SPECIAL FEATURE: PERSPECTIVE

The role of the genome in experience-dependentplasticity: Extending the analogy of thegenomic action potentialDavid F. Claytona,b,1, Ina Anreitera,c, Maria Aristizabala,c,d,e, Paul W. Franklanda,f,g,h,i, Elisabeth B. Bindera,j,k,and Ami Citria,l,m,1

Edited by Gene E. Robinson, University of Illinois at Urbana–Champaign, Urbana, IL, and approved April 24, 2019 (received for reviewJanuary 23, 2019)

Our past experiences shape our current and future behavior. These experiences must leave some enduringimprint on our brains, altering neural circuits that mediate behavior and contributing to our individualdifferences. As a framework for understanding how experiences might produce lasting changes in neuralcircuits, Clayton [D. F. Clayton, Neurobiol. Learn. Mem. 74, 185–216 (2000)] introduced the concept of thegenomic action potential (gAP)—a structured genomic response in the brain to acute experience. Similarto the familiar electrophysiological action potential (eAP), the gAP also provides a means for integratingafferent patterns of activity but on a slower timescale and with longer-lasting effects. We revisit thisconcept in light of contemporary work on experience-dependent modification of neural circuits. We re-view the “Immediate Early Gene” (IEG) response, the starting point for understanding the gAP. We discussevidence for its involvement in the encoding of experience to long-term memory across time and biolog-ical levels of organization ranging from individual cells to cell ensembles and whole organisms. We exploredistinctions between memory encoding and homeostatic functions and consider the potential for perpet-uation of the imprint of experience through epigenetic mechanisms. We describe a specific example of agAP in humans linked to individual differences in the response to stress. Finally, we identify key objectivesand new tools for continuing research in this area.

genome | plasticity |memory

Accounts of information processing and integrationin the brain often focus on membrane potentials, ionchannels, and synaptic signaling. As captured inthe classical observation of the electrophysiologicalaction potential (eAP), neurons can integrate signalsfrom many synaptic inputs to generate discrete func-tional outputs that propagate forward in time. However,the activity of every brain cell is ultimately governed bythe genome in the cell’s nucleus, and since the 1980s, ithas been clear that the genome responds dynamicallyto signals important for information processing (1, 2).

Across many organismal models of behavior, contextsassociated with learning and/or stress have been ob-served to trigger discrete time-limited pulses of neuralgene expression, which analogously, have been termeda genomic action potential (gAP) (3). Mechanistically,the gAP, like the eAP, has cascading consequencesthat spread across both time and space in the nervoussystem (Table 1). By altering the landscape of proteins,RNA, and chromatin structure in the cell, a gAP caninfluence how a cell responds to a subsequent activa-tion event and how it communicates with other cells.

aProgram in Child and Brain Development, Canadian Institute for Advanced Research, Toronto, ON M5G 1M1, Canada; bSchool of Biological andChemical Sciences, Queen Mary University of London, London E1 4NS, United Kingdom; cDepartment of Ecology and Evolutionary Biology,University of Toronto, Toronto, ON M5S 3B2, Canada; dCentre for Molecular Medicine and Therapeutics, British Columbia Children’s Hospital,Vancouver, BC V6H 3N1, Canada; eDepartment of Medical Genetics, University of British Columbia, Vancouver, BC V6H 3N1, Canada; fProgram inNeurosciences & Mental Health, Hospital for Sick Children, Toronto, ON M5G 0A4, Canada; gDepartment of Physiology, University of Toronto,Toronto, ON M5S 1A8, Canada; hDepartment of Psychology, University of Toronto, Toronto, ON M5S 3G3, Canada; iInstitute of Medical Sciences,University of Toronto, Toronto, ON M5S 1A8, Canada; jDepartment of Translational Research in Psychiatry, Max Planck Institute of Psychiatry,Munich 80804, Germany; kDepartment of Psychiatry and Behavioral Sciences, Emory University School of Medicine, Atlanta, GA 30322; lEdmondand Lily Safra Center for Brain Sciences, Hebrew University of Jerusalem, Jerusalem 91904, Israel; and mAlexander Silberman Institute of LifeSciences, Hebrew University of Jerusalem, Jerusalem 91904, IsraelAuthor contributions: D.F.C., I.A., M.A., P.W.F., E.B.B., and A.C. wrote the paper.The authors declare no conflict of interest.This article is a PNAS Direct Submission.Published under the PNAS license.1To whom correspondence may be addressed. Email: d.clayton@qmul.ac.uk or ami.citri@mail.huji.ac.il.

www.pnas.org/cgi/doi/10.1073/pnas.1820837116 PNAS Latest Articles | 1 of 9

SPECIA

LFEATURE:

PERSPECTIV

E

Dynamic Genome in Every Brain CellSince the original analogy was formulated in 2000 (3), a wealth ofnew information has accumulated about epigenetic mechanismsand the ways in which brains adapt to experience. With this newinformation, we revisit the gAP analogy and extend it further byconsidering several frames of reference or levels of biologicalorganization. These include (i ) the level of individual cells; (ii ) thelevel of neural circuits, where we propose that different gAPsmay have distinctive roles in homeostasis vs. engram encoding(Hebbian plasticity); (iii) potential mechanisms propagating theimpact of the gAP; and (iv) the potential for gAP mechanisms tocontribute to adaptive whole-organism physiology specifically inthe context of stress responses. Across all levels, we observe howthe gAP may serve effectively as a salience filter, determining whichpatterns of cellular, neural, or behavioral activity drive lasting func-tional changes in the organism.

gAP in the Cell: Molecular ElementsThe primary genomic response in the brain to immediate expe-rience (the gAP) involves acute time-limited changes in expressionof a core set of Immediate Early Genes (IEGs) within individualcells (3, 4). Activation of intracellular signal transduction—in par-ticular, the MAPK (mitogen-activated protein kinase) pathway andCREB (cAMP-responsive element binding protein)—leads to in-creased transcription of IEGs, such as cFOS, EGR1, and ARC, withtheir RNA levels peaking after ∼30–60 min followed by a peak oftheir proteins after ∼90 min (5). IEGs encode a variety of proteins,including transcription factors (e.g., c-Fos, Egr1), components ofintracellular signal transduction pathways (e.g., MAPK phospha-tases), and epigenetic readers/writers [e.g., DNA methyltransferase3 (DNMT3)]. In addition to protein codingmessenger RNAs (mRNAs),microRNAs and other noncoding RNAs may also participate in theinitial genomic response, and the response may include decreases asopposed to increases in some RNAs (6–8). Underlying or correlatedwith these transcriptional changes are chromatin-based mecha-nisms of targeted gene regulation, including histone acetylation (9–11), DNA methylation (12), and even introduction of DNA breaks(13). Initial functional consequences may be directed towardmodulation of synaptic transmission as suggested for ARC, an IEG-encoded protein that has been shown to depress glutamatergictransmission through endocytosis of synaptic AMPARs (α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptors) (14, 15) orpossibly even by direct intercellular transfer of RNAs as suggestedby the recent discovery that ARC can form virus-like capsids thattraffic across synapses (16, 17).

