Introduction

Chromatin is a plastic entity, which adapts to external stimuli such as a changing environment and stress [1, 2]. This feature shapes the endogenous response to a changing environment via long-term modulation of gene expression [3, 4]. Psychosocial stress causes an epigenetic remodeling of stress-responsive DNA and chromatin regions, and associated histone proteins in susceptible brain areas such as the ventral hippocampus [1, 5], amygdala [6], nucleus accumbens (NAc) [7], and the prefrontal cortex (PFC) [8] – dorsal raphe nucleus loop [9,10,11]. Similarly, and in an opposite fashion, a remodeling of chromatin regions involved with synaptic plasticity, neurotransmission, neurogenesis, and neuroinflammation is required in infralimbic and prelimbic neurons to elicit antidepressant effects [1, 9,10,11,12,13]. Hence, drugs which directly or indirectly affect the epigenetic control of chromatin regions linked to neurotransmission, neurogenesis, and neuroinflammation, represent promising epigenetics-based therapeutics.

Rapid-acting antidepressants (referred herein as RAADs) are being investigated as novel therapeutics in psychiatry [14,15,16]. These compounds are generally classified as serotonin (5-HT)2A receptor agonists such as psilocybin, lysergic acid diethylamide (LSD), N, N-Dimethyltryptamine (DMT) and 2,5-Dimethoxy-4-iodoamphetamine (DOI) [14]; N-Methyl-D-Aspartate (NMDA) antagonists such as ketamine [17]; and empathogens such as 3,4-methylenedioxymethamphetamine (MDMA), which act mainly through neurotransmitter transporters [18]. The long-term improvements in psychiatric symptoms and neuropsychological function elicited by RAADs [19,20,21,22,23,24] are accompanied by enduring serotonergic (such as a peripheral increase of serotonin -5-HT- platelets uptake sites in long-term Ayahuasca users) [25], and neuromorphological changes (such as thinner precuneus and posterior cingulate cortex and thicker anterior cingulate cortex in long-term Ayahuasca users) [26], which suggest the entrainment of epigenetic mechanisms.

Preliminary evidence indicates that the therapeutic effects of RAADs, akin to conventional antidepressants, are accompanied by a remodeling of DNA methylation, histone post-translational modifications (PTMs), and non-coding RNAs (ncRNAs) dynamics in stress-responsive brain areas. Interestingly, repeated, binge, or high-dose administration of RAADs elicit side effects mediated by overlapping yet opposite epigenetic mechanisms, such as hippocampal apoptosis, reduced neurotrophic signaling, increased neuroinflammation, and cognitive impairments (see Involvement of Epigenetic Mechanisms in Potential Side Effects Elicited by RAADs). These effects, contrasting with those of clinically relevant doses and resembling stress-induced epigenetic responses, underscore the importance of identifying the optimal dosages and regimens for clinical RAADs applications to maximize desirable epigenetic outcomes and minimize undesirable ones.

Epigenetic Control in Stress-Related Disorders, and the Effects of Conventional and Rapid-Acting Antidepressants

DNA methylation, stress, and rapid-acting antidepressants

DNA methylation consists in the reversible addition of methyl groups to cytosines in CpG dinucleotides by DNA methyltransferases (DNMTs). This modification generally modulates transcription by inhibiting the binding of transcription factors or recruiting proteins involved in gene repression [27]. Stress and psychosocial trauma modulate the methylation status of CpG dinucleotides in DNA sequences involved with the stress response, and synaptic and neuronal plasticity [28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43]. For example, the methylation status of genes encoding the glucocorticoid receptor, arginin-vasopressin, and the brain-derived neurotrophic factor (BDNF) can be modulated by early-life adversities as early as 1 day after birth [28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43] and specific cell types such as astrocytes carryover the epigenetic marks of stress [35, 44]. Epigenetically-active compounds with DNA demethylase activity [45] have antidepressant effects [46, 47], confirming that the methylation state and the associated psychiatric symptoms can be reversed pharmacologically [48]. Stressors also affect the regulation of DNA-methylating enzymes such as DNMT3a and 3b [49] in stress-susceptible brain areas such as the PFC and hippocampus [35, 44, 50]. Therefore, stress exposure is physically “remembered” in specific cell types and brain areas, causing altered gene expression and ultimately lasting effects on behavior. The most studied example of epigenetic programming by stress, and its normalization by antidepressants, is the BDNF gene. Prenatal and life adversities decrease the promoter IV-mediated BDNF expression via increasing its methylation [36, 43, 49, 51,52,53,54,55,56,57,58], and psychiatric disorders largely share increased BDNF promoter methylation, which can be reversed by conventional antidepressants [36, 54,55,56,57,58,59,60,61].

