Modulation of Chromatin Architecture Influences the Neuronal Nucleus through Activity-Regulated Gene Expression

Abstract

The disruption of chromatin-regulating genes is associated with many neurocognitive syndromes. While most of these genes are ubiquitously expressed across various cell-types, many chromatin regulators act upon activity regulated genes (ARGs) that play central roles in synaptic development and plasticity. Recent literature suggests a link between ARG expression disruption in neurons with the human phenotypes observed in various neurocognitive syndromes. Advances in chromatin biology have demonstrated how chromatin structure, from nucleosome occupancy to higher-order structures such as topologically associated domains, impacts the kinetics of transcription. This review discusses the dynamics of these various levels of chromatin structure and their influence on the expression of ARGs.

Introduction:

Chromatin regulators have emerged as a major genetic contributor to neurocognitive diseases such as intellectual disability and Autism Spectrum Disorder [1–3]. Studies over the last decade have identified that many chromatin regulators are required for the accurate expression of ARGs. Experience or activity-dependent transcription during mammalian postnatal development is required to direct synapse formation and subsequent pruning of extra synapses. Indeed a major hypothesis of autism pathogenesis relates to failed synaptic pruning during neurodevelopment, leading to an improper balance between excitation and inhibition in the mature brain [4,5].

Activity-regulated gene expression refers to the process whereby cells respond to external stimuli through a series of cellular signals that results in new gene transcription that enacts adaptation or growth. This process occurs in all cells, but in the post-mitotic neuron, these stimuli usually represent synaptic stimulation via neurotransmitters. This leads to depolarization via ion channels, calcium influx, and the initiation of signaling cascades that ultimately induce ARG transcription. ARGs are often further divided into immediate early genes (IEG) that are transcribed within minutes of stimulation and late response genes (LRG) that are transcribed within hours of stimulation. IEGs include ubiquitously expressed genes such as Fos, Jun, or Nr4a1. Many of these IEGs are themselves transcription factors that induce the transcription of cell- or context-specific LRGs, such as Bdnf in the neuron. Most IEGs are not necessarily cell-specific, but their actions can enact cell-specific functions through cooperation with cell-specific transcription factors or by acting upon cell-specific enhancers [6–8].

Chromatin regulators tend to be ubiquitously expressed throughout the body, and they function in many transcriptional processes in the cell. Nonetheless, it appears that genetic disruption of chromatin regulators frequently results in neurocognitive phenotypes. Scientists have therefore searched for common pathways affected to explain this predilection. Among other proposed mechanisms, chromatin regulation of ARGs is a common pathogenetic mechanism widely studied in neurodevelopmental disorders. As an example of how this hypothesis has been explored, a recent study depleted five chromatin regulators with diverse functions (ASH1L, CHD8, CREBBP, EHMT1, and NSD1) that are all associated with autism in neuron cultures [9]. In this work, the authors identified the downregulation of neurotransmitters and ARGs as a common trend across these different types of chromatin regulators.

Many chromatin regulators have been described as orchestrating key steps in ARG expression, and these components are implicated in neurological disorders [7]. Additionally, earlier work has identified characteristic changes in histone modifications that correlate with the activation of ARGs [8,10]. In the last decade, the chromatin field has increasingly focused on higher-order chromatin structure within the nucleus. New technologies, such as Hi-C, single-cell sequencing, and super-resolution microscopy techniques have become accessible to query the function of these higher order structures.

Other excellent reviews have characterized how chromatin regulators modulate the function of brain-specific transcription factors at ARG promoters and enhancers [11–14]. In this review, we will discuss recent literature on how chromatin regulators modulate nucleosomal and higher-order chromatin structure to regulate neuronal ARG expression. Figure 1 illustrates examples of the multiple levels of chromatin reorganization following neuronal depolarization. Table 1 highlights the key literature reviewed and the model systems used.