Just as neuronal subtypes exhibit characteristic differences intheir eAPs, they may also vary in the set of genes that is inducedby acute activity. Moreover, even for a single cell type, a givenphysiological or behavioral context may result in different patterns

of spatiotemporal input to the cell, and there is some evidencethat this could affect different subsets of genes. Direct evidencefor sensitivity to stimulus structure comes from recent studies ofcultured neurons, where different temporal patterns of electricalor chemical depolarization were found to trigger different generesponse patterns (18, 19). Within the brain, where neurons aresynaptically connected, gAPs induced by somatic calcium influx(i.e., neuronal firing) may differ from ones driven by localizeddendritic activity (see below). Interestingly, a recent study of long-term potentiation in the rat hippocampus demonstrated thatsome memories can be formed in the absence of somatic cellfiring altogether (20); in this case, the gAP and associated memoryprocesses may proceed independently from the eAP. Thus, thegAP can be viewed as a “salience filter,” integrating a variety ofdifferent cellular inputs, which on exceeding a threshold level (ofsalience), trigger the induction of the gAP.

Although the gAP has been described as an “all or none” re-sponse within an individual cell (3), there are several factors thatconfound a simple interpretation of this analogy, especially whenconsidering populations of cells. IEG responses are typicallymeasured as aggregate activity in extracts or sections of complexmulticellular brain tissue (e.g., analogous to electrophysiologicalfield potential or functional MRI signals). Different responsemagnitudes could reflect either differential recruitment of cellsubpopulations or differential activation levels within individualcells (ref. 21 is an early example where differential recruitment wasdocumented). A recent application of single-cell transcriptomicsshows that brain tissue may also contain cells at different temporalstages in the response trajectory to stimulation (22). How se-quential gAPs may interact within a single cell is also unclear: dothey summate, or might there be refractory phases? In somecontexts, sustained repeated behaviors may sum to increase orprolong IEG expression, whereas in other contexts, sequentialexposures to a stimulus can result in habituation of responses (23,24). Temporal interactions could emerge within individual cells(e.g., through epigenetic mechanisms; see below) or throughsystems-level modulations as originally outlined in 2000 (1). Ap-proaches, such as catFISH (cellular compartment analysis oftemporal activity by fluorescent in situ hybridization) and TAI-FISH(tyramide-amplified immunohistochemisty fluorescent in situhybridization) (25, 26), may provide a means to follow the longi-tudinal integration of gAP induced by sequential stimuli at single-cell resolution. Other tests of these questions may be developedin mice using increasingly sophisticated optogenetic and phar-macogenetic manipulations to control both timing and location ofspecific IEG expression.

Attempts to relate IEG expression to the intensity or magnitudeof a behavior may be further confounded if the brain region understudy does not directly and exclusively determine the magnitude of

Table 1. Comparison of the eAP with the gAP

eAP vs. gAP eAP gAP

Intracellular integration of inputs Yes YesSite of integration Axon initial segment Cell nucleusTime course Milliseconds MinutesRefractory phase Yes (Yes?)Intracellular phenotype Cascading changes in membrane

potentialCascading changes in specific protein

amountsEffects Increased probability of transmitter

releaseSynaptic and cellular remodeling,

neurotrophin releaseConsequence Propagation of information Information filtering for long-term encoding

2 of 9 | www.pnas.org/cgi/doi/10.1073/pnas.1820837116 Clayton et al.

the behavioral output being measured. If the behavioral readout ismodulated by additional circuit elements, salience at the level ofindividual neurons (i.e., the gAP) may not necessarily correspond tothe absolute salience of the experience of the organism. Wherecontrol of a measured behavior has been closely mapped to adiscrete brain area, such as, for example, in the dedicated aviansong control circuit, linearities have been reported between mag-nitude of behavior (singing) and magnitude of gene expression (ofegr1/ZENK) (27).

gAP Within Different Neural ContextsTo understand the precise functional significance of a particularinstance of the gAP, it is important to consider the specific neuralcontext in which it occurs. In behaving organisms, different ex-periences clearly trigger different patterns of genomic activationthroughout the brain. This was demonstrated with formal rigor inthe study of Mukherjee et al. (24, 28), where patterns of IEG ex-pression in the mouse brain were compared across 13 differentexperiences, ranging from reinstatement of feeding and mild footshock to different regiments of cocaine exposure and volitionalsucrose consumption. Each experience was found to be repre-sented by a unique transcriptional signature to the extent that aminimal expression profile of 4 IEGs across 7 brain regions wassufficient to decode an individual mouse’s recent experience withnearly 100% accuracy. Expression changes of the same gene can alsosupport opposite behavioral changes depending on where in thebrain they occur. For example, in cichlid fish, both ascent anddescent in social hierarchy are associated with expression changesof a subset of IEGs in the brain, but these are found in differentanatomical patterns (29).

Within a single brain region or circuit, there may be variationsin gAP function depending on the organization of the circuit andthe roles of the different cell types involved. One obvious functionis the encoding of learned events into memory. Putatively, an“encoding gAP” could be induced by stimuli that drive Hebbianplasticity of synaptic transmission. In this case, the gene expres-sion programs induced by the initiating experience will culminatein transcription of the building blocks for specific synaptic modi-fications aimed at supporting the long-term maintenance of in-formation storage at synapses within the network while maintainingsynapse specificity. A second and possibly distinct function for thegAP would be in mediating homeostatic rebalancing, wherebythe induced gene expression programs support the adaptationof neural networks to substantial shifts in activity (a “homeostaticgAP”). These functionally distinct experience-dependent gAPsmay comprise different transcription programs, supporting differentconsequences for the synapses, neurons, and circuits in which theyare encoded. Homeostatic mechanisms may be recruited morerobustly in situations in which networks are establishing the ap-propriate gain on incoming stimuli and network activity, as ob-served during development, or after large shifts in neuronal activity(convulsions, seizures, large shifts in ionic balance, and large tran-sitions in sensory input) as well as during sleep (19, 30–36). A“homeostatic gAP” is, therefore, anticipated to recruit large pro-portions of principal neurons as well as inhibitory interneurons,which robustly control network gain (37, 38). In contrast, an“encoding gAP,” induced within a neuronal ensemble recruited toencode a specific memory, is expected to recruit a fraction of theprincipal neurons in a given brain region (39). While most well-studied IEGs (ARC, FOS, EGR1, NPAS4, etc.) have been found tobe induced by gAPs that could be classified as either putativelyhomeostatic or putatively encoding (40–42), there are indications

that different transcriptional programs are induced by transient vs.prolonged stimuli, with transient stimuli potentially mimickingsingle-trial encoding of experience (19). Furthermore, NPAS4 hasbeen associated with gAPs that may be classified as either ho-meostatic (37, 43) or encoding (41) and has been found to activatedistinct programs of late response genes in inhibitory and excitatoryneurons (37). A further speculative distinction between homeostaticand encoding gAPs is the expectation that neuropeptides (such asBDNF) may be recruited by homeostatic gAPs to propagate a signalto a large network of neurons, compromising synapse specificity,while encoding gAPs may retain synapse specificity in communi-cation between the neurons in the network (44).

Interestingly, multiple waves of IEG transcription induced byexperience have been reported (45–47). Potentially, the first waveof transcription may be encoding an individual experience, whilethe delayed wave of transcription could be mediating a globalhomeostatic shift in network function induced by the accumula-tion of experiences over time, possibly corresponding to theproposed role of sleep in memory consolidation (48). During“real-life” induction of gAPs, homeostatic and encoding gAPsmay both be recruited, with the relative contribution of either typeof gAP defined by the circumstances: everyday events (49) may berepresented primarily by encoding gAPs, while unusual or especiallysalient events, which evoke substantial release of neuromodulatorsor stress hormones, may evoke a larger ratio of homeostatic gAPs.Speculatively, the recruitment of a homeostatic gAP may serve to“increase the gain” of the representation of events occurring withinthe same time window (of a few hours), thus creating a temporalwindow for the coupling of events in time. This idea may provide amolecular framework for “behavioral tagging,” whereby experi-ences of significant novelty (putatively inducing homeostaticgAPs) prolong the retention of everyday experiences (putativelyinducing encoding gAPs) (49). “Behavioral tagging” is an exten-sion of the “synaptic tagging” concept, demonstrating the essentialrole of inducible transcription and translation in the formation ofproteins that are captured by recently active synapses to supportlong-term plasticity and long-term memory (50).

