfected with luciferase reporter constructs, beta-gal control plasmid and miRNA mimics with Lipofectamine 2000 according to the manufacturer's instructions (Invitrogen, NY, USA). The miRNA mimics that were transfected into the cells were purchased from GenePharma. Twenty-four hours later, luciferase activity was measured using the Dual-Luciferase Reporter Assay System (Promega, USA). The experiments were performed in triplicate.
2.4. ChIRP-qPCR Four 15 cm dishes of cells were used per ChIRP-qPCR experiment. Cell harvesting, lysis, disruption, and ChIRP were essentially performed according to the manufacturer's instructions (Millipore, USA). Briefly, cells were cross-linked in 3% formaldehyde for 10 min, followed by 10 glycine quenching for 5 min. And then the cross-linked DNA was sheared to approximately 100 to 500 base pairs in length using sonication. For a typical ChIRP reaction, we used 100 pmol of total probe (even and odd probes) (Ribobio, Guangzhou, China) per 1 mL sonicated cell lysate is used. Proteinase K and QIAGEN miRNeasy® Mini Kit were used to isolate the RNA. The retrieved RNA was reversely transcribed to cDNA by TaqMan™ Advanced miRNA cDNA Synthesis Kit (ABI, USA). Expression of miRNAs was quantified by PCR using TaqMan® MicroRNA Assay Kit (ABI, USA). The expression of miRNAs was normalized against U6.
1 107 EC109 cells were injected subcutaneously into the 4 weeks old female nude BALB/c mice. Tumor size was measured every few days after one week of the injection. The tumor volume was calculated as length width2/2. All mice were killed on day 25, and the average tumor size and weight were estimated. 2.10. DNA extraction and methylation analysis Genomic DNAs from frozen tissues and cell lines were extracted using DNA Extraction Kit (Promega). DNAs were treated with sodium bisulfite using the EZ DNA Methylation kit (Zymo Research, CA, USA) and subjected to PCR using primer sets designed to amplify the CpG island of GAS5. Methylation-specific PCR (MSP) primers and bisulfite sequencing (BSP) primers were designed using Methyl Primer Express Software v1.0. For MSP, PCR products were run on a 2.5% agarose gel and visualized with ChemiDoc Touch Imaging System (Bio-Rad, USA). For bisulfite sequencing, PCR products were subcloned in T vectors (Promega) and then sequenced in Sangon Biotech (Shanghai, China). All primers used in the study are provided in Supplementary Table 1. 2.11. Statistical analysis Statistical analysis was performed by SPSS software (SPSS Inc., USA). Student's t-test was used for continuous variables, and the x2 test was used for categorical variables. A p value of less than 0.05 was considered as statistically significant.
K. Wang et al. / Biochemical and Biophysical Research Communications 495 (2018) 1151e1157
3. Results 3.1. Identification of lncRNAs regulated by miR-196a using RT-qPCR The strategy of choosing lncRNAs is shown in Fig. 1A. First, we used bioinformative tool (http://starbase.sysu.edu.cn/index.php) to choose miR-196a-targeted lncRNAs. Ultimately, we chose 5 cancerrelated lncRNAs (HCG18, GAS5, H19, Hoxa-as2, XIST) as the candidate targets of miR-196a. To determine whether miR-196a can regulate these lncRNAs, reverse transcription-polymerase chain reaction (RT-qPCR) was conducted to detect lncRNAs expression after transfecting miR-196a mimics or inhibitors into esophageal cancer cells. First, we increased the expression of miR-196a on EC109 and Kyse450 cells with miRNA mimics or decreased its expression with inhibitor (Fig. S2). We found that miR-196a mimics could reduce the expression of GAS5 significantly in both EC109 and Kyse450 cells (Fig. 1B). However, there was no obvious change in other lncRNAs expression after miR-196a mimics or inhibitors transfection (Fig. S3). These results indicated that GAS5 may be regulated by miR-196a. 3.2. GAS5 is frequently down-regulated in human esophageal cancer tissues and esophageal cancer cell lines We examined the expression of GAS5 and miR-196a in esophageal cancer cell lines firstly. As compared with the normal esophageal epithelial cell Het-1A, GAS5 was downregulated whereas miR-196a was upregulated in esophageal cancer cell lines (Fig. 2A). The expression level of GAS5 and miR-196a was tested in 86 pairs of esophageal cancer tissues and matched with the adjacent normal tissues. Detailed clinical characteristics are presented in Table S2. As illustrated in Fig. 2B, the expression of GAS5 in about 62.8% (54 out of 86) tumor tissues was expressed at lower levels as compared with adjacent normal tissues. Additionally, GAS5 in cancer tissues is downregulated comparing with normal tissues (Fig. 2C). Furthermore, we analyzed the GAS5 expression in patients with different clinical features and found that GAS5 expression was lower in stage Ⅲ and Ⅵ ESCC tissues as compared with stageⅠandⅡtissues (Fig. 2D). And the expression of miR-196a in about 83.7% (72 out to 86) tumor tissues was overexpressed in compared with adjacent normal tissues(Fig.S4A). And miR-196a in cancer tissues is upregulated as compared with normal tissues (Fig.S4B). MiR-196a expression was lower in stageⅠandⅡ ESCC tissues as compared withⅢ and Ⅵ tissues (Fig.S4C). In addition, the Pearson
1153
correlation analysis showed that GAS5 expression was inversely correlated with miR-196a (R2 ¼ 0.073,P ¼ 0.011,Fig. 2E). These results suggest that GAS5 might be dysregulated in ESCC, which maybe the result of miR-196a. 3.3. GAS5 inhibits esophageal cancer cell growth in vitro and in vivo Due to the abnormal expression of GAS5 in ESCC tissues, we speculated that it might have some potential roles in ESCC development. First, we successfully increased the expression of GAS5 with overexpression vector and decreased its expression using Gas5 siRNAs (Fig. S5). In order to determine the role of GAS5 on cell growth in vitro, CCK8 cell viability assay was performed. We found that GAS5 could decrease the cell populations of EC109 and Kyse450 cells. And GAS5 siRNAs could increase the cell population of EC109 and Kyse450 cells (Fig. 3A). The EDU assay revealed that GAS5 could inhibit cell proliferation of EC109 and Kyse450 cells, whereas inhibition of GAS5 expression could increase the cell growth of EC109 and Kyse450 cells (Fig. 3B and C). To further investigate whether GAS5 was involved in tumorigenesis, we examined whether downregulation of GAS5 in ESCC cells could promote tumorigenicity in vivo. EC109 cells transfected with GAS5 siRNA or negative control were injected subcutaneously into the nude mice. After 25 days, the 4 mice from each group were sacrificed and the tumors were weighed. As observed in Fig. 3D, the growth curve and the average weight of the tumors from the mice injected with GAS5 siRNA group were significantly larger than those of control group(p ¼ 0.015 and p ¼ 0.013, respectively). These results were consistent with the effects of GAS5 in vitro and strongly suggest that GAS5 could inhibit the growth of ESCC cells. 3.4. Posttranscriptional regulation of GAS5 by miR-196a with the RISC complex We designed a set of probes to pull down GAS5 and its binding RNAs. Both the even and odd probes of GAS5 could retrieve the lncRNA GAS5 (Fig. S6). Using ChIRP-RT-qPCR, we found that miR196a could bind to GAS5 (Fig. 4A). To further determine whether miR-196a could bind to the 7th exon of GAS5, we constructed 2 reporter vectors containing either GAS5 exon 7 or mutant exon 7, named Gas5-WT and Gas5-MU respectively (Fig. 4B). We then transfected the 2 vectors into EC109 cell along with either miR196a mimics or its inhibitor. As shown in Fig. 4(C) miR-196a mimics decreased the luciferase activity significantly as compared
Fig. 1. (A) Flowchart of screening for candidate lncRNAs. (B) EC109 or Kyse450 cells transfected with miR-196a mimics, inhibitor or NC. Forty-eight hours after transfection, cells were harvested for measurement of mRNA expression of lncRNAs using RT-qPCR. GAPDH served as a control. * indicates p < 0.05.
