The Effects of Hallucinogens on Gene Expression

This paper is a narrative review synthesising experimental data from animal and cellular studies rather than reporting a new empirical experiment. The extracted text does not describe a systematic search strategy or explicit inclusion criteria, so the review appears curated from representative molecular, cellular, and behavioural studies in the literature. Primary source data discussed include: immunohistochemistry and in situ hybridisation mapping of IEGs and other transcripts in rat and mouse brain after drug administration; dose‑ and time‑course studies using agents such as DOI, LSD and psilocin; northern blots, RNase protection assays, microarray screens and RNA sequencing (RNA‑seq) for transcriptome-level analyses; cell culture experiments using 5-HT2A-expressing HEK293 cells and primary neuronal cultures; pharmacological manipulations with receptor antagonists and signalling inhibitors; genetic models including 5-HT2A knockout mice and conditional receptor re‑expression in cortical lineages; and recently developed neurocytometry to isolate transcriptionally responsive cell populations. Where relevant, methodological caveats reported by the authors—such as antibody specificity problems and limitations of microarray validation—are noted in the Results and Discussion summaries because they bear on interpretation.

Immediate early genes: Psychedelics robustly induce IEGs such as c-Fos, arc, NGF1C (egr4), and NR4A family members in a dose‑ and time‑dependent manner. In rats given DOI, c-Fos protein in cortex and several subcortical nuclei was first detectable at ~30 minutes, peaked around 3 hours and returned to baseline by 6 hours. Dose–response studies with DOI showed increases in cortical c-Fos above background at ~2 mg/kg with a plateau by 12 mg/kg. Northern blot and in situ hybridisation studies reported increases of transcription factor mRNAs (e.g. c-Fos, ngf1c, tis1/NR4A1) of 200–300% in cortex 90 minutes after 4 mg/kg DOI. Arc and BDNF were also regulated: arc mRNA rose with DOI doses from 0.2 to 2.0 mg/kg across multiple cortical areas, and BDNF increased in parietal cortex while decreasing in the dentate gyrus, effects blocked by 5-HT2 antagonists. Cellular and regional specificity: Early immunohistochemical work suggested poor colocalisation between IEG protein and 5-HT2A receptor immunoreactivity, but more recent fluorescence in situ hybridisation (FISH) studies show that psychedelic‑induced IEGs occur almost exclusively in cells positive for 5-HT2A mRNA. The discrepancy likely reflects differences between mRNA and protein detection and problems with certain 5-HT2A antibodies. Co‑labelling studies indicate that both excitatory pyramidal neurons and subsets of inhibitory interneurons (including somatostatin and parvalbumin subclasses) become transcriptionally active; one study using neurocytometry found that only ~5% of cortical neurons directly increase transcription of IEGs after psychedelic administration, with ~5–10% of inhibitory neurons activated as well. Responses are regionally heterogeneous: for example, somatostatin interneurons were recruited in somatosensory cortex but not medial prefrontal cortex. Post‑genomic profiling and functional selectivity: Microarray screens of rat prefrontal cortex after 1.0 mg/kg LSD (90 min post‑dose) identified upregulated genes including sgk, Ijb‑a, nor1 (nr4a3), ania3 (Homer1 splice variant), and krox‑20 (egr2); subsequent screens added mkp1, C/EBP‑β and arrdc2. Most expression changes peaked at ~90 minutes and returned toward baseline within hours, though some transcripts (e.g. nor1) remained elevated longer. The authors emphasise substantial validation challenges: only about 1 in 4–5 microarray hits validated by RNase protection and some expected IEG signals (c-Fos, arc) were not detected by the arrays despite independent confirmation by other methods. Comparative work revealed ligand‑specific transcriptional signatures: LSD and DOI (psychedelic agonists) produced different gene induction patterns from non‑hallucinogenic 5-HT2A agonists such as lisuride, supporting functional selectivity at the receptor. Signalling pathways: Multiple downstream effectors of 5-HT2A are implicated. The canonical Gaq→PLCβ pathway (leading to intracellular Ca2+ and PKC activation) is involved, but so are PLA2/arachidonic acid pathways, Gi/o-associated Gβγ→Src signalling, Ga12/13→Rho, β‑arrestin→pERK and PLD/ARF pathways. Pharmacological blockade studies suggest that both PLCβ and Gi/o–Src pathways contribute to psychedelic-relevant gene induction (for example, PLCβ inhibition suppressed responses to LSD and lisuride, whereas pertussis toxin selectively reduced LSD‑induced transcriptional magnitude). Ionotropic glutamate receptors are also important: AMPA antagonists (GYKI 52466) and NMDA antagonists (MK‑801) attenuated DOI‑induced Arc in many cortical regions, implicating glutamate transmission as a necessary element for IEG induction. mGluR2 activity modulates these effects: mGluR2/3 agonists can dose‑dependently prevent DOI‑induced BDNF upregulation in cortex, while mGluR2/3 antagonists potentiate it; positive allosteric modulation of mGluR2 attenuated cortical c‑Fos responses to psychedelics. In vitro versus in vivo: In HEK293 cells transfected with 5-HT2A, ligands produced distinct transcriptional profiles, but some drugs (LSD, lisuride) that are active in vivo produced little induction in this heterologous system. Primary neuronal cultures showed ligand- and cell‑type dependent responses and could exhibit transcriptional induction without action potential firing, indicating direct 5-HT2A→transcription coupling in some contexts. Chronic effects: Long‑term LSD exposure in rats (0.16 mg/kg every other day for 90 days) produced persistent behavioural alterations (increased locomotion, reduced sucrose preference, social changes) that in some cases endured after drug cessation. RNA‑seq of medial prefrontal cortex four weeks after stopping the 90‑day protocol revealed several hundred modest (≈two‑fold) but significant transcriptional changes clustered around neurotransmission, synaptic plasticity and metabolism. Notably, altered transcripts included homologues of genes implicated in schizophrenia (dopamine D1/D2 receptors, BDNF, ERBB4, NMDA and GABA receptor subunits). Peripheral effects: The 5-HT2A receptor is widely expressed outside the CNS and psychedelics at very low doses were reported to inhibit TNF‑α‑mediated inflammatory gene expression in cultured cells and whole animals. Nebulised (R)-DOI prevented allergic‑asthma development in a mouse model, and peripheral 5-HT2A activation reduced mRNA levels of several proinflammatory cytokines/chemokines. The authors propose involvement of PKC isoforms and inhibition of NF‑κB signalling as potential mechanisms. Validation and technical caveats: The review highlights technical issues that affect interpretation, including antibody specificity problems for 5-HT2A, differences between mRNA and protein localisation, the high false positive/negative rates of early microarray screens, and discrepancies between in vitro and in vivo models.

