The NOTCH1 Driven Long Non-Coding RNA Repertoire in T-Cell Acute Lymphoblastic Leukemia

Introduction: T-cell acute lymphoblastic leukemia (T-ALL) is an aggressive cancer resulting from the malignant transformation of T-cell precursors. Genetic studies in T-cell acute lymphoblastic leukemia (T-ALL) have uncovered a remarkable heterogeneity of genetic defects. Amongst these, NOTCH1activating mutations are the most frequently occurring events (>50%) in T-ALL. Since long non-coding RNAs (lncRNAs) are emerging as important players in oncogenesis, we decided to decode the NOTCH1 driven lncRNA transcriptional landscape in T-ALL and normal T-cell development.

Methods and Results: RNA-sequencing was performed following pharmacological inhibition (GSI) of the NOTCH1 mutant and gamma secretase inhibitor (GSI) sensitive T-ALL cell line CUTLL1 in a time series experiment as well as for human CD34+ thymic progenitor T-cells cultured on an OP9 feeder layer with or without DL1-triggered NOTCH1 stimulation. First, we validated both model systems by confirming robust regulation of multiple canonical known protein coding NOTCH1 target genes including DTX1, NOTCH3 and NRARP. Next, we identified distinct subsets of NOTCH1 regulated lncRNAs in both experiments with an overlap of 27 commonly regulated NOTCH1 driven and previously annotated lncRNAs. An even larger number of novel, unannotated T-ALL/T-cell specific lncRNAs was found to be NOTCH1 regulated. Next, we took advantage of publically available ChIP-sequencing data for ICN1 and enhancer specific chromatin marks in CUTLL1 (Wang et al., PNAS, 2013), allowing the selection for direct regulated lncRNAs with enhancer properties. Amongst these lncRNAs, the recently described LUNAR1 lncRNA (Trimarchi et al., Cell, 2014) was present as top candidate in our dataset, thus validating this approach for further selecting bona fide NOTCH1 regulated lncRNAs. In a first step towards functional annotation of this subset of selected lncRNAs, we performed so-called guilt-by-association analysis through correlating expression levels of the individual lncRNAs with transcriptome data for all protein coding genes followed by gene set enrichment analysis in a large cohort of primary T-ALL patients. Subsequent enrichment mapping of significant gene sets yielded markedly different gene set clustering patterns for each of the individual analyzed lncRNAs, as well as distinct annotated related functionalities such as cytokine signaling, TCA-cycle, DNA replication and repair and translation. Prioritarization of lncRNAs for further functional validation was performed by measuring their expression in an extended panel of GSI-treated T-ALL cell lines (HPB-ALL, DND-41, T-ALL1 and ALL-SIL), sorted subsets of CD34+ and CD4+/CD8+ double positive thymocytes and an independent T-ALL patient cohort.

Conclusion: We present the landscaping of an integrated lncRNA network acting downstream of NOTCH1 signaling in T-ALL and normal T-cells. These data pave the way towards the development of novel therapeutic strategies impacting on hyperactive NOTCH1 signaling.