Selective Targeting Of Inv(16)+ AML Stem Progenitor Cells By Inhibiting HDAC8

Chromosomal inversion inv(16)(p13.1q22) which leads to the fusion of the transcription factor gene CBFb and the MYH11 gene, occurs in over 8% of acute myeloid leukemia (AML) cases. The fusion product CBFβ-SMMHC (CM) inhibits differentiation of hematopoietic stem and progenitor cells (HSPCs) and creates pre-leukemic populations predisposed to acute myeloid leukemia (AML) transformation. The mutations of tumor suppressor p53 occur in approximately half of all cases of human cancer, but TP53 mutations are relatively rare in inv(16) AML. We have previously shown that CM expression leads to reduced acetylation of p53 and impaired p53 target gene activation through formation of aberrant protein complex with p53 and HDAC8 (Blood, 2012,120: A772.). Here, we showed that CM interacts with p53 both in CM transformed mouse primary bone marrow cells as well as in AML stem and progenitor cells from inv(16) patients. When HDAC8 selective pharmacological inhibitor 22d directed against its catalytic sites (ChemMedChem 2012, 7:10, 1815-24;) were used to treat inv(16) mouse primary bone marrow progenitor cells and inv(16)+ CD34+ stem progenitor cells from patients, Ac-p53 levels were remarkably increased as shown by western blot. We further assessed the p53 target genes expression after HDAC8 inhibitor 22d treatment by qRT-PCR assay in inv(16)+ CD34+ stem progenitor cells (n=8), and observed variable levels of activation in p53 targets (Fold activation: p21:2.25-fold, hdm2:1.17-fold, 14-3-3σ: 3.12-fold, puma: 2.39-fold), indicating p53 was re-activated. Similar results were also shown in CM transformed mouse bone marrow progenitor cells. Importantly, we found that 22d treatment significantly inhibit the growth of inv(16)+ AML CD34+ cells (n=9) rather than normal CD34+ cells (n=7) , (AML IC50= 6.509 μM, vs Normal cells IC50=13.83 μM, p=0.0003). Meanwhile, 22d selectively induces apoptosis of inv(16)+ AML stem and progenitor cells while sparing normal HSPCs (AML LD50= 10.24 μM, vs NL LD50= 46.36 μM, p=0.001). To evaluate whether the effect of HDAC8i is mediated by p53, we knocked down p53 with a lentiviral vector expressing shRNA against p53 (or non-silencing shRNA) in AML CD34+ cells, and treated the cells with HDAC8 inhibitor 22d (5-20 µM). We showed that despite the inter-sample variability, knocking down p53 expression in all AML samples tested (n=3) led to reduced HDAC8i-induced apoptosis, suggesting that p53 contributes to the apoptosis effect induced by HDAC8i (22d) in inv(16)+ AML cells. Importantly, by taking advantage of our conditional knock-in mouse model (Cbfb^56M/+/Mx1-Cre), which develops AML under induced expression of CBFß-SMMHC (Cancer Cell, 2006, 9:1, 57-68), we were able to perform the ex vivo treatment assay by treating primary leukemic cells (marked with dTomato) with either DMSO (as vehicle control) or with HDAC8 inhibitor 22d (10μM) for 48h, followed by transplantion into congenic mice (control group n=8, treatment group n=7). We observed reduced short-term engraftment of leukemic cells that are treated with 22d (10 μM) at 4 weeks post-transplantation in the peripheral blood (Donor cell%: control group=5.99%, treatment group=0.178%, P=0.0093). Interestingly, engraftment of cord blood CD34⁺ cells at 16 weeks post-bone marrow transplantation was not reduced after treatment with 22d (10 μM) (human CD45+ %: control=66.2% versus treatment=63.4%, p=0.9), indicating the effect by HDAC8 inhibition is selective for leukemic cells. In conclusion, we have identified a novel mechanism whereby CBFβ-SMMHC inhibits p53 fucntion, and may further implicate inhibition of HDAC8 as a promising approach to selectively target inv(16)⁺ AML stem and progenitor cells.