CARM1/PRMT4 As a Novel Therapeutic Target for AML

Chromatin modifying enzymes, and specifically the protein arginine methyltransferases (PRMTs) have emerged as important targets in cancer. PRMT4, also known as CARM1, is overexpressed in a number of cancers, including breast, prostate, pancreatic, and lung cancer. We have reported the overexpression of PRMT4 in the context of acute myeloid leukemia (AML), showing that more than 70% of cytogenetically normal AML patients have up-regulation of PRMT4. Here, we investigated the role of PRMT4 in normal hematopoiesis and leukemia development.

In order to study the role of PRMT4 in normal hematopoiesis, Prmt4 -floxed mice were crossed with Vav1-cre mice purchased from the Jackson Laboratory. Inducible Prmt4 conditional KO mice were generated by crossing Prmt4 -floxed mice with Mx1-Cre mice and Prmt4 gene excision was induced by poly(I:C), given by three i.p. injections at a dose of 10 mg/kg. Using this hematopoietic specific knockout system, we show that loss of PRMT4 has little effect on normal hematopoiesis, but promotes the differentiation of hematopoietic stem and progenitor cells. Next we evaluated the role of PRMT4 in leukemia initiation and maintenance using leukemia transplantation models driven by fusion oncoproteins. Strikingly, the knockout of PRMT4 completely abrogates leukemia initiation and maintenance in fetal liver transplantation models driven by the AML1-ETO or MLL-AF9 fusion proteins.

We further characterized the mechanism for the leukemia-specific dependence on PRMT4 using leukemia cell lines lentiviral vectors that express shRNAs directed against PRMT4. We knocked down PRMT4 in four leukemia cell lines or normal human cord-blood derived CD34⁺ cells and found that knockdown of PRMT4 impairs cell cycle progression, induces apoptosis, and down-regulates E2F target genes, identifying several mechanisms for the leukemia-specific dependence on PRMT4. Gene set enrichment analysis showed that all four leukemia cell lines with knockdown of PRMT4 showed a significant down-regulation of MYC and E2F target genes compared to the scrambled control. Differentially regulated pathways included Cell cycle, DNA replication, and Homologous recombination. Finally, we utilized inhibitors of the Type I arginine methyltransferase family in order to validate the selective inhibition of PRMT4 in leukemia cells and patient cell cells.

The PRMT4 conditional knockout mice, PRMT4 knockdown in leukemia cell lines, and chemical inhibition of PRMT4, all suggest that loss of PRMT4 protein levels or activity has a selective effect on leukemia cells compared to normal hematopoietic stem and progenitor cells. Collectively this work suggests that targeting PRMT4 activity may be an effective therapeutic strategy for AML, with minimal effects on normal hematopoiesis.