While the field is developing thanks to contributions of groupsstudying both large shifts in homeostasis or sensory environment(19, 37, 38, 43, 51, 52) and more subtle, everyday experiences (3,4, 24, 36, 49, 53–57), it is possible that these two categories ofgAP may differ in their mechanisms and consequences. Awarenessof the potential distinctions between these 2 forms of experience-dependent plasticity may promote the resolution of fundamentalquestions regarding the computational significance of the gAP andits role in encoding information as discussed next.

gAP and the Encoding of EngramsEncoding of a particular memory is likely to involve only a fractionof neurons that otherwise share similar broad afferent and efferentconnectivity (58, 59). Coincident activity in such a subset of neuronsdefines a “neuronal ensemble,”which is thought to represent partof the “engram” or physiological representation of a particularmemory. The idea of encoding in neuronal ensembles was firstarticulated by Hebb (60) almost 70 years ago and has been sub-stantiated by a large number of studies that used IEGs to identifycollections of neurons that are coactive during the encoding of anexperience (Fig. 1). Recent studies have provided direct evidencelinking gAPs in individual neurons and membership in putativeengrams. These studies have used IEG-based tagging systemsto express optogenetic and chemogenetic actuators in neuronalensembles that were active at the time of encoding. Subsequent

Clayton et al. PNAS Latest Articles | 3 of 9

manipulation of these ensembles provides direct evidence thatthey constitute a component of the engram. Activation of tag-ged ensembles can induce memory retrieval in the absence ofexternal sensory cues, whereas inhibition of tagged ensem-bles can prevent memory retrieval in the presence of externalsensory cues (61, 62). These studies further show that only frac-tions of principal neurons are recruited to ensembles encodinglearned experiences (39, 63), with interneurons participatingmore broadly in defining the allocation of principal neurons toensembles (64, 65).

New research suggests that the recent activation history of aneuron, represented by the status of gene activity in the cell (i.e.,the gAP), determines whether it becomes allocated to an en-gram (56). These studies have used a fear-conditioning paradigmin which rodents learn an association between a conditionedstimulus (CS; e.g., a tone) and an unconditioned stimulus (e.g.,a foot shock). Using viral vectors to increase excitability in arandom, sparse population of excitatory neurons in the lateralamygdala showed that neurons with increased excitability at thetime of an event were more likely to be allocated to the engramsupporting the memory of that event. Conversely, expressingconstructs that decrease neuronal excitability decreased thisprobability (66–68).

More recent studies have also examined how excitabilitychanges influence the interaction between engrams (55). In ex-periments where mice were conditioned to fear 2 different-tonedCSs, memories were encoded in overlapping neuronal ensembleswhen the 2 training events occurred close together in time (<6 h).Moreover, they were linked behaviorally—extinguishing oneCS led to extinction of fear to the other. In contrast, when the2 training events were separated in time (24 h apart), there was noevidence for memory linking, and the engrams supporting thesedistinct memories did not overlap.

These results suggest that memories acquired in close tem-poral proximity are coallocated to overlapping engrams (thuslinking the 2 memories), whereas memories acquired at moredistal times are disallocated to distinct engrams (and the memo-ries remain distinct). Similar to initial allocation, coallocation is alsomediated by a winner-take-all mechanism in which more excitableneurons win the competition for recruitment to the second en-gram. Importantly, after they are allocated, neurons remain moreexcitable for a time period that overlaps with the “coallocationwindow” (typically 6 h). Broadly, these studies raise the possi-bility that dynamic gene activity governs how information is or-ganized in neural circuits, with a neuron’s recent activation

history (captured in the temporal dynamics of a gAP) defining thecoupling of memories and their coallocation to neuronal ensembles.

Cellular Perpetuation of the gAPAlthough longer lasting than eAPs, gAPs are still transient bynature. The average mRNA transcript has a half-life of 6–10 h,while the mRNAs encoding regulatory transcription factors, in-cluding IEGs, are among those with the shortest half-lives (69).Nevertheless, there is growing evidence that molecular mecha-nisms help sustain the effects of gAPs for extended periods oftime (70) at multiple levels—cellular, multicellular, generational,and potentially, even intergenerationally (Fig. 2). For instance, IEGsthat form the core of the gAP (like Fos and Jun) exhibit epigeneticfeatures that support their fast activation rates: bivalent promoterscontaining activating [histoneH3 lysine 4 trimethylation (H3K4me3)]and repressing (histone H3 lysine 27 trimetrylation) chromatinmarks(71). In addition, these genes are often bound by “poised” RNApolymerases, fully initiated transcriptionally competent complexesthat can be quickly released into productive transcription (72).Furthermore, these genes also display high levels of histone acet-ylation even before stimulation, which is thought to promote open,transcriptionally competent chromatin. Interestingly, these molec-ular features are also characteristic of genes described as express-ing “transcriptional memory” (73), whereby an initial event driveslocal epigenetic modifications, leading to accelerated and morepronounced expression in response to subsequent activation.Like IEGs, genes expressing transcriptional memory are deco-rated with H3K4me3. In addition, their promoters are enrichedfor transcription factors bound to “memory recruitment sequences”as well as the histone variant H2A.Z. Importantly, molecularfeatures of transcriptional memory can survive DNA replicationsuch that, in dividing cells, increased activation potential can bemaintained up to 8 cellular generations. Whether transcriptionalmemory per se is a robust feature of genes recruited by the gAP(such as IEGs) remains unclear, but given the similarities betweenIEG regulation and genes expressing transcriptional memory, itseems likely that at least some features of transcriptional memorysupport the rapid transcription of IEGs and may be subject toregulation based on past experience and activity patterns.

Neuronal activity also results in global increases in histoneacetylation, which is important for learning and memory (9). His-tone acetylation, occurring concurrently with the unfolding of thegAP or following closely behind it, relaxes chromatin, perpetuat-ing the impact of the gAP (74–77). Interestingly, changes in his-tone acetylation in response to neuronal activity are linked tohistone phosphorylation (75) and DNA methylation (74), sug-gesting the involvement of multiple epigenetic mechanisms inthe gAP. Furthermore, specific acetylated histone residues havebeen differentially associated with the occurrence of salient ex-periences (i.e., resulting in memory formation) vs. nonsalient ex-periences (76).

An example illustrating epigenetic mechanisms that may un-derlie the perpetuation of a gAP is the positive feedback loopinvolving the histone deacetylase 2 (HDAC2) in the regulation ofBDNF expression. BDNF is a neuronal growth factor involved inthe activity-dependent decision regarding the elimination ormaintenance of synapses (78).* Sustained BDNF up-regulationthrough histone acetylation in the prefrontal cortex has been

Coherent

Egr1 cFosArc Npas4

Bi-stable - Experience + Experience

Recruited cell subset - Experience + Experience

Indu

ctio

n

Time

Indu

ctio

n

Time

A B C

Fig. 1. Assumptions regarding the relationship of the gAP toneuronal ensembles encoding an experience. (A) A salient experienceis anticipated to induce a gAP in only a fraction of the principalneurons within a defined brain structure. (B) Within an ensemble, IEGexpression is assumed to be bistable—maintained at low levels in theabsence of stimuli and induced to its full extent after experience,rapidly returning to baseline. (C) Ensembles are presumed to becoherent—such that the ensemble defined by the expression of asingle IEG will overlap to a large extent with the ensemble defined byother induced IEGs.

