Molecular signatures underlying the tumourigenic transformation in a CEBPA mutated model of leukaemia

Abstract

Comprehensive genetic profiling and advances in sequencing studies have enabled the identi- fication of mutations and chromosomal rearrangements frequently found in Acute Myeloid Leukaemia (AML). However, understanding the individual and synergistic impacts of these aberrations on the development of disease is challenging due to the vast clinical and bio- logical heterogeneity of leukaemia. The CEBPA gene is found to be commonly mutated in AML patients. Nevertheless, questions remain in terms of the mechanism of leukaemic transformation and diagnostic implications. Mutations in CEBPA are thought to arise early in tumourigenesis as patients retain the mutations between the original and relapsed tu- mour. Therefore, capturing the effects of CEBPA mutation on the haematopoietic system could provide insight into early cellular and molecular mechanism of leukaemia development. In this work a cellular model of CEBPA mutated leukaemia was generated focusing on a C-terminal N321D mutation of CEBPA that was previously reported to cause AML in mouse models. The mutated version of the gene was introduced into the Hoxb8-FL cell line as cell of origin. The Hoxb8-FL cells are multipotent and can differentiate into dendritic cells with the supplementation of Flt3 ligand (Flt3L). However, Hoxb8-FL cells transduced with the CEBPA N321D construct failed to terminally differentiate upon addition of Flt3L, thereby retaining self-renewal properties. Flow cytometry and transcriptomic profiling showed that the mutant cells exhibited a monocyte progenitor-like phenotype with expression of several plasmacytoid dendritic cell surface markers. Molecular characterisation of different batches of CEBPA N321D transformed cells confirmed that the generation of the CEBPA N321D cell line was reproducible. In vivo transplantation of the CEBPA N321D cell line into lethally irradiated mouse re- cipients verified their leukaemogenic potential. All injected mice developed leukaemia, albeit with a long latency of 1-year post-transplantation. Exome sequencing of bone marrow cells from diseased animals revealed the presence of secondary mutations in the RAS-RTK pathway, including mutations in the Kras, Nras, Ptpn11, Flt3 and Cbl genes. This result demonstrated that in this model, a single C-terminal CEBPA mutation alone is not sufficient to induce disease and requires secondary drivers. Intriguingly, secondary mutations recapitu- lated the most common RAS mutations found in human cancer patients. Therefore, the cells from the developed tumours could provide an attractive source for the testing of potential RAS inhibitors that are currently an unmet clinical target. Finally, this work also presents a comprehensive characterisation of the transcription factor (TF) network in the preleukaemic CEBPA N321D cell line. The network was formed by a CRISPR/Cas9 knock-out of 80 TFs that was followed by a transcriptomic readout using RNA-sequencing analysis. The constructed network consists of 45,553 gene interactions and 10,313 gene targets. Data analysis identified gene modules that divide TFs according to the overlap in their gene expression patterns revealing possible molecular mechanisms underlying the functionalities of the CEBPA N321D cells. Moreover, the analysis uncovered possible vulnerabilities of the preleukaemic cell line that could be explored experimentally. Furthermore, the developed comprehensive network of gene expression in the preleukaemic state can be used as a reference tool in comparison to the leukaemic state to define the key genetic changes that drive the progression to full blown disease. Overall, the work within this thesis used a cellular model of leukaemia to reveal new insights into the molecular mechanisms of CEBPA-mutant driven disease. The model pro- vides characterisation along the different stages of tumourigenic transformation as it captures the cell of origin in the form of the Hoxb8-FL parental cells, the preleukaemic state of the CEBPA N321D cell line and the developed leukaemic state from the diseased bone marrow.