The Expression And Bioinformatics Analysis Of MiR-449a In Different Molecular Subtypes Of Breast Cancer

OBJECTIVE:To assess the expression of miR-449a and antiapoptotic protein Bcl-2 and their clinical value in different molecular subtypes of breast cancer tissue specimens. Bioinformatics methods were used to analysis the expression of miR-449a in breast cancer tissues.The target genes of miR-449a were predicted and enrichment analysis were acted to further explore the biological function and regulation mechanism of miR-449a, and further to investigate Bcl-2 protein and risk factors in breast cancer with a systematic review.METHODS:Five hundred and five cases of breast cancer with the clinical parameters information of the expression of miR-449a datas were downloaded from TCGA database. Paired tissue specimens from 58 breast cancer patients and two breast cancer cells were selected for qRT-PCR analysis of miR-449a expression. Gene ontology, KEGG analyses were performed to predict miR-449a-targeting genes and pathways with different web-based tools. The correlations between miR-449a expression with clinical characteristics parameters were analyzed. One hundred and twenty-three female breast cancer (all FFPE) tissues and adjacent normal tissues were collected. According to the results of immunohistochemistry, all samples were histologically distributed into four molecular subtypes:luminal A(30), luminal B(31), HER2 overexpression(33) and triple negative breast cancer(29). The expressions of Bcl-2 proteins were examined by immunohistochemistry. The published studies were searched in the CNKI database, Weipu database, Wanfang database and other relevant journals were also hand searched to identify all the relevant case-control trials. Then the quality of the included trials was assessed and meta-analysis was performed by Stata 11.0.RESULTS:TCGA datas statistical analysis showed that the miR-449a is higher expression in breast cancer tissue than normal, and the difference was statistically significant (P<0.05).The miR-449a expression in cancer of breast was in significant correlation with tumor node metastasis staging (P<0.05) and closely related to lymph node metastasis (P<0.05). However, miR-449a expression in breast cancer was not associated with other parameters such as age, tumor size, histological grade, pathologic types and ER status, PR status, HER2 status (P>0.05). The expression of miR-449a was significantly down-regulated in breast cancer tissues when compared to corresponding adjacent normal breast tissues(P<0.05). The relative expression level was 0.36±0.35 and no expression differences between the four different subtypes (P>0.05). Bioinformatics results showed that the function of the genes were enriched in the molecular processes of cell proliferation, transcription and metabolic process, sequence-specific DNA binding, enzyme binding, cytoskeleton et al. In KEGG pathway, the predicted target genes involved in MAPK and P53 signal transduction pathways, and the pathway in cancer disease such as chronic myeloid leukemia and small cell lung cancer, as well as cell cycle and endocytosis. The network analysis of target gene encoding protein showed that the core genes were BCL2, SRC, NOTCH 1, HDAC1, CCND1 and CDK6, and they play vital roles of stabilizing structures in the interaction network. The difference of positive expression rates of Bcl-2 protein expression among breast cancer tissues and adjacent normal tissues were 77.2%(95/123),60.2%(74/123), and different subtypes including luminal A, luminal B, HER2 over-expression and triple-negative were 86.7%(26/30),90.3%(28/31),72.7%(24/33),58,6% (17/29), and the difference was statistically significant (x2=10.634, P=0.014). The positive Bcl-2 expression rate of TNBC was lower than luminal A and luminal B. The analysis revealed that a high expression of Bcl-2 positively associated with TNM staging (P<0.05). However, no statistically significant correlations were observed between Bcl-2 expression and other clinicopathological features, including age, tumor size, histological grade, lymph node metastasis were found. The expression levels of miR-449a showed statistically significant correlation with the expression of Bcl-2 with the following Spearman’s correlation coefficients:-0.299 (P=0.019). And low levels of miR-449a were more likely to be observed in BC with high expression of Bcl-2. Meta analysis showed:as for the positive rate of Bcl-2 expression, no significant differences were tested between breast cancer vs. normal tissues (OR=0.956,95%CI=0.469-1.950, P=0.902). There were significant differences between highly differentiation and moderately differentiated vs. low-grade differentiation (P<0.05). No significant difference was tested between age, tumor size, TNM staging, lymph node metastasis (P>0.05). Bcl-2 expression was positively correlated with ER and PR (r=0.363,0.254; P<0.01).CONCLUSION:The expression of miR-449a was significantly down-regulated in breast cancer tissues. The predicted target genes of miR-449a participate in the regulation of multiple biological processes and signal transduction pathways. There were significant differences of Bcl-2 expression in different molecular subtypes of breast cancer, and meanwhile substantial correlation is confirmed between Bcl-2 and ER, PR. Bcl-2 expression was negative correlation with the expression of miR-449a. The Bcl-2 may act a potential target gene of miR-449a and participate in the process of breast cancer.