Figure 1:

Neuronal depolarization leads to changes to chromatin structure at multiple levels. 1: At the smallest order, nucleosomes are deposited and ejected at specific loci in response to stimulation. This includes the deposition of histone variants, including H3.3, in post-mitotic neurons. 2: Chromatin is remodeled in response to stimulation, particularly at ARG promoters and enhancers. BAF cooperates with a variety of factors to coordinate the release of repression by factors such as NuRD and activation of loci through activity-inducible and cell-specific transcription factors. 3: Activity induces changes in short and long-range intrachromosomal contacts. BAF coordinates with cohesin to establish activity-dependent loops. Immediate early genes (IEGs) tend to make shorter loops, while LRGs tend to make longer and more complex loops. 4: At the highest level, activity induces changes in TAD and LAD structure for facilitated transcription of genes necessary for activity-induced synaptic plasticity. Chromatin regulators, including SATB2, facilitate changes of position within the nucleus, including association with the nuclear lamina.

Table 1.

Summary of Key Literature

Ref #. Authors (First, Last); Journal, Year; PMID	Key Point	Model System	Level of Chromatin Organization
15. Maze, Allis; Neuron, 2015; 26139371	Neuronal activity induces HIRA-mediated H3.3 deposition in ARG gene bodies.	Mouse and human brain tissue; mouse neuron culture	Histone deposition/turn- over
22. Ibrahim, Hamiche; Science, 2021; 34324427.	MeCP2 is a chromatin organizer that blocks nucleosome invasion through CA-repeat binding.	Various mouse cell lines and mouse brain tissue.	Histone deposition/turn-over
25. Wenderski, Gleeson; PNaS, 2020. 32312822	nBAF serves to repress ARG expression (particularly AP1-mediated factors) at rest. This study also highlights the human genetics of nBAF (BAF53b) loss.	Cultured mouse neurons; nBAF KO mouse model; Human patient samples	Chromatin Remodeling
27. Yang, Bonni; Science, 2016; 27418512	By analyzing actively translating mRNAs in vivo, they show NuRD inactivates ARGs through H2AZ deposition. NuRD is required for normal dendritic development and learning.	Mouse in vivo studies using electroporation to deplete NuRD	Histone deposition; chromatin remodeling
33. Vierbuchen, Greenberg; Mol Cell, 2017; 29272704	A genetic screen for enhancer selection identified AP-1, which collaborates with BAF and cell-specific transcription factors during differentiation.	Mouse cell culture (mostly fibroblasts, but also neurons)	Chromatin remodeling
38. Kim, Wu; Cell Rep, 2021; 34260936.	Activity-mediated phosphorylation of BRG1 coordinates a switch in co-factor binding which facilitates ARG induction.	Cultured mouse neurons; BRG1-mutant mice	Chromatin Remodeling and enhancer-promoter looping
39. Calderon, Merkenschlager; eLife, 2022; 35471149	By comparing long-term and acute cohesin depletion, ARGs require looping for full expression throughout neuronal life.	Mouse neuron cultures (conditional knock out models)	Enhancer-promoter looping
51. Winick-Ng, Pombo; Nature, 2021. 34789882	They used a novel method to map chromatin topology in brain tissues to identify specific neuron-cell-type-specific TADs.	Mouse brain tissue	TADs
52. Feurle, Dechant. EMBO, 2020; 33319920	This study identified changes to nuclear membrane morphology following stimulation that are mediated by SATB2-LEMD2 interactions.	Mouse brain cortex	Nuclear membrane and LADs

Nucleosome Deposition and Histone Turnover

At the nucleosome level, early chromatin studies identified combinatorial patterns of post-translational histone modifications that are associated with specific transcriptional states. However, recent work has focused on the nucleosome particle itself and how its placement or exclusion in specific loci regulates ARG expression. Interestingly, the studies described in this section have teased apart neuronal activity-driven histone deposition from the enrichment of characteristic histone post-translational modifications (PTMs), such as activating marks (e.g. methylation at H3K4). The study by Maze, et al., suggested that activity-dependent histone turnover is an independent process from the placement of histone PTMs [15].

