Search

Most patients with acute promyelocytic leukemia (APL) can be cured with combined all-trans retinoic acid (ATRA) and arsenic trioxide therapy, which induces the destruction of PML-RARA, the initiating fusion protein for this disease. However, the underlying mechanisms by which PML-RARAinitiates and maintains APL cells are still not clear. Therefore, we identified genes that are dysregulated by PML-RARAin mouse and human APL cells and prioritized GATA2for functional studies because it is highly expressed in preleukemic cells expressing PML-RARA, its high expression persists in transformed APL cells, and spontaneous somatic mutations of GATA2occur during APL progression in mice and humans. These and other findings suggested that GATA2may be upregulated to thwart the proliferative signal generated by PML-RARAand that its inactivation by mutation (and/or epigenetic silencing) may accelerate disease progression in APL and other forms of acute myeloid leukemia (AML). Indeed, biallelic knockout of Gata2with CRISPR/Cas9-mediated gene editing increased the serial replating efficiency of PML-RARA–expressing myeloid progenitors (as well as progenitors expressing RUNX1-RUNX1T1, or deficient for Cebpa), increased mouse APL penetrance, and decreased latency. Restoration of Gata2expression suppressed PML-RARA–driven aberrant self-renewal and leukemogenesis. Conversely, addback of a mutant GATA2R362Gprotein associated with APL and AML minimally suppressed PML-RARA–induced aberrant self-renewal, suggesting that it is a loss-of-function mutation. These studies reveal a potential role for Gata2as a tumor suppressor in AML and suggest that restoration of its function (when inactivated) may provide benefit for AML patients.

NPM1and DNMT3Aare among the most commonly mutated genes in Acute Myeloid Leukemia (AML), affecting 25-30% of patients. Importantly, these mutations co-occur much more frequently than expected by chance, with ~15% of AML samples containing mutations in both genes.This synergy is recapitulated in mouse models, where mice with either Dnmt3aR878Hor Npm1cAmutations develop leukemia with a long latency and low penetrance, whereas mice with both Dnmt3aR878Hand Npm1cAmutations develop leukemia in 8-12 months with nearly 100% penetrance (Dovey et al., Blood, 2017; Loberg et al., Leukemia, 2019).To better understand the mechanisms underlying this synergy, and the process of leukemic transformation, we have characterized pre-leukemic bone marrow and 9 independent spontaneous AMLs that have arisen in these mice. These spontaneous AMLs rapidly cause fatal leukemias in secondary recipients; whole genome sequencing of all tumors revealed one or more human AML-like cooperating mutations in each, including canonical mutations in Ptpn11(E69K, A72V and S506L), an activating KitD818Ymutation, a Cblexon 9 deletion, a CblY369Hmutation,and an amplification of Flt3associated with >20 fold overexpression of WT Flt3. Characterization of the pre-leukemic transcriptomes and methylomes demonstrated little change in the pre-leukemic state for several months, followed by dramatic and canonical shifts in transcription and methylation after transformation. Remarkably, 8 of the 9 AMLs contained an amplification of the entirety of murine chromosome 7 as the sole structural variant in each genome. We analyzed panel sequencing data from a set of published AMLs arising in this model (Loberg et al., Leukemia, 2019), and also exome sequencing data from a different Npm1 cAdriven tumor model (Dovey et al., Blood, 2017); these AMLs also developed complete or partial amplification of chromosome 7 in 7 of 8 and 10 of 15 cases, respectively. Through integration of these datasets, we identified a minimally amplified 8.9 Mbp region on chromosome 7, containing 209 genes, including two candidate genes ( Gab2, Pak1) that show overexpression in both murine and human AMLs with DNMT3Aand NPM1mutations. Retroviral constructs expressing each cDNA (with an IRES-GFP tag) were transduced into lineage depleted bone marrow cells from preleukemic WT, Dnmt3aR878H, Npm1cAor Dnmt3aR878Hx Npm1cAmice, and transplanted into recipient mice. Overexpression of GAB2(but not “Empty Vector” or PAK1) induced significant expansion of hematopoietic cells from the Dnmt3aR878Hx Npm1cAmice (and to a lesser extent, from Npm1cAmice),but not Dnmt3aR878Hmice or WT mice ( Figure 1A), suggesting that the mutations in Npm1and Dnmt3asynergize to account for this phenotype. Strikingly, overexpression of GAB2in Dnmt3aR878Hx Npm1cAbone marrow led to development of AML in 5 of 5 engrafted mice in less than 5 months (p<10 -6compared to untransduced marrow; Figure 1B). 0 of 3 engrafted mice with “Empty Vector” (GFP only) in Dnmt3aR878Hx Npm1cAbone marrow and 0 of 3 engrafted mice with overexpression of PAK1in Dnmt3aR878Hx Npm1cAbone marrow developed leukemia during this 5 month period. Taken together, these data suggest that AML-associated mutations in DNMT3Aand NPM1allow for the selective expansion of cells overexpressing GAB2as an important step towards leukemic transformation. In human myeloid neoplasms, amplification of chromosome 11q (containing human GAB2) is a recurrent cytogenetic event, observed in 1.5% of cases (Papaemmanuil, NEJM, 2016), and expression of GAB2across AML samples containing NPM1mutations is increased ~50% compared to healthy CD34+ cells or promyelocytes. GAB2 is known to bind to GRB2 and SHP2/PTPN11 to facilitate signaling from receptor tyrosine kinases (including FLT3) to downstream pathways (including RAS/RAF/MEK/ERK, PI3K and AKT, in some contexts). These data nominate GAB2overexpression as a factor that may be relevant for the progression of founding clones with DNMT3Aand NPM1mutations to overt AML. Additional studies to define its protein interactome in preleukemic cells, its mechanism of action, and how leukemia-causing mutations in DNMT3Aand NPM1shape the fitness landscape to select for GAB2overexpression are underway.