Like conventional antidepressants, but only requiring one or few administrations, ketamine restores some of the stress-induced aberrant DNA methylation, including on the BDNF gene (Fig. 1 and Table 1). Delving into the molecular mechanisms, ketamine induces a Extracellular Signal-Regulated Kinase 1 (ERK1) / nuclear receptor binding protein 1 (NRBP1)-mediated microglial decrease of methyl-CpG binding protein 2 (MeCP2), which upregulates the cAMP response-binding protein (CREB)-mediated BDNF exon IV transcription [62]. The enhancement of fear extinction elicited by ketamine is also accompanied by demethylation of PFC and hippocampal BDNF exon IV, and increased BDNF transcription [49]. Similarly, fluoxetine dissociates the MeCP2–CREB complex from BDNF promoter IV through the protein kinase A (PKA)-mediated phosphorylation of MeCP2, thus increasing BDNF transcription [55]. Resembling chronic imipramine’s effects in the PFC [50], acute ketamine reversed the nerve injury-induced increase in the hippocampal DNMT1, 3a, and 3b. It also restored the decrease in hippocampal total BDNF and BDNF exon I transcription and protein levels [52]. In rodents, chronic fluoxetine, imipramine, and desipramine blunted the stress-induced increase in hippocampal DNA methylation, and restored the decrease elicited in the PFC [50, 63, 64]. Antidepressant use increased the peripheral methylation level of the 5-HT transporter [65], and the interleukin (IL)-6 gene [66], as well as telomerase activity in individuals with PTSD [67]. While it has been speculated that RAADs such as psilocybin might increase telomerase activity and telomere length, this hypothesis remains to be tested [68].

Fig. 1: Converging epigenetic mechanisms of rapid-acting and conventional antidepressants.
figure 1

Abbreviations: 5-HTRs, Serotonin Receptors; SIGMAR1, Sigma-1 Receptor; SERT, Serotonin Transporter; NET, Norepinephrine Transporter; TRKB, Tropomyosin Receptor Kinase B; BDNF, Brain-Derived Neurotrophic Factor; NMDA, N-Methyl-D-Aspartate; PLC, Phospholipase C; IP3, Inositol trisphosphate; DAG, Diacylglycerol; MAPK, Mitogen-Activated Protein Kinase; ERK, Extracellular Signal-Regulated Kinase; PI3K, Phosphoinositide 3-Kinase; Akt, AKT Serine/Threonine Kinase 1; PKC, Protein Kinase C; GSK-3, Glycogen Synthase Kinase 3; NF-kB, Nuclear Factor Kappa-B; CaMKs, Calcium Calmodulin Dependent Protein Kinases; Wnt, Proto-Oncogene Wnt-1; JAK, Janus Kinase; STAT, Signal Transducer and Activator of Transcription; PLCγ, Phospholipase C Gamma 1; SSRI, Selective Serotonin Reuptake Inhibitors; SNRI, Serotonin-Norepinephrine Reuptake Inhibitors; MAOI, Monoamine Oxidase Inhibitors; H3, Histone 3; H4, Histone 4; H3K4, Histone 3 Lysine 4; H3K9, Histone 3 Lysine 9; H3K12, Histone 3 Lysine 12; H3K27, Histone 3 Lysine 27; PKA, Protein Kinase A; Mecp2, Methyl-Cpg Binding Protein 2; CREB, cAMP Response Element-Binding Protein; G9a, Histone Methyltransferase G9a; HDAC, Histone Deacetylase; DNMT, DNA Methyltransferase; lncRNA, Long non-Coding RNA; circRNA, Circular RNA; miRNA, micro RNA; NR3C1, Glucocorticoid Receptor; CRH31, Corticotropin Releasing Hormone; PTSD, Post-Traumatic Stress Disorder; LSD, lysergic acid diethylamide.

Table 1 Effects of rapid-acting antidepressant on DNA methylation, histone post-translational modifications, and ncRNAs.

