Contributions
Abstract: S877
Type: Oral Presentation
Presentation during EHA23: On Saturday, June 16, 2018 from 16:15 - 16:30
Location: Room A7
Background
How the identity of the cells targeted by oncogenic lesions impacts on the clinical picture of the resulting cancer has not been established.
In myelodysplastic syndrome (MDS), all mutations in a malignant clone can be tracked back to the hematopoietic stem cells (HSCs) (Woll et al., Cancer Cell 2014). This suggests that MDS originates within and is propagated by the malignant counterparts of normal HSCs. However, the human phenotypically-defined HSC compartment also contains substantial numbers of non-HSCs, and it remains to be established to what extent also non-HSCs may be targeted and transformed by MDS-associated mutations. As well as in MDS (Papaemmanuil et al., Blood 2013), co-occurrence of inactivating mutations in EZH2 and RUNX1 is frequently observed in early thymic progenitor (ETP) leukemia, a clinically and molecularly distinct subtype of T cell acute lymphoblastic leukemia (Zhang et al., Nature 2012).
Aims
To determine the distinct impacts of inactivation of Ezh2 and Runx1 in different hematopoietic stem and progenitor populations in mice.
Methods
We used Mx1Cre, Flt3Cre and Rag1Cre to target conditional inactivation of Ezh2 and Runx1 to HSCs, multipotent progenitors (MPPs) or early lymphoid progenitors respectively.
Results
In agreement with previous studies using inactivation of Ezh2 combined with expression of a dominant-negative Runx1 mutant (Sashida et al., Nat Commun 2014), we found that targeting of Ezh2 and Runx1 inactivation to HSCs using Mx1Cre induced an MDS phenotype. Competitive transplantation experiments demonstrated a clonal advantage of mutant cells over wild-type cells, combined with an extrinsic suppression of wild-type HSCs in recipient bone marrow.
Surprisingly, targeting of the same mutations to MPPs using Flt3Cre resulted in development of the same MDS phenotype with a similar latency to that induced by Mx1Cre. Absolute numbers of Lin-Kit+Sca1+CD48-CD150+ HSCs were reduced in these mice, while Lin-Kit+Sca1+CD48+CD150- MPPs were expanded compared to wild-type controls, suggesting that MPPs rather than HSCs could represent the MDS-propagating malignant stem cells in this model. We confirmed this using transplantation of purified E13.5 fetal liver MPPs. Unlike wild-type MPPs, MPPs from Flt3Cre-induced Ezh2 and Runx1-inactivated embryos were able to engraft and reconstitute long-term hematopoiesis and an MDS phenotype in recipient mice, while transplanted MPPs from Flt3Cre-negative littermate controls showed no engraftment.
In striking contrast, when Ezh2 and Runx1 inactivation was targeted to early lymphoid progenitors using Rag1Cre, we did not observe development of an MDS phenotype. Rather, absolute numbers of ETPs within the thymuses of these mice were markedly expanded compared to wild-type controls. This expanded ETP population showed transcriptional signatures characteristic of human ETP leukemia. Addition of Flt3-ITD, a constitutively activating RAS signalling mutation recurrently found in ETP leukemia, induced an acute T cell/myeloid leukemia which could be propagated by the expanded ETP population.
Conclusion
Our findings reveal that targeting the same clinically relevant mutations to different hematopoietic stem and progenitor populations can induce distinct and clinically relevant malignant phenotypes. This provides experimental evidence that the identity of the target cell of specific oncogenic lesions might be a crucial factor for the phenotype, prognosis and therapeutic response of the resulting malignancy.
-Adam Mead and Sten Eirik Jacobsen contributed equally to this work.
Session topic: 9. Myelodysplastic syndromes – Biology & Translational Research
Keyword(s): Hematopoietic stem and progenitor cells, Myeloid malignancies, T cell acute lymphoblastic leukemia
Abstract: S877
Type: Oral Presentation
Presentation during EHA23: On Saturday, June 16, 2018 from 16:15 - 16:30
Location: Room A7
Background
How the identity of the cells targeted by oncogenic lesions impacts on the clinical picture of the resulting cancer has not been established.
In myelodysplastic syndrome (MDS), all mutations in a malignant clone can be tracked back to the hematopoietic stem cells (HSCs) (Woll et al., Cancer Cell 2014). This suggests that MDS originates within and is propagated by the malignant counterparts of normal HSCs. However, the human phenotypically-defined HSC compartment also contains substantial numbers of non-HSCs, and it remains to be established to what extent also non-HSCs may be targeted and transformed by MDS-associated mutations. As well as in MDS (Papaemmanuil et al., Blood 2013), co-occurrence of inactivating mutations in EZH2 and RUNX1 is frequently observed in early thymic progenitor (ETP) leukemia, a clinically and molecularly distinct subtype of T cell acute lymphoblastic leukemia (Zhang et al., Nature 2012).
Aims
To determine the distinct impacts of inactivation of Ezh2 and Runx1 in different hematopoietic stem and progenitor populations in mice.
Methods
We used Mx1Cre, Flt3Cre and Rag1Cre to target conditional inactivation of Ezh2 and Runx1 to HSCs, multipotent progenitors (MPPs) or early lymphoid progenitors respectively.
Results
In agreement with previous studies using inactivation of Ezh2 combined with expression of a dominant-negative Runx1 mutant (Sashida et al., Nat Commun 2014), we found that targeting of Ezh2 and Runx1 inactivation to HSCs using Mx1Cre induced an MDS phenotype. Competitive transplantation experiments demonstrated a clonal advantage of mutant cells over wild-type cells, combined with an extrinsic suppression of wild-type HSCs in recipient bone marrow.
Surprisingly, targeting of the same mutations to MPPs using Flt3Cre resulted in development of the same MDS phenotype with a similar latency to that induced by Mx1Cre. Absolute numbers of Lin-Kit+Sca1+CD48-CD150+ HSCs were reduced in these mice, while Lin-Kit+Sca1+CD48+CD150- MPPs were expanded compared to wild-type controls, suggesting that MPPs rather than HSCs could represent the MDS-propagating malignant stem cells in this model. We confirmed this using transplantation of purified E13.5 fetal liver MPPs. Unlike wild-type MPPs, MPPs from Flt3Cre-induced Ezh2 and Runx1-inactivated embryos were able to engraft and reconstitute long-term hematopoiesis and an MDS phenotype in recipient mice, while transplanted MPPs from Flt3Cre-negative littermate controls showed no engraftment.
In striking contrast, when Ezh2 and Runx1 inactivation was targeted to early lymphoid progenitors using Rag1Cre, we did not observe development of an MDS phenotype. Rather, absolute numbers of ETPs within the thymuses of these mice were markedly expanded compared to wild-type controls. This expanded ETP population showed transcriptional signatures characteristic of human ETP leukemia. Addition of Flt3-ITD, a constitutively activating RAS signalling mutation recurrently found in ETP leukemia, induced an acute T cell/myeloid leukemia which could be propagated by the expanded ETP population.
Conclusion
Our findings reveal that targeting the same clinically relevant mutations to different hematopoietic stem and progenitor populations can induce distinct and clinically relevant malignant phenotypes. This provides experimental evidence that the identity of the target cell of specific oncogenic lesions might be a crucial factor for the phenotype, prognosis and therapeutic response of the resulting malignancy.
-Adam Mead and Sten Eirik Jacobsen contributed equally to this work.
Session topic: 9. Myelodysplastic syndromes – Biology & Translational Research
Keyword(s): Hematopoietic stem and progenitor cells, Myeloid malignancies, T cell acute lymphoblastic leukemia