*J. M. George et al., Acute social isolation alters neurogenomic state in songbirdforebrain. Proc. Nat. Acad. Sci. U.S.A., submitted.

4 of 9 | www.pnas.org/cgi/doi/10.1073/pnas.1820837116 Clayton et al.

associated with both consolidation and extinction of fear condi-tioning (79) (Fig. 2). Mechanistically, increased neuronal activitytriggers the release of the repressivemethyl-CpG (cytosine–guaninedinucleotide)-binding protein 2 (MECP2) from the BDNF promoter,resulting in an initial activity-dependent increase in BDNF expres-sion (80). The action of BDNF at its receptors then triggers the re-lease of HDAC2 from chromatin, increasing histone acetylation atneurotrophin-dependent gene promoters, including BDNF itself(81, 82). BDNF expression is also negatively regulated in humans byan overlapping long noncoding antisense RNA (BDNF-AS), which isitself negatively regulated by neuronal activity (83). Other epige-netic mechanisms, such as DNA methylation, may further con-tribute to the regulation of BDNF after salient experiences (84),indicating that a complex set of molecular mechanisms mayfunction in the gAP to fine tune BDNF expression, potentiallybroadly impacting the expression of additional genes.

Although a nascent field, evidence is also mounting in supportof the intriguing prospect that epigenetic mechanisms are in-volved in the transmission of the impact of experience acrossgenerations (85–89). Furthermore, although certain epigeneticmarks correlate tightly with gene expression changes, in manycases the relationship between epigenetic marks and gene ex-pression is still unclear (90).

gAP in the Organism: Modulating Responses to StressResponding to stress is a critical organismal challenge in which thegAP plays central roles, recording stressful associations andshaping future responses to similar circumstances. In this section,we extend the analogy of the classical gAP to gene expressionchanges induced by environmental/stressful stimuli via nuclearhormone receptors. Such a broader gAP may be induced bystress-driven activation of the hypothalamus–pituitary–adrenal(HPA) axis and release of glucocorticoids (GCs): GCs trigger rapidtranscriptomic responses via intracellular receptors, which act astranscription factors [e.g., the glucocorticoid receptor (GR)]. GRactivation propels a wave of downstream gene regulation (i.e., aGC gAP) starting within 30–60 min after stress exposure andpeaking 3–6 h thereafter. A GC gAP can occur across differenttissues beyond the brain and may be analog in nature (rather thandigital/all or none), as GCs elicit graded transcriptional responsesdepending on timing and dose. In this context, it is worth men-tioning the concept of the “neuroendocrine action potential,”(nAP) which refers to the propagation of a gAP, induced by ex-perience in the brain, to a neuroendocrine response throughoutthe body, which can be propagated over hours to months (91).The nAP can support stable changes in the behavior of an or-ganism corresponding, for example, to adaptations to changingseasons or a shift in social hierarchy. In this sense, the nAP pro-vides a conceptual framework for considering how some gAPsmay propagate to impact behavior over long periods of time.

In addition to inducing a stress transcriptome profile, GR ac-tivation also induces local epigenetic changes at its DNA bindingsites, the glucocorticoid response elements (GREs). This has beendescribed for GR in the context of transcription factor binding-mediated DNA demethylation at GREs (92), which subsequentlyfacilitates the transcriptional effects of the GR on the targetgene (92, 93). In this issue, Provençal et al.† demonstrate lasting

changes in DNAmethylation at GREs and bivalent enhancers afterGC exposure in hippocampal progenitor cells, suggesting thatearly life GC exposure could lead to the establishment of poisedstates that alter the gAP induced by subsequent stress and in turn,the ensuing behavioral responses.

The set point of the magnitude of the gAP induced by GR mayalso be altered at the level of the GR gene itself. Reception ofmaternal care by rat pups was shown to be sensed and transducedinto lasting changes in (HPA axis) stress-dependent GC release viaa mechanism that is initiated by binding of an IEG protein (EGR1)to regulatory sites in the GR gene in the hippocampus. This bindinghas lasting consequences, as it suppresses DNAmethylation at thesesites, sustaining higher levels of GR gene expression in the hippo-campus of the pups after they mature and tempering the HPA re-sponse to subsequent stressors (94). Increasedmethylation of theGRpromoter after reduced maternal care seems to be commonly ob-served in 7 of 10 studies of this model of early life adversity (95).Recent studies in macaques suggest that social context, such associal status, may also alter the transcriptional response toGC (96): inthis case, in peripheral blood cells. These studies bring up interestingfuture research questions of (i) whether more complex social stimulican also influence gAPs, (ii) whether environmental factors may in-fluence gAP in a cross-tissue manner, and (iii) how changes in the setpoint of the gAP induced by the same environmental challengemapbetween different target tissues.

The FKBP5 gene is among the transcripts most strongly in-duced by GR activation in a number of tissues, including the brain(79, 97). Studies in both animals and humans are providing

A

B

C

Fig. 2. Epigenetic mechanisms for activity-dependent perpetuationof gene expression states. (A) Poised transcriptional states allow forfast transcription on experience. (B) Transcriptional memory allowsfor the faster and stronger reactivation of previously active genes. (C)The BDNF positive feedback loop. In its inactive state, BDNFtranscription is repressed by methylation of the promoter, HDAC2-dependent nucleosomal suppression, and binding of MECP2 as wellas BDNF antisense long noncoding RNA (BDNF-AS). Neuronal activitytriggers demethylation of the promoter, release of MECP2, andrepression of BDNF-AS. The resulting expression of BDNF triggersthe release of HDAC2 from chromatin, perpetuating an active genetranscription state. MRS, memory recruitment sequences; TTS,transcription start site.

†N. Provençal et al., Glucocorticoid exposure during hippocampal neurogenesisprimes future stress response by inducing changes in DNA methylation. Proc.Nat. Acad. Sci. U.S.A., submitted.

Clayton et al. PNAS Latest Articles | 5 of 9

fascinating evidence for complex interactions between stress,genotype, and FKBP5 expression, which may influence suscepti-bility to psychiatric disease. Overexpression of the FKBP5 gene inseveral limbic brain regions has been associated with increasedanxiety and decreased stress coping behaviors in laboratory ani-mals (98). In humans, functional genetic haplotypes moderate theinduction of FKBP5 by GR activation. The haplotype associatedwith increased transcriptional activation by GCs has been shown,in combination with exposure to early life adversity, to predictrisk for a number of psychiatric disorders, including depression,PTSD (post-traumatic stress disorder), and psychosis (97). Thisgene × environment interaction is likely additionally mediatedby allele-specific DNA demethylation at GREs of FKBP5 (96),further derepressing FKBP5 transcription in individuals exposedto adversity. Only the combined genetic and epigenetic tran-scriptional effects would increase FKBP5 protein levels beyondthe threshold for initiating a cascade of events that may lead tothe development of psychiatric disorders. The stress-inducedepigenetic and transcriptional changes of FKBP5 are illustratedin Fig. 3. An interesting point is that FKBP5, via its effects as acochaperone of the GR itself, also influences the neuroendo-crine response to stress. States of higher FKBP5 expression havebeen associated with a prolonged cortisol response to stress inlaboratory animals and humans (97), suggesting that alterationsof the gAP of FKBP5 would have consequences on neuroen-docrine set points.

