Differential Effects of HDAC Inhibitors on CD34+ Hematopoietic Progenitor Lineage Choice Decisions

Epigenetic modifications play an important role in the pathogenesis of myeloid malignancies of which alterations in cellular acetylation profiles have proven to contribute substantially. The clinical use of chromatin modulating drugs, such as histone deacetylase (HDAC) inhibitors, has more recently been evaluated in several clinical trials. Nonetheless, little is known concerning the effects of HDAC inhibitors on myelopoiesis or their specific targets in human hematopoietic stem cells. In order to investigate this, we utilized a differentiation system in which umbilical cord blood derived CD34+ cells were differentiated ex-vivo in liquid and semi-solid cultures. We investigated the effect on lineage development, colony-forming potential, proliferation and differentiation using trichostatin A (TSA; group I/II/IV, 5–25nM), sodium butyrate (SB; group I/IIa, 100–500μM) and valproic acid (VPA; group I/IIa, 100–500μM). In our liquid culture system we observed that TSA increased the numbers of CD34+ hematopoietic progenitors resulting in a dose dependent delay in granulocyte/macrophage progenitor (GMP) differentiation. Furthermore, TSA promoted differentiation towards megakaryocyte/erythroid progenitors (MEP) in the absence of thrombopoietin (TPO) and erythropoietin (EPO). SB promoted a shift towards the granulocyte/macrophage lineage, and inhibited proliferation and differentiation of CD34+ progenitors. VPA increased the number of CD34+ hematopoietic progenitors in a dose dependent manner, with increased numbers of MEP and a delay in GMP differentiation. To investigate the colony-forming potential of hematopoietic progenitors and further define the effect on lineage choice decisions in the presence of HDAC inhibitors, we also used a semi-solid culture system. In agreement with the results of the progenitor analysis, 100–200mM VPA stimulated granulocyte/macrophage colony formation which was also observed after 100mM SB treatment, although colonies were smaller and less differentiated. All three HDAC inhibitors showed a dose dependent inhibition of erythroid colony-forming potential and differentiation. We also used the liquid culture system to examine proliferation during neutrophil development. TSA moderately increased CD34+ progenitor expansion in a dose dependent manner, while 100–500mM SB and 500mM VPA abrogated expansion, eventually leading to decreased cell survival. In contrast, 100–200mM VPA increased cell expansion without affecting differentiation potential. Neutrophil differentiation was improved, although not significantly, by TSA, while SB inhibited neutrophil differentiation showing dysplastic features and signs of early apoptosis. Treatment with 500μM VPA also inhibited neutrophil differentiation reflected by an increased amount of common myeloid progenitors (CMP), a differentiation block in immature neutrophils and decreased cell survival. In order to determine the effects of HDAC inhibition on histone acetylation we performed Western Blot analysis on protein lysates from hematopoietic progenitors. SB and VPA strongly increased acetylation of total H4 with hyperacetylation of H4K16 and this effect was observed to a lesser extent with TSA. During myeloid differentiation, all three HDAC inhibitors clearly increased total H3 acetylation as represented by H3K9 hyperacetylation. Taken together, we observed comparable effects on histone acetylation profiles, but distinct effects on myeloid progenitor function and neutrophil development. These data suggest that the effects observed in hematopoietic progenitor function are likely to include effects of HDAC inhibitors on non-histone targets. Furthermore, the differences between the effects of TSA, SB and VPA suggest inhibitor specific effects (HDAC preference) or cell specific responses. We are currently investigating the molecular mechanisms underlying these observations.