Valproic Acid Expands the Numbers of HSCs from Cord Blood CD34⁺ Cells By Linking Epigenetic Modifications to Mitochondrial Remodeling and p53 Upregulation

Umbical cord blood (UCB) is an important source of hematopoietic stem cells (HSCs) for allogeneic transplantation of patients suffering from hematological malignancies and genetic disorders. Its use, however, as an HSC graft for adults has been restricted due to the limited numbers of HSCs within a single UCB unit. We have previously reported that UCB CD34⁺ cell numbers can be expanded over a seven-day period of incubation by ex vivo treatment with a cytokine combination and the histone deacetylase inhibitor, valproic acid (VPA) (Chaurasia et al. J Clin Invest. 2014:124(6): 2378-2395). In this study, we addressed the mechanism underling this ex-vivo expansion of UCB-derived CD34⁺cells. We found that VPA triggered two distinct phases of HSC behavior. The first phase (days 1-4) was characterized by a prompt elevation of the expression of stem cell phenotypic markers (CD90 and CD49f), which was associated with acetylation of histone H3, and expression of pluripotency genes including OCT4, NANOG, and SOX2. During this period, the cells underwent a limited number of cell divisions. The second phase (days 4-7) was characterized by a greater number of cell divisions that resulted in a drastic increase in the absolute number of phenotypically defined HSCs. Both acetylation of H3 and expression of the pluripotency genes were transient events since they were no longer evident during the second phase. During the entire incubation period, VPA-induced HSC cells maintained a distinct mitochondrial metabolic profile characterized by low mitochondrial potential, ROS levels and mass. Remarkably, the p38 activity, a downstream target of ROS, was also suppressed in VPA-expanded HSCs. Given that both ROS and p38 activity lead to exhaustion and limit the lifespan of the pool of HSCs, our findings suggest that the VPA-expanded HSCs are equipped with mechanisms that monitor and ensure their fitness.In addition, removal of VPA from the culture after 24, 48 and 72hrs of incubation reversed the acetylation of H3 and the expression of the pluripotency genes. These events were further accompanied by the rapid loss of stem cell marker expression (24hrs after removal) leading to decreased HSC numbers. Importantly, upon VPA removal, the expanded cells exhibited an increase in mitochondrial ROS, mass and potential suggesting that the epigenetic reprograming and mitochondrial remodeling triggered by VPA were linked and were both reversible events.

Both transcript and protein levels of p53 were also increased during the first phase but not during the second phase of the ex-vivo expansion of CD34⁺cell. Importantly, removal of VPA diminished the up-regulation of p53 during the first phase. We further assessed the role of p53 in VPA stem cell expansion with pifithrin-α, a small molecule inhibitor of p53 function. P53 inhibition led to a significant elevation of the mitochondrial ROS without affecting cellular viability. Moreover, treatment with pifithrin-α reduced both the percentage and the absolute number of HSCs induced by VPA. Conversely, co-treatment with VPA and a low concentration of nutlin (1μM), an inhibitor of p53 degradation decreased further mitochondrial ROS levels and augmented the percentage of HSCs induced by VPA. By contrast, treatment with higher concentrations of nutlin (10μM) led to HSC cell death. These findings raise the possibility that VPA induces a moderate up-regulation of p53 as a stress-response mechanism to cope with oxidative stress. It seems plausible that p53 along with alterations in mitochondrial mass and potential are among the mechanisms required to repress oxidative stress and thus, are implicated in the HSC expansion by VPA.

Collectively, our data establishes that VPA expands ex-vivo the pool of HSCs from CD34⁺ CB cells by triggering epigenetic modifications that are tightly linked to mitochondrial remodeling. Notably, these events are reversible and do not result in the malignant transformation of the CD34⁺ cells. Moreover, these data indicate that VPA modulates and maintains a critical threshold of p53 that is required for HSCs expansion but is not sufficient to trigger the exhaustion and death of HSCs. In this context, VPA also represses p38 activity that if up-regulated can limit the lifespan of HSCs. These studies demonstrate that effective ex vivo HSC expansion requires the coordination and retention of a number of critical biological events that define an HSC