To identify microRNAs (miRNAs) associated with estrogen receptor (ESR1) status, we profiled luminal A, ESR1+ breast cancer cell lines versus triple negative (TN), which lack ERα, progesterone receptor and Her2/neu. Although two thirds of the differentially expressed miRNAs are higher in ESR1+ breast cancer cells, some miRNAs, such as miR-222/221 and miR-29a, are dramatically higher in ESR1- cells (~100- and 16-fold higher, respectively). MiR-222/221 (which target ESR1 itself) and miR-29a are predicted to target the 3′ UTR of Dicer1. Addition of these miRNAs to ESR1+ cells reduces Dicer protein, whereas antagonizing miR-222 in ESR1- cells increases Dicer protein. We demonstrate via luciferase reporter assays that these miRNAs directly target the Dicer1 3′ UTR. In contrast, miR-200c, which promotes an epithelial phenotype, is 58-fold higher in the more well-differentiated ERα+ cells, and restoration of miR-200c to ERα- cells causes increased Dicer protein, resulting in increased levels of other mature miRNAs typically low in ESR1- cells. Together, our findings explain why Dicer is low in ERα negative breast cancers, since such cells express high miR-221/222 and miR-29a levels (which repress Dicer) and low miR-200c (which positively affect Dicer levels). Furthermore, we find that miR-7, which is more abundant in ERα+ cells and is estrogen regulated, targets growth factor receptors and signaling intermediates such as EGFR, IGF1R, and IRS-2. In summary, miRNAs differentially expressed in ERα+ versus ERα- breast cancers actively control some of the most distinguishing characteristics of the luminal A and TN subtypes, such as ERα itself, Dicer, and growth factor receptor levels.
Bibliographical noteFunding Information:
Acknowledgements This work was supported by Department of Defense Breast Cancer Research Program Idea Award BC084162, Susan G Komen Foundation KG090415 (J. Richer), and the National Institutes of Health R01CA74285 (D. Yee).We thank Aik-Choon Tan (University of Colorado, Division of Medical Oncology) for bio-informatics and Dorraya El-Ashry and Phil Miller (University of Miami, Miller School of Medicine) for help with bioinformatics and thoughtful comments on the manuscript. We also thank the University of Colorado Cancer Center Cores for DNA Sequencing and Analysis, Tissue Procurement and Pathology, supported by the NIH/National Cancer Institute Cancer Core Support Grant P30 CA046934.
- Breast cancer
- Epithelial to mesenchymal transition