Acute myeloid leukemia (AML) is characterized by malignant proliferation of myeloid progenitors in peripheral blood and bone marrow and is the most common and aggressive among acute leukemias. The aim of the project was to uncover novel epigenetic vulnerabilities in AML by means of loss-of-function forward genetics screens and to investigate the molecular mechanisms behind them. For this, I established a recently reported CRISPR interference (CRISPRi) technology for targeted gene repression in the lab and initially focused on improving its efficiency. By systematic profiling of sgRNA effectiveness, I identified several critical parameters of CRISPRi activity (i.e. position of sgRNAs relative to the transcription start site of a target gene and sgRNA sequence). I then devised a set of rules for efficient CRISPRi sgRNA design and used them to generate a pooled CRISPRi library targeting 1046 genes, all the known and putative chromatin-associated proteins. With this library, I performed screens in two MLL-AF9 rearranged human AML cell lines (THP1 and MOLM13), which uncovered several potential regulators of leukemogenesis: PRMT5 and KANSL2. I confirmed the essentiality of these three factors in several human MLL-rearranged cell lines using both CRISPR interference and CRISPR knockout approaches and performed rescue experiments with sgRNA-resistant cDNAs to demonstrate specificity of the observed phenotypes. PRMT5 is an arginine methyltransferase with known oncogenic roles in many cancers. Using a novel proteomics approach from Cell Signaling Technology, I identified the proteins that lose arginine methylation in response to PRMT5 knockdown and, thus, represent potential PRMT5 substrates. My current work is focused on investigating which of these substrates are essential for AML propagation and could be used as targets for AML therapy. KANSL2 is a component of the NSL histone acetyltransferase complex. Using CRIPSR knockout approach, I discovered that most of the subunits of this complex are essential for the proliferation of leukemic cells. By global proteome profiling and targeted mutagenesis, I found that KANSL2 represents a key subunit for structural integrity of the complex, and its depletion leads to downregulation of various transcriptional regulators. Using a combination of ChIP-sequencing and RNA-sequencing, I identified the essential target genes of the complex. Current research is focused on understanding the molecular mechanism of target gene regulation by this complex.
Another direction of research was to establish mouse models of normal karyotype AML and perform CRISPR screens in those. This part of the project was performed in collaboration with other researchers in the lab. We have combined the most frequently mutated genes in AML, i.e. Npm1c, Flt3-ITD and either Tet2 or Dnmt3a knockouts. This resulted in aggressive leukemia in mice, which is transplantable and can be maintained in vitro. We have performed CRISPR knockout screens in these models and are currently validating the identified hits in vivo.