FKBP5 also interacts with a number of other proteins highlyrelevant for key processes in neuronal function. It could thusrepresent a molecular hub, initiating multiple cellular processeson stress that would ultimately alter synaptic strength of specificneuronal ensembles. These processes include autophagy, cellproliferation, migration, apoptosis, and DNA methylation (99).Interestingly, FKBP5 also directly interacts with CDK5, a kinasethat is known to phosphorylate and activate DNMT1. FKBP5 wasfound to decrease the interaction of CDK5 with DNMT1, reducingglobal DNA methylation via reduced phosphorylation and enzy-matic activity of DNMT1 (100) as well as specific effects on epi-genetic regulation of BDNF. By this, FKBP5 could also contributeto altering the gAP set point for other targets. Alterations of theFKBP5 gAP in response to stress would thus be propagated toother targets both via direct protein–protein interactions as well aspossibly by interfering with key epigenetic mechanisms. An ad-ditional regulation on FKBP5 has been described via binding of

the microRNA miR-15a to the 3′ end of FKBP5, decreasing FKBP5translation (101). In animal models, the stress-induced up-regulationof this noncoding RNA was shown to buffer behavioral effectsof stress.

In summary, exposure to stress or threat may lead to lastingalterations of gAPs by epigenetic effects in promoter and enhancerregions. Prior stressful experiences could alter future stress-inducedgAP and by this, change the set point for the saliency filterfor subsequent adverse experience. Furthermore, a “neuroen-docrine action potential” may also translate into lasting co-ordinated physiological responses to stress across tissues.

Roadmap for Future InvestigationWe have argued that gAPs are at the core of dynamic adaptationsto experience, helping to filter, encode, and assimilate informa-tion over time. Effects of gAPs can be observed and studied acrossmultiple levels of biological organization—molecular, cellular,systems, and organismal. Here, we consider some of the majorquestions, challenges, and opportunities for further research,building up from molecular through cellular and organismal toevolutionary perspectives.

Establishing the Causal Role of gAP Components in Encoding

Experience. A major challenge for the field is in defining the func-tional role of the different molecular components comprising thegAP in encoding experience. This challenge is exacerbated by thecomplexity of transcriptional networks and the intrinsic redundanciesand compensation mechanisms that support their robustness (102).The development of temporally regulated, cell-specific genomeediting tools that can target genes (and their regulatory elements) invarious combinations is anticipated to assist in resolving the func-tions mediated by the products of activity-dependent expressionwith cellular and temporal specificity (103, 104).

Decoding the Activity Transcription Transfer Function. An im-plicit assumption of the approach represented here is that an“encoding gAP” is a transcriptional relay of specific circuit-levelevents driven by experience. Thus, a major objective should bedecoding the “activity transcription transfer function” definingthe transformation of specific synaptic activity patterns to specifictranscriptional programs. By defining the principles underlyingthe spatiotemporal association of activity required for inductionof a gAP, we expect to also gain insight into the thresholds for

Post-stress baseline/new set point Pre-stress baseline

enhancer enhancer promoter

Glucocorticoid receptor Glucocorticoid response element Unmethylated CpG sites

Methylated CpG sites Transcriptional response Glucocorticoids

A B GRE GRE GRE GRE

GRE GRE GRE GRE

GRE

GCs

GCs

GCs

Fig. 3. Lasting epigenetic effects after GC exposure change subsequent stress gAP. (A) Prestress baseline. GR activation triggers a transcriptionalresponse and also, leaves lasting epigenetic changes, here in the form of reduced DNAmethylation within GREs. Whether these changes in DNAmethylation status are lasting is likely influenced by the developmental timing of the stress exposure and genetic factors. (B) Poststress/new setpoint. The epigenetic landscape after a prior priming stress exposure now shows reduced DNA methylation within GRE. A subsequent GRactivation now causes an enhanced transcriptional response.

6 of 9 | www.pnas.org/cgi/doi/10.1073/pnas.1820837116 Clayton et al.

neuronal salience filtering and define boundary conditions fortransition to a commitment to long-term encoding of informationin the brain. The development of an “algorithmic” definition ofthe rules governing the transformation of activity to transcriptionwill support a major aim of the burgeoning field of “BehavioralTranscriptomics,” which is to associate defined experiences withtheir corresponding brain-wide gAPs, resolving their activitytranscription transfer functions (24).

Associating Neuronal gAP History with Ensemble Selection.

Memories that occur close in time (or are related in content) areexpected to be coallocated to overlapping neuronal ensembles,while memories that are separated in time or unrelated in contentare expected to be encoded in nonoverlapping ensembles. Evi-dence is accumulating that the neural space is not an even playingfield and that a neuron’s recent history determines whether itparticipates in information coding (56). While current studies havedemonstrated the allocation of temporally adjacent events tooverlapping neuronal ensembles, allocation to a memory trace isanticipated to work over multiple timescales, with epigeneticchanges possibly having very long-term consequences for whereand how information is encoded in neural circuits. Tests of theseideas may be developed in mice using increasingly sophisticatedoptogenetic and pharmacogenetic manipulations to control bothtiming and location of specific IEG expression: for example, inhippocampus during fear memory formation and recall (105).

Analysis of the gAP Within Diverse Cell Populations. The re-cent development and rapid improvement of approaches forsingle-cell analysis (scRNAseq, snRNAseq, and smFISH) (22, 30,106) are supporting the transition from population-based in-vestigation to sensitive analysis with cell-specific resolution, en-abling accurate measurement even of low-abundance transcripts,such as transcription factors. One of the contributions of thistechnical development has been the recognition that the geno-mic response to stimuli extends beyond neurons and is observedto be distributed across all cell types found in the brain (30, 106,107). Disambiguating the gAP response in different cell types (andto different patterns of cellular stimulation) is expected to providenovel insight into how information is encoded in the brain.

Individual Differences in the gAP. Here, we have outlinedhow the gAP may contribute to the filtering of experience, the

formation and persistence of memory, and stress reactivity. Indi-viduals differ greatly in these attributes—could variations in thegAP contribute to these individual differences? If so, what sort ofgene × environment × development interactions contribute tovariations in the gAP? Animal studies already suggest that thesensitivity or structure of the gAP can vary with developmentalstage as seen, for example, in songbirds (108) and rats (109), andhuman studies have provided evidence for genetic influences ongAP reactivity (97). Might variations in the sensitivity and temporalparameters of the gAP have consequences on how information isfiltered and stored either in different individuals or in differentphysiological or developmental states?

Translational Relevance of the gAP. How might an under-standing of the gAP be used to provide immediate benefits tohuman society? Direct measurement of the gAP in individualscould easily have clinical applications, but this is impossible withcurrent techniques, as it would require multiple invasive mea-surements of brain tissue. Ultimately, it may be feasible to developnoninvasive markers of IEG expression and epigenetic mark de-position in the human brain: for example, using MRI or intravitalfluorescence for detection (110, 111). Alternatively, there may bea link between responses in the blood and responses in the brain,allowing traces of a brain gAP to be identified in blood (112, 113).Such a connection has been reported in the context of stress, butcurrent research suggests that it may extend more broadly toeveryday life events (114). If these approaches can be refined andvalidated, they may herald a new quantitative psychiatry based onmeasuring and perturbing components of the gAP that mediatemaladaptive memory processes (e.g., PTSD, drug addiction).

