P466: the intrinsically disordered transcriptional regulator mn1 depends on the chaperone dnajb6b for its full transforming potential

Background: Overexpression of the Meningioma-1 (MN1) gene drives leukemia development in mice and is also a poor prognostic marker in human acute myeloid leukemia (AML). Expression of MN1 blocks myeloid differentiation and promotes AML development through several mechanisms, usually involving BAF complex stabilization and strong dependency on the MEIS1/AbdB-like HOX protein complex. We demonstrated in a genetic homogeneous subtype of AML that lentiviral overexpression (OE) of MN1 in primary human APL cells (n=5 patient samples) allowed in vivo engraftment using xenograft mouse models with infiltration of secondary organs, which was not observed in controls. Flow cytometry analysis revealed that MN1-engrafted mice displayed increased levels of APL blasts in comparison to controls, which instead had a large proportion of mature myeloid cells. These results reinforced the idea that MN1 drives leukemia transformation through blocking myeloid differentiation. Aims: Using a large cohort of AML samples (n=625), we found that MN1 is highly expressed in more immature AML subtypes such as patients carrying mutations in TP53 and RUNX1, whereas low MN1 expression was generally observed in more progenitor-like AMLs, such as FLT3-ITD and NPM1 mutant AML. In AML models where MN1 is lowly expressed, such as the FLT3-ITD+ AML, lentiviral MN1-OE resulted in reduced cell survival. These intriguing observations clearly indicate that MN1 has opposing roles in different AML subtypes. MN1 is an intrinsically disordered protein containing a long polyQ-stretch that is required for transformation. We questioned whether the expression of chaperone proteins would be necessary for the appropriate functioning of MN1. Methods: We performed label-free quantitative proteome analysis on sorted CD34+ AML cells (n=42) and identified the DNAJB6 isoform 2 (DNAJB6b) as one of the proteins differentially expressed in patients with high versus low MN1 protein levels. We validated our observations in a cohort of adult AML samples (n=173), where DNAJB6b expression was correlated with MN1 expression (r2=0.49). Furthermore, DNAJB6b expression was highest in immature AML subtypes while expression was low in FLT3-ITD+ AML. Since the DNAJB6b expression was correlated with MN1 levels, we therefore hypothesized that the removal of DNAJB6b in these cells would result in loss of oncogenic functions of MN1. Results: shRNA-induced knockdown (KD) of DNAJB6b in several AML cell models resulted in cytosolic retention of MN1, associated with reduced cell survival and increased drug-induced myeloid differentiation. Chromatin immunoprecipitation analysis showed loss of H3K4me3 and MN1 binding at the MEIS1 locus upon KD of DNAJB6b. Additionally, overexpression of DNAJB6b in FLT3-ITD+ cells rescued the effects on cell viability and survival that were observed in MN1-OE cells. Finally, differential gene expression analysis of AML patients with high versus low DNAJB6b showed increased MEIS1 expression in DNAJB6bhigh patients, which displayed transcriptome associated with the terms “LSC_UP” and “HSC_UP” while patients with DNAJB6blow was correlated with “CELL_CYCLE_CONTROL” and “APOPTOSIS_BY_CDKN1A_VIA_TP53”. Summary/Conclusion: Together, our data demonstrate that MN1 can only act as oncogene in cells that express sufficient levels of DNAJB6b. Downregulation of DNAJB6b results in cytoplasmic retention of MN1 and consequently loss of its transcriptional control, and MN1-driven transformation. Our findings suggest that targeting chaperones such as DNAJB6b can be leveraged as a therapeutically actionable mechanism for driving cell death in MN1high AMLs. Keywords: Proteomics, Xenotransplantation, Myeloid differentiation, Acute myeloid leukemia