Abstract 1945: Identification of a miRNA/mRNA network driving non-small cell lung cancer (NSCLC) dissemination

Abstract

Abstract Background: Non-small cell lung cancer (NSCLC) is one of the most aggressive tumor entities and first data indicate that microRNAs (miRNAs) are central regulators of NSCLC dissemination. Since each miRNA is able to modulate the expression of several transcripts, they are promising targets for the development of drugs that cause efficient antitumor effects and low resistance. However, the relevant network of miRNA/mRNA driving NSCLC metastasis has not been identified yet. Methods: The differential expression of miRNAs was compared between NSCLC samples from patients with and without lymph node metastasis (N1, N2 and N3 vs. N0) in a cohort of The Cancer Genomic Atlas (TCGA) database (n = 449). The dysregulation of the miRNAs in tumors versus normal lung samples (n = 39) was also analyzed. For validation, fresh-frozen samples from an independent patient cohort (n = 108) were analyzed by qRT-PCR. The role of selected miRNAs in tumor dissemination was assessed by migration and invasion experiments after transfection of respective antagomirs and agomirs in NSCLC cells (time-lapse microscopy). The novel algorithm “miRlastic” was used to identify potential miRNA targets through the integration of miRNA-mRNA expression data by negative multiple linear regression analysis. Moreover, differential methylation of the miRNA genomic locations was studied as a possible mechanism of miRNA dysregulation by analyzing Illumina Infinium 450 k DNA methylation TCGA data (n = 29). Results: By using a stringent selection process, we identified 135 miRNAs differentially induced or reduced in NSCLCs with lymph node metastasis (p≤0.05). Interestingly, 22/135 (16.3%) of the selected miRNAs were located in the chromosomal cluster 14q32.31. Elevated expression of miR-323b, miR-487a and miR-539, which are located in 14q32.31, significantly correlated with poor patient survival. Time-resolved and quantitative analysis of lateral migration illustrated that these miRNAs increased tumor migration without affecting cell viability. Moreover, miRlastic identified several metastasis-related genes as potential downstream targets of these miRNAs. The connection between miRNAs encoded in 14q32.31 and candidate targets was confirmed in NSCLC cell lines (e.g. Pumilio RNA-Binding Family Member-2; PUM2). Lastly, hypomethylation of the 14q32.31 cluster in tumor tissues might explain increased expression of these miRNAs. Conclusions: Our results demonstrate that miRNAs located in the chromosomal cluster 14q32.31 are driving NSCLC dissemination. Therefore, we hypothesize that the coordinated overexpression of these miRNAs is part of a genetic network supporting cancer progression and that they represent promising cancer biomarkers and therapeutic targets. Citation Format: Margarita González-Vallinas, Marco Albrecht, Adriana Pitea, Manuel Rodríguez-Paredes, Damian Stichel, Steffen Sass, Julian Gutekunst, Jennifer Schmitt, Thomas Muley, Michael Meister, Arne Warth, Peter Schirmacher, Fabian J. Theis, Nikola S. Müller, Franziska Matthäus, Kai Breuhahn. Identification of a miRNA/mRNA network driving non-small cell lung cancer (NSCLC) dissemination. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 1945.