1154
K. Wang et al. / Biochemical and Biophysical Research Communications 495 (2018) 1151e1157
Fig. 2. Expression of GAS5 in cancer tissues and adjacent non-tumor tissues of 86 ESCC patients. (A) Expression of GAS5 on three ESCC cell lines (EC109, KYSE150 and KYSE450) and an epithelial cell line (HET-1A). (B) The relative expression level of GAS5 was measured by RT-qPCR. LogC2 (cancer)/N (non-cancer) relative expression ratios between normalized values were calculated. (C) Expression of GAS5 was determined by RT-qPCR. Data were analyzed using a 2-△CT method. n ¼ 86, Ca: cancer tissues. N: non-tumor tissues. (D) The expression of GAS5 related to GAPDH in different clinical stages of ESCC patient tissues. GAPDH served as a control. (E) Scatter plots show the inverse association between miR-196a and GAS5 expression.
to NC. Furthermore, miR-196a inhibitor increased the luciferase activity significantly. MiRNAs regulate gene expression by binding to the 3 prime untranslated region (UTR) of its target genes with RISC complex. We reasoned that miRNAs regulate the gene expression always through RISC complex. Therefore, we performed RNA immunoprecipitation using an antibody against to AGO2 which is a key component of the RISC complex. As expected, we detected GAS5 in the AGO2 pellet and found that GAS5 had over a 20-fold enrichment (AGO2 antibody versus IgG, Fig. 4D). Then we used miRNA Inhibitors to block miR-196a, and we performed the RIP experiment to AGO2 which showed a significantly decreased enrichment for GAS5 (Fig. 4E). Thus, these results indicate that miR196a could regulate the expression of GAS5 through the RISC complex. 3.5. The decrease of GAS5 in ESCC is not related to DNA methylation Hypermethylation of the CpG islands in the gene regulatory region is a mechanism that can account for the down-regulation of lncRNA in many tumors. GAS5 is located in Chromosome 1 and the promoter region of GAS5 contains a CpG island of approximately 745bp. Therefore, we explored whether hypermethylation of this CpG island might cause reduced GAS5 expression in ESCC tissues. In order to examine the methylation status, paired methylation primers, unmethylation primers and BSP primers were designed for
GAS5 (Fig. S7A). Firstly, the methylation status of GAS5 in ESCC cells and normal esophageal epithelial cell line Het-1A was detected by BSP. Every CpG site detected was unmethylated in the Het-1A cells and in the esophageal cancer cell lines (Fig. S7B). Additionally, the 12 paired ESCC tissues were used for MSP examination. This region showed an unmethylation pattern in both ESCC tissues and the adjacent normal tissues(Fig. S7C). In conclusion, GAS5 in ESCC were unmethylated and the reduced expression of GAS5 in ESCC may be not related to DNA methylation. 4. Discussion There are many reports about miR-196a upregulation in cancers such as gastrointestinal stromal tumor [1], pancreatic cancer [13] and head and neck squamous cell carcinoma [3]. In our previous study, we found that a single nucleotide polymorphism (SNP) in pre-miR-196a was associated with ESCC [14]. However, some studies have found that miR-196a was downregulated in cancers such as malignant melanoma [6]. MiRNAs are involved in carcinogenesis by binding to the 30 UTR of the target genes. MiR-196a promotes pancreatic cancer progression by targeting NFKBIA [2]. And miR-196a could increase the cell proliferation and migration of cervical cancer cells by targeting netrin4 [15]. However, miR-196a could suppress cell proliferation, apoptosis and migration of renal cell carcinoma [16]. Which
K. Wang et al. / Biochemical and Biophysical Research Communications 495 (2018) 1151e1157
1155
Fig. 3. GAS5 inhibits ESCC cell growth in vitro and in vivo. (A) EC109 and Kyse450 cells transfected with GAS5 overexpression vector, GAS5 siRNA and their negative control respectively were analyzed by CCK-8 assay. n ¼ 3, error bars indicate standard deviation (SD). (B) Edu assay was used to determine cell proliferation of EC109 cells after transfected with GAS5 overexpression vector or GAS5 siRNA. (C) Edu assay was used to determine cell proliferation of Kyse450 cells after transfected with GAS5 overexpression vector or GAS5 siRNA. (D) GAS5 inhibits ESCC cell growth in vivo. The growth curve and average weight of the tumors derived from GAS5 siRNA1 group and the control group in nude mice. Data presented as mean±SD,n ¼ 4 per group.