In addition to replacing histones (e.g. following transcription by RNA or DNA Polymerase), histone variant exchange and nucleosome turnover contribute to transcriptional regulation [16]. Nucleosome turnover occurs through replication-dependent and independent mechanisms. Post-mitotic cells, such as neurons, have been used as a tool to study the role of replication-independent nucleosome turnover over the cellular lifespan. H3.3 is a replication-independent histone variant that, unlike replication-dependent H3.1 or H3.2, may be incorporated into the nucleosome at any point in the cell cycle. As neurons exit the cell cycle, they continue to exhibit dynamic nucleosome turnover through replication-independent histone deposition mechanisms. Overtime with ongoing nucleosome exchange, H3.3 accordingly increases in abundance throughout neuronal life in the mammalian brain [15,17].

Multiple studies have identified that depolarization induces H3.3 deposition into the genome [15,18]. Distinct histone chaperones mediate this process and act at specific genomic positions. The histone chaperone DAXX deposits H3.3 at the promoters of ARGs upon neuronal activation, but only at selected ARGs such as c-Fos and Bdnf [18]. Another study performed RNA-Seq and H3.3 ChIP-Seq at time points after depolarization and found that neuronal activity leads to HIRA-mediated gene body deposition of H3.3 [15]. This effect was specific to LRGs as opposed to the IEGs. They also characterized in vivo effects of H3.3 loss which led to a failure of LRG expression, a decrease in CA1 dendritic spine formation, and a failure of mice to perform correctly in hippocampal learning & memory paradigms such as novel object recognition or contextual fear conditioning [15]. With both examples of histone chaperones depositing H3.3 at only selected ARGs, there are likely other molecules to be identified that contribute to activity-dependent histone variant exchange.

In addition to histone turnover and deposition, factors involved in nucleosome exclusion can modulate ARG expression. MeCP2 is a well-known methyl-cytosine binder that is associated with Rett syndrome. There has been much work on the role of MeCP2 in dampening gene expression through methylated-DNA binding, which has been previously reviewed [7]. MeCP2 is highly abundant in neurons such that it nearly matches the expression of histone octamers [19]. MeCP2 has been shown to compete with linker histone H1, which is evidence of its role in shaping global chromatin architecture [19–21]. A recent study found that when MeCP2 is bound to CA-repeats, nucleosomes are consequently excluded [22]. MeCP2 KO cells have defects in ARG expression that are correlated with CA content, suggesting MeCP2 binding out-competes nucleosomes to fine-tune gene expression. Ubiquitous factors such as MeCP2 are emerging as major structural remodelers, influencing nucleosome positioning and exclusion throughout the genome. Interestingly, a critique was published that re-analyzed mouse brain arrays in several independent studies and failed to find MeCP2 enrichment at CA repeats, relative to other types of cytosine methylation [23]. Further work will be required to resolve these conflicting data.

Chromatin Remodelers

Chromatin remodelers broadly refer to enzymes that modulate chromatin architecture. In this section, we specifically highlight ATP-dependent chromatin remodeling enzymes that function by moving or restructuring nucleosomes. Following neuronal stimulation, ATP-dependent chromatin remodelers access chromatin on the timescales required for ARG expression. Additionally, several chromatin remodelers have been identified with specialized neuronal functions, many of which are also associated with risk genes for neurodevelopmental disease [24,25].

One well-known chromatin remodeler is the Nucleosome Remodeling and Deacetylase (NuRD) complex. While many studies focus on ARG activation, NuRD activity has been found to mediate ARG inactivation. Under basal conditions, the NuRD complex maintains ARGs in an off state through deposition of the histone H2A variant, H2A.Z. In an in vivo mouse contextual fear stimulus paradigm, NuRD is evicted, and H2A.Z is depleted in ARG transcription start site +1 nucleosomes [26]. NuRD subsequently returns to ARGs to inactivate them following stimulus cessation. One group studied cerebellar cells at time points following cessation of a mouse rotarod activity stimulus and found that CHD4, the core subunit of the NuRD complex, deposits H2A.Z at promoters shortly following the stimulation cessation [27]. NuRD-deficient cells failed to inactivate ARGs, which led to transcriptomic hyper-sensitivity of ARGs and was associated with ineffective dendritic spine pruning. In both studies, disruption of H2A.Z or NuRD, respectively, led to abnormalities in learning and memory behavioral assays [26,27].

