MiR-99b-5p and miR-203a-3p Function as Tumor Suppressors by Targeting IGF-1R in Gastric Cancer

MicroRNAs (miRNAs) have been explored in many critical cellular processes, including proliferation and apoptosis. The purpose of this study was to detect the biological function and regulation of miR-99b-5p and miR-203a-3p in gastric cancer (GC). Here, we demonstrated that miR-99b-5p/203a-3p were downregulated in both GC tissues and cell lines. MiR-99b-5p/203a-3p overexpression reduced GC cell proliferation and cell cycle progression in vitro. Notably, we combined bioinformatics tools with biological validation assays to demonstrate that insulin-like growth factor 1 receptor (IGF-1R) is a direct co-target and functional mediator of miR-99b-5p/203a-3p in GC cells. Mechanistically, the AKT pathway, which is downstream of IGF-1R, is essential for the functional roles of miR-99b-5p/203a-3p in GC cells. Taken together, our data revealed that IGF-1R is a direct co-target of miR-99b-5p/203a-3p, and miR-99b-5p/203a-3p may function as tumor suppressive miRNAs by negatively regulating IGF-1R expression in GC cells.

Gastric cancer (GC) is one of the most common malignant diseases and the second cause of cancer mortality worldwide 1,2 . Despite of the multiple advances in clinical and experimental cancer treatment, the prognosis of GC remains poor, with a terrible 5-year survival. Numerous studies have shown that the progression of GC might be a multistep process, involving the interaction between oncogenes and tumor suppressor genes 3,4 . Consequently, a better understanding of molecular mechanisms and signaling pathways is indispensable for identification of therapeutic targets for GC.
MiRNAs are a class of highly conserved (18-23 nucleotides), non-coding RNAs that modulate target genes expression through binding to 3′-untranslated regions (3′UTR) of mRNAs, which may result in mRNA degradation and translational repression 5,6 . MiRNAs play a significant role in regulation of biological processes, including cellular apoptosis, proliferation and differentiation [7][8][9] . Many reports have suggested that more than half of the miRNAs are located in cancer-related genomic regions and have been identified as oncogenes or tumor suppressors, suggesting that lots of miRNAs contribute to the progression of human cancers 10-12 .
MiR-99b belongs to the members of the miR-125a-let-7e cluster and has been reported to be involved in the differentiation, migration and proliferation of cancer cells. Studies have shown that miR-99b may function as modulators within a complex network of factors regulating TGF-β induced breast epithelial to mesenchymal transition 13 . Moreover, miR-99b inhibits cervical cancer cell invasion and proliferation by targeting mTOR signaling pathway 14 . MiR-203, is located on chromosome 14q32.33, which plays an important role in the regulation of human cancers. For example, miR-203a regulates proliferation and migration by targeting glycogen synthase kinase-3β in renal cell carcinoma 15 . In GC, aberrant expression of miRNAs, such as miR-338-3p and miR-145, has been shown to regulate tumor process by targeting downstream genes 16,17 . However, the role of miR-99b-5p/203a-3p and their cellular signaling pathway in GC are less reported and need further study.
Here, we investigated the roles of miR-99b-5p and miR-203a-3p in GC, which might play as suppressors, and demonstrated that IGF-1R was a co-target of miR-99b-5p/203a-3p to regulate the AKT signaling pathway. Hence, our findings facilitate better understanding of the miRNAs network control mechanism in the progression of GC.

Results
MiR-99b-5p and miR-203a-3p are downregulated in GC. To determine whether miR-99b-5p and miR-203a-3p were aberrantly expressed in GC, we re-analyzed RNA-seq data that was downloaded from The Cancer Genome Atlas website (TCGA) and found that miR-99b and miR-203a levels were lower in GC tissues than in normal tissue controls (P < 0.01) (Fig. 1A). We then performed qRT-PCR analysis in a set of 30 GC tissues and 30 matched adjacent normal gastric tissues. Significantly, a remarkable decrease of miR-99b-5p and miR-203a-3p expression was observed in the GC tissues compared with that in the matching non-tumor tissues. Based on the miR-99b-5p and miR-203a-3p expression levels measured by RT-qPCR, the 30 patients were divided into low (<1) and high (>1) miR-99b-5p/203a-3p expression groups according to the non-tumor tissues using the relative quantification. (Fig. 1C). The correlation between clinicopathologic factors and miR-99b-5p/203a-3p levels was examined in GC samples as shown in Table 1. The data showed that the expression was not associated with gender (P = 0.227/0.227), age (P = 0.169/ 0.612), lymphatic metastasis histology (P = 0.499/0.501), TNM stage (P = 0.543/0.743) and histology (P = 0.213/0.103), probably owing to the limited number of samples. In addition, as shown in Fig. 1B, the survival time was significantly longer in patients with miR-203a-high expression than that in patients with miR-203a-low expression (P < 0.05). However, the survival time did not differ between miR-99b-high and miR-99b-low expression.