suggests that miR-196a may exert different functions in different cancers. This might be due to the states of different tumor cells, or the cell-specific function of miR-196a by targeting of different genes. However, these reported targets of miR-196a are all protein coding genes. In the present study, we found that miR-196a could regulate the expression of GAS5 which is a long non-coding RNA. Thus, the identification of GAS5 as a miR-196a target expands the repertoire of miR-196a targets. GAS5 was first identified from a subtraction cDNA library enriched in growth arrested mouse NIH 3T3 fibroblasts [17]. Many studies have shown that GAS5 mainly functions as a tumor suppressor involved in cell proliferation, cell apoptosis, cell migration and invasion [18]. Decreased expression of GAS5 indicates a poor prognosis and promotes cell proliferation in gastric cancer [19]. Yang et al. showed that GAS5 is downregulated in oral squamous cell carcinoma and overexpression of GAS5 inhibited cell proliferation, migration and the invasion potential of cancer cells [20]. However, there are no studies about this lncRNA in ESCC. In the resent study, we found that GAS5 was downregulated in ESCC tissues compared with the adjacent normal tissues. And the level of
GAS5 was associated with the clinical stages of ESCC. In addition, GAS5 could suppress the cell growth of ESCC in vitro and in vivo. These results suggest that GAS5 could function as a tumor suppressor gene in ESCC by decreasing cell growth. Some evidence has revealed that the interaction between miRNAs and lncRNAs participate in the cancer development. LncRNAs could act as competing endogenous RNA (ceRNA) influencing the miRNAs pathway [21]. These lncRNAs have miRNA responsive elements (MRE) localized within them and compete with miRNAs for binding to mRNAs. GAS5 could inhibit liver fibrogenesis through a mechanism as a ceRNA for miR-222 [22]. However, GAS5 could be regulated by miRNA and miR-21 may inhibit GAS5 expression [23]. In this study, miR-196a regulates GAS5 through RISC. The reduction of several lncRNAs is the result of hypermethylation of the CpG islands in the gene regulatory region. GAS5 is downregulated in gastric cancer cells by promoter hypermethylation [24]. However, in this study, we found that GAS5 were unmethylated in both esophageal squamous cancer cell lines and primary ESCC tissues. Thus, GAS5 reduction in ESCC species is likely not related to promoter hypermethylation but rather to the high
1156
K. Wang et al. / Biochemical and Biophysical Research Communications 495 (2018) 1151e1157
Fig. 4. Posttranscriptional regulation of GAS5 by miR-196a with the RISC complex. (A) ChIRP was performed using EC109 cell lysate and either GAS5 lncRNA Probe Set even odd or Negative Control Probe Set (LacZ). Purified RNA was then analyzed by RT-qPCR using miRNA Primers and U6 (Negative Target). (B) Schematic of the human GAS5 containing the wild-type or mutant miR-196a binding sequence. (C) MiR-196a mimics, inhibitor or its negative control (NC) was cotransfected with the pMIR-REPORT constructs containing the wild-type or the mutant allele into EC109 cell lines. Data shown are the mean fold increase ±SD from two independent experiments. * means p < 0.05. (D) RIP lysate prepared from EC109 cells were subjected to immunoprecipitation using 5 ul of either a normal mouse IgG, or 5 ul of anti-Ago2 antibody. Immunoprecipitation of Ago2 associated RNAs was verified by RT-qPCR. FOS and GAPDH served as the positive control and negative control respectively. (E) Transfected with miR-196a inhibitor or inhibitor NC into EC109 cell for 24 h. Then RIP lysate prepared from the transfected EC109 cells were subjected to immunoprecipitation using 5 ul of either a normal mouse IgG, or 5 ul of anti-Ago2 antibody. Immunoprecipitation of Ago2 associated RNAs was verified by RT-qPCR. FOS and GAPDH served as the positive control and negative control respectively. The symbol ** means p < 0.01.
expression of miR-196a in ESCC. Our study presents the first data regarding the DNA methylation status in ESCC tumor tissues. DNA methylation might regulate tissue specific expression of GAS5. In conclusion, our findings demonstrate that lncRNA-Gas5 acting as a tumor suppressor in ESCC by inhibiting cell growth. Our study is the first to show that miR-196a regulates the expression of GAS5 by binding to its exon through RISC. Furthermore, our results show that the GAS5 reduction in ESCC is likely not due to the promoter hypermethylation but instead to the high level of miR196a in ESCC. Thus, our findings may provide insight for new research avenues in ESCC diagnostics and treatment. Funding This work was supported by the National Natural Science Foundation of China [31100936], Natural Science Foundation Project of CQ CSTC [cstc2016jcyjA0482] and Young Start-up Talent Fund of Third Military Medical University (2016).
Conflicts of interest The authors have declared that no conflict of interest exists. Appendix A. Supplementary data Supplementary data related to this article can be found at https://doi.org/10.1016/j.bbrc.2017.11.119. References [1] T. Niinuma, H. Suzuki, M. Nojima, K. Nosho, H. Yamamoto, H. Takamaru, E. Yamamoto, R. Maruyama, T. Nobuoka, Y. Miyazaki, T. Nishida, T. Bamba, T. Kanda, Y. Ajioka, T. Taguchi, S. Okahara, H. Takahashi, Y. Nishida, M. Hosokawa, T. Hasegawa, T. Tokino, K. Hirata, K. Imai, M. Toyota, Y. Shinomura, Upregulation of miR-196a and HOTAIR drive malignant character in gastrointestinal stromal tumors, Cancer Res. 72 (2012) 1126e1136. [2] F. Huang, J. Tang, X. Zhuang, Y. Zhuang, W. Cheng, W. Chen, H. Yao, S. Zhang, MiR-196a promotes pancreatic cancer progression by targeting nuclear factor kappa-B-inhibitor alpha, PLoS One 9 (2014) e87897.
K. Wang et al. / Biochemical and Biophysical Research Communications 495 (2018) 1151e1157 [3] L. Darda, F. Hakami, R. Morgan, C. Murdoch, D.W. Lambert, K.D. Hunter, The role of HOXB9 and miR-196a in head and neck squamous cell carcinoma, PLoS One 10 (2015) e0122285. [4] X. Fang, J.H. Jeong, X. Long, S.J. Park, D. Wang, M. Shuai, R. Wei, C. Li, S. Li, S. Zhang, M.B. Duran, K.W. Lo, S.W. Tsao, R. Glaser, Z. Luo, X. Feng, Y. Tian, J.L. Luo, IKKalpha-mediated biogenesis of miR-196a through interaction with Drosha regulates the sensitivity of cancer cells to radiotherapy, Cell Death Differ. 23 (2016) 1471e1482. [5] A.L. Pin, F. Houle, P. Fournier, M. Guillonneau, E.R. Paquet, M.J. Simard, I. Royal, J. Huot, Annexin-1-mediated endothelial cell migration and angiogenesis are regulated by vascular endothelial growth factor (VEGF)-induced inhibition of miR-196a expression, J. Biol. Chem. 287 (2012) 30541e30551. [6] S. Braig, D.W. Mueller, T. Rothhammer, A.K. Bosserhoff, MicroRNA miR-196a is a central regulator of HOX-B7 and BMP4 expression in malignant melanoma, Cell Mol. Life Sci. 67 (2010) 3535e3548. [7] D.W. Mueller, A.K. Bosserhoff, MicroRNA miR-196a controls melanomaassociated genes by regulating HOX-C8 expression, Int. J. Cancer 129 (2011) 1064e1074. [8] V. Raman, S.A. Martensen, D. Reisman, E. Evron, W.F. Odenwald, E. Jaffee, J. Marks, S. Sukumar, Compromised HOXA5 function can limit p53 expression in human breast tumours, Nature 405 (2000) 974e978. [9] K. Wang, J. Li, H. Guo, X. Xu, G. Xiong, X. Guan, B. Liu, X. Chen, K. Yang, Y. Bai, MiR-196a binding-site SNP regulates RAP1A expression contributing to esophageal squamous cell carcinoma risk and metastasis, Carcinogenesis 33 (2012) 2147e2154. [10] T. Chiyomaru, S. Fukuhara, S. Saini, S. Majid, G. Deng, V. Shahryari, I. Chang, Y. Tanaka, H. Enokida, M. Nakagawa, R. Dahiya, S. Yamamura, Long noncoding RNA HOTAIR is targeted and regulated by miR-141 in human cancer cells, J. Biol. Chem. 289 (2014) 12550e12565. [11] C. Braconi, T. Kogure, N. Valeri, N. Huang, G. Nuovo, S. Costinean, M. Negrini, E. Miotto, C.M. Croce, T. Patel, microRNA-29 can regulate expression of the long non-coding RNA gene MEG3 in hepatocellular cancer, Oncogene 30 (2011) 4750e4756. [12] H. Meng, K. Wang, X. Chen, X. Guan, L. Hu, G. Xiong, J. Li, Y. Bai, MicroRNA330-3p functions as an oncogene in human esophageal cancer by targeting programmed cell death 4, Am. J. Cancer Res. 5 (2015) 1062e1075.