The BRG1/BRM-Associated Factor (BAF) complex is another multi-subunit chromatin remodeler shown to be important for ARG expression by cooperating with other activity-induced transcription factors. During neural differentiation, the BAF complex undergoes a subunit exchange from ACTL6A/BAF53a in neural progenitor cells to ACTL6B/BAF53b in mature neurons [28,29]. These neuron-specific “nBAF” components are important for establishing neuronal transcriptomic identity and regulating neuronal developmental processes such as normal dendritic outgrowth [30,31]. More recent work on BAF/nBAF has shown its role in ARG regulation. One study used a BAF53b-CRE (pan-neuronal) model to conditionally knock out the BAF ATPase catalytic subunit, BRG1. They found a defect in the induction of a subset of ARGs that significantly overlapped with the targets of a transcription factor, MEF2C [32]. This defect in transcriptional activation could be overcome with a targeted MEF2C-VP16 construct suggesting that BRG1 acts as a coactivator.

Other activity-induced transcription factors, including AP-1, are known to utilize BAF/BRG1 to access chromatin. BAF remodeling is particularly important during development as AP-1 interacts with cell-specific transcription factors to select and activate cell-specific enhancers [33]. Following terminal differentiation, the association between BAF and AP-1 is important for ARG regulation at both the basal and activated states. Cultured neurons with deletion of the nBAF-defining factor, Baf53b, resulted in the de-repression of AP-1 and NR4A1-activated ARGs in the basal state [25]. They further identified that basal de-repression of IEGs in Baf53b-KO cells led to increases in LRG expression, mimicking an activity response in these cells. Upon activation, Baf53b-KO cells exhibited abnormally high ARG expression, indicating that BAF53B suppresses ARG expression at rest and following activation [25].

The above studies highlight opposite results of nBAF loss on ARG expression, so it appears that nBAF remodeler activity depends on its co-factor association. Figure 2 summarizes BAF interactions at rest and following activation. As cells access chromatin in response to stimuli, activity-inducible transcription factors use BAF/nBAF as a co-factor. Future work could compare canonical versus neuronal BAF in activating ARGs, but it appears that the change in subunits may fine-tune the expression of ARGs as the brain undergoes postnatal development.

Figure 2:

The BAF complex at rest and following stimulation at ARG promoters and enhancers. At rest, the BAF complex maintains ARGs in an inactive state, partly through association with the NuRD complex. The NuRD complex represses loci through deposition of H2AZ variant histones at the TSS +1 site. Activity leads to a sequence of calcium-mediated events that allow for ARG expression through modulation of BAF co-factor association. Phosphorylation of BRG1-Serine-1382 leads to dissociation of the BAF complex with NuRD. Transcription factors, such as AP1, engage with the BAF complex at selected enhancers. Cohesins also associate with the BAF complex to mediate enhancer-promoter looping, all enabling expression of ARGs.

Long-Range Chromatin Contacts

Recent work has characterized changes in long-range intrachromosomal contacts within nuclei during various cellular states and transitions. Borders between topologically associated domains (TADs, described in the next section) are demarcated by the CTCF insulator protein, and structural protein complexes, known as cohesins. Cohesins play an important role in genome organization through loop extrusion that, in combination with CTCF and other factors, helps define long-range contacts and TAD boundaries [34]. Interestingly, these two chromatin organization proteins are genetically associated with intellectual disability [35–37]. Various studies that have depleted CTCF and cohesin components show transcription defects in ARGs and other genes important for neurodevelopment that overlap with known autism risk genes [35,37].