Amongst psychedelics, preliminary evidence indicates that LSD and Ayahuasca affect DNA methylation (Fig. 1 and Table 1). Repeated LSD largely increased the methylation level of 635 CpG sites in the PFC of adult male mice receiving a prosocial, anxiolytic, and synaptoplastic repeated regimen of LSD [69,70,71,72]. The genes interested were involved with neurotropic-, neurotrophic-, and neuroplasticity-related signaling. Proteomics profiling and qRT-PCR validation suggested specificity and functional significance of the changes observed in DNA methylation [69]. Given that cytosine methylation can affect the likelihood of mutations [73, 74] future studies are warranted to investigate the potential effects of the increased DNA methylation elicited by LSD on genomic stability and mutation rates. Despite the fact that the use of RAADs has been associated with decreased likelihood of suicide [75, 76], DNA hypermethylation is also observed in the blood of individuals who attempted suicide [77] and in the PFC of suicide completers [78]. Hence, future studies are required to better characterize the relationships between the epigenetic effects elicited by psychedelics and suicidality. Participation in a ceremonial Ayahuasca retreat in the Amazon consisting of multiple administrations increased the saliva methylation level of 5 CpG sites within the Sigma-1 receptor gene promoter (in a greater fashion in individuals with greater childhood trauma), but did not affect the methylation level of one CpG site within the FKBP5 gene, which is involved with trauma and the stress response [79]. These changes were accompanied by improved depression and anxiety scores for up to 6 months [79]. While they are preliminary findings, previous biochemical and neurostructural evidence supports the existence of epigenetic mechanisms accompanying the regular consumption of Ayahuasca: increased platelets 5-HT uptake sites [25], thinner precuneus and posterior cingulate cortex, and thicker anterior cingulate cortex [26] were identified in long term Ayahuasca users. Speculations exist that Ayahuasca might extinguish the fear response via an epigenetic mechanism mediated by the DMT-activated Sigma-1 receptor [80], although this hypothesis remains to be tested. The β-carbolines contained in the Ayahuasca brew might also inhibit histone deacetylase activity, given that engineered derivatives containing β-carbolines motifs are potent HDACs inhibitors [81, 82]. However, no studies have insofar investigated the epigenetic effects elicited by β-carboline, or those elicited by the combination of β-carboline with DMT-containing plants (as in the case of Ayahuasca).

MDMA-assisted psychotherapy led to peripheral epigenetic changes correlated to therapeutic improvement [83]. The increase in saliva methylation level of one CpG site (cg08276280) in the corticotropin-releasing factor receptor 1 gene and one (cg01391283) within the glucocorticoid receptor gene was correlated with symptom reduction in individuals with severe PTSD receiving MDMA-assisted psychotherapy [83]. Further studies should characterize the epigenetic effects of psychotherapy coupled to RAADs, and if those treatments synergistically or additively interact to produce determinate epigenetic outcomes. Putative prophylactic or acute effects of RAADs on attenuating stress-induced DNA methylation changes remain to be elucidated.

Histone post-translational modifications, stress, and RAADs

Histones form protein cores around which DNA wraps. Stress and antidepressants affect histone post-translational modifications (PTMs) such as acetylation, methylation, and phosphorylation in stress-responsive chromatin areas, altering chromatin accessibility and transcription. For example, stress elicits hypoacetylation of hippocampal histone 3 (H3) at BDNF III and IV promoters, and increases the levels of histone deacetylases (HDACs) including HDAC5 and Sirtuin 2, ultimately decreasing BDNF expression [1, 84]. Indeed, specific HDACs such as HDAC2 and HDAC5, which are critical regulators of adult neurogenesis [85] and cognition [86], are altered in depression [87].

Some of the aberrant PTMs dynamics elicited by stress are restored by chronic administration of conventional antidepressants such as fluoxetine, paroxetine, reboxetine, citalopram, imipramine and mirtazapine. These drugs rescue the H3-H4 (i.e., H3 lysine (K)9 and H3K27) hypoacetylation of the BDNF promoters, reverse the stress-induced decrease in BDNF transcription and protein level, promote HDAC5 phosphorylation and nuclear export, and attenuate the stress-induced increase in HDACs in stress-susceptible brain areas [1, 55, 84, 88,89,90,91,92]. Delving into the mechanisms, fluoxetine was shown to disassociate the MeCP2-CREB-BDNF promoter IV complex via Protein Kinase A (PKA)- mediated phosphorylation of MeCP2, leading to CREB-mediated BDNF transcription [55]. Fluoxetine also inhibited the binding of delta-FosB to the Calcium/calmodulin-dependent protein kinase IIa (CaMKIIa) promoter by reducing its acetylation and increasing its H3K9 dimethylation, ultimately decreasing CaMKIIa expression in the NAc [93]. Eight-week escitalopram treatment in individuals with depression decreased H3K27 histone trimethylation, and these changes were negatively correlated with increased BDNF levels [58]. Importantly, the stress-induced hippocampal BDNF promoter-associated H3K9 hypoacetylation and HDAC2 overexpression can also be restored by non-pharmacological strategies such as acupuncture [94].

The epigenetic outcomes elicited by a single administration of ketamine largely overlap those elicited by chronic conventional antidepressants. Indeed, one or few ketamine administrations are sufficient to reverse the stress-induced histone hypoacetylation and downregulated neurotrophic-related transcription (Fig. 1 and Table 1). In a mouse model of Gulf War Illness, ketamine decreased the hippocampal HDAC1 and HDAC5 levels and increased the H3K9 acetylation of BDNF promoter IV, restoring BDNF levels and neuroplasticity [95]. Similarly, in rats exposed to early life adversities, ketamine decreased the depressive-like behaviors and restored NAc HDAC activity [7]. Moreover, high-dose ketamine had a prophylactic effect on attenuating the stress-induced increase in hippocampal H3K9 methylation [96]. Similarly to conventional antidepressants, ketamine increases HDAC5 phosphorylation and nuclear export, resulting in enhanced H3-H4 acetylation. This triggers the myocyte enhancer factor 2-mediated transcription of the plasticity-related genes Eukaryotic Translation Initiation Factor 4E Binding Protein 1 (eIF4EBP1) and CREB, and their target genes, and ultimately the onset of antidepressant-like effects [97].