Evolutionary Perspectives on the gAP. It is worth consideringhow the time course of the gAP may have evolved in response tothe temporal demands of natural experience (115). Might the ge-nomic response time course place a limit on behavioral plasticity?How might this align with the requirements of different ecologicalniches? Is the gAP itself subject to evolutionary plasticity?

These are exciting times to be involved in the investigationof the role of the genome in experience-dependent plasticity.The explosion of technical innovation across neuroscience andbiology now supports deep investigation of the hypotheses andunifying principles described herein, with potential to impacthealth care.

1 M. Sheng, M. E. Greenberg, The regulation and function of c-fos and other immediate early genes in the nervous system. Neuron 4, 477–485(1990).2 J. I. Morgan, T. Curran, Stimulus-transcription coupling in neurons: Role of cellular immediate-early genes. Trends Neurosci. 12, 459–462 (1989).3 D. F. Clayton, The genomic action potential. Neurobiol. Learn. Mem. 74, 185–216 (2000).4 D. F. Clayton The genomics of memory and learning in songbirds. Annu. Rev. Genomics Hum. Genet. 14, 45–65 (2013).5 D. N. Barry, S. Commins, Temporal dynamics of immediate early gene expression during cellular consolidation of spatial memory. Behav. Brain Res. 327, 44–53(2017).

6 P. H. Gunaratne et al., Song exposure regulates known and novel microRNAs in the zebra finch auditory forebrain. BMC Genomics 12, 277 (2011).7 S. Dong et al., Discrete molecular states in the brain accompany changing responses to a vocal signal. Proc. Natl. Acad. Sci. U.S.A. 106, 11364–11369 (2009).8 S. M. Eacker, M. J. Keuss, E. Berezikov, V. L. Dawson, T. M. Dawson, Neuronal activity regulates hippocampal miRNA expression. PLoS One 6, e25068 (2011).9 J. Gräff, L. H. Tsai, Histone acetylation: Molecular mnemonics on the chromatin. Nat. Rev. Neurosci. 14, 97–111 (2013).

10 J. M. Levenson, J. D. Sweatt, Epigenetic mechanisms in memory formation. Nat. Rev. Neurosci. 6, 108–118 (2005).11 I. B. Zovkic, B. S. Paulukaitis, J. J. Day, D. M. Etikala, J. D. Sweatt, Histone H2A.Z subunit exchange controls consolidation of recent and remote memory.Nature

515, 582–586 (2014).12 G. A. Kaas et al., TET1 controls CNS 5-methylcytosine hydroxylation, active DNA demethylation, gene transcription, and memory formation. Neuron 79, 1086–

1093 (2013).13 R. Madabhushi et al., Activity-induced DNA breaks govern the expression of neuronal early-response genes. Cell 161, 1592–1605 (2015).14 E. M. Rial Verde, J. Lee-Osbourne, P. F. Worley, R. Malinow, H. T. Cline, Increased expression of the immediate-early gene arc/arg3.1 reduces AMPA receptor-

mediated synaptic transmission. Neuron 52, 461–474 (2006).15 S. Chowdhury et al., Arc/Arg3.1 interacts with the endocytic machinery to regulate AMPA receptor trafficking. Neuron 52, 445–459 (2006).16 J. Ashley et al., Retrovirus-like gag protein Arc1 binds RNA and traffics across synaptic boutons. Cell 172, 262–274.e11 (2018).

Clayton et al. PNAS Latest Articles | 7 of 9

17 E. D. Pastuzyn et al., The neuronal gene arc encodes a repurposed retrotransposon gag protein that mediates intercellular RNA transfer. Cell 172, 275–288.e18(2018).

18 P. R. Lee, J. E. Cohen, D. A. Iacobas, S. Iacobas, R. D. Fields, Gene networks activated by specific patterns of action potentials in dorsal root ganglia neurons. Sci.Rep. 7, 43765 (2017).

19 K. M. Tyssowski et al., Different neuronal activity patterns induce different gene expression programs. Neuron 98, 530–546.e11 (2018).20 J. I. Rossato et al., Silent learning. Curr. Biol. 28, 3508–3515.e5 (2018).21 C. V. Mello, D. S. Vicario, D. F. Clayton, Song presentation induces gene expression in the songbird forebrain. Proc. Natl. Acad. Sci. U.S.A. 89, 6818–6822 (1992).22 B. Lacar et al., Nuclear RNA-seq of single neurons reveals molecular signatures of activation. Nat. Commun. 7, 11022 (2016).23 C. V. Mello, D. F. Clayton, Song-induced ZENK gene expression in auditory pathways of songbird brain and its relation to the song control system. J. Neurosci.

14, 6652–6666 (1994).24 D. Mukherjee et al., Salient experiences are represented by unique transcriptional signatures in the mouse brain. eLife 7, 1–20 (2018).25 J. Xiu et al., Visualizing an emotional valence map in the limbic forebrain by TAI-FISH. Nat. Neurosci. 17, 1552–1559 (2014).26 J. F. Guzowski, B. L. McNaughton, C. A. Barnes, P. F. Worley, Environment-specific expression of the immediate-early gene Arc in hippocampal neuronal

ensembles. Nat. Neurosci. 2, 1120–1124 (1999).27 E. D. Jarvis, F. Nottebohm, Motor-driven gene expression. Proc. Natl. Acad. Sci. U.S.A. 94, 4097–4102 (1997).28 V. Sagar, T. Kahnt, Genetic signatures of memories. eLife 7, e36064 (2018).29 K. P. Maruska, L. Becker, A. Neboori, R. D. Fernald, Social descent with territory loss causes rapid behavioral, endocrine and transcriptional changes in the brain.

J. Exp. Biol. 216, 3656–3666 (2013).30 S. Hrvatin et al., Single-cell analysis of experience-dependent transcriptomic states in the mouse visual cortex. Nat. Neurosci. 21, 120–129 (2018).31 E. Nedivi, D. Hevroni, D. Naot, D. Israeli, Y. Citri, Numerous candidate plasticity-related genes revealed by differential cDNA cloning. Nature 363, 718–722

(1993).32 J.I. Morgan, D.R. Cohen, J.L. Hempstead, Curran T, Mapping patterns of c-fos expression in the central nervous system after seizure. Science 237, 192–197

