Dnmt3a DELETION COOPERATES WITH THE Flt3-ITD MUTATION TO DRIVE LEUKEMOGENESIS IN A MURINE MODEL
Embargo until
Date
2016-03-31
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Johns Hopkins University
Abstract
The advent of next generation sequencing has facilitated the establishment of an exhaustive catalog of recurrent mutations in Acute Myeloid Leukemia (AML). We know that FLT3 is the most commonly mutated gene in the disease, it is activated most commonly through internal tandem duplication (ITD) within the juxtamembrane domain, and at a lesser frequency via point mutations in the kinase domain. Flt3ITD/+ knock-in mice develop a fatal myeloproliferative neoplasm (MPN) but fail to fully transform, indicating that additional mutations are necessary for leukemogenesis. Genetically engineered mouse strains provide a powerful platform to investigate what these cooperating mutations might be, and identify signaling differences in these mice.
While ITD and mutations within the kinase domain (most commonly D835Y) both constitutively activate the kinase, these mutations confer distinct prognoses and signaling differences according to survival analyses and in vitro studies, respectively. In addition to our Flt3ITD/+ model, we generated a Flt3D835Y/+ knock-in and observed several differences between the two mutant strains, including differences in the stem cell compartments and disease spectra. As in human subjects, Flt3D835Y/+ conferred a more indolent disease with longer median survival than Flt3ITD/+. Additionally, in agreement with previous in vitro studies Stat5 was preferentially phosphorylated in progenitor cells from Flt3ITD/+ bone marrow compared with Flt3D835Y/+. To further characterize signaling difference between these genotypes, we performed expression arrays using progenitor cells from each genotype, and identified a number of differentially expressed genes, and dysregulated pathways. A number of these candidate genes were validated by qPCR.
As previously stated, studies in Flt3ITD/+ mice suggest that additional mutations are necessary for transformation. Pairwise comparisons of large datasets have identified a number of potentially cooperating mutations, including DNMT3A and FLT3-ITD, which co-occur in a significant proportion of patients, portending a poor prognosis. We examined the potential cooperativity by breeding a substrain of our Flt3ITD/+ mice with a conditional knock-out of Dnmt3a and find that the two mutations do, indeed cooperate to drive leukemia development, expansion of multiple progenitor pools, and enhanced self-renewal. Interestingly, we found that Dnmt3a dosage significantly affected a number of these parameters, indicating the importance of Dnmt3a stoichiometry in hematopoiesis, and transformation in the context of Flt3ITD/+.
Among DNMT3Amut;FLT3-ITD patients, it was recently discovered that a significant subset of these patients also harbor a mutant NPM1 allele (NPM1c+). We hypothesized that the addition of a mutant Nucleophosmin to our Flt3ITD/+;Dnmt3af/f model might lead to a more aggressive and uniform disease, since patients with all three mutations cluster together based on a number of other molecular parameters. While data is very preliminary, we find that the additional of NPMc+ doesn’t shorten survival, but increases disease aggressiveness, with a larger percentage of blasts in the bone marrow at the time of sacrifice.
This work underscores the power and utility in discerning functionally characterizing mutations relevant to human disease. Using isogenic mouse strains, we have begun to elucidate signaling differences underlying various activating FLT3 mutations, demonstrated for the first time that mutant DNMT3A and FLT3 cooperate to drive leukemia development, and the addition of mutant NPMc+ enhances disease severity. Taken together, these mice can be used a powerful tool to discover underlying disease mechanisms, and a platform for transplantation studies to test novel therapeutics.
Description
Keywords
leukemia, genetics