Dnmt3a Haploisufficiency Cooperates with Oncogenic Kras to Promote an Early-Onset T-Cell Leukemia

Recent whole genome/exome sequencing efforts in myeloid malignancies identified that mutations in DNA methyltransferase 3A (DNMT3A) are prevalent in acute myeloid leukemia (AML). In addition, DNMT3A mutations are also identified in various T cell malignancies. Of note, DNMT3A mutations are typically heterozygous and some WT DNMT3A functions thus remain in this state. However, the predominant DNMT3A R882 mutations, which locate in the catalytic domain, seem to inhibit the methyltransferase activity of the remaining WT DNMT3A due to its dominant-negative function (Yang L, Rau R, Goodell MA, Nat. Rev. Cancer 15: 152-165, 2015). COSMIC database analysis reveals different prevalence of DNMT3A R882 mutations in various hematopoietic malignancies. Approximately 60% of DNMT3A mutations in AML are R882 mutations, while the frequency of R882 mutations drops to ~40% in myelodysplastic syndrome (MDS) and myeloproliferative neoplasm (MPN). In contrast, the frequency of R882mutations is less than 25% in T-cell acute lymphoblastic leukemia (T-ALL). The significantly different frequencies of DNMT3A R882 mutations in AML versus T-ALL inspired us to investigate whether downregulation of DNMT3A regulates malignancies of different lineages in a dose-dependent manner.

We previously showed that Dnmt3a^-/- promotes MPN progression in Kras^G12D/+ mice and ~1/3 compound mice develop AML-like disease (Chang et al. Leukemia 29: 1847-1856, 2015). Here, we generated Kras^G12D/+; Dnmt3a^fl/+; Mx1-Cre mice to determine how Dnmt3a haploisufficiency affects Kras^G12D/+-induced leukemogenesis. After pI-pC injections to induce Mx1-Cre expression, primary Kras^G12D/+; Dnmt3a^+/- mice died quickly as primary Kras^G12D/+ mice; the survival rates of these two groups of animals were not significantly different. However, in a competitive transplant setting, recipients transplanted with Kras^G12D/+; Dnmt3a^+/- bone marrow cells displayed a significantly shortened survival than recipients with Kras^G12D/+ cells. Moreover, all of the recipients with Kras^G12D/+; Dnmt3a^+/- cells developed a lethal T-ALL without significant MPN phenotypes, while ~20% of recipients with Kras^G12D/+ cells developed MPN with or without T-ALL. This is in sharp contrast to the recipients with Kras^G12D/+; Dnmt3a^-/- cells, in which ~60% developed a lethal myeloid malignancy (MPN or AML). Our data suggest that in the context of oncogenic Kras, loss of Dnmt3a promotes myeloid malignancies, while Dnmt3a haploisufficiency induces T-ALL. This dose-dependent phenotype is highly consistent with the prevalence of DNMT3A R882 mutations in AML versus T-ALL in human. We are currently investigating the underlying mechanisms.