As mentioned above, nBAF binds to activity-inducible transcription factors like AP-1 and MEF2C. However, recent studies have also shown nBAF to associate with cohesin to mediate enhancer-promoter contacts. A recent study showed that neuronal activity leads to cohesin binding at the c-Fos promoter by ChIP-qPCR and decreased enhancer-promoter looping, using 3C assay in neuron cultures and P5 mouse cortices in vivo [38]. They further found that BRG1 deletion in cultured neurons leads to decreased cohesin binding to ARG enhancers and promoters. This study found that neuronal activity leads to the phosphorylation of a BRG1 serine (S1382). This phosphorylation does not change genome localization but rather alters the interactions between BAF and other factors. At rest, they found that BAF binds the NuRD complex, but this interaction is released following depolarization and BRG1 phosphorylation [38]. BAF works in a context-dependent manner and coordinates multiple activators, repressors, and structural molecules, including cohesins allowing transition between basal and activated transcriptional states (Figure 2).

Cohesins, as a structural chromatin protein, also contribute directly to ARG regulation. Many neurodevelopmental genetic studies are complicated by the fact that early insults to neuronal maturation and connectivity can lead to persistent transcriptional defects, including failure of proper ARG expression in mature neurons. Calderon et al. completed an elegant set of experiments to address whether the ARG expression deficiencies seen in chromatin structure disruption are the cause or consequence of neuronal maturation deficits [39]. They compared the consequences of acute cohesin depletion using a proteolytic system in post-mitotic neuron cultures to long-term depletion using a Neurod6-Cre system (which acts upon neurons as they exit the cell cycle and terminally differentiate) and found a highly significant overlap of differential gene expression [39]. Therefore, they concluded that ARG expression is affected by acute changes in chromatin structure and looping independent of neuronal maturation or connectivity changes. Their study also found differential effects of the cohesin-mediated effect between IEGs and LRGs. Both groups are down-regulated in cohesin-deficient neurons, but only a subset of LRGs expression upon stimulation is mediated by cohesin. In particular, the necessity of cohesin in ARG expression depends on the length of the chromatin loop [39]. Interestingly, another study found that IEGs tend to form shorter loops whereas LRGs form longer and more complex loops [40]. Besides cohesins, the role of chromatin regulators in establishing and maintaining complex loops has not been studied. Future work to identify additional structural factors for long-distance contacts may also require single-cell techniques.

Higher-Order Chromatin Structure: TADs and LADs

Since the development of HiC and related methods, changes in higher-order structure due to chromatin disruption have been shown across many disease states. Recent work has used chromatin conformation capture techniques to profile changes in neuronal nuclei following activation and during neurodevelopment [40–42].

Early studies examining neuronal nuclei used FISH to show that following seizure induction, Bdnf alleles moved from the nuclear lamina toward the center of the nucleus [43]. This change in nuclear localization was also associated with decreased ChIP binding to lamina proteins and increased co-localization of active RNA Polymerases (pSer-2 and pSer-5) by FISH. This finding was an early clue that changes in transcriptional state are associated with changes in nuclear localization.

Since this discovery, other promoters have been characterized that are repressed when localized to the nuclear lamina (i.e. organized within lamina-associated domains or LADs) but become active upon recruitment into the nuclear interior [44,45]. LADs are repressive nuclear compartments enriched with heterochromatic histone PTMs such as H3K9-methylation. The study described in an earlier section found that MeCP2 deletion leads to increased nucleosome density mostly in LADs [22]. Within LADs, MeCP2 appears to reduce nucleosome occupancy around transcription start sites and CTCF sites [46]. These findings support the idea that MeCP2 plays a role in higher-order nuclear chromatin structure, particularly in the regulation of repressive or heterochromatic regions, including LADs.

