Molecular Characterisation and Therapeutic Targeting of NPM1-mutant Acute Myeloid Leukaemia

Abstract

Mutations in the NPM1 gene are found in 30% of acute myeloid leukaemia (AML) cases. Despite their prevalence, the molecular consequences of NPM1 mutations (NPM1c) are incompletely understood and this hinders progress in designing effective therapeutic strategies. This thesis describes my work to characterise NPM1-mutant cells at the molecular level using different omics approaches with the key goal of identifying those NPM1c-driven changes that can be exploited for translational applications. We first examined the effects of isolated NPM1c mutations on the proteome of preleukaemic haematopoietic progenitors from knock-in Npm1cA/+ mice and discovered that many proteins involved in ribosome biogenesis were significantly depleted. Importantly, many ribosome biogenesis factors were also depleted in human NPM1- mutant AMLs. In line with this, pre-leukaemic Npm1cA/+ progenitors displayed increased sensitivity to RNA polymerase I inhibitors, such as Actinomycin D (ActD). The combination of ActD and Venetoclax inhibited the growth and colony forming ability of pre-leukaemic and leukaemic NPM1c+ cells and low-dose ActD treatment was able to re-sensitise Venetoclax-resistant NPM1c+ models. Next, using data from clustered regularly interspaced palindromic repeats (CRISPR) -Cas9 drop-out screens, we identified and validated TSR3, a 40S ribosome maturation factor whose knock-out preferentially inhibited the proliferation of NPM1c+ AML cells by activating a p53-dependent apoptotic response. Similarly to low-dose ActD treatment, TSR3 depletion could partially restore Venetoclax sensitivity to previously Venetoclax-resistant NPM1c+ AML models. To explore the protein interactors of NPM1c, we designed and initiated the development of a proximity-labelling platform in NPM1-mutant AML cells. However, a series of functional assays revealed that lentivirally-delivered NPM1c fusion proteins do not mimic the behaviour of the native protein; results with potential implications for future studies interrogating the NPM1c function. Finally, to unmask the impact of NPM1c mutations at the single-cell level, we performed Cellular Indexing of Transcriptomes and Epitopes by Sequencing (CITE-Seq) in pre-leukaemic Npm1cA/+ haematopoietic progenitors. We found only minor changes in cell type abundance between Npm1cA/+ and Npm1+/+ controls, but widespread transcriptional changes across most cell clusters, including an upregulation of several mitochondria-related genes in Npm1c+ early progenitors. Reflecting this, pre-leukaemic Npm1cA/+ progenitors displayed increased basal and maximal mitochondrial respiration and enhanced sensitivity to mitochondrial complex I inhibitors. Furthermore, inhibition of the mitochondrial polymerase POLRMT preferentially reduced the proliferation of NPM1c+ AML cells and synergised with Venetoclax to enhance cytotoxicity. Overall, our work uncovers novel molecular characteristics and phenotypic traits induced by NPM1c mutations in haematopoietic cells and highlights new potential therapeutic avenues for future investigation.