YAP/TAZ-dependent expression of lncRNAs [HLF]

Body fluids from cancer patients are informative regarding disease conditions in tumor tissues. Thus, biomarkers detected by liquid biopsy can guide clinicians in designing personalized therapies and may serve as a proxy for treatment success. However, biomarkers that predict the existence of druggable target structures cannot be reliably defined in blood samples due to genetic tumor heterogeneity and presence of molecules from non-tumorous cells. To test the applicability of RNA signatures as liquid biopsy biomarkers, expression data from hepatocellular carcinoma (HCC) cells after inhibition of Hippo pathway effectors is investigated. We show that the oncogene yes-associated protein (YAP) transcriptionally controls a panel of long non-coding RNAs (lncRNAs), which support HCC progression via tumor cell-intrinsic mechanisms. These lncRNAs are detectable and overexpressed in YAPS127A transgenic mouse livers as well as in a subgroup of human HCC tissue and serum samples. Using a machine learning algorithm, a 4 gene lncRNA signature is defined that correlates with the nuclear abundance of YAP in HCC tissues (CYTOR, SNHG1, SNHG17, MIR4435-2HG). Importantly, YAP accumulation in human HCCs is significantly associated with lncRNA signature serum abundance in two independent patient cohorts. Evaluation of expression data and confirmatory experiments illustrate that the lncRNA signature is a robust predictor for YAP activity in other tumor types such as lung adenocarcinoma. Liquid biopsy-based detection of lncRNAs in cancer patients is informative for the activity of transcriptional regulators and can serve as diagnostic tool that guides clinicians in the design of targeted therapies. 2 different siRNA combinations targeting YAP and TAZ with 4 biological replicates