Elucidating the role of mTOR complexes (mTORC1 and mTORC2) in normal haemopoiesis and in Chronic Lymphocytic Leukaemia

Mechanistic target of rapamycin (mTOR) functions within a complex signalling cascade, through its activity in two unique complexes mTORC1 and mTORC2, to promote a multitude of different cellular functions including autophagy, protein synthesis and survival. The exact role of these complexes during leukaemia initiation/maintenance remains to be elucidated. Here, using transgenic knockout (KO) mouse models, we determine the individual roles of mTORC1 (targeting Raptor) and mTORC2 (targeting Rictor) in normal haemopoiesis and in CLL initiation/maintenance. Our results demonstrate that mice carrying a targeted KO of Raptor at the haemopoietic stem cell (HSC) stage (Vav-Raptor KO) do not survive post birth. This is due to anaemia resulting from a significant decrease in Ter119+ population, a significant decrease in Klf1 and Klf2 gene expression, and a significant increase in the megakaryocyte-erythroid progenitor (MEP) population, suggesting a block at the MEP stage in Vav-Raptor KO foetal liver (FL). While mTORC1 plays a fundamental role in RBC development, we show that mTORC2 plays a potential role in RBC regulation, as Rictor-deficient HSPCs exhibit an increase in RBC colony formation ex vivo. Conditional KO (cKO) of Raptor (Mx1-Raptor cKO) in adult mice results in splenomegaly accompanied by increased spleen organ cellularity. Furthermore, there is a significant decrease in B cell lineage commitment, with a block in B cell development at the Lin-Sca-1+CD117+ (LSK) stage in the BM. mTORC2, on the other hand regulates late B cell maintenance as indicated by a significant decrease in transitional B cells (T1/T2), marginal zone progenitor (MZP), and follicular 1 (fol1) cells in Vav-Rictor KO mice compared to controls. To address the role of mTORC1 and mTORC2 in CLL initiation/maintenance in vitro, BM-derived haemopoietic progenitor cells (HPCs) isolated from control (cre-), Raptor-deficient (Mx1-Raptor cKO) or Rictor-deficient (Vav-Rictor KO) mice were retrovirally-transduced with a kinase dead PKCα (PKCαKR) construct to induce an aggressive CLL-like disease. Raptor-deficient BM progenitors exhibited reduced proliferation and failed to generate a CLL-like disease, due to a block in B cell lineage commitment in vitro. However, there was an increase in cell cycling and migration in PKCαKR CLL-like cells with Rictor-deficiency suggesting a role of mTORC2 in disease maintenance. To determine a role for mTORC1 in disease maintenance in vivo, NSG mice were transplanted with Mx1-Raptor control or Mx1-Raptor cKO PKCαKR transduced BM cells. Once disease was established in vivo, cKO was induced and disease load and progression was monitored. Our data demonstrate a decrease in disease load with Raptor cKO, together with a significant increase in survival. Additionally, host mice transplanted with CD19-Raptor KO PKCαKR cells exhibited a significant increase in survival. However, these mice eventually died of disease due to limitations of the KO model. Lastly, to test the translational capacity of mTOR inhibitors, efficiency of AZD2014 (dual mTOR inhibitor), ibrutinib and a combination of the two drugs was assessed in reducing PKCαKR CLL-like disease load in host mice. AZD2014 was as efficient at reducing disease load as ibrutinib, however combination therapy of these drugs was not as efficient compared to single agents. Interestingly, we demonstrate that a more aggressive PKCαKR CLL-like disease (in secondary transplants) is more mTORC1 dependent than in primary transplants, as indicated by the superiority of rapamycin (allosteric mTORC1 inhibitor) in markedly decreasing disease load as compared to AZD2014 in host mice. Taken together, mTORC1 plays an essential role in haemopoiesis, with Raptor-deficiency causing a block in RBC and B cell development at the MEP and LSK stage respectively. In comparison, Rictor-deficiency regulates later B cell lineages and promotes RBC colony formation, potentially through mTORC1 activation. Importantly, CLL-like cells lacking mTORC2 have increased cell cycling and migration whereas mTORC1 deficiency causes a decrease in disease load. Therefore, mTORC1 and mTORC2 play distinct/complementary roles in haemopoietic development and leukaemia initiation/progression. These studies provide a strong foundation for further studies testing novel mTOR inhibitors for CLL in our models