| Popis: |
Background: Fetal haematopoiesis has clear functional and molecular differences to adult haematopoiesis. Some leukaemias, such as juvenile myelomonocytic leukaemia (JMML), are believed to originate in fetal cells, and frequently have a poor prognosis. The majority of patients with JMML have a RAS activating mutation, and where mouse models exist, these primarily involve post-natal induction of mutations, which may not be suitable given the fetal origin of the disorder. Children with JMML develop recurrent mutations involving loss of function in the enhancer of zeste homolog 2 (EZH2), which is known to control developmental pathways, either through direct mutations or through loss of chromosomal material (monosomy 7). These are particularly associated with Kras mutations. It is not clear whether these mutations are gained in post-natal or adult life, however they worsen the prognosis. In other mouse models of myeloid disorders Ezh2 loss leads to reactivation of fetal pathways, but their role in JMML remains unclear. Aim: The overarching aim of this project was to investigate the fetal origins of JMML by creating a Kras mutant mouse model derived from fetal cells. This would allow me to study whether fetal versus adult induction of mutations affected the phenotype of the leukaemia and the effect of secondary Ezh2 mutation in this model. My specific aims were: 1. To develop a mouse model to investigate the fetal origins of JMML; 2. To investigate the effect of the acquisition of leukaemogenic mutations at differing stages of murine development; 3. To identify if mutation of Ezh2 alters or enhances leukaemogenesis in a JMML model system. Results: 1. Developing a mouse model to study the fetal origins of JMML: Initial experiments with Vav-iCre induction of mutations demonstrated the proof of principle that KrasG12D mutant cells from the fetal liver (FL) can cause a myelomonocytic leukaemia. Ezh2 knock-out (KO) in combination with KrasG12D, however, led to a cellular defect with no engraftment. Therefore, spontaneous induction in a small clone with Mx1Cre was used. Transplantation experiments showed a possible combinatorial effect of Kras and Ezh2 mutations, but there was a high rate of T-cell leukaemia as well as difficulty in proving a fetal origin of mutation. Fgd5CreERT2 induction was optimised, showing no spontaneous deletion in adult or fetal life, and Fgd5 surface expression on yolk sac (YS) erythro-myeloid progenitors (EMPs), as well as FL, and adult bone marrow (aBM) haematopoietic stem cells (HSCs). Optimised dosing allowed delivery of induced FL mice, as well as induction in the aBM for comparison. 2. The effect of acquisition of mutations at differing stages of ontogeny: Kras mutation in the FL created a good model for JMML, with splenomegaly, thrombocytopenia, and monocytosis with evidence of lung infiltration, as seen in children with JMML. KrasG12D mutant FL mice had a significantly shortened latency of disease to adult mice, accompanied by a change in phenotype. The fetal induction allowed an almost universally penetrant myelomonocytic leukaemia, however on adult induction these mice developed a range of disorders, including a T-cell leukaemia not seen in the fetal cohort. Further, where the adult mice had phenotypic HSC depletion, the FL-induced mice did not. 3. Does Ezh2 alter or enhance leukaemogenesis? Induction of Ezh2 in adult and fetal mice altered the phenotype of disease, with adult mice developing a more ‘fetal-like’ increase in myelomonocytic leukaemia, whereas fetal loss of Ezh2 led to a diminishing of the penetrance as well as alteration of the phenotype. On analysis of the HSPC compartment, Ezh2 loss in adult leukaemic mice appeared to rescue the stem cell phenotype, but in fetal life depleted the stem cell compartment. This suggests a clear differential role of Ezh2 and points towards a post-natal acquisition of Ezh2 loss. Conclusion: This project demonstrates a novel model for JMML with a clear fetal origin. Mice induced with KrasG12D in the FL develop a JMML phenotype with an alteration in phenotype with mutational gain in adult life. This suggests an important role of ontogeny in the timing of these mutations, and the model is the first model to allow study of fetal versus adult gain of mutations. The loss of Ezh2 suggests that this may be a post-natal acquired secondary mutation, but clearly changes the phenotype of both adult and fetal mice, whilst accelerating disease. This project has been impacted by the COVID-19 pandemic, particularly due to cancellation of RNA sequencing, which is being planned to further investigate differences in the gene expression programs in the fetal and adult leukaemic mice. |
