MLL-AF6 fusion oncogene sequesters AF6 into the nucleus to trigger RAS activation in myeloid leukemia.

A rare location, t(6;11)(q27;q23) (MLL-AF6), is associated with poor outcome in childhood acute myeloid leukemia (AML). The described mechanism by which MLL-AF6, through constitutive self-association and in cooperation with DOT-1L, activates aberrant gene expression does not explain the biological differences existing between t(6;11)-rearranged and other MLL-positive patients nor their different clinical outcome. Here, we show that AF6 is expressed in the cytoplasm of healthy bone marrow cells and controls rat sarcoma viral oncogene (RAS)-guanosine triphosphate (GTP) levels. By contrast, in MLL-AF6-rearranged cells, AF6 is found localized in the nucleus, leading to aberrant activation of RAS and of its downstream targets. Silencing MLL-AF6, we restored AF6 localization in the cytoplasm, thus mediating significant reduction of RAS-GTP levels and of cell clonogenic potential. The rescue of RAS-GTP levels after MLL-AF6 and AF6 co-silencing confirmed that MLL-AF6 oncoprotein potentiates the activity of the RAS pathway through retention of AF6 within the nucleus. Exposure of MLL-AF6-rearranged AML blasts to tipifarnib, a RAS inhibitor, leads to cell autophagy and apoptosis, thus supporting RAS targeting as a novel potential therapeutic strategy in patients carrying t(6;11). Altogether, these data point to a novel role of the MLL-AF6 chimera and show that its gene partner, AF6, is crucial in AML development.