Brain-targeted epigenetic effects of two emerging psychoplastogens: ketamine & MDMA

Semple and colleagues introduce ketamine and MDMA as two emerging psychoplastogens, meaning compounds thought to promote structural and functional neural plasticity. They note that both agents show clinical promise for depression- and trauma-related symptoms, but the biological mechanisms underlying their benefits remain uncertain. In particular, although earlier research links these drugs to neuroplastic, neurotransmitter, neuroimmune, and stress-related effects, it is still unclear how they may produce longer-lasting therapeutic change after only a few treatments. The authors frame DNA methylation as a plausible mechanism and a potential peripheral biomarker of treatment response. They explain that prior work has suggested methylation changes in blood or saliva may track psychiatric outcomes, but that ketamine and MDMA had not yet been examined comprehensively using a brain-relevant, discovery-oriented approach. The study therefore aims to evaluate treatment-associated DNA methylation changes for both compounds using a Brain-Epigenome-Wide Association Study (BEWAS), focusing on genes relevant to brain health and function rather than the entire epigenome.

The researchers analysed DNA methylation data from participants enrolled in clinical trials of ketamine or MDMA. The study used pre- and post-treatment peripheral biospecimens: blood from the ketamine cohort and saliva from the MDMA cohort. The text presents this as a secondary analysis of trial samples, rather than a new intervention study. The ketamine cohort included 20 adults with moderate to severe major depressive disorder (MDD), post-traumatic stress disorder (PTSD), or both. The reported demographics were 15 women and 5 men, with a mean age of 40.24 years. Participants received six ketamine infusions at 0.5 mg/kg over two to three weeks. Blood was collected 24-48 hours before the first dose and again 10 days after treatment. The MDMA cohort was a convenience subsample of 16 adults with severe PTSD drawn from a Phase III trial. The reported demographics were 7 women and 9 men, with a mean age of 43.49 years. Participants received three oral MDMA doses, escalating from 80 mg to 180 mg, approximately four weeks apart, alongside psychotherapy. Saliva was collected at baseline and after treatment, although the extracted text does not clearly specify the exact post-treatment timing in this section. Methylation was assayed on the Infinium HumanMethylationEPICv1 BeadChip and preprocessed in R using minfi. The authors estimated cell composition with EpiDISH and included epithelial cell proportions for the MDMA cohort and neutrophil proportions for the ketamine cohort as covariates, to account for differences in peripheral cell mixtures. They then applied repeated-measures linear mixed-effects models, with age, sex, and estimated cell proportions as fixed effects, to assess treatment-associated methylation change. CpG-level results were summarised at the gene level using inverse-variance-weighted fixed-effects meta-analysis, and genes with false discovery rate-adjusted q values below 0.05 were considered significant. Nominally significant CpG sites were also entered into functional enrichment analyses using missMethyl, querying Gene Ontology and KEGG pathways while correcting for probe-number and annotation bias.

At the CpG level, ketamine treatment was associated with 8,344 nominally significant sites at p<0.05, with 43.38% showing decreased methylation and 56.62% increased methylation. However, no individual CpG sites remained significant after false discovery rate correction. For MDMA, 11,544 CpG sites were nominally significant, with 50.65% showing decreased methylation and 49.34% increased methylation, and again no single CpG site survived false discovery rate correction. Gene-level aggregation revealed stronger signals. After false discovery rate correction, ketamine was associated with 405 genes showing significant treatment-related methylation change, with roughly half showing positive and half negative beta coefficients (48.65% positive, 51.35% negative). MDMA was associated with 346 significant genes, again split fairly evenly between positive and negative effects (45.37% positive, 54.62% negative). Functional enrichment analyses identified 169 significant pathways for ketamine and 183 for MDMA. For ketamine, enriched terms were mainly related to neuronal projection, synaptic organisation, and neuroplasticity, with notable KEGG pathways including neuroactive ligand signalling, glutamatergic synapse, and dopaminergic synapse. For MDMA, enriched terms also clustered around neuronal projection and synaptic plasticity, with KEGG pathways including neurotrophin signalling and dopaminergic synapse. The two treatments showed partial overlap. In total, 521 CpG sites were altered by both compounds, but only 48.7% shared the same direction of change and the CpG-level effect estimates were not correlated (ρ=0.047, p=0.28). At the gene level, 47 genes overlapped, 27% shared directionality, and the gene-level effects were also not clearly correlated (ρ=-0.26, p=0.07). Although overlap at the site and gene level was modest, the authors report that many functional pathways converged across the two compounds.

The authors interpret the findings as evidence that peripheral DNA methylation changes occur after ketamine and MDMA treatment in brain-relevant gene systems. They argue that these changes are consistent with the broader idea that psychoplastogens engage coordinated biological processes linked to neuroplasticity, and that peripheral epigenetic signatures may act as translational markers of treatment-related biology. They relate the ketamine results to prior evidence of large-scale transcriptional remodelling after repeated ketamine exposure and to the drug’s known pharmacology, particularly NMDA receptor and glutamatergic signalling. For MDMA, they emphasise methylation changes in genes such as BDNF, NTRK3, SLC6A3, and GALR1, which they view as compatible with MDMA’s known effects on neurotrophin and monoaminergic systems. The authors also highlight EFNA5, a locus implicated in PTSD genetic risk, because both compounds were associated with increased methylation there. Beyond synaptic and plasticity-related pathways, they suggest that neuroimmune, metabolic, and stress-response signalling may also be involved. They mention enrichment of pathways such as chemokine signalling, FoxO signalling, IgSF CAM signalling, and adipocytokine signalling, and connect these to prior work on immune and inflammatory processes in depression and trauma-related disorders. However, they caution that the brain-enriched filtering strategy was not designed to systematically capture immune pathways, so these observations should be treated carefully. The authors further note that shared overlap between ketamine and MDMA was most apparent at the pathway level rather than for individual CpGs or genes. They interpret this as convergence on synaptic scaffolding, postsynaptic organisation, excitability, calcium-dependent signalling, vesicle release machinery, and potentially RNA-editing-related processes, all of which they link to circuit adaptation and psychiatric biology. In their view, the broader pattern supports the idea that distinct pharmacological agents can converge on similar plasticity-related molecular substrates. They also discuss several limitations. Sample sizes were modest, the tissues and post-treatment timepoints differed between cohorts, and the ketamine sample had more diagnostic heterogeneity, which could increase baseline epigenetic variability. They note that peripheral methylation cannot be assumed to represent central nervous system function and that functional effects on gene expression or neuroplasticity were not measured. They also state that smoking status and ethnicity were not included as covariates, and that medication discontinuation rules in the ketamine cohort may have contributed to variability. The authors conclude that larger, longitudinal studies with multimodal measures are needed, and they suggest that BEWAS is a useful but complementary strategy alongside broader epigenome-wide and cell-type-specific approaches. They also propose extending the framework to other psychedelics and psychoplastogens to compare shared and distinct molecular pathways.