Abstract 2472

RKH, YK, and LZ all contributed equally to this work

Inv(16) is found in nearly all patients with acute myeloid leukemia (AML) subtype M4Eo. Inv(16) results in the fusion of the transcription factor gene CBFB, and the MYH11 gene, which encodes Smooth Muscle Myosin Heavy Chain (SMMHC). This results in the fusion gene CBFB-MYH11, which encodes CBFβ-SMMHC. Previously we showed that knock-in mice with a single allele of Cbfb-MYH11 (Cbfb+/MYH11) have severe differentiation defects in primitive hematopoiesis and a total block in definitive hematopoiesis. In addition, chimeric mice generated from Cbfb+/MYH11 ES cells consistently developed leukemia within a few months after treatment with the mutagen N-ethyl-N-nitrosourea (ENU).

It is currently not clear which functional domains of CBFβ-SMMHC are responsible for its activity in differentiation and leukemogenesis. In vitro experiments have indicated that CBFβ-SMMHC can form multimeric complexes via the C terminal domain. It has been postulated that this multimerization may be important for the function of CBFβ-SMMHC by resulting in large macromolecular complexes and/or sequestration of its binding partner, the transcription factor RUNX1. To determine the importance of this domain in vivo, we generated knock-in mice expressing a mutant Cbfb-MYH11 allele with a deletion of the 95 C-terminal amino acids (Cbfb+/MYH11ΔC95). In analysis of primitive hematopoiesis, we found that Cbfb+/MYH11ΔC95 and CbfbMYH11ΔC95/MYH11ΔC95 mice had no or very mild differentiation defects, statistically significantly less severe (p<.05) than seen in embryos expressing full-length Cbfb-MYH11. During definitive hematopoesis, there were no observable defects in Cbfb+/MYH11ΔC95 mice, but CbfbMYH11ΔC95/MYH11ΔC95 embryos showed a complete block in definitive hematopoiesis, as seen in mice expressing a single allele of full length Cbfb-MYH11. This indicates that Cbfb-MYH11ΔC95 is less effective in blocking differentiation than the full length fusion gene. Interestingly, both the primitive and definitive embryonic blood phenotypes of the CbfbMYH11ΔC95/MYH11ΔC95 were similar to that observed in embryos lacking functional Cbfb (Cbfb−/−), implying that Cbfb-MYH11ΔC95 may act as a null allele. To test this possibility we used gene expression microarrays to compare gene expression profiles in the peripheral blood from embryonic day 12 CbfbMYH11ΔC95/MYH11ΔC95, Cbfb−/−, and Cbfb+/MYH11, as well as their Cbfb+/+ littermates. Surprisingly, CbfbMYH11ΔC95/MYH11ΔC95 embryos showed deregulated expression of a distinct gene set as compared to both Cbfb−/− and Cbfb+/MYH11 embryos. This implies that Cbfb-MYH11ΔC95 is not a null allele of Cbfb, and likely retains some, but not all, of the neomorph properties of full length Cbfb-MYH11. Consistent with this finding, we observe the accumulation of abnormal myeloid cells in some adult Cbfb+/MYH11ΔC95 mice after ENU treatment, which has not been reported in Cbfb+/− mice. However, we found that Cbfb-MYH11ΔC95 has not retained the most critical of the fusion gene's activities: the ability to induce leukemogenesis. Importantly, none of the Cbfb+/MYH11ΔC95 mice developed leukemia after treatment with ENU. This is in contrast to mice expressing full length Cbfb-MYH11, which all develop leukemia under these conditions. Together, these results indicate that the 95 C-terminal amino acids of CBFβ-SMMHC are required for both embryonic hematopoietic defects and leukemogenesis.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.