The function and regulation of histone deacetylase-4 in neurons

Reversible acetylation is a post-translational protein modification of central importance in many cellular processes, and therefore study of the enzymes that regulate acetylation is an important topic in biology. Here we describe the function and regulation of histone deacetylase-4 (HDAC4), a member of the class IIa HDAC family. HDAC4 undergoes phosphorylation-dependent shuttling between the cytoplasm and nucleus during muscle differentiation. We have sought to expand the understanding of HDAC4 function to the brain, where HDAC4 is highly expressed. We find that HDAC4 is primarily cytoplasmic in neurons both in vivo and in culture. However, in response to conditions that induce neuron death, HDAC4 rapidly translocates into the nucleus. Treatment with other factors or inhibitors already known to affect neuron death also alters HDAC4 subcellular distribution. Moreover, ectopic expression of nuclear-localized HDAC4 promotes neuronal apoptosis. In contrast, inactivation of HDAC4 by siRNA or HDAC inhibitors suppresses neuronal cell death. To address the mechanism of HDAC4-mediated neuron death, we showed that HDAC4 represses the transcriptional activities of MEF2 and CREB, survival factors in neurons. MEF2 can be modified by sumoylation, and HDAC4 has previously been shown to enhance MEF2 sumoylation. Because MEF2 is degraded in conditions of neuron death, we hypothesized that sumoylation of MEF2 contributes to its degradation. We show that SUMO-deficient mutants of MEF2 are more stable and more potent at reducing neuron death than wild-type HDAC4. Knockdown of HDAC4 also stabilizes MEF2. We also implicate HDAC4 and MEF2 in neurodegenerative disease. They both associate with the polyglutamine disease protein ataxin-1, and localize to ataxin-1 nuclear inclusions. However, HDAC4 and MEF2 association with a less-toxic mutant of ataxin-1, S776A, is greatly reduced. Their role in pathogenesis is unclear, but ataxin-1 represses MEF2-dependent transcription, though the S776A mutant does not do so as effectively. Intriguingly, HDAC4 is also localized to postsynaptic structures in neurons. Although it is not known whether HDAC4 has a synaptic function, a number of candidates for acetylated synaptic proteins were identified. All together, this shows that HDAC4 has important functions in the brain, and it is a reasonable therapeutic target for preventing neuron loss in various pathologies.