Understanding leukaemogenesis: towards identification of leukaemia enhancers and suppressors through zebrafish genetics

Abstract

Zebrafish runx clones showing high sequence identity with mammalian Runx genes have been isolated. The orthology of the runx genes was confirmed by phylogenetic analysis, and supported by a conservation of synteny for runx1 and runx2 with their human counterparts. The runx2 gene in zebrafish is duplicated, runx2b (examined in this work) maps to LG 20, while its paralog runx2a (cloned by S. Fisher laboratory) maps to LG 17. Preliminary analysis of runx2a and runx2b showed that their expression patterns have diverged, suggesting that subfunctionalisation has occurred for runx2 paralogs in zebrafish. Expression analysis of runx1 and runx3 suggested a role in haematopoietic and neuronal development. Both genes were expressed in blood-forming areas during zebrafish embryogenesis. Runx1 co-localised with scl in individual cells of the lateral plate mesoderm, suggesting a role in haemangioblast differentiation, runx1 was also expressed in the intermediate cell mass and cells located in the ventral wall of the dorsal aorta. Haematopoietic expression of runx3 was first seen at 24 hpf in the ventral tail and later co-localised with spil and lyz in circulating blood cells. Zebrafish runx1 and runx3 were coexpressed in Rohon-Beard cells. Transcripts of runx1 were also present in putative VIII cranial nerve ganglia, while runx3 was detected in trigeminal ganglia. In addition, runx3 was expressed in the developing gut and cartilaginous tissues. To further characterise the function of the zebrafish runx genes, their expression was analysed in haematopoietic mutants and overexpression studies were undertaken. Haematopoietic expression of both runx1 and runx3 was reduced in cloche embryos, indicating involvement in a haematopoietic pathway downstream of the cloche gene. Transient expression of runx1 resulted in partial rescue of the cloche haematopoietic defect. In contrast, overexpression of runx3 failed to effect a response in cloche embryos. When overexpression was performed in wild-type embryos, runx1 produced ectopic blood, while runx3 increased numbers of blood cells. Furthermore, overexpression of runx3 enhanced runx1 expression in both the aortic region and Rohon-Beard cells, suggesting that these genes can cooperate in haematopoietic and neural development in zebrafish. Depletion of Runx1 with antisense morpholino oligonucleotides abrogated the development of both blood and vessels, as demonstrated by loss of circulation, abnormal development of vasculature and the accumulation of immature haematopoietic precursors. The block in definitive haematopoiesis was similar to that observed in Runx1 knock-out mice, implying that zebrafish Runx1 has a function equivalent to that in mammals. Depletion of Runx3 caused a reduction in the number of mature blood cells, implicating Runx3 in proliferation or survival of haematopoietic progenitors/stem cells. Furthermore, the knockdown of runx3 interfered with the development of definitive haematopoiesis. In addition to the haematopoietic abnormalities, neuronal defects were also observed in runx1- and runx3-morpholino injected embryos. Knock-down of runx1 disturbed development of the otic vesicle and Rohon-Beard cells, while knock-down of runx3 caused hydrocephaly and abnormalities in Rohon-Beard cells and trigeminal ganglia development. To provide a base for examining the role of Runx1 in leukaemogenesis, the effects of transient expression of a human RUNX1-ETO transgene [product of the t(8;21) translocation in acute myeloid leukaemia] was investigated in zebrafish embryos. Expression of RUNX1-ETO caused disruption of normal haematopoiesis, aberrant circulation, internal haemorrhages and cellular dysplasia. These defects reproduced those observed in Runx1-depleted embryos and RUNX1-ETO knock-in mice. The phenotype obtained validated the zebrafish as a model system to study t(8;21)-mediated leukaemogenesis. Furthermore, these abnormalities were reproduced when RUNX1-ETO was driven from the zebrafish heat shock protein 70 promoter. Raising embryos at 20°C, when the temperature was reduced slowly, prevented specific abnormalities and resulted in normal embryo survival. Attempts to tag the RUNX1-ETO fusion with the EGFP reporter were not successful. As such, this work created a platform to generate a stable transgenic zebrafish line that expresses RUNX1-ETO under the control of an inducible promoter. In addition, genomic clones containing runx1 promoter sequences were isolated to advance future haematopoietic studies.