Abstract 1173: A genomics approach to understanding DICER1's role in tumorigenesis

Proceedings: AACR 102nd Annual Meeting 2011‐‐ Apr 2‐6, 2011; Orlando, FL DICER1 is required for the generation of mature microRNAs (22-24 nt miRNAs) and other short noncoding RNAs. Several lines of evidence point to DICER1's role as a tumor suppressor. In particular, reduced DICER1 levels are associated with adverse outcomes in a variety of human and mouse cancers. To date the effects of reduced DICER1 on mRNA transcript abundance in tumor cells remain largely unknown. Furthermore, DICER1 may have non-canonical (unrelated to miRNA activity) functions in tumorigenesis. In model organisms, small RNAs initiate transcriptional silencing. DICER1 may have a similar role in transcriptional silencing and DNA methylation in mammalian cells. To begin to determine how reduced DICER1 levels contribute to tumor phenotypes we used shRNA to stably knock down DICER1 in endometrial cancer cell lines for extended periods (up to 25 passages). Reduced DICER1 levels (measured by Western blot) had no effect on cell proliferation but were associated with enhanced cell migration. We performed microRNA, mRNA, and DNA methylation profiling experiments in the KLE endometrial cancer cell line. MiRNA levels were decreased overall and the fold reductions for individual miRNAs were highly correlated in two different DICER1 shRNA knockdowns. RNAseq analysis (Illumina) revealed a 4-fold increase in the TARBP2 (TRBP) and MOV-10 transcripts. TARBP2 and MOV-10 are integral components of the DICER1 complex and the observed upregulation of their transcripts could reflect selection to stabilize DICER1 in the knockdowns. Furthermore, miR-103 and miR-107, which target DICER1 mRNA, were downregulated in the knockdown cells, possibility reflecting selection to maintain DICER1 activity. Consistent with decreases in miRNA levels, many mRNA transcripts were upregulated with DICER1 knockdown (average 1.1 fold increase across 8077 informative transcripts). Most striking was the number of interferon responsive genes that showed upregulation. Among the 8077 informative transcripts, 28 were classified as interferon responsive, and for those the average and median fold increase was 3.1 and 2.1 respectively (range 0.9 – 12.9). The majority of those interferon responsive genes with increased transcript levels lack miRNA target sequences. Previous work has shown that shRNAs do not induce a general antiviral interferon response. We are currently performing studies to confirm that the effect observed is DICER1 dependent. Technical and biologic validations for those interferon responsive genes that show upregulation based on RNAseq are underway. Increases in IFI6 and IFI44L transcript levels have been confirmed by qPCR. We hypothesize a transcription-based mechanism of DICER1 control of interferon responsive transcripts and are attempting to determine if there is such an effect and whether it is a direct or indirect effect. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 1173. doi:10.1158/1538-7445.AM2011-1173