Psychedelics such as LSD and DOI also have histone-acetylating properties, which might be involved in their therapeutic effects (Fig. 1 and Table 1). Early studies reported that LSD acutely increases histone acetylation in the midbrain and cortex, but not the cerebellum of rabbits [98]. The increased histone acetylation activity elicited by LSD suggests gene activation, and indeed, LSD activates the transcription [99] of immediate early genes and genes involved with synaptic potentiation and neurotropism in the hippocampus, cortex, midbrain [100], and brainstem [99]. Given that enhancing hippocampal and PFC histone acetylation ameliorates fear extinction [101] and the consolidation of cued fear extinction [102] via a permissive effect over transcription [101], the increased histone acetylation elicited by LSD could be involved in its therapeutic action for alcohol addiction, and distress associated with a life-threatening condition [103,104,105,106].

A single administration of the LSD analog DOI also elicited a sustained modulation of the acetylation level of the transcriptional enhancer histone H3K27 in neurons of the mouse PFC [107]. This sustained effect had highly specific spatio-temporal dynamics, was accompanied by neurotrophic-related transcriptional shifts, and was associated to enhanced synaptic plasticity and antidepressant-like activity [107]. Lastly, the recently identified histone PTMs serotonylation and dopaminylation [108,109,110] might potentially be involved in the therapeutic effects of psychedelics, although further investigations are required. Together, the therapeutic effects of RAADs, similarly to conventional antidepressants, are accompanied by PTMs opposite to those elicited by stress, leading to increased accessibility of genes involved with neurotrophic signaling and antidepressant response.

Non-coding RNAs, stress, and RAADs

Non-coding RNAs (ncRNAs) are portions of the genome which are transcribed but not translated, with roles in tissue-specific selective or cooperative regulation of splicing, transcription, and translation [111,112,113,114]. ncRNAs, including stress-responsive ones, are intricately linked to neuroplasticity, psychiatric disorders, and suicide, offering potential as biomarkers and therapeutics for psychiatric disorders [115,116,117,118,119,120,121,122]. Early-life and chronic stressors affect the PFC, amygdalar and hippocampal regulatory RNA network, influencing neurotransmission, neuroplasticity, neurogenesis, and behavior [118, 123,124,125,126,127]. Among these networks, microRNAs (miRNAs), approximately 22 nucleotides in length, stand out for their tissue-specific regulation of a significant proportion of protein-coding genes expression by binding mRNA untranslated regions. Dysregulated miRNA dynamics are observed in depression and suicide, with studies pinpointing alterations in the PFC and hippocampus during depressive states [121, 122, 128, 129]. Conventional antidepressants influence miRNA expression patterns. For instance, miRNAs such as miR-16, miR-124, miR-132, and miR-135 are implicated in antidepressant response, with drugs like fluoxetine and venlafaxine, imipramine, sertraline, and citalopram modulating their expression, as well as the expression of miR-18a, miR-34a, miR-326, miR-1202 and miR-1971 [117, 130,131,132,133,134,135,136,137,138,139,140,141].

RAADs such as ketamine also alter miRNA pathways, with PFC miR-29b-3p, miR-98-5p, and miR-132-5p playing important roles in mediating the upregulation of BDNF leading to ketamine’s antidepressant-like effects (Fig. 1 and Table 1) [142,143,144]. In preclinical studies, ketamine modulated some of the stress-dysregulated hippocampal miRNAs (i.e., miR-598), which are also regulated by other antidepressant strategies such as fluoxetine and electroconvulsive therapy (ECT) [124]. Ketamine also decreased miR-451 levels and increased those of the RNA-binding protein hur-6, a de-repressor miRNA “sponge” for deleterious miRNAs involved in inflammatory responses [124, 145, 146]. ECT and ketamine in rodents exposed to early-life adversities affected 43 common miRNA targets, 7 of which were reversals of stress-induced changes. Interestingly, ketamine rescued some of the stress-induced miRNAs alterations which were not reversed by fluoxetine [124], suggesting that some of the epigenetic mechanisms engaged by conventional and RAADs diverge. A three-day ketamine regimen upregulated 23 miRNAs and downregulated 15 including miR-206 [147] (a negative modulator of BDNF) in vivo and in vitro, while increasing BDNF and decreasing apoptosis [147,148,149]. Ketamine time-dependently modulated the transcription of a cluster of hippocampal (but not PFC) miRNAs (764-5p, 1912-3p, 1264-3p, 1298-5p, and 448-3p) which are hosted in the 5-HT2C gene locus, mediated by the inhibition of glycogen synthase kinase 3 [150]. Acutely, ketamine increased the prefrontal level of miR-148a-3p and decreased miR-128-3p, miRNAs involved with the ubiquitin proteasome system [151]. Ketamine also elicited immunomodulatory prophylactic effects in lipopolysaccharide (LPS)-treated mice through miR-149 [152], indicating that some of the immunomodulatory effects of ketamine involved in its antidepressant effects [153] might be mediated by ncRNAs.