(1987).33 K. B. Hengen, A. Torrado Pacheco, J. N. McGregor, S. D. Van Hooser, G. G. Turrigiano, Neuronal firing rate homeostasis is inhibited by sleep and promoted by

wake. Cell 165, 180–191 (2016).34 F. J. L. Arnold et al., Microelectrode array recordings of cultured hippocampal networks reveal a simple model for transcription and protein synthesis-dependent

plasticity. J. Physiol. 564, 3–19 (2005).35 M. Miyatake, M. Narita, M. Shibasaki, A. Nakamura, T. Suzuki, Glutamatergic neurotransmission and protein kinase C play a role in neuron-glia communication

during the development of methamphetamine-induced psychological dependence. Eur. J. Neurosci. 22, 1476–1488 (2005).36 T. Miyashita, S. Kubik, N. Haghighi, O. Steward, J. F. Guzowski, Rapid activation of plasticity-associated gene transcription in hippocampal neurons provides a

mechanism for encoding of one-trial experience. J. Neurosci. 29, 898–906 (2009).37 I. Spiegel et al., Npas4 regulates excitatory-inhibitory balance within neural circuits through cell-type-specific gene programs. Cell 157, 1216–1229 (2014).38 A. R. Mardinly et al., Sensory experience regulates cortical inhibition by inducing IGF1 in VIP neurons. Nature 531, 371–375 (2016).39 A. Vazdarjanova, B. L. McNaughton, C. A. Barnes, P. F. Worley, J. F. Guzowski, Experience-dependent coincident expression of the effector immediate-early

genes arc and Homer 1a in hippocampal and neocortical neuronal networks. J. Neurosci. 22, 10067–10071 (2002).40 J. D. Shepherd, M. F. Bear, New views of Arc, a master regulator of synaptic plasticity. Nat. Neurosci. 14, 279–284 (2011).41 Ramamoorthi K, et al., Npas4 regulates a transcriptional program in CA3 required for contextual memory formation. Science 334, 1669–1675 (2011).42 E. L. Yap, M. E. Greenberg, Activity-regulated transcription: Bridging the gap between neural activity and behavior. Neuron 100, 330–348 (2018).43 B. L. Bloodgood, N. Sharma, H. A. Browne, A. Z. Trepman, M. E. Greenberg, The activity-dependent transcription factor NPAS4 regulates domain-specific

inhibition. Nature 503, 121–125 (2013).44 P. Kowianski et al., BDNF: A key factor with multipotent impact on brain signaling and synaptic plasticity. Cell. Mol. Neurobiol. 38, 579–593 (2018).45 D. F. Marrone, M. J. Schaner, B. L. McNaughton, P. F. Worley, C. A. Barnes, Immediate-early gene expression at rest recapitulates recent experience. J. Neurosci.

28, 1030–1033 (2008).46 S. Ribeiro et al., Novel experience induces persistent sleep-dependent plasticity in the cortex but not in the hippocampus. Front. Neurosci. 1, 43–55 (2007).47 Ramırez-Amaya V, et al., Spatial exploration-induced arc mRNA and protein expression: Evidence for selective, network-specific reactivation. J. Neurosci. 25,

1761–1768 (2005).48 B. Rasch, J. Born, About sleep’s role in memory. Physiol. Rev. 93, 681–766 (2013).49 S.-H. Wang, R. L. Redondo, R. G. M. Morris, Relevance of synaptic tagging and capture to the persistence of long-term potentiation and everyday spatial

memory. Proc. Natl. Acad. Sci. U.S.A. 107, 19537–19542 (2010).50 M. Nomoto, K. Inokuchi, Behavioral, cellular, and synaptic tagging frameworks. Neurobiol. Learn. Mem. 153 (Pt A), 13–20 (2018).51 P. L. Greer, M. E. Greenberg, From synapse to nucleus: Calcium-dependent gene transcription in the control of synapse development and function. Neuron 59,

846–860 (2008).52 H. Stroud et al., Early-life gene expression in neurons modulates lasting epigenetic states. Cell 171, 1151–1164.e16 (2017).53 D. J. Cai et al., A shared neural ensemble links distinct contextual memories encoded close in time. Nature 534, 115–118 (2016).54 J. Yokose, et al., Overlapping memory trace indispensable for linking, but not recalling, individual memories. Science 355, 398–403 (2017).55 A. J. Rashid, et al., Competition between engrams influences fear memory formation and recall. Science 353, 383–387 (2016).56 S. A. Josselyn, P. W. Frankland, Memory allocation: Mechanisms and function. Annu. Rev. Neurosci. 41, 389–413 (2018).57 F. Ballarini, D. Moncada, M. C. Martinez, N. Alen, H. Viola, Behavioral tagging is a general mechanism of long-term memory formation. Proc. Natl. Acad. Sci.

U.S.A. 106, 14599–14604 (2009).58 E. T. Rolls, A. Treves, Neural Networks and Brain Function (Oxford University Press, Oxford, England, UK 1998).59 B. L. McNaughton, R. G. M. Morris, Hippocampal synaptic enhancement and information storage within a distributed memory system. Trends Neurosci. 10, 408–

415 (1987).60 D. O. Hebb, The Organization of Behaviour: A Neuropsychological Theory (John Wiley and Sons, New York, NY, 1949).61 S. A. Josselyn, S. Köhler, P. W. Frankland, Finding the engram. Nat. Rev. Neurosci. 16, 521–534 (2015).62 S. Tonegawa, X. Liu, S. Ramirez, R. Redondo, Memory engram cells have come of age. Neuron 87, 918–931 (2015).63 A. Vazdarjanova et al., Spatial exploration induces ARC, a plasticity-related immediate-early gene, only in calcium/calmodulin-dependent protein kinase II-

positive principal excitatory and inhibitory neurons of the rat forebrain. J. Comp. Neurol. 498, 317–329 (2006).64 T. Stefanelli, C. Bertollini, C. Lüscher, D. Muller, P. Mendez, Hippocampal somatostatin interneurons control the size of neuronal memory ensembles.Neuron 89,

1074–1085 (2016).65 D. J. Morrison et al., Parvalbumin interneurons constrain the size of the lateral amygdala engram. Neurobiol. Learn. Mem. 135, 91–99 (2016).66 J.-H. Han, et al., Neuronal competition and selection during memory formation. Science 316, 457–460 (2007).67 J.-H. Han, et al., Selective erasure of a fear memory. Science 323, 1492–1496 (2009).68 A. P. Yiu et al., Neurons are recruited to a memory trace based on relative neuronal excitability immediately before training. Neuron 83, 722–735 (2014).69 L. V. Sharova et al., Database for mRNA half-life of 19 977 genes obtained by DNA microarray analysis of pluripotent and differentiating mouse embryonic stem

cells. DNA Res. 16, 45–58 (2009).

8 of 9 | www.pnas.org/cgi/doi/10.1073/pnas.1820837116 Clayton et al.

70 M. Kundakovic, F. A. Champagne, Early-life experience, epigenetics, and the developing brain. Neuropsychopharmacology 40, 141–153 (2015).71 M. Rye et al.; FANTOM consortium, Chromatin states reveal functional associations for globally defined transcription start sites in four human cell lines. BMC

Genomics 15, 120 (2014).72 A. D’Urso et al., Set1/COMPASS and Mediator are repurposed to promote epigenetic transcriptional memory. eLife 5, e16691 (2016).73 A. D’Urso, J. H. Brickner, Mechanisms of epigenetic memory. Trends Genet. 30, 230–236 (2014).74 C. Maharana, K. P. Sharma, S. K. Sharma, Depolarization induces acetylation of histone H2B in the hippocampus. Neuroscience 167, 354–360 (2010).75 C. Crosio, E. Heitz, C. D. Allis, E. Borrelli, P. Sassone-Corsi, Chromatin remodeling and neuronal response: Multiple signaling pathways induce specific histone

H3 modifications and early gene expression in hippocampal neurons. J. Cell Sci. 116, 4905–4914 (2003).76 O. Bousiges et al., Detection of histone acetylation levels in the dorsal hippocampus reveals early tagging on specific residues of H2B and H4 histones in

response to learning. PLoS One 8, e57816 (2013).77 E. Palomer, J. Carretero, S. Benvegnù, C. G. Dotti, M. G. Martin, Neuronal activity controls Bdnf expression via Polycomb de-repression and CREB/CBP/

JMJD3 activation in mature neurons. Nat. Commun. 7, 11081 (2016).78 C. R. Bramham, E. Messaoudi, BDNF function in adult synaptic plasticity: The synaptic consolidation hypothesis. Prog. Neurobiol. 76, 99–125 (2005).79 T. W. Bredy et al., Histone modifications around individual BDNF gene promoters in prefrontal cortex are associated with extinction of conditioned fear. Learn.

