AUTHORS
Reina Takeda, MD, PhD, George Wan2, Hannah E Rice3, Matteo Mazzocca, PhD4, Wallace A. Bourgeois, MD5, Christian Marinaccio, MSc, PhD6, Meaghan Boileau, PhD7, Laura C Godfrey, PhD8, Pablo R Freire, PhD9, Sarah Naomi Olsen, PhD10, Emily B Heikamp, MD, PhD11, Andrei Krivtsov, PhD12, Jared Brown, PhD13, Charlie Hatton14, Anders Sejr Hansen, PhD15, Scott A Armstrong, MD, PhD16
AFFILIATIONS
1 Dana-Farber Cancer Institute, Boston, Massachusetts, United States
2 Dana-Farber Cancer Institute, Boston, Massachusetts, United States
3 Dana-Farber Cancer Institute, Boston, Massachusetts, United States
4 Massachusetts Institute of Technology, Cambridge, Massachusetts, United States
5 Dana-Farber Cancer Institute, Boston, Massachusetts, United States
6 Dana-Farber Cancer Institute, Boston, Massachusetts, United States
7 Dana-Farber Cancer Institute, Boston, Massachusetts, United States
8 Dana-Farber Cancer Institute, Boston, Massachusetts, United States
9 Dana-Farber Cancer Institute, Boston, Massachusetts, United States
10 Dana-Farber Cancer Institute, Boston, Massachusetts, United States
11 Dana-Farber Cancer Institute, Boston, Massachusetts, United States
12 Dana-Farber Cancer Institute, Boston, Massachusetts, United States
13 Dana-Farber Cancer Institute, Boston, Massachusetts, United States
14 Dana-Farber Cancer Institute, Boston, Massachusetts, United States
15 Massachusetts Institute of Technology, Cambridge, Massachusetts, United States
16 Dana-Farber Cancer Institute, Boston, Massachusetts, United States
ABSTRACT
Nucleophosmin1 (NPM1) is one of the most frequently mutated genes in adult acute myeloid leukemia (AML). NPM1-mutant AML is characterized by aberrant expression of HOXA/B cluster genes and MEIS1 and shares transcriptional programs with KMT2A-rearranged (KMT2A-r) AML. Mutant NPM1 (NPM1c) colocalizes with menin-KMT2A at HOX and MEIS1 oncogenes to drive oncogenic transcription, and menin inhibitors, which exploit this chromatin mechanism, are now FDA-approved for the treatment of relapsed/refractory NPM1-mutant AML. Despite this dependency on menin-KMT2A interaction, how KMT2A mechanistically contributes to NPM1c-driven oncogenic transcription remains unknown.
To elucidate the regulatory machinery controlling NPM1c-driven transcription, we generated an endogenous dual fluorescent HOXA9/MEIS1 transcriptional reporter system using CRISPR-Cas9-mediated HDR knock-in genome editing in OCI‑AML3 (NPM1-mutant) and MOLM13 (KMT2A‑r) cells, followed by parallel genome-scale CRISPR-Cas9 screens. Despite convergent upregulation of HOXA9/MEIS1 expression, NPM1-mutant AML and KMT2A-r AML rely on partially overlapping but distinct regulatory circuitry to maintain these critical oncogenic programs.
KMT2A emerged as a top regulatory hit for both HOXA9 and MEIS1 expression in OCI-AML3 cells, along with the driver oncogene, NPM1. In multiple NPM1-mutant AML patient-derived xenograft (PDX) models, genetic depletion of KMT2A reduced both HOXA9 and MEIS1 mRNA levels, demonstrating that KMT2A is required to sustain transcription of critical leukemogenic programs in NPM1-mutant AML. Acute depletion of KMT2A protein using the dTAG-degron system significantly reduced HOXA/B and MEIS1 expression, leading to myeloid differentiation and apoptosis. Integrated RNA-seq and ChIP-seq analyses showed that genes most affected by acute KMT2A loss substantially overlapped with previously defined NPM1c target genes.
Mechanistically, acute KMT2A degradation resulted in eviction of RNA polymerase II (RNA pol II) together with NPM1c, CRM1, and menin-KMT2A complex-associated chromatin regulators from target loci. Notably, we observed a locus-specific reduction in chromatin occupancy of the preinitiation complex, including TBP, in addition to transcriptional elongation factors within 24 hours of KMT2A degradation. Importantly, nuclear protein levels of these factors remained unchanged, indicating that KMT2A orchestrates chromatin recruitment of NPM1c, NPM1c- and KMT2A-associated complexes to NPM1c target loci. These findings identify KMT2A as a transcriptional scaffold that organizes and maintains the active transcriptional machinery at critical oncogenic loci in NPM1-mutant AML.
Next, to determine the immediate transcriptional consequences of KMT2A loss, we conducted nascent RNA transcription analysis with chromatin profiling over a 0 to 180-minute time course following KMT2A protein depletion. Nascent transcription of HOXA9 and MEIS1 was significantly reduced within 15 minutes of KMT2A degradation, coinciding with rapid loss of KMT2A and RNA pol II chromatin occupancy. Although KMT2A possesses a C-terminal SET domain and functions as an H3K4 methyltransferase, we found that H3K4me3 and other active histone modifications, such as H3K27ac, were unchanged at these early time points. Furthermore, catalytic inhibition of the KMT2A SET domain or restoration of H3K4me3 using KDM5 inhibition failed to recapitulate or rescue the drastic transcriptional attenuation caused by KMT2A degradation. Finally, a methyltransferase-deficient KMT2A mutant reduced H3K4me3 levels at specific regions while maintaining target gene expression, whereas acute degradation of this mutant significantly reduced gene expression and phenocopied acute loss of wild-type KMT2A.
Collectively, our findings define a previously unrecognized, methyltransferase-independent function of KMT2A as a transcriptional scaffold that is indispensable for maintaining active RNA pol II transcriptional machinery at NPM1c target loci. Rather than acting primarily through H3K4 methylation, KMT2A has a structural role in organizing RNA pol II and NPM1c-associated regulatory complexes at loci that sustain the leukemogenic transcriptional program. These findings establish the non-catalytic scaffolding function of KMT2A as a therapeutic vulnerability in NPM1-mutant AML and provide a rationale for developing strategies that disrupt KMT2A-dependent transcriptional hub assembly beyond menin-KMT2A interaction blockade.
DISCLOSURES
Reina Takeda, MD, PhD: No relevant conflicts of interest to declare.