Similarly to what observed with conventional antidepressants [140, 154,155,156,157,158,159] the miRNome might predict the therapeutic response elicited by RAADs. For example, the non-responder status to ketamine was predicted by lower pretreatment level of miR-548d-5p and miR-605 in individuals with neuropathic pain [160, 161]. However, in a study in individuals with treatment-resistant depression receiving a ketamine infusion, no significant effects on whole blood miRNA levels were detected 24 hours after [162]. Further studies are required to identify miRNAs that can predict favorable treatment outcome in response to RAADs to increase therapeutic precision.

Circular RNAs (circRNAs), ncRNAs generated by back-splicing, have extensive complementarity to target mRNAs and can encode proteins or increase the expression of the target gene(s) [163]. A regimen of repeated ketamine in rats increased the hippocampal expression of 4, and decreased the expression of 1 circRNA, with predicted downstream effects on genes involved in calcium signaling, G protein signaling, protein phosphorylation, Mammalian Target of Rapamycin (mTOR) signaling, transcription, alternative splicing, and neuroplasticity [164]. Long non-coding RNAs (lncRNAs) are a class of ncRNAs longer than 200 nucleotides with brain area-specific expression patterns involved with neural differentiation and plasticity [165]. lncRNA dynamics are stress- and antidepressant-responsive, are altered in several neuropsychiatric disorders including Major Depressive Disorder (MDD), and might predict the antidepressant response [118, 127, 166]. For example, greater decreases of the lncRNA FEDORA predicted the decrease in depression severity following ketamine in individuals identifying as female who experience MDD [167].

Potential mechanisms of rapid-acting antidepressants-induced epigenetic changes

Neuronal activity-mediated epigenetic effects

One mechanism through which psychedelics might affect the epigenetic regulation of neurotrophic-related gene expression is via neuronal activity-induced modification of chromatin structure and accessibility. The membrane depolarization of cortical neurons results in chromatin remodeling and enhanced BDNF gene accessibility via a) decreased methylation of BDNF promoter III and IV [168, 169], b) H3K4 dimethylation of BDNF promoter IV, and c) HDAC1 and mammalian Switch-independent 3 A promoter dissociation [169]. Additionally, upon neuronal depolarization, CaMKII-mediated phosphorylation and release of the methyl-binding protein MeCP2 from the BDNF promoter III permits BDNF promoter III-dependent transcription, causing dendritic plasticity in neurons [168, 170]. Additionally, the ketamine-induced loss of somatostatin-expressing interneurons dendritic inhibition leads to heightened synaptic calcium transients in the dendritic spines of pyramidal neurons in the PFC. This process might play a role in the epigenetic effects observed following ketamine administration through altered calcium signaling, leading to increased activity-dependent synaptic plasticity [171,172,173]. DOI activates specific subsets of 5-HT2A- and metabotropic glutamate receptor 2 (mGlu2)-expressing glutamatergic neurons and interneurons, as well as astrocytes of deep-layer prefrontal and somatosensory cortices, the claustrum, and the insula [174], and this activation might be involved in the sustained epigenetic and antidepressant-like effects elicited [107]. A single administration of DOI induced prolonged alterations in the acetylation level of neuronal transcriptional enhancers, causing sustained effects over axonogenesis, synapse organization, assembly, and function, and receptor internalization and activity [107]. Despite this preliminary correlative evidence, given that physiological neuronal activation itself can profoundly modulate the epigenetic landscape de novo [175], this calls into questions whether the epigenetic effects elicited by RAADs are causally involved in the therapeutic effects, or neuronal activity-driven changes. Future studies are encouraged to address this research question.

