Targeting BRD9 in KMT2A-rearranged Acute Myeloid Leukemia to circumvent MEN1 inhibitor resistance

Abstract

Abstract Introduction: Acute myeloid leukemia (AML) with rearrangement of the KMT2A gene (KMT2Ar) is a blood malignancy with adverse clinical outcome. Many of these patients develop resistance to chemotherapies and are not eligible for stem cell transplantation. Protein interaction between KMT2A and its cofactor MEN1 activates a key oncogenic pathway in these leukemias. Hence, MEN1 inhibitors (MEN1i) disrupting this interaction have been designed to treat the KMT2Ar patients with relapsed and refractory disease. However, ~40% of the patients receiving MEN1i treatment develop drug resistance due to the acquisition of sporadic MEN1 mutations. These patients urgently need effective therapies to reduce the morbidity and mortality. In addition to the KMT2A/MEN1 pathway, AML cells with KMT2Ar are maintained by a small set of transcription factors (TFs) including IRF8, MEF2D, and MYC, which is defined as the TF addiction pathway. BRD9 is a subunit of the BRG1/BRM-associated factor (BAF) chromatin remodeling complex that controls chromatin accessibility of regulatory elements such as enhancers. In this study, we investigated the functions of BRD9 in the TF addiction pathway. Moreover, we hypothesize that the KMT2A/MEN1 pathway and the TF addiction pathway are regulatory independent. Therefore, we also explored the therapeutic potential of anti-BRD9 therapy in KMT2Ar cell lines that are resistant to MEN1i. Methods: A BRD9 degrader (dBRD9A) and a MEN1i (revumenib) were used to deplete BRD9 and inhibit MEN1 in AML cells, respectively. We applied MTT and CellTiter-Glo assays to measure cell viability in AML cell lines and primary AML samples, respectively, and used flow cytometry to quantify CD11b expression as an indicator of myeloid differentiation. ATAC-seq and RNA-seq were employed to record chromatin accessibility and transcriptomic changes of MOLM-13 cell line and primary AML samples before and after drug treatment(s). To reconstitute the expression of IRF8 and MYC, we lenti-virally transduced the two genes into MOLM-13 cell line, in which the expression of the genes was induced by doxycycline treatment. Results: The Depmap CRISPR screen data revealed that BRD9 is essential in myeloid cell lines carrying KMT2Ar or KMT2A mutations. In line with the data, chemical depletion of BRD9 in 11 myeloid cell lines confirmed that BRD9 loss induces growth defect and myeloid differentiation of the KMT2Ar and KMT2A mutated cell lines. Subsequently, transcriptomic analysis revealed rapid downregulation of IRF8 and MYC after 3 h BRD9 depletion in MOLM-13 cell line (KMT2Ar). Concomitantly, we observed reduced chromatin accessibility of IRF8 and MYC super enhancers. Motif analysis identified the enrichment of MYB consensus binding sequences in these super enhancers. Importantly, reconstitution of IRF8 or MYC expression reduced myeloid differentiation caused by BRD9 depletion. Next, we studied the effects of BRD9 depletion in 14 primary AML samples carrying KMT2Ar. Based on the expression level of IRF8, the transcriptomic profiles of these samples are clustered into two groups. The samples with high IRF8 expression are significantly more sensitive to BRD9 depletion than the samples with low IRF8 expression in CellTiter-Glo assay. CD34+ hematopoietic progenitors isolated from 3 healthy donors are insensitive to BRD9 loss. Moreover, the expression of the TF addiction genes (IRF8, MYC and MEF2D) is reduced in the IRF8-high but not IRF8-low AML samples after BRD9 degradation. Furthermore, we compared the transcriptomic effects of BRD9 depletion and MEN1 inhibition in MOLM-13 cell line. As expected, BRD9 and MEN1 independently regulate the downstream genes of the TF addiction pathway and KMT2A/MEN1 pathway, respectively. We further tested viability of the wild-type and the MEN1-mutated MV4-11 cell lines (KMT2Ar) before and after drug treatment. BRD9 depletion reduced the viability of both cell lines, while MEN1 inhibition only affected the growth of the wild-type cell line. Conclusion: Our data support that BRD9 cooperates with MYB to activate the TF addiction genes and sustain the growth of AML cells carrying KMT2Ar. Translationally, the TF addiction pathway is regulatory independent from the KMT2A/MEN1 pathway. Anti-BRD9 agents remain effective in AML cells harboring the MEN1 mutations. After further development, these novel drugs may be used to treat KMT2Ar patients who are resistant to MEN1i.