CBFβ-SMMHC Inhibits G1 to S Progression in Primary Murine and Human Myeloid Progenitors Dependent upon Its AML1-Interaction and Assembly Competence Domains.

CBFβ-SMMHC, encoded by the inv(16) or t(16;16) translocations in approximately 8% of acute myeloid leukemia (AML) cases, is a fusion protein containing amino acids 1-165 of the 182 residue core binding factor β (CBFβ) and the rod domain of smooth muscle myosin heavy chain (SMMHC). The CBFβ domain of CBFβ-SMMHC retains the ability to interact with AML1/RUNX1. The SMMHC domain both mediates multimerization and interacts directly with corepressors, including mSin3A. CBFβ-SMMHC inhibits the expression of AML1-regulated genes, by sequestering AML1 in multimeric complexes and by directly repressing AML1-regulated genes. CBFβ-SMMHC was previously found to slow G1 to S cell cycle progression in hematopoietic cell lines, reflecting repression of AML1-regulated genes required for cell cycle, including cyclin D3. This effect was overcome be exogenous c-Myc or cdk4. In this study, murine marrow or human CD34+ cells were transduced with retroviral or lentiviral vectors, respectively, expressing CBFβ-SMMHC or two mutant variants. CBFβ-SMMHC reduced murine or human myeloid cell proliferation 3- to 4-fold in liquid culture, during a period when control murine cells accumulated 5-fold and human cells 20-fold. CBFβ-SMMHC decreased the formation of myeloid, but not erythroid, colonies 2- to 4-fold, and myeloid colonies expressing CBFβ-SMMHC were markedly reduced in size. Lack of effect on erythroid colonies reflects their lack of expression of AML1. The mutant variant CBFβ-SMMHC(Δ2-11) does not bind AML1 due to a deletion near its N-terminus, and CBFβ-SMMHC(ΔACD) does not multimerize or efficiently bind corepressors due to a 28 residue deletion near its C-terminus. Neither of these mutants, which were expressed at levels similar to wild-type, slowed proliferation or reduced myeloid colonies. CBFβ-SMMHC increased the G1/S ratio in wild-type murine and human progenitors. Proliferation was still slowed in p15(−/−) murine marrow cells transduced with CBFβ-SMMHC, suggesting that additional mutations, such as activation of growth factor receptors and consequent c-myc induction, are required in primary AMLs to allow enhanced proliferation. AML1-ER(T), which contains full-length AML1 and accelerates G1 to S progression in cell lines when activated by 4HT, had an effect opposite to CBFβ-SMMHC, stimulating proliferation of murine or human myeloid progenitors 2-fold. In summary, CBFβ-SMMHC inhibits the proliferation of myeloid progenitors dependent upon inhibition of AML1 and integrity of its Assembly Competence Domain. Targeting the CBFβ-SMMHC ACD or its CBFβ domain may uncover novel therapeutics useful for AML cases expressing this oncoprotein. Furthermore, these findings support a model we have proposed previously which states that mutations which accelerate G1 are required during leukemogenesis by CBFβ-SMMHC and other CBF oncoproteins. Finally, our results lend support to the conclusion that AML1 participates in the regulation of normal myeloid stem-progenitor cell proliferation. Exogenous AML1 may therefore be useful for expansion of hematopoietic stem-progenitor cells.