1157
[13] M. Liu, Y. Du, J. Gao, J. Liu, X. Kong, Y. Gong, Z. Li, H. Wu, H. Chen, Aberrant expression miR-196a is associated with abnormal apoptosis, invasion, and proliferation of pancreatic cancer cells, Pancreas 42 (2013) 1169e1181. [14] K. Wang, H. Guo, H. Hu, G. Xiong, X. Guan, J. Li, X. Xu, K. Yang, Y. Bai, A functional variation in pre-microRNA-196a is associated with susceptibility of esophageal squamous cell carcinoma risk in Chinese Han, Biomarkers 15 (2010) 614e618. [15] J. Zhang, F. Zheng, G. Yu, Y. Yin, Q. Lu, miR-196a targets netrin 4 and regulates cell proliferation and migration of cervical cancer cells, Biochem. Biophys. Res. Commun. 440 (2013) 582e588. [16] Y. Li, L. Jin, D. Chen, J. Liu, Z. Su, S. Yang, Y. Gui, X. Mao, G. Nie, Y. Lai, Tumor suppressive miR-196a is associated with cellular migration, proliferation and apoptosis in renal cell carcinoma, Mol. Med. Rep. 14 (2016) 560e566. [17] C. Schneider, R.M. King, L. Philipson, Genes specifically expressed at growth arrest of mammalian cells, Cell 54 (1988) 787e793. [18] M.R. Pickard, M. Mourtada-Maarabouni, G.T. Williams, Long non-coding RNA GAS5 regulates apoptosis in prostate cancer cell lines, Biochim. Biophys. Acta 1832 (2013) 1613e1623. [19] M. Sun, F.Y. Jin, R. Xia, R. Kong, J.H. Li, T.P. Xu, Y.W. Liu, E.B. Zhang, X.H. Liu, W. De, Decreased expression of long noncoding RNA GAS5 indicates a poor prognosis and promotes cell proliferation in gastric cancer, BMC Cancer 14 (2014) 319. [20] Meng Yang, Longcong Chen, Li Yang, Xian Li, Identification and validation long non-coding RNAs of oral squamous cell carcinoma by bioinformatics method, Oncotarget (2017), https://doi.org/10.18632/oncotarget.18178. [21] L. Salmena, L. Poliseno, Y. Tay, L. Kats, P.P. Pandolfi, A ceRNA hypothesis: the Rosetta Stone of a hidden RNA language? Cell 146 (2011) 353e358. [22] X. Zhao, P. Wang, J. Liu, J. Zheng, Y. Liu, J. Chen, Y. Xue, Gas5 exerts tumorsuppressive functions in human glioma cells by targeting miR-222, Mol. Ther. 23 (2015) 1899e1911. [23] Z. Zhang, Z. Zhu, K. Watabe, X. Zhang, C. Bai, M. Xu, F. Wu, Y.Y. Mo, Negative regulation of lncRNA GAS5 by miR-21, Cell Death Differ. 20 (2013) 1558e1568. [24] N. Zhang, A.Y. Wang, X.K. Wang, X.M. Sun, H.Z. Xue, GAS5 is downregulated in gastric cancer cells by promoter hypermethylation and regulates adriamycin sensitivity, Eur. Rev. Med. Pharmacol. Sci. 20 (2016) 3199e3205.
Contents lists available at ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
Negative regulation of lncRNA GAS5 by miR-196a inhibits esophageal squamous cell carcinoma growth Kai Wang a, Juan Li a, Gang Xiong b, Gang He a, Xingying Guan a, Kang Yang b, **, Yun Bai a, * a b
Department of Medical Genetics, Department of Basic Medicine, Third Military Medical University, Chongqing 400038, PR China Department of Thoracic and Cardiac Surgery, Southwest Hospital, Third Military Medical University, Chongqing 400038, PR China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 16 November 2017 Accepted 19 November 2017 Available online 21 November 2017
MiR-196a could play important roles in carcinogenesis by targeting many protein coding genes. However, little is known about whether miR-196a can target any long non-coding RNAs (lncRNAs). In the present study, we screen lncRNAs which are regulated by miRNA-196a in human esophageal squamous cell carcinoma (ESCC). We found that miR-196a could suppress the expression of lncRNA growth arrestspecific 5(GAS5). GAS5 is frequently down-regulated in 86 paired human ESCC tissues. Importantly, there was lower GAS5 expression in the late stage of ESCC patients. The reduced expression of GAS5 in ESCC may not be related to DNA methylation but related to the high expression of miR-196a. In vitro and in vivo studies indicated that GAS5 could inhibit the growth of ESCC cells. Using Chromatin Isolation by RNA PurificationeqPCR, we found that miR-196a could bind to GAS5. The Luciferase Reporter Assay indicated that miR-196a could bind to the seventh exon of GAS5. Additionally, both GAS5 and miR-196a could bind to Ago2 which is a key component of the RNA-induced silencing complex (RISC). Together, these results suggest that GAS5 functions as a tumor suppressor gene in ESCC and is regulated by miR196a involved in RISC. © 2017 Elsevier Inc. All rights reserved.
Keywords: ncRNA miR-196a GAS5 Esophageal squamous cell carcinoma
1. Introduction MicroRNAs are small non-protein-coding RNAs. Increasing evidence has indicated that miRNAs play important roles in carcinogenesis through post-transcriptional gene silencing. Many miRNAs are dysregulated in cancers and function as tumor suppressors or oncogenes. Among them, miR-196a has been found to function as an oncogene or tumor suppressor gene in many types of cancers. For example, miR-196a is upregulated in many cancers including gastrointestinal stromal tumor [1], pancreatic cancer [2] and head and neck squamous cell carcinoma [3]. In our previous miRNA microarray study (GSE61047), miR-196a was one of the upregulated miRNAs in ESCC tissues (fold change ¼ 3.5, p < 0.01). As a cancer related gene, miR-196a regulates many genes involved in cell proliferation, cell apoptosis, migration and invasion [2,4,5]. MiR-196a functions as a cancer related gene by inhibiting target genes. Some studies have found that HOXC8 and HOXB7 are targets
* Corresponding author. ** Corresponding author. E-mail addresses: [email protected] (K. Yang), [email protected] (Y. Bai). https://doi.org/10.1016/j.bbrc.2017.11.119 0006-291X/© 2017 Elsevier Inc. All rights reserved.
of miR-196a [6,7]. Many studies indicated that the HOX genes were abnormally expressed in a number of cancers [8]. Our previous study suggested that miR-196a could participate in ESCC migration and invasion by targeting RAP1A [9]. Currently, there are at least 10 experimentally validated targets of miR-196a and the number of miR-196a targets continues to grow. However, the targets of miR196a are all protein coding genes. It has been estimated that only 2% of the genome is transcribed into protein-coding RNAs. The majority of the transcribed RNAs are non protein coding RNAs. LncRNAs are non protein coding RNAs larger than 200bp. Therefore, more and more studies suggested that miRNAs could regulate the expression of lncRNAs [10,11]. However, whether there are any lncRNAs regulated by miR-196a is unknown. Currently, several protein coding genes have been identified as miR-196a targets. While, many lncRNAs have been characterized and shown to be associated with cancer development. In the present study, we asked whether miR-196a can regulate cancer related lncRNAs. Our results indicated that lncRNA GAS5 is a novel target of miR196a. Moreover, GAS5 is frequently down-regulated in ESCC tissues and is associated with clinical stages. The reduced expression of GAS5 in ESCC may not be related to DNA methylation of the GAS5 promoter but instead to the high expression of miR-196a. GAS5 can
1152
K. Wang et al. / Biochemical and Biophysical Research Communications 495 (2018) 1151e1157
suppress the tumor growth of ESCC in vitro and in vivo. Finally, miR196a negatively regulates the expression of GAS5 possibly through the RNA-induced silencing complex (RISC).
