Identifying transcription factor dysregulation in RUNX1-ETO leukaemia

Acute myeloid leukaemia (AML) results from clonal expansion of primitive myeloid cells incapable of terminal differentiation, giving rise to an accumulation of ‘blast’ cells at various stages of maturation within the bone marrow niche. AML is a heterogenous disease with multiple morphological, immunophenotypic and genetic features. This includes the t(8;21) which results in the expression of RUNX1-ETO, and occurs in 12% of AML cases. To understand the role of RUNX1-ETO in the pathogenesis of AML, our group previously ectopically expressed RUNX1-ETO in normal human haematopoietic stem progenitor cells (HSPC). This resulted in a block in granulocytic differentiation and was associated with increased self-renewal - hallmarks of leukaemia. A subsequent study analysed the transcriptome of these cells and identified 380 differentially expressed genes using an unsupervised approach. This current study has now refined this analysis to determine the most significant changing transcription factors (TFs). Using Pathway Analysis programme (Metacore™), this study identified ZNF217 to be significantly overexpressed compared to control (1.5-fold; p=0.003). ZNF217 is a TF responsible for binding to the promotors of several target genes, such as E-cadherin, as well as cooperating in transcriptional silencing programs by recruiting chromatin modifiers. This study determined that ZNF217 overexpression, as single abnormality, induced myeloid differentiation of HSPC, particularly within the monocytic population, suggesting that it is unlikely that this TF possesses a role in leukemogenesis on its own. Additionally, ZNF217 was found to be dispensable for myeloid differentiation, as knockdown (KD) of this TF failed to inhibit this process. Whilst studies have determined the transcriptomic changes observed in cells expressing RUNX1-ETO, there is a paucity of studies quantitating proteomic changes. Therefore, this study also aimed at analysing the proteomic profile of RUNX1-ETO expressing HSPC using quantitative proteomics by SWATH-MS (on different subcellular structures including cytosolic or nuclear fractions). 4,635 proteins were quantified, of which 2,787 were detected in the cytoplasm, and 1,848 in the nucleus. Statistical analysis identified 257 significantly differentially expressed proteins in RUNX1-ETO compared to controls; of which 71% were detected in cytoplasm and 29% in the nucleus. RUNX1-ETO significantly downregulated the expression of C/EBPβ protein and mRNA vs control suggesting transcriptional suppression by RUNX1-ETO. Knocking-down C/EBPβ expression in HSPC, however, failed to induce significant changes in both monocytic and granulocytic development. Interestingly, KD of C/EBPβ in the RUNX1-ETO-expressing cell line, SKNO-1, completely suppressed myeloid cell surface marker expression and gave a concomitant increase in cell proliferation. In nont(8;21) cells lines (HEL and U937), on the other hand, KD of C/EBPβ ablated cell growth and increased apoptotic frequency, suggesting that the effects of C/EBPβ KD are context dependent. In conclusion, both transcriptomic and proteomic analysis proved to be useful tools for the identification of potential mediators of the block in terminal differentiation observed in RUNX1-ETO-expressing cells. Subsequently, ZNF217 and C/EBPβ were identified as targets of interest in the context of t(8;21). Whilst it is unlikely that ZNF217 overexpression contributes to leukaemogenic development, additional studies would be necessary to fully determine the role of C/EBPβ in this process