TADs are thought of as 3D genomic regions with increased contact frequency and bounded by insulators, such as CTCF. Initial Hi-C studies additionally identified multi-TAD compartments, whereby the “A” compartment is enriched with active chromatin compared to the “B” compartment enriched with inactive chromatin [47–50]. While TAD organization is largely conserved across cell types and even across species, recent studies have identified neuron-specific genome organization that influences ARG expression. One study analyzed genome folding in mouse brain tissue slices and characterized cell-type specific TADs in oligodendrocytes, pyramidal glutamatergic neurons, and dopaminergic neurons [51]. They found cell-type specific TAD boundaries and reorganization of A/B compartments. In a specific example, they found a TAD specific to pyramidal glutamatergic neurons (compared to dopaminergic neurons) that facilitated contact between Egr1 (an ARG) and transcription factor Neurod1/2 putative binding sites. Therefore, they hypothesized that these cell-type-specific TADs allow for coordinated transcription of ARGs by neuronal transcription factors [51]. Figure 3 summarizes changes in TAD structure seen across brain cell types.

Figure 3:

A representation of the data described in Figure 4 of Winick-Ng et al [51]. A chromosomal region containing ARG, Egr1, is depicted where in Pyramidal Glutamatergic Neurons (PGNs), there are relatively increased contacts formed relative to Dopaminergic Neurons. Deeper red color corresponds to strength of long-distance contact. In PGNs, the authors found that this correlates with an increased frequency of NeuroD2 binding sites by ChIP, as shown by a simplified depiction of ChIP data. As NeuroD2 is a key cell-specific transcription factor in PGNs, the authors hypothesized that this leads to interconnected hubs where NeuroD2 can facilitate the transcription of distal genes with cell-specific functions, such as synaptic plasticity for PGNs.

Specific chromatin remodelers associated with neurodevelopmental disorders have also been identified to change higher-order chromatin structures that modulate ARG expression. SATB2 is a chromatin scaffold protein that interacts with an inner-nuclear membrane protein, LEMD2. One study demonstrated that neuronal activity leads to SATB2-LEMD2-mediated nuclear membrane remodeling that permits enhanced bursting of IEGs such as c-FOS [52]. SATB2 depletion in primary hippocampal cultures leads to improper nuclear membrane folding and defective ARG expression [52]. From the micro to the macro scale within the nucleus, chromatin regulators alter the architecture for the fine-tuned activation and subsequent de-activation of ARGs.

New Technologies for Studying Activity-Dependent Chromatin Dynamics

Next-generation sequencing technologies are increasingly accessible, including ATAC-Seq as a method to evaluate chromatin states. One early study used ATAC-Seq following neuronal stimulation and showed increases in chromatin accessibility that correlated with gene expression changes and known AP-1 binding sites [53]. While most of the studies described in this review focused on neuronal changes, multiple brain cell types respond to activity, and recent work has shown that single-cell ATAC-Seq can make accurate inferences about cellular state by linking putative enhancers to known gene expression [54].

As single-cell technologies become more common, transcriptional bursting has become an important concept. Transcriptional bursts refer to the stochastic process that ultimately leads to RNA synthesis on a single-cell level. Various factors influence burst kinetics including chromatin remodeling and histone modifications, particularly histone acetylation [55]. Nucleosome occupancy at the promoter also influences burst frequency, such that increased nucleosome density is associated with less frequent bursting [56]. ARGs also undergo burst transcription, and one group was able to manipulate the bursting kinetics by using a CRISPR-Cas9 tool to direct chromatin regulators to ARG enhancers. They altered H3K27ac levels using dCas9-p300 and dCas9-HDAC constructs and thereby modulated ARG burst kinetics [57]. Both single-cell- and CRISPR-based methods have the potential to further dissect the steps that chromatin regulators take to enable ARG expression.

