MYC Influences mRNA Processing and Nuclear Export to Initiate Acute Myeloid Leukemia

Background: Although MYC is a potent oncogene that is frequently associated with poor outcomes in many solid tumors, its precise role in Acute Myeloid Leukemia (AML) pathogenesis and outcomes are not clear. We recently discovered that intermediate-risk AML patients with long first remissions with standard chemotherapy have an unexpectedly high frequency of pathogenic MYC mutations in the dominant leukemia clone at diagnosis (6 of 26 [23%] cases, versus 1.8% of all AML patients, p=0.002). An analysis of all MYC mutations reported in AML, revealed that they are nearly always single base substitutions that localize to a hotspot in the MYC Box I domain (MBI - AAs hotspot 73-74); this stands in distinct contrast to other hematopoietic malignancies, where MYC mutations are scattered throughout the entire gene.

In this study, we engineered mouse hematopoietic stem and progenitor cells (HSPCs) to express AML-associated MYC missense mutations and characterized their in-vitro and in-vivo cellular and molecular phenotypes.

Results: We observed that constitutive expression of MYC MBI mutant proteins in mouse HSPCs significantly prolonged Myc half-life and conferred different degrees of growth advantage over both wild type MYC overexpression and overexpression of non-MBI mutant proteins.

To determine whether these mutations are AML drivers, HSPCs expressing MYC^WT or the MBI mutants MYC^P74Q and MYC^T73N were transplanted into lethally irradiated syngeneic recipient mice, which were then evaluated for the development of leukemia. Primary recipient mice transplanted with MYC^WT-expressing cells developed leukemia with a median survival of 58 days for whereas mice transplanted with MYC^P74Q- or MYC^T73N-expressing cells displayed a median survival of 42 and 41.5 days, respectively (p<0.001). Furthermore, secondary recipient mice transplanted with MYC^WT-, MYC^P74Q- or MYC^T73N-expressing AML cells from primary recipient mice also developed AML, indicating that expression of WT and MBI-MUT Myc leads to the development of frank leukemia in mice.

Next, we investigated the impact of these point mutations on the transcriptional targets of Myc by comparing single-cell RNA-seq profiles of murine HSPCs expressing wild type, P74Q, or T73N Myc versus an empty vector control. From these analyses, we found that while each of these mutants retain many of the transcriptional activities of wild type Myc, they also display unique transcriptional signatures.

To better understand how these mutations affect the transcriptional output of MYC, we performed ChIP-seq analysis of primary murine HSPCs expressing MYC^WT-, MYCc^P74Q- or MYC^T73N. Despite MBI mutations causing persistent MYC protein expression (due to prolonged half-life), we detected an overall decrease in chromatin binding of MBI-MUT MYC, with preserved binding to transcriptional start sites of a core sets of genes involved in splicing and translation. Importantly, MBI mutations did not affect the canonical E-box Myc binding site, nor create novel binding sites, indicating that these mutations lead to an overall decrease in Myc occupancy at target promoters.

Immunofluorescent staining for WT vs. MBI mutant MYC showed that despite their overall decrease in chromatin binding, Myc mutant proteins localized predominantly to the nucleus. However, a substantial number of nuclei showed a ring-like morphology of MBI-MUT MYC, suggesting that MBI mutations might enrich protein localization at the nuclear periphery. To determine whether the changes in staining pattern observed for MBI-MUT were associated with alterations in the ability of MYC to interact with its regulatory proteins, we employed a modified BioID proximity labeling system, "TurboID", to screen for differential protein interactions in murine HSPCs. This analysis showed that MBI mutations resulted in several gained interactions over WT MYC - most prominently with components of the nuclear pore complex (NUP98, NUP107, NUP133, XPOT, RAE1) supporting the idea that MBI-MUT might accelerate transformation by controlling translation and mRNA nuclear export.

Conclusions: Collectively, these data indicate that AML-associated MYC mutations support leukemogenesis through non-canonical (transcription-independent) MYC proprieties. More broadly, these data suggest that aberrant mRNA processing and nuclear export might play an important role in the initiation of AML by MYC mutations.

No relevant conflicts of interest to declare.