Skip to content

Review: The Role of Histone Deacetylases in Autism Spectrum Disorder

Key findings

  • This review reports what is known about associations between autism spectrum disorder (ASD) and histone deacetylases (HDACs), enzymes that can suppress gene transcription
  • HDACs might interact with environmental risk factors for ASD, have a role in core ASD symptoms, underlie genetic disorders that are related to ASD, and explain why certain conditions are common comorbidities of ASD
  • Most studies have been preclinical, but human research is underway to investigate how HDACs are involved in the pathophysiology of ASD
  • Another major research goal is to determine whether HDAC inhibition can mitigate ASD symptoms

The causes of autistic spectrum disorder (ASD) remain largely unknown. Still, it's increasingly clear that prenatal environmental exposures can produce epigenetic marks—chemical modifications to DNA and histone proteins that modulate gene expression.

In Biological Psychiatry, researchers at Massachusetts General Hospital review relationships between ASD and histone deacetylases (HDACs). These epigenetic enzymes lead to a change in the chromatin structure of a cell through tighter wrapping of the DNA on histones, resulting in reduced gene transcription. Therefore, altered HDAC expression/activity can impact gene transcription.

The authors are Chieh-En Jane Tseng, PhD, research fellow, Nicole R. Zürcher, PhD, assistant professor in the Department of Radiology, Jacob M. Hooker, PhD, scientific director of the Lurie Center for Autism and director of radiochemistry at the Athinoula A. Martinos Center for Biomedical Imaging, and Christopher J. McDougle, MD, director of the Lurie Center for Autism in the Department of Psychiatry at Mass General.

HDACs and ASD Environmental Risk Factors

Some of the environmental factors linked to ASD—parental age, maternal stress, suboptimal maternal nutrition, maternal obesity, prenatal or early-life immune challenges, and in utero exposure to toxins—have also been shown to alter HDAC expression/activity.

The evidence that most strongly implicates HDACs in ASD etiology comes from studies of prenatal exposure to HDAC inhibitors. Most of these have been conducted in mouse models. However, a population-based study reported in JAMA showed children of women who used valproic acid, a weak HDAC inhibitor, during pregnancy had a significantly increased risk of ASD.

HDACs and Social Behavior

Likewise, most data that implicate HDACs in social behavior come from animal models:

  • Prenatal HDAC inhibition—Numerous research groups have reported associations between prenatal exposure to valproic acid and diminished social behavior in rodents and nonhuman primates. Examples include reductions in play, sociability, and social communication. The long-term effects of prenatal HDAC inhibition have not been investigated
  • Postnatal HDAC inhibition—In rodent models of ASD, postnatal treatment with HDAC inhibitors frequently improves social deficits. It seems counterintuitive that postnatal HDAC inhibition improves symptoms, whereas prenatal inhibition leads to social deficits, but regional HDAC levels may differ at different developmental phases. Alternatively or in addition, changes in behavior may be compensatory effects

Ketogenic diets have been associated with improved social behavior in small-scale studies of children with ASD and animal models. The major product of ketogenic diets, β-hydroxybutyrate, is transported into the central nervous system and inhibits class I HDACs, suggesting these diets deserve further study.

HDACs and Restricted/Repetitive Behaviors

Rodents exposed prenatally to valproic acid exhibit restricted/repetitive behavior patterns such as repeated entry into the same arm in a Y maze and increased time spent grooming and digging. Findings are mixed about how postnatal treatment with an HDAC inhibitor affects these behaviors.

Postnatally, restricted/repetitive behaviors in rodents can be produced by either conditional knockout or overexpression of Hdac, depending on the timing of the intervention.

So far there are no human studies of how HDAC levels relate to restricted/repetitive behaviors.

HDACs and Genetic Disorders Linked to ASD

Emerging evidence implicates chromatin-modifying genes in the pathophysiology of ASD. A notable example is a whole-exome sequencing study reported in Cell, which found several risk genes for ASD that coded for proteins or histone-modifying enzymes that interact with HDACs.

Furthermore, in multiple genetic disorders associated with ASD, such as Rett syndrome and fragile X syndrome, a genetic alteration leads to aberrant HDAC expression or loss of HDAC recruitment. Others, including Angelman syndrome and Prader–Willi syndrome, are genomic imprinting disorders that cause genes to be expressed from only one parental allele, possibly due to differences in histone acetylation of the maternal and paternal alleles.

HDACs and Comorbidities of ASD

Common comorbidities of ASD include epilepsy, anxiety disorders, intellectual disability, attention-deficit/hyperactivity disorder (ADHD), sleep disorders, and gastrointestinal disorders. There is strong evidence for HDAC alterations in epilepsy, anxiety, and intellectual disability, and preclinical evidence of links between HDACs and ADHD, sleep disorders, and GI disorders.

Looking Ahead

Additional human research is underway to investigate how epigenetic alterations are involved in the pathophysiology of ASD. These studies should also reveal whether therapeutically targeting epigenetic marks, including treatment with HDAC inhibitors, can mitigate ASD symptoms.

Learn more about research in the Department of Psychiatry

Learn more about the Martinos Center for Biomedical Imaging

Related topics


Massachusetts General Hospital researchers have reported the first in vivo human evidence of a role for histone deacetylases in the pathophysiology of bipolar disorder as well as in two of its hallmark clinical features, emotional dysregulation and attention deficits.


Daniel Lee, MD, and Ziv M. Williams, MD, of the Department of Neurosurgery, and colleagues have demonstrated that changes in social behavior due to disruption of the Shank3 gene in mice relate to the ability of neurons to encode information about other individuals—and recovery of Shank3 gradually increases sociability.