HDAC inhibitors dysregulate neural stem cell activity in the postnatal mouse brain

• HDAC1 and 2 are in distinct progenitors within different postnatal neurogenic niches.
• Postnatal proliferation rates vary across CNS neurogenic niches.
• Valproate decreases progenitor proliferation and postnatal neurogenesis in vivo.
• Proliferation in the DG and RMS is differentially sensitive to nutritional status.
• Treatment with HDAC inhibitors in vitro dysregulates neural stem cell activity.

The mammalian central nervous system (CNS) undergoes significant expansion postnatally, producing astrocytes, oligodendrocytes and inhibitory neurons to modulate the activity of neural circuits. This is coincident in humans with the emergence of pediatric epilepsy, a condition commonly treated with valproate/valproic acid (VPA), a potent inhibitor of histone deacetylases (HDACs). The sequential activity of specific HDACs, however, may be essential for the differentiation of distinct subpopulations of neurons and glia. Here, we show that different subsets of CNS neural stem cells (NSCs) and progenitors switch expression of HDAC1 and HDAC2 as they commit to a neurogenic lineage in the subventricular zone (SVZ) and dentate gyrus (DG). The administration of VPA for only one week from P7–P14, combined with sequential injections of thymidine analogs reveals that VPA stimulates a significant and differential decrease in the production and differentiation of progeny of NSCs in the DG, rostral migratory stream (RMS), and olfactory bulb (OB). Cross-fostering VPA-treated mice revealed, however, that a postnatal failure to thrive induced by VPA treatment had a greater effect on DG neurogenesis than VPA action directly. By one month after VPA, OB interneuron genesis was significantly and differentially reduced in both periglomerular and granule neurons. Using neurosphere assays to test if VPA directly regulates NSC activity, we found that short term treatment with VPA in vivo reduced neurosphere numbers and size, a phenotype that was also obtained in neurospheres from control mice treated with VPA and an alternative HDAC inhibitor, Trichostatin A (TSA) at 0 and 3 days in vitro (DIV). Collectively, these data show that clinically used HDAC inhibitors like VPA and TSA can perturb postnatal neurogenesis; and their use should be carefully considered, especially in individuals whose brains are actively undergoing key postnatal time windows of development.