2. Materials and methods 2.1. Tumor tissues and cell lines The ESCC patients were histopathologically diagnosed and conformed at the Southwest Hospital, the Third Military Medical University. Informed consent was obtained from the subjects, and the study was performed with the approval of the ethical committee of the Third Military Medical University. Tumors and adjacent non-cancerous tissues were collected from patients who underwent surgery at the Thoracic and Cardiac Surgery of Southwest Hospital. Three esophageal squamous cancer cell lines (EC109, KYSE150 and KYSE450) were purchased from the Cell Bank of the Chinese Academy of Science. A human esophageal epithelial cell (HET-1A) was purchased from ATCC. All the cells were genotyped for identity by STR method at Key Laboratory of Birth Defects and Reproductive Health (Fig. S1).
2.2. Plasmid construction The entire GAS5 sequence was amplified with cDNA and then cloned into the expression vector pcDNA3.1. Two step PCR was performed to clone of GAS5 with mutations at the putative miR196a binding site. We used two sets of primers for the wild-type GAS5 gene as a template to generate two overlapped PCR products. Then we mixed the two PCR products as a template to generate a mutated product and then cloned it into the pMIRREPORT Firefly Luciferase reporter vector. All constructed vectors were verified by DNA sequencing. All the primer sequences for vector construction are presented in Table S1.
2.5. LncRNA quantitative reverse transcription-polymerase chain reaction (qRT-PCR) Total RNA was isolated using the RNAiso Plus Kit. Then, 100 ng of RNA from each sample was reverse-transcribed into cDNA and subjected to conventional polymerase chain reaction (PCR). Primer sequences for PCR were presented in supplementary table S1. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used for normalization. The analysis was performed in triplicate. 2.6. RNA immunoprecipitaion (RIP) RIP was performed by using the Magna RIP RNA-Binding Protein Immunoprecipitation Kit and the Ago2 antibody according to the manufacturer's protocol. After the antibody was recovered by protein beads, qRT-PCR was performed to detect GAS5 and miR196a in the precipitates. 2.7. Cell viability assay The EC109 cells were planted in 96 well plates one day prior to transfection. When the cells reached 50% to 70% confluency, plasmids or siRNA were transfected. After incubation for 24 h, 48 h, 72 h, 96 h and 130 h, 100 ml fresh medium and 10 ul Cell Counting Kit-8 (Dojindo, Japan) were incubated together for 2 h. Absorbance readings were detected at 450 nm. The cell viability assay was performed for three times. 2.8. Cell proliferation assay Cell-Light Edu Apollo DNA in vitro Kit was used to detect the ability of cell proliferation. The detailed procedure has been described previously [12]. 2.9. In vivo tumorigenicity
2.3. Luciferase reporter assay For luciferase activity analysis, EC109 cells were co-transfected with luciferase reporter constructs, beta-gal control plasmid and miRNA mimics with Lipofectamine 2000 according to the manufacturer's instructions (Invitrogen, NY, USA). The miRNA mimics that were transfected into the cells were purchased from GenePharma. Twenty-four hours later, luciferase activity was measured using the Dual-Luciferase Reporter Assay System (Promega, USA). The experiments were performed in triplicate.
2.4. ChIRP-qPCR Four 15 cm dishes of cells were used per ChIRP-qPCR experiment. Cell harvesting, lysis, disruption, and ChIRP were essentially performed according to the manufacturer's instructions (Millipore, USA). Briefly, cells were cross-linked in 3% formaldehyde for 10 min, followed by 10 glycine quenching for 5 min. And then the cross-linked DNA was sheared to approximately 100 to 500 base pairs in length using sonication. For a typical ChIRP reaction, we used 100 pmol of total probe (even and odd probes) (Ribobio, Guangzhou, China) per 1 mL sonicated cell lysate is used. Proteinase K and QIAGEN miRNeasy® Mini Kit were used to isolate the RNA. The retrieved RNA was reversely transcribed to cDNA by TaqMan™ Advanced miRNA cDNA Synthesis Kit (ABI, USA). Expression of miRNAs was quantified by PCR using TaqMan® MicroRNA Assay Kit (ABI, USA). The expression of miRNAs was normalized against U6.
1 107 EC109 cells were injected subcutaneously into the 4 weeks old female nude BALB/c mice. Tumor size was measured every few days after one week of the injection. The tumor volume was calculated as length width2/2. All mice were killed on day 25, and the average tumor size and weight were estimated. 2.10. DNA extraction and methylation analysis Genomic DNAs from frozen tissues and cell lines were extracted using DNA Extraction Kit (Promega). DNAs were treated with sodium bisulfite using the EZ DNA Methylation kit (Zymo Research, CA, USA) and subjected to PCR using primer sets designed to amplify the CpG island of GAS5. Methylation-specific PCR (MSP) primers and bisulfite sequencing (BSP) primers were designed using Methyl Primer Express Software v1.0. For MSP, PCR products were run on a 2.5% agarose gel and visualized with ChemiDoc Touch Imaging System (Bio-Rad, USA). For bisulfite sequencing, PCR products were subcloned in T vectors (Promega) and then sequenced in Sangon Biotech (Shanghai, China). All primers used in the study are provided in Supplementary Table 1. 2.11. Statistical analysis Statistical analysis was performed by SPSS software (SPSS Inc., USA). Student's t-test was used for continuous variables, and the x2 test was used for categorical variables. A p value of less than 0.05 was considered as statistically significant.