Heterozygous loss-of-function mutations in the DNMT3Agene are the most common cause of clonal hematopoiesis, and among the most common initiating events for Acute Myeloid Leukemia (AML). A reduction of DNMT3A activity causes a canonical, focal hypomethylation phenotype in hematopoietic cells, which is associated with immortalization of hematopoietic stem cells, and blockage of myeloid differentiation. The methylation defect can be partially reversed with restoration of physiologic levels of DNMT3Aexpression over a period of several months (Ketkar et al., PNAS, 2020; PMID 35313694).To identify a more efficient remethylation strategy, we first characterized the DNA methylation phenotypes of bone marrow cells from mice with hematopoietic cell deficiency of Dnmt3a, Dnmt3b(or both enzymes), or expressing the dominant negative Dnmt3aR878Hmutation (equivalent to R882H in humans; the most common mutation found in AML patients). Using these different mouse models as substrates, we then defined the patterns and completeness of DNA remethylation after “adding back” supraphysiologic levels of wild type DNMT3A1, DNMT3B1, DNMT3B3 (an inactive splice isoform of DNMT3B), or DNMT3L (a catalytically inactive “chaperone” that augments the activity of DNMT3A and DNMT3B in early embryogenesis), with MSCV-based retroviruses transduced into primary mouse hematopoietic stem/progenitor cell s. Overexpression of WT DNMT3A for 2 weeks in vitro(~3 fold) can accurately reverse the differentially methylated regions (DMRs, which are >99% hypomethylated) of Dnmt3adeficient hematopoietic cells, or cells expressing the R878H mutation. The DMRs of Dnmt3bdeficient mouse bone marrow cells can be corrected by overexpression of DNMT3A, DNMT3B1, or DNMT3B3; however, DNMT3B3 failed to remethylate the DMRs in Dnmt3a/Dnmt3bdouble knockout mouse bone marrow cells. Since DNMT3B3 is an inactive enzyme that does not cause remethylation in the absence of DNMT3A, these data suggest that DNMT3B3 functions in a DNMT3A-dependent manner in hematopoietic cells. Importantly, both DNMT3B3 and DNMT3L have been shown to facilitate DNA methylation by acting as a chaperone for DNMT3A (Duymich et al., Nat. Communication, 2016; PMID 27121154). In vitro, DNMT3L copurified with DNMT3A leads to an increase in DNMT3A methyltransferase activity that is ~5 times greater than DNMT3B3 copurified with DNMT3A. Remarkably, overexpression of DNMT3L (which is not expressed in adult hematopoietic cells, or AML cells) can completely correct the hypomethylation phenotype of Dnmt3aR878H/+bone marrow cells within 2 weeks of in vitrooverexpression, or 4 weeks of in vivooverexpression ( Figure 1A and 1B), probably by augmenting the activity of WT DNMT3A encoded by the residual WT allele in these heterozygous mutant cells. Two weeks of overexpression of DNMT3A or DNMT3L in Dnmt3aR878H/+bone marrow cells (cultured in vitrowith IL-3, SCF, FLT3L, and TPO to maintain HSPC populations) can induce a differentiation response, increasing the fraction of mature myeloid cells in the cultures by ~4-fold. This finding was recapitulated in vivowhere one month of overexpression of DNMT3Aor DNMT3Lin transplanted Dnmt3aR878H/+mouse bone marrow cells increased the fraction of mature myeloid cells by 2-fold by DNMT3L and 4-fold by DNMT3A. Together,these data show that the focal, canonical DNA hypomethylation phenotype of Dnmt3a R878H/+hematopoietic cells can be accurately and efficiently corrected by overexpressing WT DNMT3A or DNMT3L for several weeks. We have also shown that the restoration of methylation in Dnmt3a R878H/+mutant cells alters the developmental fate of progenitors, increasing the proportion of mature myeloid cells. These data suggest that DNMT3Lexpression may represent a novel approach for restoring DNMT3A activity in AMLs initiated by DNMT3Amutations, and could possibly alter AML cell fate. Additional studies are underway to determine whether the correction of DNMT3A activity in Dnmt3aR878H/+-initiated AML cells with retroviral addback of DNMT3A or DNMT3L will cause a differentiation response that alters AML cell fate.

Self-renewal is a key feature of the cells that initiate acute myeloid leukemia. To identify the mechanisms involved in PML-RARA (PR)-driven self-renewal, we made use of Ctsg-PRmice, which have PRknocked into the UTR of the first exon of Ctsg. Ctsg-PRdrives the expression of PRin myeloid progenitor cells and gives these cells the ability to serially replate in methylcellulose-based colony assays. Most Ctsg-PRmice develop acute promyelocytic leukemia (APL) with an average latency of ~300 days. To identify target genes regulated by Ctsg-PR, we performed single cell RNA-seq (scRNA-seq) on whole bone marrow from young, preleukemic Ctsg-PRmice or age-matched littermates. We identified 959 differentially expressed genes (DEGs) within myeloid progenitors (546 upregulated by PR,and 413 downregulated). Gata2was identified as a DEG in this analysis, and we confirmed this phenotype with bulk RNA-seq of purified promyelocytes from young, preleukemic WT vs. Ctsg-PRmice. All APLs derived from Ctsg-PRmice also expressed high levels of Gata2.