The authors interpret the assembled evidence to indicate that activation of 5-HT2A receptors engages a multilayered transcriptional programme beginning with induction of immediate early genes and progressing to late response genes that together can alter synaptic structure and function. Such gene expression changes are positioned as plausible molecular mediators of the persistent psychological and therapeutic effects reported after one or a few psychedelic sessions, as well as of maladaptive outcomes seen after repeated exposure. Martin and colleagues emphasise that different ligands at the 5-HT2A receptor can recruit distinct intracellular effector pathways (functional selectivity), producing divergent transcriptional signatures that may explain why structurally related drugs differ in behavioural effects. They also stress that only a small subset of cortical neurons—what they term a ‘trigger population’ of roughly 5%—appear to mount robust transcriptional responses in vivo, and that activation of this minority population could be sufficient to propagate network‑level changes such as default mode network disintegration and altered connectivity. Key limitations acknowledged include methodological issues that complicate interpretation: antibody specificity problems for 5-HT2A protein detection, discordances between mRNA and protein labelling, the limited quantitative reliability of early microarray screens, and important differences between cell culture systems and intact brain tissue. The authors also note that causal links between particular transcriptional changes and long‑term therapeutic or adverse outcomes remain to be established. In terms of implications, the review suggests that mapping which genes and cell types are engaged by psychedelics, and understanding the signalling pathways that produce those transcriptional programmes, are important next steps. The peripheral anti‑inflammatory effects observed raise the possibility of non‑psychiatric therapeutic applications. Finally, the authors propose that finer-grained, cell‑type specific transcriptomic studies and functional validation of candidate genes will be necessary to determine which molecular changes are responsible for clinically relevant long‑term effects.

Clinical interest in psychedelic compounds has resurged, with early trials showing promising long‑lasting benefits of psilocybin in mood and addiction contexts and renewed exploration of LSD as an adjunct to psychotherapy. Martin and colleagues conclude that psychedelic-induced changes in gene expression provide a plausible mechanistic bridge between acute receptor activation and durable changes in mood, behaviour and brain connectivity. They argue that further work to identify the specific gene expression changes that mediate therapeutic efficacy is warranted and would be ‘‘very exciting’’ should larger clinical studies confirm clinical benefits.