K. Wang et al. / Biochemical and Biophysical Research Communications 495 (2018) 1151e1157
3. Results 3.1. Identification of lncRNAs regulated by miR-196a using RT-qPCR The strategy of choosing lncRNAs is shown in Fig. 1A. First, we used bioinformative tool (http://starbase.sysu.edu.cn/index.php) to choose miR-196a-targeted lncRNAs. Ultimately, we chose 5 cancerrelated lncRNAs (HCG18, GAS5, H19, Hoxa-as2, XIST) as the candidate targets of miR-196a. To determine whether miR-196a can regulate these lncRNAs, reverse transcription-polymerase chain reaction (RT-qPCR) was conducted to detect lncRNAs expression after transfecting miR-196a mimics or inhibitors into esophageal cancer cells. First, we increased the expression of miR-196a on EC109 and Kyse450 cells with miRNA mimics or decreased its expression with inhibitor (Fig. S2). We found that miR-196a mimics could reduce the expression of GAS5 significantly in both EC109 and Kyse450 cells (Fig. 1B). However, there was no obvious change in other lncRNAs expression after miR-196a mimics or inhibitors transfection (Fig. S3). These results indicated that GAS5 may be regulated by miR-196a. 3.2. GAS5 is frequently down-regulated in human esophageal cancer tissues and esophageal cancer cell lines We examined the expression of GAS5 and miR-196a in esophageal cancer cell lines firstly. As compared with the normal esophageal epithelial cell Het-1A, GAS5 was downregulated whereas miR-196a was upregulated in esophageal cancer cell lines (Fig. 2A). The expression level of GAS5 and miR-196a was tested in 86 pairs of esophageal cancer tissues and matched with the adjacent normal tissues. Detailed clinical characteristics are presented in Table S2. As illustrated in Fig. 2B, the expression of GAS5 in about 62.8% (54 out of 86) tumor tissues was expressed at lower levels as compared with adjacent normal tissues. Additionally, GAS5 in cancer tissues is downregulated comparing with normal tissues (Fig. 2C). Furthermore, we analyzed the GAS5 expression in patients with different clinical features and found that GAS5 expression was lower in stage Ⅲ and Ⅵ ESCC tissues as compared with stageⅠandⅡtissues (Fig. 2D). And the expression of miR-196a in about 83.7% (72 out to 86) tumor tissues was overexpressed in compared with adjacent normal tissues(Fig.S4A). And miR-196a in cancer tissues is upregulated as compared with normal tissues (Fig.S4B). MiR-196a expression was lower in stageⅠandⅡ ESCC tissues as compared withⅢ and Ⅵ tissues (Fig.S4C). In addition, the Pearson
1153
correlation analysis showed that GAS5 expression was inversely correlated with miR-196a (R2 ¼ 0.073,P ¼ 0.011,Fig. 2E). These results suggest that GAS5 might be dysregulated in ESCC, which maybe the result of miR-196a. 3.3. GAS5 inhibits esophageal cancer cell growth in vitro and in vivo Due to the abnormal expression of GAS5 in ESCC tissues, we speculated that it might have some potential roles in ESCC development. First, we successfully increased the expression of GAS5 with overexpression vector and decreased its expression using Gas5 siRNAs (Fig. S5). In order to determine the role of GAS5 on cell growth in vitro, CCK8 cell viability assay was performed. We found that GAS5 could decrease the cell populations of EC109 and Kyse450 cells. And GAS5 siRNAs could increase the cell population of EC109 and Kyse450 cells (Fig. 3A). The EDU assay revealed that GAS5 could inhibit cell proliferation of EC109 and Kyse450 cells, whereas inhibition of GAS5 expression could increase the cell growth of EC109 and Kyse450 cells (Fig. 3B and C). To further investigate whether GAS5 was involved in tumorigenesis, we examined whether downregulation of GAS5 in ESCC cells could promote tumorigenicity in vivo. EC109 cells transfected with GAS5 siRNA or negative control were injected subcutaneously into the nude mice. After 25 days, the 4 mice from each group were sacrificed and the tumors were weighed. As observed in Fig. 3D, the growth curve and the average weight of the tumors from the mice injected with GAS5 siRNA group were significantly larger than those of control group(p ¼ 0.015 and p ¼ 0.013, respectively). These results were consistent with the effects of GAS5 in vitro and strongly suggest that GAS5 could inhibit the growth of ESCC cells. 3.4. Posttranscriptional regulation of GAS5 by miR-196a with the RISC complex We designed a set of probes to pull down GAS5 and its binding RNAs. Both the even and odd probes of GAS5 could retrieve the lncRNA GAS5 (Fig. S6). Using ChIRP-RT-qPCR, we found that miR196a could bind to GAS5 (Fig. 4A). To further determine whether miR-196a could bind to the 7th exon of GAS5, we constructed 2 reporter vectors containing either GAS5 exon 7 or mutant exon 7, named Gas5-WT and Gas5-MU respectively (Fig. 4B). We then transfected the 2 vectors into EC109 cell along with either miR196a mimics or its inhibitor. As shown in Fig. 4(C) miR-196a mimics decreased the luciferase activity significantly as compared
Fig. 1. (A) Flowchart of screening for candidate lncRNAs. (B) EC109 or Kyse450 cells transfected with miR-196a mimics, inhibitor or NC. Forty-eight hours after transfection, cells were harvested for measurement of mRNA expression of lncRNAs using RT-qPCR. GAPDH served as a control. * indicates p < 0.05.
1154
K. Wang et al. / Biochemical and Biophysical Research Communications 495 (2018) 1151e1157
Fig. 2. Expression of GAS5 in cancer tissues and adjacent non-tumor tissues of 86 ESCC patients. (A) Expression of GAS5 on three ESCC cell lines (EC109, KYSE150 and KYSE450) and an epithelial cell line (HET-1A). (B) The relative expression level of GAS5 was measured by RT-qPCR. LogC2 (cancer)/N (non-cancer) relative expression ratios between normalized values were calculated. (C) Expression of GAS5 was determined by RT-qPCR. Data were analyzed using a 2-△CT method. n ¼ 86, Ca: cancer tissues. N: non-tumor tissues. (D) The expression of GAS5 related to GAPDH in different clinical stages of ESCC patient tissues. GAPDH served as a control. (E) Scatter plots show the inverse association between miR-196a and GAS5 expression.