To investigate miR-99b-5p/203a-3p function in vitro, we examined their expression in two GC cell lines. As shown in Fig. 1D, SGC-7901 and MKN-45, were characterized with lower expression of miR-99b-5p/203a-3p in mRNA levels compared with the immortalized gastric epithelial cell lines (GES-1) (P < 0.05). Together, these data reveal that the downregulation of miR-99b-5p and miR-203a-3p may be involved in GC development.
Overexpression of miR-99b-5p and miR-203a-3p inhibit proliferation of GC cells. To study the biological role of miR-99b-5p/203a-3p in tumor cells, cell growth was evaluated by MTT and colony formation assays. GC cells were transfected with pre-miR-99b or pre-miR-203a plasmid and controls. The transfection efficiency was monitored with GFP-labeled oligo, and an average of 80% efficiency was observed ( Supplementary  Fig. 1). Successful increase of miR-99b-5p and miR-203a-3p expression in GC cells was confirmed by qRT-PCR (P < 0.05) ( Fig. 2A). Overall survival analysis showed that there was no statistically significance between miR-99b high expression and low expression tumors, but miR-203a higher expression tumors had a better prognosis than lower expression tumors (P < 0.05). (C) Relative expression of miR-99b-5p and miR-203a-3p in GC tissue (n = 30) and matched adjacent normal tissues were determined by quantitative RT-PCR (qRT-PCR). U6 was used as loading control. (D) qRT-PCR analysis of miR-99b-5p/203a-3p expression in normal gastric cell (GES) and GC cell lines (MKN-45 and SGC-7901). Each sample was analyzed in triplicate (*P < 0.05, **P < 0.01, Student's t test).
SCiENTifiC RepoRts | (2018) 8:10119 | DOI:10.1038/s41598-018-27583-y MTT and colony formation assays showed that significant inhibition of cell proliferation in the miR-99b or miR-203a transfected cells compared with the control vector (P < 0.05, Fig. 2B,C). To further address whether the observed changes in proliferation were due to cell cycle and apoptosis, PI staining assay was used. Overexpression of miR-99b-5p/203a-3p repressed cell cycle arrest at G1-S in GC cells (Fig. 2D) and induced cell apoptosis (Fig. 2E).
To further explore the possible molecular mechanisms of miR-99b-5p/203a-3p arrested cell cycle and resulting apoptosis, we detected the expression of cell cycle related regulators and apoptosis-related proteins. After overexpression of miR-99b-5p/203a-3p in GC cells, the results of western blot analysis demonstrated that miR-99b-5p/203a-3p expression reduced the expression of cyclin D1, CDK4/6 ( Fig. 2D) and Bcl-2 ( Fig. 2F), as well as upregulated Bax. Together, these findings indicate that miR-99b-5p and miR-203a-3p may act as tumor suppressors in GC.
Moreover, we found that the IGF-1R level was overexpressed in GC tissues than compared to normal tissue controls from TCGA database (P < 0.01). Kaplan-Meier survival curves are shown in Fig. 4C (P > 0.05). We then performed qRT-PCR in a set of 30 primary GC tissues and 30 paired non-tumor gastric tissues from the same patients and found that IGF-1R was upregulated in GC tissues. Meanwhile, miR-99b-5p and miR-203a-3p levels were inversely correlated with IGF-1R expression by qRT-PCR assay. Likewise, the same results IGF-1R mRNA and protein overexpressed in MKN-45/SGC-7901 cell lines, in which both miR-99b-5p and miR-203a-3p were downregulated (Fig. 4D). The analysis of the relationship between IGF-1R expression level and clinical features of GC patients is shown in Table 1.