5-HT receptors signaling-mediated epigenetic effects

Psychedelics might activate similar epigenetic mechanisms to those activated by endogenous 5-HT signaling through various 5-HT receptors. 5-HT triggers a transient remodeling of chromatin structure mediated by 5-HT1A and BDNF-TRKB signaling which decreases HDAC5 expression, and increases cortical H3K9 acetylation of the BDNF promoter and BDNF levels, reinstating plasticity in the adult nervous system [176]. 5-HT signaling also induces changes in the methylation status of the promoter of CREB2, a plasticity-related transcription factor involved in long-term facilitation [177, 178]. The properties of psychedelics resemble those of 5-HT: they interact with 5-HT receptors such as the 5-HT1A and 5-HT2A receptors, triggering biased signaling and the transcription of genes involved in neurotransmission, neuroplasticity, and neuroimmunomodulation [14,15,16]. Given that these 5-HT receptors are fundamentally involved in governing cortical circuits mediating cognitive and executive functions [179, 180], the activation of these receptors by RAADs might result in specific epigenomic fingerprints and chromatin architecture changes, ultimately leading to structural and functional synaptic changes. The epigenetic effects elicited in discrete cell types through activation of 5-HT receptors, might alter circuit and network activities, putatively mediating therapeutic improvement at the circuit-, network-, and system-levels.

TRKB signaling-mediated epigenetic effects

Like classical antidepressants [181], a mechanism that might mediate the epigenetic effects of RAADs is the activation of endogenous BDNF signaling through direct binding to and activation of the TRKB receptor [182,183,184,185]. Activation of TRKB signaling results in the activation of other cascades such as the CaMKII, RAS/MAPK and phosphoinositide 3 kinase pathways [14, 17, 186,187,188]. Moreover, TRKB signaling leads to altered chromatin structure of the BDNF gene, and transcriptional upregulation of the BDNF gene and its downstream effector target AKT Serine/Threonine Kinase 1 (Akt)-mTOR. These events ultimately cause an increase in neuroplasticity, synaptic potentiation, and antidepressant effects. The outcome is the reopening of a period of structural and functional plasticity within the CNS that can be harnessed therapeutically [182,183,184, 189, 190]. Supporting a fundamental role of the reopening of critical periods of plasticity, RAADs lead to the degradation of the extracellular matrix in the NAc, creating the prerequisites to enable metaplasticity [190]. Whether epigenetic mechanisms are involved in this process remains unknown.

Interestingly, the epigenetic changes elicited by LSD resemble those observed during critical periods of neurodevelopment and synaptic plasticity, thus seemingly re-creating in the adult nervous system the epigenetic correlates observed during neurodevelopment and learning [69]. Indeed, CCAAT enhancer binding protein beta (CEBP2), a co-activator and target gene of CREB, was a “top-hit” in the PFC in terms of differential DNA methylation following repeated LSD [69]. In accordance, CEBP2 together with several other neuroplasticity-related genes is dose-dependently transcriptionally modulated in the PFC following psilocybin administration [191]. Accordingly, a single administration of psilocybin following chronic stress during adolescence in rats was sufficient to normalize depressive-like and cognitive-like behaviors [192]. Lastly, given that (i) psychedelics interact with TRKB, 5-HT2A, and mGlu2 receptors, that (ii) TRKB interacts with the 5-HT2A [193] and the mGlu2 [194] receptors, and that (iii) the latter two receptors interact with each other [195], these interactions may be functionally relevant from an epigenetic standpoint.

Direct interaction with DNA, histones, or chromatin remodeling complexes

Another intriguing mechanism supported by several lines of evidence is that RAADs may directly interact with DNA, histones, or chromatin remodeling complexes, thus acting as direct epigenetic modulators. Indeed, the β-carbolines contained in the brew Ayahuasca directly interact with DNA in vitro [196], and early studies observed that LSD directly binds DNA or histone proteins [197,198,199]. For psychedelics to interact with DNA or histones in vivo, they would first need to reach the intracellular space and the nucleus. Indeed, membrane permeability is essential for the neuroplastic effects of psychedelics, and ligand-bound 5-HT2A receptors are internalized in vivo and in vitro [200,201,202,203]. Additionally, neurotrophic factors are uptaken and transported to the cell body through retrograde axonal transport. Given that RAADs bind the TRKB receptor [186], they might similarly be internalized by nerve terminals. In support of this hypothesis, DMT is internalized via a three-step process requiring active transport: 1) it is actively transported across the blood-brain barrier, 2) acts as a 5-HT transporter substrate on the neuronal plasma membrane, and 3) it is internalized by neuronal cells, and stored into synaptic vesicles by the neuronal vesicle monoamine transporter 2 for up to 1 week [204,205,206]. Whether the sequestered DMT in synaptic vesicles or the DMT-activated Sigma-1 receptor interact with DNA or chromatin remodeling complexes remains to be assessed [206, 207]. Recently, it was reported that 5-HT can covalently bind to glutamine 5 on the trimethylated H3K4 histone, eliciting a permissive transcriptional influence that modulates the interaction of histone PTMs with chromatin readers during neuronal differentiation [109]. Due to the structural similarity of 5-HTergic psychedelics to 5-HT, it remains to be assessed whether they modulate the relationship between histone PTMs and chromatin readers.

