Analysis of the miRNA–mRNA–lncRNA networks in ER+ and ER− breast cancer cell lines

Recently, rapid advances in bioinformatics analysis have expanded our understanding of the transcriptome to a genome‐wide level. miRNA–mRNA–lncRNA interactions have been shown to play critical regulatory role in cancer biology. In this study, we discussed the use of an integrated systematic approach to explore new facets of the oestrogen receptor (ER)‐regulated transcriptome. The identification of RNAs that are related to the expression status of the ER may be useful in clinical therapy and prognosis. We used a network modelling strategy. First, microarray expression profiling of mRNA, lncRNA and miRNA was performed in MCF‐7 (ER‐positive) and MDA‐MB‐231 cells (ER‐ negative). A co‐expression network was then built using co‐expression relationships of the differentially expressed mRNAs and lncRNAs. Finally, the selected miRNA–mRNA network was added to the network. The key miRNA–mRNA–lncRNA interaction can be inferred from the network. The mRNA and non‐coding RNA expression profiles of the cells with different ER phenotypes were distinct. Among the aberrantly expressed miRNAs, the expression levels of miR‐19a‐3p, miR‐19b‐3p and miR‐130a‐3p were much lower in the MCF‐7 cells, whereas that of miR‐148b‐3p was much higher. In a cluster of miR‐17‐92, the expression levels of six of seven miRNAs were lower in the MCF‐7 cells, in addition to miR‐20b in the miR‐106a‐363 cluster. However, the levels of all the miRNAs in the miR‐106a‐25 cluster were higher in the MCF‐7 cells. In the co‐expression networking, CD74 and FMNL2 gene which is involved in the immune response and metastasis, respectively, had a stronger correlation with ER. Among the aberrantly expressed lncRNAs, lncRNA‐DLEU1 was highly expressed in the MCF‐7 cells. A statistical analysis revealed that there was a co‐expression relationship between ESR1 and lncRNA‐DLEU1. In addition, miR‐19a and lncRNA‐DLEU1 are both located on the human chromosome 13q. We speculate that miR‐19a might be co‐expressed with lncRNA‐DLEU1 to co‐regulate the expression of ESR1, which influences the occurrence and development of breast cancer cells with different levels of ER expression. Our findings reveal that the status of ER is mainly due to the differences in the mRNA and ncRNA profile between the breast cancer cell lines, and highlight the importance of studying the miRNA–mRNA–lncRNA interactions to completely illustrate the intricate transcriptome.