Discussion
MiRNAs have been regarded as critical regulators in cancer-related processes 29,30 . Increasing evidence have demonstrated that tumor-targeting therapies using miRNAs is becoming a novel diagnostic and therapeutic tool 31 . miR-223, miR-21 and miR-218 have been identified as novel potential biomarkers for gastric cancer detection. In previous studies, miR-99b-5p and miR-203a-3p have been reported to be involved in several cancers [32][33][34] . For example, miRNA-99b-5p suppresses liver metastasis of colorectal cancer by down-regulating mTOR, and miR-203a suppresses cell metastasis and angiogenesis through VEGFR by targeting HOXD3 in human hepatocellular carcinoma cells 35,36 . Besides, Liu W et al. have reported that downregulation of miR-203a by promoter methylation contributes to the invasiveness of gastric cardia adenocarcinoma. However, the underlying mechanisms responsible for decreased expression of miR-99b-5p and miR-203a-3p in GC remain to be determined.
In the present study, we observed that miR-99b-5p and miR-203a-3p were markedly down-expressed in GC tissues and cell lines, which is consistent with the analysis of TCGA. However, the overlap between nontumoral and tumor tissues may be because a large difference in the number of groups. Functionally, based on gain-or loss-of-function assays, our study showed that miR-99b-5p/203a-3p suppressed the cell growth and colony formation, leading to cell cycle arrest and apoptosis in GC cells. Moreover, exogenous miR-203a-3p expression inhibited cell proliferation by decreasing its viability and inhibiting its colony forming capacity more efficiently than miR-99b-5p. In addition, miR-99b-5p/203a-3p altered the expression of the apoptosis/cell cycle-related proteins, including Bcl-2, Bax, CDK4/6, and CyclinD1. Therefore, these results suggest that miR-99b-5p and miR-203a-3p functioned as tumor suppressors for the cell growth in GC. To this end, validation by in vivo assays is needed for further research.
To further determine the mechanisms and target mRNAs that were responsible for the suppressive role of miR-99b-5p/203a-3p, we observed that IGF-1R was a target gene of these two miRNAs. Previous studies have showed that IGF-1R is frequently overexpressed in human myeloma 37 , oral squamous cell carcinoma 38 and breast cancer [39][40][41][42] and that lots of miRNAs coordinate to regulate IGF-1R at multiple levels 43,44 . There were also two reasons for choosing IGF-1R. Firstly, IGF-1R was a co-target gene of miR-99b-5p and miR-203a-3p. Secondly, in our previous study that miR-302b-3p suppresses cell proliferation by targeting IGF-1R in GC 44 .
In this study, using bioinformatic analyses and a luciferase reporter assay, we confirmed that IGF-1R as a direct co-target gene of miR-99b-5p and miR-203a-3p. However, there is no significant inverse correlation between miR-99b-5p/203a-3p and IGF-1R, may be due to the limited number of GC tissue samples. Both mRNA and protein of IGF1R were significantly decreased in miRNAs overexpression group when compared to controls in GC cells. In contrast, we used siRNA to knockdown the expression of IGF-1R and showed that silencing IGF-1R inhibited the cell proliferation and induced cell apoptosis, which is similar to the effect of miR-99b-5p/miR-203a-3p overexpression in GC cells. Interestingly, silencing of IGF-1R in GC cells partly rescued the cells from the effects of anti-miR-99b-5p/203a-3p on regulation of IGF-1R expression and cell proliferation. These results strongly suggested that miR-99b-5p/203a-3p suppress the expression of IGF1R through directly targeting its 3′UTR.
Numerous studies showed that activation of the AKT signaling pathway was essential to the development and progression of cancer. Notably, IGF-1R exerted its function by activating the AKT signaling pathway followed by activation of target genes. In this study, we found that AKT signal pathway was suppressed by miR-99b-5p/203a-3p. Additionally, knockdown of IGF-1R by siRNA could induce similar inhibitory effects with miR-99b-5p/203a-3p on IGF-1R and AKT signal pathway.
In summary, our study showed that aberrant expression of miR-99b-5p/miR-203a-3p could affect cell proliferation of GC cells, probably through IGF-1R and its downstream signal pathway. We hope that our findings for the miR-99b-5p/203a-3p/IGF-1R/AKT signaling pathway will provide valuable information for the development of therapies against GC.

Materials and Methods
Gastric tissue samples and cell lines. Human GC tissue samples were obtained from patients undergoing surgical gastric resection at the First Affiliated Hospital of Xi'an Jiaotong University and People's Hospital of Shannxi Province. The matched non-tumorous tissues were taken from at least 5 cm distance from the edge of tumor tissues. To protocol used in the study was in accordance with the approved guidelines by the ethics committee, Xi'an Jiaotong University, and informed consent was obtained from all individuals. No patients (UICC I-II stages) received preoperative chemo-or radiotherapy before surgery. For UICC III and IV stages, patients agreed to be adjuvant with intraperitoneal chemotherapy to achieve better therapeutic effects. In addition, SGC-7901, MKN-45 and GES-1 cell lines were grown in RPMI-1640 medium (Thermo Scientific HyClone, USA) supplemented with 10% Biological Industries (BI), 10 mg/ml streptomycin (1% P/S) and incubated at 37 °C under a 5% CO 2 condition.
