Long non-coding RNA (lncRNA) transcriptional landscape in breast cancer identifies LINC01614 as non-favorable prognostic biomarker regulated by TGFβ and focal adhesion kinase (FAK) signaling

Radhakrishnan Vishnubalaji; Hibah Shaath; Eyad Elkord; Nehad M Alajez

doi:10.1038/s41420-019-0190-6

Long non-coding RNA (lncRNA) transcriptional landscape in breast cancer identifies LINC01614 as non-favorable prognostic biomarker regulated by TGFβ and focal adhesion kinase (FAK) signaling

Cell Death Discov. 2019 Jun 24:5:109. doi: 10.1038/s41420-019-0190-6. eCollection 2019.

Authors

Radhakrishnan Vishnubalaji¹, Hibah Shaath¹, Eyad Elkord¹, Nehad M Alajez¹

Affiliation

¹ Cancer Research Center, Qatar Biomedical Research Institute (QBRI), Hamad Bin Khalifa University (HBKU), Qatar Foundation (QF), PO Box 34110 Doha, Qatar.

Abstract

Long non-coding RNAs (lncRNAs) represent a class of epigenetic regulators implicated in a number of physiological and pathological conditions. Herein, we characterized the lncRNA expression portrait from 837 patients with invasive breast cancer and 105 normals from the cancer genome atlas (TCGA), which revealed eighteen upregulated and forty-six downregulated lncRNAs. Clustering analysis revealed distinct lncRNA profile for the triple negative breast cancer (TNBC) and normal breast tissue, while less separation was observed among the HER2⁺HR⁺, HER2⁺HR^-, HER2^-HR⁺ molecular subtypes. LINC01614, and LINC01235 correlated with worse disease-free survival (DFS), while the expression of lnc-LRR1-1, lnc-ODF3B-2, AC015712.5, lnc-LAMB3-1, lnc-SPP2-3, and lnc-MAP9-2 correlated with better DFS. The expression of LINC01235 correlated with worse overall survival (OS), while the expression of MIR205HG, lnc-MAP2K6-5, FGF14-AS2, lnc-SPP2-3 correlated with better OS. Highest expression of LINC01614 was observed in progesterone receptor (PR)⁺, Estrogen receptor (PR)⁺, and HER2⁺ tumors, while lowest expression was in TNBC. Concordantly, LINC01614 was highly expressed in the luminalB/HER2⁺ subtype from the SRP062132 dataset. Elevated expression of LINC01614 was subsequently validated in primary breast cancer tissue and breast cancer cell lines. Bioinformatics and pathway analyses on LINC01614^high vs. LINC01614^low BC tissue revealed TGFβ1 and ECM as the most activated networks in LINC01614^high tumors. Concordantly, strong correlation between the expression of LINC01614 and COL10A1 (R ² = 0.6929), SPOCK1 (R ² = 0.5156), ZEB1 (R ² = 0.3372), TGFBI (R ² = 0.2978), TGFB1 (R ² = 0.1985), ACTA2 (R ² = 0.1833), and TAGLN (R ² = 0.1909) was observed. Mechanistically, exogenous TGFB1 induced LINC01614 expression in the BT474 triple positive BC model, while small-molecule inhibition of transforming growth factor β (TGFβ, SB-431542) or focal adhesion kinase (FAK, PF-573228) abrogated LINC01614 expression. Our data revealed the lncRNA transcription landscape in breast cancer and its molecular subtypes. Our data provide novel insight implicating LINC01614 as unfavorable prognostic marker in BC, its association with the HR⁺/HER2⁺ BC molecular subtype and its regulation by TGFβ and FAK signaling.