to NC. Furthermore, miR-196a inhibitor increased the luciferase activity significantly. MiRNAs regulate gene expression by binding to the 3 prime untranslated region (UTR) of its target genes with RISC complex. We reasoned that miRNAs regulate the gene expression always through RISC complex. Therefore, we performed RNA immunoprecipitation using an antibody against to AGO2 which is a key component of the RISC complex. As expected, we detected GAS5 in the AGO2 pellet and found that GAS5 had over a 20-fold enrichment (AGO2 antibody versus IgG, Fig. 4D). Then we used miRNA Inhibitors to block miR-196a, and we performed the RIP experiment to AGO2 which showed a significantly decreased enrichment for GAS5 (Fig. 4E). Thus, these results indicate that miR196a could regulate the expression of GAS5 through the RISC complex. 3.5. The decrease of GAS5 in ESCC is not related to DNA methylation Hypermethylation of the CpG islands in the gene regulatory region is a mechanism that can account for the down-regulation of lncRNA in many tumors. GAS5 is located in Chromosome 1 and the promoter region of GAS5 contains a CpG island of approximately 745bp. Therefore, we explored whether hypermethylation of this CpG island might cause reduced GAS5 expression in ESCC tissues. In order to examine the methylation status, paired methylation primers, unmethylation primers and BSP primers were designed for
GAS5 (Fig. S7A). Firstly, the methylation status of GAS5 in ESCC cells and normal esophageal epithelial cell line Het-1A was detected by BSP. Every CpG site detected was unmethylated in the Het-1A cells and in the esophageal cancer cell lines (Fig. S7B). Additionally, the 12 paired ESCC tissues were used for MSP examination. This region showed an unmethylation pattern in both ESCC tissues and the adjacent normal tissues(Fig. S7C). In conclusion, GAS5 in ESCC were unmethylated and the reduced expression of GAS5 in ESCC may be not related to DNA methylation. 4. Discussion There are many reports about miR-196a upregulation in cancers such as gastrointestinal stromal tumor [1], pancreatic cancer [13] and head and neck squamous cell carcinoma [3]. In our previous study, we found that a single nucleotide polymorphism (SNP) in pre-miR-196a was associated with ESCC [14]. However, some studies have found that miR-196a was downregulated in cancers such as malignant melanoma [6]. MiRNAs are involved in carcinogenesis by binding to the 30 UTR of the target genes. MiR-196a promotes pancreatic cancer progression by targeting NFKBIA [2]. And miR-196a could increase the cell proliferation and migration of cervical cancer cells by targeting netrin4 [15]. However, miR-196a could suppress cell proliferation, apoptosis and migration of renal cell carcinoma [16]. Which
K. Wang et al. / Biochemical and Biophysical Research Communications 495 (2018) 1151e1157
1155
Fig. 3. GAS5 inhibits ESCC cell growth in vitro and in vivo. (A) EC109 and Kyse450 cells transfected with GAS5 overexpression vector, GAS5 siRNA and their negative control respectively were analyzed by CCK-8 assay. n ¼ 3, error bars indicate standard deviation (SD). (B) Edu assay was used to determine cell proliferation of EC109 cells after transfected with GAS5 overexpression vector or GAS5 siRNA. (C) Edu assay was used to determine cell proliferation of Kyse450 cells after transfected with GAS5 overexpression vector or GAS5 siRNA. (D) GAS5 inhibits ESCC cell growth in vivo. The growth curve and average weight of the tumors derived from GAS5 siRNA1 group and the control group in nude mice. Data presented as mean±SD,n ¼ 4 per group.
suggests that miR-196a may exert different functions in different cancers. This might be due to the states of different tumor cells, or the cell-specific function of miR-196a by targeting of different genes. However, these reported targets of miR-196a are all protein coding genes. In the present study, we found that miR-196a could regulate the expression of GAS5 which is a long non-coding RNA. Thus, the identification of GAS5 as a miR-196a target expands the repertoire of miR-196a targets. GAS5 was first identified from a subtraction cDNA library enriched in growth arrested mouse NIH 3T3 fibroblasts [17]. Many studies have shown that GAS5 mainly functions as a tumor suppressor involved in cell proliferation, cell apoptosis, cell migration and invasion [18]. Decreased expression of GAS5 indicates a poor prognosis and promotes cell proliferation in gastric cancer [19]. Yang et al. showed that GAS5 is downregulated in oral squamous cell carcinoma and overexpression of GAS5 inhibited cell proliferation, migration and the invasion potential of cancer cells [20]. However, there are no studies about this lncRNA in ESCC. In the resent study, we found that GAS5 was downregulated in ESCC tissues compared with the adjacent normal tissues. And the level of
GAS5 was associated with the clinical stages of ESCC. In addition, GAS5 could suppress the cell growth of ESCC in vitro and in vivo. These results suggest that GAS5 could function as a tumor suppressor gene in ESCC by decreasing cell growth. Some evidence has revealed that the interaction between miRNAs and lncRNAs participate in the cancer development. LncRNAs could act as competing endogenous RNA (ceRNA) influencing the miRNAs pathway [21]. These lncRNAs have miRNA responsive elements (MRE) localized within them and compete with miRNAs for binding to mRNAs. GAS5 could inhibit liver fibrogenesis through a mechanism as a ceRNA for miR-222 [22]. However, GAS5 could be regulated by miRNA and miR-21 may inhibit GAS5 expression [23]. In this study, miR-196a regulates GAS5 through RISC. The reduction of several lncRNAs is the result of hypermethylation of the CpG islands in the gene regulatory region. GAS5 is downregulated in gastric cancer cells by promoter hypermethylation [24]. However, in this study, we found that GAS5 were unmethylated in both esophageal squamous cancer cell lines and primary ESCC tissues. Thus, GAS5 reduction in ESCC species is likely not related to promoter hypermethylation but rather to the high
1156
K. Wang et al. / Biochemical and Biophysical Research Communications 495 (2018) 1151e1157
Fig. 4. Posttranscriptional regulation of GAS5 by miR-196a with the RISC complex. (A) ChIRP was performed using EC109 cell lysate and either GAS5 lncRNA Probe Set even odd or Negative Control Probe Set (LacZ). Purified RNA was then analyzed by RT-qPCR using miRNA Primers and U6 (Negative Target). (B) Schematic of the human GAS5 containing the wild-type or mutant miR-196a binding sequence. (C) MiR-196a mimics, inhibitor or its negative control (NC) was cotransfected with the pMIR-REPORT constructs containing the wild-type or the mutant allele into EC109 cell lines. Data shown are the mean fold increase ±SD from two independent experiments. * means p < 0.05. (D) RIP lysate prepared from EC109 cells were subjected to immunoprecipitation using 5 ul of either a normal mouse IgG, or 5 ul of anti-Ago2 antibody. Immunoprecipitation of Ago2 associated RNAs was verified by RT-qPCR. FOS and GAPDH served as the positive control and negative control respectively. (E) Transfected with miR-196a inhibitor or inhibitor NC into EC109 cell for 24 h. Then RIP lysate prepared from the transfected EC109 cells were subjected to immunoprecipitation using 5 ul of either a normal mouse IgG, or 5 ul of anti-Ago2 antibody. Immunoprecipitation of Ago2 associated RNAs was verified by RT-qPCR. FOS and GAPDH served as the positive control and negative control respectively. The symbol ** means p < 0.01.
expression of miR-196a in ESCC. Our study presents the first data regarding the DNA methylation status in ESCC tumor tissues. DNA methylation might regulate tissue specific expression of GAS5. In conclusion, our findings demonstrate that lncRNA-Gas5 acting as a tumor suppressor in ESCC by inhibiting cell growth. Our study is the first to show that miR-196a regulates the expression of GAS5 by binding to its exon through RISC. Furthermore, our results show that the GAS5 reduction in ESCC is likely not due to the promoter hypermethylation but instead to the high level of miR196a in ESCC. Thus, our findings may provide insight for new research avenues in ESCC diagnostics and treatment. Funding This work was supported by the National Natural Science Foundation of China [31100936], Natural Science Foundation Project of CQ CSTC [cstc2016jcyjA0482] and Young Start-up Talent Fund of Third Military Medical University (2016).