Conclusions:

From nucleosome exchange and chromatin remodeling to TADs and association with the nuclear lamina, multiple levels of chromatin structure affect ARG expression. These structural changes are important not only for initial IEG induction but also for subsequent phases of LRG expression and even ARG inactivation following cessation of the stimulus. Although many of these transcriptional and epigenetic characteristics are not unique to neurons, much work has characterized the extensive diversity of chromatin structural mechanisms that a post-mitotic neuron must use to encode various stimuli.

Altogether, these studies emphasize several important principles in studying chromatin regulators and ARGs in the context of human neurocognitive disease. First, these studies show the importance of analyzing chromatin dynamics and transcriptional output at both basal and stimulated states. Second, chromatin regulators play crucial roles in various steps in neuronal differentiation and in regulating post-mitotic cellular plasticity. Therefore, the timing of genetic manipulation of chromatin regulators may lead to differing results, as shown in many studies including some reviewed here [17,39]. Finally, chromatin factors have variable effects on different ARGs. Indeed, different types of neuronal stimulation induce distinct gene expression programs [58]. While the studies described in this review cover several chromatin regulators, combinatorial involvement of additional, unstudied chromatin factors is likely required to generate specific ARG expression changes in response to varied stimuli. The specificity required from chromatin regulators is even more complex given the many different types of neurons, circuits, and brain regions.

Model systems for the study of activity-dependent chromatin regulation are another important consideration. Many of the studies described here use in vivo stimulation paradigms and collect tissue for genomic and chromatin profiling. However, many studies also use neuron culture models to be able to manipulate the systems studied. Since neuron culture systems do not mature in the same way as in vivo systems, particularly regarding synaptic pruning, it may be difficult to extrapolate conclusions from these experiments [59].

While this review focused on neuronal chromatin regulation of ARGs, non-neuronal brain cells also undergo dynamic ARG regulation in response to stimuli. One study reviewed here describes how astrocytes use similar mechanisms of H3.3 histone deposition as neurons following stimulation; however, there are different loci regulated based on astrocyte-specific transcription factors and enhancers [15]. Certainly, ARGs are not specific to brain cells either, as cells throughout the body use many of the same ARGs to respond to stimuli but with cell-specific transcription factors and enhancers. Calderon et al., in their study about the role of cohesins in neuronal ARG expression, found a similar result in macrophages in studying interferon stimulation for induction of inflammatory responses [39,60]. Interestingly, cohesion-deficient macrophages fail to induce interferons, whereas most IEGs were still able to be expressed in cohesin-deficient neurons. This discrepancy suggests additional unidentified regulatory factors in neurons ARG induction or possibly cell non-autonomous effects. In summary, more work is needed to understand the complexity of ARG expression in the ensemble of neurons and non-neuronal brain cells.

Future studies seeking to understand how single cells efficiently manipulate chromatin structure for ARG expression will require interdisciplinary methods with genomics, biophysics, super-resolution microscopy, and statistical modeling. Such expertise will allow scientists to best model neurocognitive molecular pathology in a way that simultaneously accounts for spatial and temporal changes in the nucleus. There have not yet been successful genetic therapies developed for neurodevelopment disorders. However, given the critical role of ARGs throughout postnatal development, a better understanding of the pathogenesis of diseases associated with chromatin regulators will reveal new opportunities for therapeutic intervention for these patients and their families.

Perspectives:

• Chromatin regulators are implicated in neurodevelopmental disorders such as Intellectual Disability. Dysregulation of activity-regulated gene expression is a proposed common pathogenetic mechanism.

• Chromatin regulators affect ARG expression through multiple levels of structural changes within the nucleus, from nucleosome occupancy and chromatin remodeling to higher-order structural changes such as TADs.

• Future work will require interdisciplinary approaches to probe activity-induced changes to chromatin structure over space and time. Given the crucial role of ARG regulation throughout postnatal development, studies in this area may elucidate opportunities for therapeutic intervention for patients with neurodevelopmental diseases.

Abbreviations Used:

ARG

Activity-regulated gene

IEG

Immediate early gene

LRG

Late response gene

TAD

Topologically associated domain

LAD

Lamina associated domain