Epilepsy and intellectual disability often occur alongside autism spectrum disorder (ASD), and researchers have long suspected that these conditions share common biological roots¹. Two gene classes stand out as particularly strong risk factors: those encoding chromatin remodelers, which govern how genes are regulated in a given cell, and synaptic proteins, which mediate communication between neurons. Large genetic studies implicate both categories^2,3, yet how chromatin remodelers connect to synaptic function at the mechanistic level has remained unclear.

A new study published in Advanced Science addresses that question directly. Tingting Wang, a former Bridge to Independence (now Fellows-to-Faculty) Fellow, led the work alongside colleagues, including fellows in the Simons Undergraduate Research Fellows in Neuroscience (SURFiN) program.

The team focused on CHD2, a gene encoding a chromatin remodeler that is associated with ASD, epilepsy, intellectual disability, and motor problems^4,5 and is known to regulate genes encoding synaptic proteins⁶. To study its function, they targeted the Drosophila homolog of CHD2, Chd1, generating flies carrying one of several loss-of-function mutations in this gene. These mutants were more prone to seizures when exposed to external stressors and showed pronounced motor difficulties, two traits corresponding with clinical features seen in people carrying CHD2 mutations^4,5.

The researchers then investigated how loss of Chd1 affects synaptic function. For neural circuits to function properly, synaptic connections must maintain stable signaling even when components are altered or compromised, for example when receptor function is reduced. The nervous system achieves this through intrinsic compensation mechanisms such as presynaptic homeostatic potentiation (PHP). During PHP, the presynaptic neuron responds to weakened postsynaptic detection by releasing more neurotransmitters, effectively restoring the strength of the connection⁷.

Conserved across species and implicated in ASD, epilepsy, and neurodegenerative disease, PHP is thought to play a fundamental role in maintaining stable neural circuit activity^8,9. The researchers found that Chd1 mutant flies were unable to carry out PHP, pointing to a role for this chromatin remodeler in synaptic homeostasis.

PHP can occur rapidly, within minutes of a perturbation, or can be sustained over a longer period when receptor function is persistently compromised. The requirement for Chd1 varied depending on the cell type and the form of PHP. Selectively silencing Chd1 in specific tissues revealed that perineurial glia, cells that envelop peripheral nerves at the neuromuscular junction, were essential for rapid PHP. Long-term PHP, however, depended on Chd1 across multiple cell types, including motor neurons, muscle, and perineurial glia.

Restoring Chd1 specifically in perineurial glia rescued rapid PHP in mutant flies, reinforcing the central role of these cells. The findings add to growing evidence that glia, which express ASD-associated genes and influence neuronal function, play a critical and understudied role in synaptic regulation^10,11.

To understand why PHP failed, the researchers looked more closely at the cellular mechanisms involved. Without functional Chd1, synapses were unable to make the compensatory adjustments needed to restore stable signaling. Specifically, presynaptic neurons showed no increase in calcium entry when stimulated, nor did they expand their pool of neurotransmitter vesicles ready for release, two changes that normally accompany PHP.

The team then turned to identifying the specific genes through which Chd1 exerts its effects, conducting a genetic screen guided by machine learning. Among 14 candidate genes identified, the authors prioritized cadherin 74A (Cad74A), which encodes an adhesion molecule that may help mediate communication between glia and neurons. Its expression dropped by roughly 80% in flies lacking functional Chd1.

Silencing Cad74A specifically in perineurial glia disrupted PHP, and whole-animal Cad74A loss-of-function mutants showed motor impairments similar to those of Chd1 mutants. Further supporting this relationship, flies with reduced expression of both Chd1 and Cad74A showed complete loss of PHP, even though reducing the expression of either gene alone was not enough to disrupt it. These observations establish that Cad74A and Chd1 act in the same pathway.

In revealing how chromatin remodeling shapes synaptic homeostasis, this research positions glial gene regulation as a critical component of circuit stability. It also opens the door to investigating how failures in synaptic compensation contribute to ASD and related disorders, particularly if comparable mechanisms are at work in mammals. If so, the findings could open up new avenues for intervention.

1. Jeste S.S. et al. J. Child Neurol. 30, 1963–1971 (2015) PubMed
2. De Rubeis S. et al. Nature 515, 209–215 (2014) PubMed
3. Iossifov I. et al. Nature 515, 216–221 (2014) PubMed
4. Carvill G.L. et al. Nat. Genet. 45, 825–830 (2013) PubMed
5. Chénier S. et al. J. Neurodev. Disord. 6, 9 (2014) PubMed
6. Kim Y.J. et al. Neuron. 100, 1180–1193.e6 (2018) PubMed
7. Davis G.W. and Müller M. Annu. Rev. Physiol. 77, 251–270 (2015) PubMed
8. Ramocki M.B. and Zoghbi H.Y. Nature 455, 912–918 (2008) PubMed
9. Orr B.O. et al. Neuron 107, 95–111.e6 (2020) PubMed
10. Eroglu C. and Barres B.A. Nature 468, 223–231 (2010) PubMed
11. Velmeshev D. et al. Science 364, 685–689 (2019) PubMed