RNA extraction and qRT-PCR. Total RNA was prepared from the GC cells and gastric tissues using Trizol reagent (Invitrogen, USA) following the manufacturer's instruction. The RNA was quantified with a NanoDrop spectrophotometer (USA). Using PrimeScript RT Reagent Kit and SYBR Premix Ex Taq II Kit were purchased from TAKARA (Japan) for the detection of mature miRNAs expression and mRNA expression. The relative expression levels of IGF-1R and miR-99b-5p/203a-3p were respectively normalized to β-actin and U6. PCR was Figure 6. Knockdown of IGF-1R rescues anti-miR-99b-5p/203a-3p-induced cellular phenotypes in GC cells. (A-E) qRT-PCR, MTT assay, cell cycle, cell apoptosis and western blot were performed to determine the impact of GC cells treated with anti-miR-99b-5p/203a-3p plus si-IGF-1R expression vectors or related negative controls (*P < 0.05, **P < 0.01, Student's t test or Mann-Whitney test).
Dual-luciferase assay. The 3′UTR of human IGF-1R mRNA was constructed with synthetic oligonucleotides and cloned in between the Sac I and Xho I sites of the pmirGLO Dual-luciferase miRNA target expression vector (Promega). HEK293 cells were seeded in 96-well plate and allowed to settle for ~12 h. MiR-99b-5p or miR-203a-3p was co-transfected with the pmirGLO-IGF-1R-3′-UTR-WT or pmirGLO-IGF-1R-3′-UTR-WUT vector into HEK293 cells, respectively. The pmirGLO vector was used as control signals. Then, cells were measured at 24 h after transfection using the Dual-Luciferase Reporter Assay System (Promega, USA).
Cell proliferation assay. 3000 GC transfected cells were transferred to each well of a 96-well plate. At Colony formation assay. For the assessment of colony formation, transfected GC cells were seeded in 6-well plates with 1000 cells/well in triplicate and incubated for 1-2 weeks. Then, the plates were washed with PBS and stained with 0.5% crystal violet for 20 minutes. After washing 3 times, the number of colonies was counted by taking pictures from Bio-Rad.
Cell cycle and cell apoptosis analysis. Cells were cultured in 6-well plates with 2 × 10 5 cells/well in triplicate and transfected for 48 h, which were then collected and washed with PBS two times and fixed with 70% ethanol at 4 °C overnight. Then, cells were treated with RNase A and propidium iodide (PI) according to KGI Cell Cycle Detection Kit (China). After incubation, the cells were assayed by flow cytometry (Becton, USA). For cell apoptosis analysis, after 48 h transfection, cells were analyzed with Annexin-V FITC Apoptosis Detection Kit (Invitrogen, USA) and examined by using flow cytometer; the data were examined by ModFit software.
Western blot. Total protein was harvested from GC cells using RIPA buffer (Wolsen, China) after 48 h transfection and 20 μg of isolated protein lysates were separated by 10% SDS-PAGE and transferred to PVDF membrane (Millipore, USA). The membranes were probed with the following primary antibodies: IGF-1R, AKT, phospho-AKT (Ser473), CCND1, CDK4/6, CDK2, Bcl-2 and Bax (Cell Signaling Technology, diluted 1/1,000) overnight at 4 °C. The expression levels of above proteins were standardized to human β-actin using a mouse mAb anti-β-actin antibody. Then, the membranes were incubated with the HRP-conjugated goat anti-mouse or anti-rabbit IgG antibody (ZSGB-BIO, China). The blots were scanned, and the band density was measured using the Quantity One imaging software. Statistical analysis. Statistical analysis was performed using the PASW Statistics 17.0 software (SPSS, USA).
Each experiment was repeated at least three times. Independent sample Student's t-test was used if the quantitative data between groups show normal distribution. If not consistent with the normal distribution, using the Wilcoxon-Mann-Whitney test. Pearson's correlation analysis was used to analyze the correlation between two indices. The relationship between miR-99b-5p/203a-3p/IGF-1R expression level and clinical parameters was calculated by the χ 2 -test. A p-value ≤ 0.05 was considered statistically significant.