BACKGROUND: Lung cancer is the leading cause of cancer-related deaths, because its morbidity and mortality is the highest in the world. Lung adenocarcinoma (LAC) is the most common pathological type, and its morbidity number is increasing year by year, therefore, it has attracted people’s attention. A large number of patients with LAC is characterized by solitary pulmonary nodules (SPN) on CT. Because the SPNs in early stage are usually small and mostly located in the periphery of the lung, it is a great challenge for clinicians to judge its benign or malignant. Therefore, it is necessary to search some new diagnostic biomarkers. Cyclic RNAs (CircRNAs) are tissue-specific and disease-specific, and they has been identified to be candidate biomarkers in many kind of tumors. Here, we will screen a circRNA with diagnostic value in LAC.
OBJECTIVE: The aim of this study is to screen a circRNA with diagnostic value in LAC and then to study its functions and mechanisms.
METHODS：In this study, we firstly used CircRNA microarray to analysis the cancer and normal tissues of three cases of patients with LAC, one of the highest CircRNA has_circ_0013958 was selected. The PCR products were verified by Northern Blot and sequencing method. Then, 7 kind of lung adenocarcinoma cell lines, 49 lung adenocarcinoma tissues and 30 lung adenocarcinoma samples were examined by qPCR to verify the expression of has_circ_0013958. The diagnostic value of has_circ_0013958 in tissues and plasma was analyzed by using ROC curve.
In order to further study the biological function of has_circ_0013958, three siRNAs of has_circ_0013958 were designed and screened by qPCR. The effect of proliferation of has_circ_0013958 in lung adenocarcinoma cells was detected by CCK-8 and EDU experiments. The effect of has_circ_0013958 on cell apoptosis was detected by flow cytometry, and Transwell assays were used to detect metastasis and invasion functions. The model of human LAC xenografts in nude mouse was made to verify the effect of has_circ_0013958 on tumorigenesis and tumor growth in vivo. After the model was successfully made, cholesterol modified has_circ_0013958 siRNA was intratumorally injected once every 3 days, 10 nmol every time. The length and diameter of tumors were measured by vernier caliper every week for 4 weeks. After the mice were killed, the tumor tissue were removed and detected Ki67 by immunohistochemical staining, apoptosis were detected by TUNEL. In order to further investigate the mechanism of has_circ_0013958, we first detect its content in cytoplasm and nucleus by qPCR, and then it was morphologically verified by FISH method. On this basis, the bioinformatics database was used to analyze the possible downstream miRNA and mRNA molecules of has_circ_0013958, and then a network of interactions between them was constructed. The combination of has_circ_0013958 and miRNA was verified by luciferase assay, and the relationship between has_circ_0013958 and its downstream miRNA and mRNA molecules was verified by Western Blot method.
RESULTS：A total of 59 kinds of CircRNA which were differentially expressed to more than 2 times and had statistical significance. Among them, 39 tissues were highly expressed, and 20 were low expressed. Has_circ_0013958 (hsa_circRNA_100323) is one of the most obvious expression of CircRNA, and it was we choose as the candidate biomarker for verification. The qPCR product was proved as correct by Northern Blot and sequencing. On this basis, has_circ_0013958 was highly expressed in 7 lung adenocarcinoma cell lines, 49 lung adenocarcinoma tissues and 30 lung adenocarcinoma patients (P < 0.05). The ROC analysis showed that the area under (AUC) of the overall ROC curve of has_circ_0013958 was 0.815, and the sensitivity and specificity were 0.755 and 0.796 respectively, and the cutoff value was 0.00101; Stage I ROC AUC 0.750 (P < 0.001), sensitivity and specificity were 0.583 and 0.833; stage II ROC AUC is 0.766, the sensitivity and specificity were 0.813 and 0.750; stage III~IV ROC AUC 0.874, sensitivity and specificity were 0.762 and 0.857 respectively. The ROC curve of has_circ_0013958 in plasma samples showed that AUC was 0.794 (P < 0.001), and sensitivity and specificity were 0.667 and 0.933 respectively.
In order to perform functional studies, we first screened the siRNA sequence of has_circ_0013958. CCK-8 experiments showed that the lung adenocarcinoma cells were interfered by has_circ_0013958 of siRNA, and the OD450 value of the cells was significantly lower than that of the control group (P < 0.01). EDU experiments showed that after siRNA interference with has_circ_0013958, the percentage of cells in the proliferative phase decreased significantly (P < 0.01). To detect whether has_circ_0013958 can affect apoptosis, we used Annexin V FITC/PI to stain the cells and detected apoptosis by flow cytometry. The results showed that siRNA specific silencing of has_circ_0013958 significantly increased the apoptotic rate of cells (P < 0.01). Transwell experiments showed that the number of cells transferred and attacked by has_circ_0013958 was significantly decreased after silencing (P < 0.01). Tumor formation in nude mice showed that interference with has_circ_0013958 significantly inhibited tumor growth compared with the control group (P < 0.001). Compared with the control group, the expression of Ki67 in the tissues of the nude mice after the interference of has_circ_0013958 was significantly reduced (P < 0.01). TUNEL experiments were carried out on tumor tissues, and the results showed that the proportion of apoptotic cells was significantly increased after has_circ_0013958 interference (P < 0.01).
After the clinical research and functional research, we do some mechanism study. The has_circ_0013958 in the nucleus and cytoplasm were detected by qPCR, it was mainly distributed in cytoplasm, FISH experiments further confirmed this conclusion. In order to further explore the mechanism of has_circ_0013958, we use TargetScan and miRanda to predict miRNA molecules and downstream target genes can be combined with the has_circ_0013958 database, five miRNA molecules were predicted: miR-134, miR-545, miR-629, miR-509, miR-660. The dual luciferase showed that only miR-134 can bind with has_circ_0013958, so we confirm that miR-134 was the downstream target of has_circ_0013958. According to the literature, the target gene downstream of miR-134 in non-small cell lung cancer (NSCLC) is CCND1, and we hypothesize that there may be a circ_0013958/miR-134/CCND1 pathway in LAC. Western Blot experiments showed that has_circ_0013958 control of CCND1 needs to be achieved by adsorption of miR-134. Western Blot experiments showed that has_circ_0013958 can affect the expression of CCND1, but it must be achieved by adsorption of miR-134.
Taken together, this study demonstrated for the first time that has_circ_0013958 might function as competing endogenous RNA (ceRNA) for miR-134, which could contribute to proliferation through activating CCND1 signaling pathway.
CONCLUSIONS: Our data suggest that i) has_circ_0013958 was overexpressed in lung adenocarcinoma cell lines, tissues and plasma samples; ii) Has_circ_0013958 accelerated cell proliferation, migration and inhibit apoptosis in LAC; iii) Has_circ_0013958 might function as competing endogenous RNA (ceRNA) for miR-134, which could contribute to proliferation through activating CCND1 signaling pathway; iv) Has_circ_0013958 would be a promising biomarker for LAC diagnosis