The mechanisms discussed here are not necessarily mutually exclusive and may be part of a series of events responsible for reopening a period of neurodevelopmental-like neuroplasticity, providing a neurobiological substrate that can be harnessed therapeutically. Achieving this seems to require appropriate support and integration before, during, and following psychedelic-assisted psychotherapy. Given that the different mechanisms modulating chromatin structure and DNA accessibility may interact with one another, deciphering the “psychedelic epigenome” represents a new frontier in psychiatry.

Involvement of epigenetic mechanisms in the potential side effects elicited by RAADs

While preliminary evidence suggests that the epigenetic effects induced by RAADs may contribute to therapeutic improvement, uncertainties persist regarding side effects. Studies have documented epigenetic alterations following the abuse of RAADs, which appear antithetical to those elicited by clinically relevant doses, yet similar to those elicited by stress. These changes are reminiscent of neuronal, immune and cardiac dysfunction. Most of the potential epigenetic-mediated side effects associated with RAADs were observed following repeated administration of higher doses of ketamine during gestation, lactation, and the neonatal period (Table 1).

Immune-related side effects

High or repeated doses of ketamine may cause side effects through the epigenetic modulation of immune function, via (i) upregulation of permissive H3K4 trimethylation, (ii) downregulation of repressive H3K27 and H3K36 trimethylation, and (iii) DNA hypomethylation at nuclear factor-κB (NF-κB)-responsive promoters [208, 209], as well as iv) modulation of proinflammatory, neurotoxic, and oxidative signaling through RNA regulatory pathways [152, 210,211,212,213,214,215,216]. Repeated MDMA administration also causes epigenetic marks reminiscent of immune dysfunction such as hypermethylation of the Terminal Nucleotidyltransferase 5B gene, a chaperone involved with cyclooxygenase 2 maturation [217]. Future studies are warranted to further investigate potential epigenetics-mediated immune side effects arising from repeated use and abuse of RAADs.

Cardiac hypertrophy

In a model of ketamine abuse during early gestation, ketamine decreased the acetylation level of cardiac histone H3K9 at the Myosin Light Chain 2 promoter by increasing HDAC3 level and histone deacetylase activity. These effects were accompanied by altered expression of several genes related to cardiomyogenesis, resulting in enlarged heart and ventricular chamber, thinner ventricular wall, degeneration of cardiomyocytes, and reduced systolic function [218]. Repeated MDMA exposure also led to hypermethylation of cardiac DNA promoter regions resulting in cardiac hypertrophy and progressive damage [217]. Given the concern surrounding the potential impact of RAADs on cardiovascular function through cardiac 5-HT receptors [219], future investigations are recommended to explore potential deleterious effects at clinically relevant doses and regimens.

Deleterious gestational and transgenerational effects

While few studies have assessed the transgenerational effects of RAADs, maternal intake during pregnancy and lactation may similarly to conventional antidepressants lead to epigenetic modifications and altered anxiety and depressive-like behavior in the offspring [56, 220,221,222]. For example, ketamine abuse during gestation impacts the epigenetic make-up of the fetus, and can result in impaired neurocognitive function in the offspring [218, 223]. Similarly, sociability, emotional behavior, and increased PFC dopamine levels are observed in the offspring of rats following repeated Ayahuasca administration during pregnancy and lactation [224].

Potential for addiction

While there is general consensus that psychedelics have low potential for addiction, psychological dependence or reinforcing effects, the NMDA antagonist ketamine [225, 226] and the empathogen MDMA [227, 228] have been associated with drug-seeking behaviors accompanied by epigenetic changes, similarly to drugs of abuse [229, 230]. For instance, one study found that ketamine-induced conditioned place preference resulted in differential expression of 34 hippocampal miRNAs [225], while another study showed that the conditioned place preference was accompanied by altered expression of serum exosomal miRNAs associated with processes such as nervous system development, neuron generation and differentiation, and apoptosis [231]. Acute and repeated MDMA exposure altered histone PTMs in the NAc promoter regions of the opioid-related nociceptin/orphaninFQ (N/OFQ)- Nociceptin Receptor (NOP) and dynorphin (DYN)- Kappa-Opioid Receptor (KOP) systems [232]. Specifically, acute MDMA increased the H3K4 methylation at N/OFQ and DYN promoters, while increasing H3K9 acetylation and decreasing H3K9 dymethylation at the DYN promoter. Acute and repeated MDMA also decreased H3K9 acetylation at the N/OFQ promoter [232]. These opioidergic effects could have direct implications for the therapeutic effects of MDMA on trauma [233], and for MDMA’s potential for addiction and neurotoxicity [234]. Considering the known involvement of epigenetics and transcriptional changes in addiction, it is prudent to investigate through cell-type and brain area-specific approaches whether epigenetic mechanisms may play a role in putative addictive, or anti-addictive, properties of RAADs.