Conflicts of interest The authors have declared that no conflict of interest exists. Appendix A. Supplementary data Supplementary data related to this article can be found at https://doi.org/10.1016/j.bbrc.2017.11.119. References [1] T. Niinuma, H. Suzuki, M. Nojima, K. Nosho, H. Yamamoto, H. Takamaru, E. Yamamoto, R. Maruyama, T. Nobuoka, Y. Miyazaki, T. Nishida, T. Bamba, T. Kanda, Y. Ajioka, T. Taguchi, S. Okahara, H. Takahashi, Y. Nishida, M. Hosokawa, T. Hasegawa, T. Tokino, K. Hirata, K. Imai, M. Toyota, Y. Shinomura, Upregulation of miR-196a and HOTAIR drive malignant character in gastrointestinal stromal tumors, Cancer Res. 72 (2012) 1126e1136. [2] F. Huang, J. Tang, X. Zhuang, Y. Zhuang, W. Cheng, W. Chen, H. Yao, S. Zhang, MiR-196a promotes pancreatic cancer progression by targeting nuclear factor kappa-B-inhibitor alpha, PLoS One 9 (2014) e87897.
K. Wang et al. / Biochemical and Biophysical Research Communications 495 (2018) 1151e1157 [3] L. Darda, F. Hakami, R. Morgan, C. Murdoch, D.W. Lambert, K.D. Hunter, The role of HOXB9 and miR-196a in head and neck squamous cell carcinoma, PLoS One 10 (2015) e0122285. [4] X. Fang, J.H. Jeong, X. Long, S.J. Park, D. Wang, M. Shuai, R. Wei, C. Li, S. Li, S. Zhang, M.B. Duran, K.W. Lo, S.W. Tsao, R. Glaser, Z. Luo, X. Feng, Y. Tian, J.L. Luo, IKKalpha-mediated biogenesis of miR-196a through interaction with Drosha regulates the sensitivity of cancer cells to radiotherapy, Cell Death Differ. 23 (2016) 1471e1482. [5] A.L. Pin, F. Houle, P. Fournier, M. Guillonneau, E.R. Paquet, M.J. Simard, I. Royal, J. Huot, Annexin-1-mediated endothelial cell migration and angiogenesis are regulated by vascular endothelial growth factor (VEGF)-induced inhibition of miR-196a expression, J. Biol. Chem. 287 (2012) 30541e30551. [6] S. Braig, D.W. Mueller, T. Rothhammer, A.K. Bosserhoff, MicroRNA miR-196a is a central regulator of HOX-B7 and BMP4 expression in malignant melanoma, Cell Mol. Life Sci. 67 (2010) 3535e3548. [7] D.W. Mueller, A.K. Bosserhoff, MicroRNA miR-196a controls melanomaassociated genes by regulating HOX-C8 expression, Int. J. Cancer 129 (2011) 1064e1074. [8] V. Raman, S.A. Martensen, D. Reisman, E. Evron, W.F. Odenwald, E. Jaffee, J. Marks, S. Sukumar, Compromised HOXA5 function can limit p53 expression in human breast tumours, Nature 405 (2000) 974e978. [9] K. Wang, J. Li, H. Guo, X. Xu, G. Xiong, X. Guan, B. Liu, X. Chen, K. Yang, Y. Bai, MiR-196a binding-site SNP regulates RAP1A expression contributing to esophageal squamous cell carcinoma risk and metastasis, Carcinogenesis 33 (2012) 2147e2154. [10] T. Chiyomaru, S. Fukuhara, S. Saini, S. Majid, G. Deng, V. Shahryari, I. Chang, Y. Tanaka, H. Enokida, M. Nakagawa, R. Dahiya, S. Yamamura, Long noncoding RNA HOTAIR is targeted and regulated by miR-141 in human cancer cells, J. Biol. Chem. 289 (2014) 12550e12565. [11] C. Braconi, T. Kogure, N. Valeri, N. Huang, G. Nuovo, S. Costinean, M. Negrini, E. Miotto, C.M. Croce, T. Patel, microRNA-29 can regulate expression of the long non-coding RNA gene MEG3 in hepatocellular cancer, Oncogene 30 (2011) 4750e4756. [12] H. Meng, K. Wang, X. Chen, X. Guan, L. Hu, G. Xiong, J. Li, Y. Bai, MicroRNA330-3p functions as an oncogene in human esophageal cancer by targeting programmed cell death 4, Am. J. Cancer Res. 5 (2015) 1062e1075.
1157
[13] M. Liu, Y. Du, J. Gao, J. Liu, X. Kong, Y. Gong, Z. Li, H. Wu, H. Chen, Aberrant expression miR-196a is associated with abnormal apoptosis, invasion, and proliferation of pancreatic cancer cells, Pancreas 42 (2013) 1169e1181. [14] K. Wang, H. Guo, H. Hu, G. Xiong, X. Guan, J. Li, X. Xu, K. Yang, Y. Bai, A functional variation in pre-microRNA-196a is associated with susceptibility of esophageal squamous cell carcinoma risk in Chinese Han, Biomarkers 15 (2010) 614e618. [15] J. Zhang, F. Zheng, G. Yu, Y. Yin, Q. Lu, miR-196a targets netrin 4 and regulates cell proliferation and migration of cervical cancer cells, Biochem. Biophys. Res. Commun. 440 (2013) 582e588. [16] Y. Li, L. Jin, D. Chen, J. Liu, Z. Su, S. Yang, Y. Gui, X. Mao, G. Nie, Y. Lai, Tumor suppressive miR-196a is associated with cellular migration, proliferation and apoptosis in renal cell carcinoma, Mol. Med. Rep. 14 (2016) 560e566. [17] C. Schneider, R.M. King, L. Philipson, Genes specifically expressed at growth arrest of mammalian cells, Cell 54 (1988) 787e793. [18] M.R. Pickard, M. Mourtada-Maarabouni, G.T. Williams, Long non-coding RNA GAS5 regulates apoptosis in prostate cancer cell lines, Biochim. Biophys. Acta 1832 (2013) 1613e1623. [19] M. Sun, F.Y. Jin, R. Xia, R. Kong, J.H. Li, T.P. Xu, Y.W. Liu, E.B. Zhang, X.H. Liu, W. De, Decreased expression of long noncoding RNA GAS5 indicates a poor prognosis and promotes cell proliferation in gastric cancer, BMC Cancer 14 (2014) 319. [20] Meng Yang, Longcong Chen, Li Yang, Xian Li, Identification and validation long non-coding RNAs of oral squamous cell carcinoma by bioinformatics method, Oncotarget (2017), https://doi.org/10.18632/oncotarget.18178. [21] L. Salmena, L. Poliseno, Y. Tay, L. Kats, P.P. Pandolfi, A ceRNA hypothesis: the Rosetta Stone of a hidden RNA language? Cell 146 (2011) 353e358. [22] X. Zhao, P. Wang, J. Liu, J. Zheng, Y. Liu, J. Chen, Y. Xue, Gas5 exerts tumorsuppressive functions in human glioma cells by targeting miR-222, Mol. Ther. 23 (2015) 1899e1911. [23] Z. Zhang, Z. Zhu, K. Watabe, X. Zhang, C. Bai, M. Xu, F. Wu, Y.Y. Mo, Negative regulation of lncRNA GAS5 by miR-21, Cell Death Differ. 20 (2013) 1558e1568. [24] N. Zhang, A.Y. Wang, X.K. Wang, X.M. Sun, H.Z. Xue, GAS5 is downregulated in gastric cancer cells by promoter hypermethylation and regulates adriamycin sensitivity, Eur. Rev. Med. Pharmacol. Sci. 20 (2016) 3199e3205.