Novel Mechanisms of Growth Inhibition By Histone Deacetylase Inhibitors in MPN

Histone deacetylase inhibitors (HDACi) are being evaluated in several clinical trials for the treatment of hematological malignancies and 2 of them have been approved for the treatment of T cell lymphoma. We have investigated the effectiveness of using 2 HDACi - Vorinostat (SAHA) and Panobionstat for treatment of systemic mastocytosis (SM) for which no treatments are available. SAHA and Panobinostat showed concentration dependent inhibition of proliferation in mastocytosis patient derived cell lines, HMC1.1 and HMC1.2 bearing either single mutation in the juxtamembrane domain KIT^G560V or double mutation also including the activation loop KIT^G560V,D816V, respectively. Interestingly HMC1.2 cells with 2 mutations in KIT^{G560V, D816V} were more sensitive to treatment with both SAHA (IC50 ~ 260 nM) and Panobinostat (IC50 ~11 nM) as compared to HMC1.1 cells (IC50 ~ 600 nM for SAHA and IC50 ~ 23 nM with Panobionstat) that have a single mutation in KIT. The inhibition of proliferation by HDACi treatment appeared to be associated with modulation of cell cycle and differentiation rather than induction of apoptosis. We further tested the effect of combining HDACi treatment with tyrosine kinase inhibitor, Dasatinib. Combination of Dasatinib (5 - 20 nM) with either SAHA or Panbionostat was more effective than any of the single drugs in HMC1.1 cells. However such a combination effect was not seen in HMC1.2 cells. Over 90% patients with aggressive systemic mastocytosis carry KIT^D816V mutation. This particular mutation of KIT is highly resistant to therapeutics and is associated with poor survival. Thus patients carrying the KIT^D816V mutation are likely to benefit from treatment with HDACi while juxtamembrane mutation bearing patients would benefit more from a combination of HDACi with kinase inhibitors such as Dasatinib. To further understand the mechanism behind HDACi inhibition of KIT^D816V mediated oncogenic signals, we expressed KIT^D814V (murine equivalent of human D816V) in myeloid cells and treated them with SAHA and Panobinostat. Cells expressing KIT^D814V showed cytokine independent proliferation and interestingly, cytokine independent proliferation was more sensitive to inhibition by both SAHA and Panbinostat as compared to the cytokine dependent proliferation. In contrast to HMC cells, inhibition of cytokine independent proliferation by HDACi in KIT^D814V oncogene expressing cells was associated with significant increase in apoptosis. The percent cells negative for Annexin V stain decreased from 74% in control to ~46% in 24h upon treatment with 500 nM SAHA or 10 nM Panbinostat (p< 0.001; n=3). Induction of apoptosis by HDACi in KIT^D814V oncogene expressing cells was significantly suppressed in presence of cytokine. Consistent with the in vitro data, treatment with SAHA prolonged the survival of mice transplanted with myeloid cells expressing KIT^D814V mutation as compared to the vehicle treated group (Median survival - 21d (veh), 34d (SAHA), p=0.0015, n=5). In addition to increased histone acetylation, treatment of cells with HDACi also leads to increase in acetylation of chaperone proteins such as HSP90 which can mediate proteasomal degradation of oncogenes and has been suggested as one of the mechanisms by which HDACi mediates growth inhibition. In contrast, we found no significant decrease in mutant KIT protein expression or AKT expression and phosphorylation upon treatment with either of the HDACi. Though expression of Pim1 a pro survival serine/threonine kinase was down-regulated by SAHA and Panobinostat. Treatment with HDACi was also associated with differential modulation of STAT3, STAT5 and ERK pathways. Treatment with both SAHA and Panobinostat down-regulated STAT5 phosphorylation and up-regulated ERK1/2 phosphorylation with more increase in ERK1/2 phosphorylation observed in cells with higher sensitivity to inhibition by HDACi. Though increased activation of the ERK pathway has been associated with leukemogenesis; there is some evidence to indicate that hyper activation of ERK pathway in absence of additional leukemogenic signals activates a pro senescence program. Differential expression of Pim1 and along with differential modulation of STAT and ERK pathways may contribute to differential sensitivity of the cells to HDACi and are likely to be of therapeutic value in aggressive systemic mastocytosis that has so far been associated with poor prognosis.