Recent observations indicate that microRNA (miR) are deregulated in cancer and are important for disease pathogenesis; however, how miR exert direct or indirect regulatory effects on intracellular signaling pathways is poorly understood. can be attributed at least in part to delayed development of technologies to capture miRNA-target RISC complexes. Target prediction algorithms use miRNA seed sequence homology scans and RNA structural information, but problems with accuracy and failure to account for critical biological factors such as cellular context diminish their utility. Acute myeloid leukemias (AML) with HoxA gene expression signatures (e.g. NPM1c mutant, 11q23 and 11p15 cytogenetic abnormalities) have distinct gene and miRNA expression profiles compared to other subgroups of AML and normal hematopoietic cells. MLL (11q23) translocations comprise approximately 10% of adult AML, and confer a poor prognosis. Thus, this subset of AML may benefit from targeted disruption of miR function through RNA therapeutic intervention. We find significantly elevated miR-196b expression in primary patient blasts, human AML cell lines, and primary murine leukemic spleenocytes harboring HoxA signatures compared to non-Hox based leukemias and normal controls. Moreover, RNA therapeutic antagonism of miR-196b function specifically kills 11q23 AML cells but not AML cells containing other chromosomal rearrangements or normal cells. To define miR-196b targets, miR-196b-mRNA RISC complexes were purified from two different cell lines then target mRNA identities were revealed by microarray analyses. We identified 422 putative leukemia-relevant miR-196b captured targets, of which 50% were predicted by algorithms. Both predicted and novel targets were independently validated by repeated miR-target capture experiments and target site sequences were confirmed by luciferase reporter assays. When miR-196b levels were modulated over a four-day time course, we quantified differential responsiveness in captured target mRNA expression patterns over time, while protein levels of the novel targets were uniformly reduced. Bioinformatic network analyses of captured targets illustrate several biochemical pathways that might be involved in leukemogenesis. To functionally delineate the leukemia-relevant miR-196b targets, we devised an shRNA screen of miR-196b captured targets in primary murine Mll-Af9 leukemia. We identified tumor suppressor Cdkn1b (p27) as a functional miR-196b target, in that knockdown of Cdkn1b mimicked miR-196b function by accelerating Mll-Af9 leukemogenesis. Importantly, RNA therapeutic inhibition of miR-196b in Mll-Af9 leukemia cells mediates p27 upregulation concomitant with robust induction of cell death. Thus, the combination of two open-ended approaches, miRNA-target-identification and functional shRNA screen validation, revealed leukemia-context specific miR-196b functions, which are not detected by standard approaches such as target mRNA modulation or prediction algorithms.

Disclosures:

Lieberman:Alnylam Pharmaceuticals: Membership on an entity’s Board of Directors or advisory committees.

Sign in via your Institution