Neurotoxicity

Several studies have reported neurotoxic effects of RAADs arising from repeated administration or higher doses. While such effects are not triggered in vivo by clinically relevant doses, this knowledge remains relevant from a harm-reduction perspective. Repeated ketamine increased histone deacetylase and DNMT activity, in an opposite fashion to the homolog changes elicited by clinically-relevant doses [235]. Additionally, repeated high-dose ketamine in young mice led to neurotoxicity, neuronal degeneration and memory impairments during adulthood, at least partially through miR-34c [236], miR-124 [237], miR-137 [238], and miR-199a-5p [239] signaling. In adolescent rats, repeated ketamine high-doses elicited hippocampal apoptosis and cognitive impairments, putatively mediated by hippocampal miR-214 downregulation/ Phosphatase And Tensin Homolog upregulation [240] and miR-34a upregulation/ Fibroblast Growth Factor Receptor 1 downregulation [241]. Repeated high-dose ketamine also elicited schizophrenia-like behavior accompanied by modulation of the miRNome and neurotrophic-related gene expression in the PFC and hippocampus [242, 243]. In vitro studies corroborate the neurotoxic effects of ketamine at higher doses through epigenetic mechanisms involving HDAC6, miR-22, miR-124, miR-497-5p, the lncRNAs SPRY4-IT1, LINC00641, and SNHG16, and the BDNF antisense RNA [214, 215, 237, 244,245,246,247,248,249,250,251]. Together, future research exploring the neurotoxic effects of RAADs abuse through epigenetic changes is encouraged.

Cancer

Several of the epigenetic effects elicited by RAADs involve signaling cascades related to cell growth, proliferation, and cancer, such as mTOR and BDNF [252, 253]. Therefore, epigenetic changes in these genes might induce or affect the progression of cancer. Considering that psychedelic-assisted psychotherapy is a promising treatment to attenuate psychological and existential distress in individuals facing a life-threatening conditions such as cancer, it is important to assess the neoplastic potential of RAADs [254,255,256]. If RAADs affected DNA methylation and other epigenetic mechanisms indiscriminately, for example by increasing or decreasing global DNA methylation aspecifically, this could pose a risk factor by causing genomic instability or repressing the transcription of tumor-suppressor genes [257,258,259]. A recent study reported oxidative DNA damage following the administration of higher-than clinically-relevant doses of psilocybin in the PFC and hippocampus [260]. Early reports suggested that ingesting or being exposed in utero to psychedelics such as LSD might be mutagenic and cause chromosomal damage and potential delayed mutations in the offspring (reviewed in [261]). However, subsequent experimental and epidemiological studies failed to replicate the findings, even in long-term psychedelic users [261, 262]. One investigation found that ketamine elicited ferropoptosis, and may thus elicit anti-neoplastic effects through the lncRNA Pvt1 oncogene /miR-214-3p signaling [263]. Given this evidence, further studies are warranted to investigate the potential of RAADs to induce, accelerate, or perhaps counteract, cancer.

Conclusion

Preliminary correlative evidence indicates that the epigenetic mechanisms engaged by RAADs converge, like conventional antidepressants, on permissive mechanisms over chromatin areas involved in neuroplasticity within stress-vulnerable brain regions. Importantly, these epigenetic changes counteract those induced by psychosocial stress, early-life adversities, and substance abuse. Yet, the abuse of RAADs triggers epigenetic marks resembling those elicited by stress.

Significant knowledge gaps remain, such as the mechanistic relationship between RAADs and epigenetic effects, the timing of onset of these epigenetic changes and their duration, the mechanisms responsible, cell type specificity, the role of metabolites, and the effects of “microdosing” (the ingestion of 1/10th–1/20th of a “large” dose). The polypharmacological nature of RAADs implies that each compound, dose, and regimen may uniquely affect the epigenetic control of gene expression, with potential indications or contraindications for various psychiatric disorders. Understanding how the subjective experience and environment modulate the epigenetic effects accompanying the administration of RAADs is an area ripe for exploration. Determining the potential modulation by the environment on the epigenetic effects of psychedelics could inform ongoing clinical investigations and expand our understanding of epigenetics. Brain area- and cell type-specific methods, facilitated by techniques like cell-sorting approaches and single-cell omics approaches, could address these questions [264, 265]. Identifying predictors [266] and modulators (i.e., set and setting, placebo effect) [267] of epigenetic responses accompanying RAADs administration could lead to improved therapeutic precision and efficacy.

While the epigenetic outcomes may causally contribute to therapeutic benefits, causative investigations lack. Given that epigenetic changes are observed in all cell types under disparate conditions (for example following neuronal activity or neurotoxic stimuli), the observed epigenetic changes could be secondary to behavioral changes induced by psychedelics. The potential epigenetic-mediated side effects, such as immune-related effects, cardiac hypertrophy, gestational and transgenerational effects, and cancer, also require further investigation. Future studies addressing these knowledge gaps could lead the way for developing novel neuroepigenetics-based precision therapeutics.