The BAF chromatin remodeling complex is essential for normal brain development and is a frequent locus of rare genetic variation underlying neurodevelopmental disorders, including autism spectrum disorder (ASD). Despite this, the factors that direct BAF binding across the human neuronal genome and the specific neuronal gene programs regulated by BAF have been incompletely defined. The present study addresses these gaps by characterizing where BAF binds in human neurons, how binding relates to stimulus-dependent chromatin and enhancer activation, and how perturbation of BAF subunits affects expression of ASD-associated genes.
Using human neuronal models, the authors identify a set of distal regulatory regions where the BAF complex co-binds with the activity-dependent transcription factor FOS. These regions show chromatin opening and features of enhancer activation in response to neuronal stimulus. The co-occupancy of BAF and FOS at distal elements suggests a collaborative mechanism in which BAF remodels chromatin to permit or facilitate enhancer activation downstream of neuronal activity-driven FOS recruitment.
The study tested the effect of perturbing a BAF subunit, ARID1A, which has been implicated in ASD. Knock-out of ARID1A resulted in decreased chromatin accessibility at sites bound by FOS and BAF. This reduction in accessibility was accompanied by decreased expression of nearby genes that are both activity-regulated and associated with ASD. These observations link loss of a BAF subunit to impaired enhancer accessibility and blunted activity-dependent transcription of ASD-relevant genes in human neurons.
The authors report that the FOS binding motif within regions co-bound by FOS and BAF is highly constrained in the human population. High constraint implies evolutionary or functional intolerance to variation at these motif positions, consistent with an important regulatory role. This population-level signal motivated evaluation of whether rare variation in these motifs occurs in individuals affected by ASD and whether such variants have functional consequences.
Rare variants in the FOS motif within FOS/BAF-bound regions were identified in ASD-affected individuals. Experimental assessment indicated that these rare variants disrupt stimulus-dependent enhancer activation. In other words, changes within the FOS motif reduced the ability of enhancers to respond to neuronal activity, consistent with impaired recruitment or function of FOS and, by extension, disrupted cooperation between FOS and BAF at these regulatory sites.
Integrating these findings, the authors propose a model in which the cooperative recruitment of FOS and BAF to distal enhancers is required for activity-dependent activation of ASD-associated gene programs in human neurons. Two perturbations converge on this mechanism: (1) loss of a BAF subunit such as ARID1A, which reduces chromatin accessibility at FOS/BAF sites and lowers expression of nearby activity-regulated ASD genes; and (2) rare non-coding variants within constrained FOS motifs that impair stimulus-dependent enhancer activation and thus prevent appropriate transcriptional responses to neuronal activity.
This model provides a mechanistic link between chromatin remodeling, activity-dependent transcription factor binding, enhancer activation, and genetic contributions to ASD. It highlights how both coding variation affecting chromatin remodelers and non-coding variation affecting transcription factor motifs can converge on the same regulatory pathway to alter neuronal gene expression programs.
The study emphasizes a role for the BAF complex in mediating neuronal transcriptional programs downstream of FOS. It demonstrates that ARID1A loss impairs chromatin accessibility at co-bound sites and reduces expression of activity-regulated, ASD-associated genes. It further shows that the FOS motif in these regions is under population constraint and that rare motif variants in ASD-affected individuals can disrupt stimulus-dependent enhancer activation. Collectively, these results support a model in which non-coding genetic variation in FOS/BAF-bound enhancers contributes to ASD risk by preventing recruitment of FOS and BAF and thereby diminishing activity-dependent gene expression in human neurons.
The authors declare no competing interests and acknowledge funding sources reported in the source. Specific experimental details, quantitative measures, sample sizes, and methods used were reported in the original preprint but are not reproduced here in full; readers should consult the source for methodological specifics and supplementary data.