SIRT1 is a critical regulator of K562 cell growth, survival, and differentiation

Inhibition of histone deacetylases (HDACi) has emerged as a promising approach in the treatment of many types of cancer, including leukemias. Among the HDACs, Class III HDACs, also known as sirtuins (SIRTs), are unique in that their function is directly related to the cell's metabolic state through their dependency on the co-factor NAD(+). In this study, we examined the relation between SIRTs and the growth, survival, and differentiation of K562 erythroleukemia cells. Using a mass spectrometry approach we previously developed, we show that SIRT expression and deacetylase activity in these cells changes greatly with differentiation state (undifferentiated vs. megakaryocytic differentiation vs. erythroid differentiation). Moreover, SIRT1 is crucially involved in regulating the differentiation state. Overexpression of wildtype (but not deacetylase mutant) SIRT1 resulted in upregulation of glycophorin A, ~2-fold increase in the mRNA levels of α, γ, ε, and ζ-globins, and spontaneous hemoglobinization. Hemin-induced differentiation was also enhanced by (and depended on) higher SIRT1 levels. Since K562 cells are bipotent, we also investigated whether SIRT1 modulation affected their ability to undergo megakaryocytic (MK) differentiation. SIRT1 was required for commitment to the MK lineage and subsequent maturation, but was not directly involved in polyploidization of either K562 cells or an already-MK-committed cell line, CHRF-288-11. The observed blockage in commitment to the MK lineage was associated with a dramatic decrease in the formation of autophagic vacuoles, which was previously shown to be required for K562 cell MK commitment. Autophagy-associated conversion of the protein LC3-I to LC3-II was greatly enhanced by overexpression of wildtype SIRT1, further suggesting a functional connection between SIRT1, autophagy, and MK differentiation. Based on its clear effects on autophagy, we also examined the effect of SIRT1 modulation on stress responses. Consistent with results of prior studies, we found that SIRT1 silencing modestly promoted drug-induced apoptosis, while overexpression was protective. Furthermore, pan-SIRT inhibition mediated by nicotinamide pre-treatment substantially increased imatinib-induced apoptosis. Altogether, our results suggest a complex role for SIRT1 in regulating many aspects of K562 cell state and stress response. These observations warrant further investigation using normal and leukemic primary cell models. We further suggest that, ultimately, a well-defined mapping of HDACs to their substrates and corresponding signaling pathways will be important for optimally designing HDACi-based therapeutic approaches.