Mem. 14, 268–276 (2007).80 W. G. Chen et al., Derepression of BDNF transcription involves calcium-dependent phosphorylation of MeCP2. Science 302, 885–889 (2003).81 A. Nott, P. M. Watson, J. D. Robinson, L. Crepaldi, A. Riccio, S-Nitrosylation of histone deacetylase 2 induces chromatin remodelling in neurons. Nature 455,

411–415 (2008).82 J.-S. Guan et al., HDAC2 negatively regulates memory formation and synaptic plasticity. Nature 459, 55–60 (2009).83 L. Lipovich et al., Activity-dependent human brain coding/noncoding gene regulatory networks. Genetics 192, 1133–1148 (2012).84 F. D. Lubin, T. L. Roth, J. D. Sweatt, Epigenetic regulation of BDNF gene transcription in the consolidation of fear memory. J. Neurosci. 28, 10576–10586 (2008).85 F. M. Vassoler, S. L. White, H. D. Schmidt, G. Sadri-Vakili, R. C. Pierce, Epigenetic inheritance of a cocaine-resistance phenotype.Nat. Neurosci. 16, 42–47 (2013).86 B. G. Dias, K. J. Ressler, Parental olfactory experience influences behavior and neural structure in subsequent generations. Nat. Neurosci. 17, 89–96 (2014).87 K. Gapp et al., Implication of sperm RNAs in transgenerational inheritance of the effects of early trauma in mice. Nat. Neurosci. 17, 667–669 (2014).88 D. A. Dickson et al., Reduced levels of miRNAs 449 and 34 in sperm of mice and men exposed to early life stress. Transl. Psychiatry 8, 101 (2018).89 F. Ciabrelli et al., Stable Polycomb-dependent transgenerational inheritance of chromatin states in Drosophila. Nat. Genet. 49, 876–886 (2017).90 M. J. Meaney, A. C. Ferguson-Smith, Epigenetic regulation of the neural transcriptome: The meaning of the marks. Nat. Neurosci. 13, 1313–1318 (2010).91 H. A. Hofmann, The neuroendocrine action potential. Winner of the 2008 Frank Beach Award in Behavioral Neuroendocrinology. Horm. Behav. 58, 555–562

(2010).92 H. Thomassin, M. Flavin, M. L. Espinas, T. Grange, Glucocorticoid-induced DNA demethylation and gene memory during development. EMBO J. 20, 1974–

1983 (2001).93 T. Klengel et al., Allele-specific FKBP5 DNA demethylation mediates gene-childhood trauma interactions. Nat. Neurosci. 16, 33–41 (2013).94 C. Anacker, K. J. O’Donnell, M. J. Meaney, Early life adversity and the epigenetic programming of hypothalamic-pituitary-adrenal function. Dialogues Clin.

Neurosci. 16, 321–333 (2014).95 G. Turecki, M. J. Meaney, Effects of the social environment and stress on glucocorticoid receptor gene methylation: A systematic review. Biol. Psychiatry 79, 87–

96 (2016).96 N. Snyder-Mackler et al., Social status alters chromatin accessibility and the gene regulatory response to glucocorticoid stimulation in rhesus macaques. Proc.

Natl. Acad. Sci. U.S.A. 116, 1219–1228 (2019).97 N. Matosin, T. Halldorsdottir, E. B. Binder, Understanding the molecular mechanisms underpinning gene by environment interactions in psychiatric disorders:

The FKBP5 model. Biol. Psychiatry 83, 821–830 (2018).98 A. S. Zannas, T. Wiechmann, N. C. Gassen, E. B. Binder, Gene-stress-epigenetic regulation of FKBP5: Clinical and translational implications.

Neuropsychopharmacology 41, 261–274 (2016).99 T. Rein, FK506 binding protein 51 integrates pathways of adaptation: FKBP51 shapes the reactivity to environmental change. BioEssays 38, 894–902 (2016).

100 N. C. Gassen et al., Chaperoning epigenetics: FKBP51 decreases the activity of DNMT1 and mediates epigenetic effects of the antidepressant paroxetine. Sci.Signal. 8, ra119 (2015).

101 N. Volk et al., Amygdalar microRNA-15a is essential for coping with chronic stress. Cell Rep. 17, 1882–1891 (2016).102 Z. Dai, X. Dai, Q. Xiang, J. Feng, Robustness of transcriptional regulatory program influences gene expression variability. BMC Genomics 10, 573 (2009).103 K. Sakata et al., Role of activity-dependent BDNF expression in hippocampal-prefrontal cortical regulation of behavioral perseverance. Proc. Natl. Acad. Sci.

U.S.A. 110, 15103–15108 (2013).104 K. E. Savell et al., A neuron-optimized CRISPR/dCas9 activation system for robust and specific gene regulation. bioRxiv:10.1523/eneuro.0495-18.2019 (17 July

2018).105 K. Minatohara, M. Akiyoshi, H. Okuno, Role of immediate-early genes in synaptic plasticity and neuronal ensembles underlying the memory trace. Front. Mol.

Neurosci. 8, 78 (2016).106 D. Avey et al., Single cell RNAseq uncovers a robust transcriptional response to morphine by oligodendrocytes. Cell Rep. 24, 3619–3629 (2018).107 Y. E. Wu, L. Pan, Y. Zuo, X. Li, W. Hong, Detecting activated cell populations using single-cell RNA-seq. Neuron 96, 313–329.e6 (2017).108 S. E. London, S. Dong, K. Replogle, D. F. Clayton, Developmental shifts in gene expression in the auditory forebrain during the sensitive period for song learning.

Dev. Neurobiol. 69, 437–450 (2009).109 M. J. Meaney, M. Szyf, Maternal care as a model for experience-dependent chromatin plasticity? Trends Neurosci. 28, 456–463 (2005).110 B. Cohen, H. Dafni, G. Meir, A. Harmelin, M. Neeman, Ferritin as an endogenous MRI reporter for noninvasive imaging of gene expression in C6 glioma tumors.

Neoplasia 7, 109–117 (2005).111 J. Chu et al., Non-invasive intravital imaging of cellular differentiation with a bright red-excitable fluorescent protein. Nat. Methods 11, 572–578 (2014).112 N. P. Daskalakis, H. Cohen, G. Cai, J. D. Buxbaum, R. Yehuda, Expression profiling associates blood and brain glucocorticoid receptor signaling with trauma-

related individual differences in both sexes. Proc. Natl. Acad. Sci. U.S.A. 111, 13529–13534 (2014).113 V. Michopoulos, S. D. Norrholm, T. Jovanovic, Diagnostic biomarkers for posttraumatic stress disorder: Promising horizons from translational neuroscience

research. Biol. Psychiatry 78, 344–353 (2015).114 T. L. Ben-Shaanan et al., Activation of the reward system boosts innate and adaptive immunity. Nat. Med. 22, 940–944 (2016).115 C. C. Rittschof, K. A. Hughes, Advancing behavioural genomics by considering timescale. Nat. Commun. 9, 489 (2018).

Clayton et al. PNAS Latest Articles | 9 of 9