The serotonergic hallucinogen 5-methoxy-N,N-dimethyltryptamine disrupts cortical activity in a regionally-selective manner via 5-HT1A and 5-HT2A receptors

Serotonergic hallucinogens produce profound alterations in perception, thought and mood. Riga and colleagues note two chemical classes: indoleamines (for example LSD, psilocin, DMT and 5-MeO-DMT) which bind with high affinity to several serotonin receptor subtypes including 5-HT1A and 5-HT2A, and phenylalkylamines (for example mescaline and DOI) which are more selective for 5-HT2A/2C receptors. Previous preclinical work has shown that psychotomimetic agents such as PCP, DOI and 5-MeO-DMT disrupt low frequency cortical oscillations (LFCO, <4 Hz) in rodent prefrontal cortex (PFC), and that antipsychotic drugs can reverse these disruptions. However, the relative contributions of 5-HT1A and 5-HT2A receptors across different cortical regions remained unclear. The present study set out to characterise how 5-MeO-DMT affects LFCO in medial PFC and primary sensory cortices (somatosensory S1, auditory Au1 and visual V1) in anaesthetised mice, and to dissect the roles of 5-HT1A and 5-HT2A receptors. To achieve this, the investigators used a combination of genetic loss-of-function (5-HT2A receptor knockout mice, KO2A) and pharmacological tools (a selective 5-HT1A antagonist and antipsychotic drugs) together with electrophysiology and in vivo microdialysis to measure local field potentials and extracellular serotonin levels.

Male homozygous 5-HT2A knockout mice (KO2A) and wild-type (WT) littermates on a C57/BL6 background aged 9–16 weeks were used. Animal procedures followed EU regulations and stereotaxic coordinates were referenced to the mouse brain atlas. Drugs included 5-MeO-DMT, haloperidol (HAL), risperidone (RIS) and the selective 5-HT1A antagonist WAY-100635; doses are reported as free bases and most administrations were subcutaneous. For systemic electrophysiology experiments 5-MeO-DMT was given at 1 mg/kg s.c.; WAY-100635 was used at 0.5 mg/kg s.c.; HAL and RIS were used at 0.6 mg/kg and 1 mg/kg respectively. For local effects in mPFC, 5-MeO-DMT was applied via reverse microdialysis in artificial cerebrospinal fluid (aCSF). Electrophysiological recordings were performed under chloral hydrate anaesthesia (initial 400 mg/kg i.p., maintenance infusion 50–70 mg/kg/h). Local field potentials (LFPs) were recorded in medial PFC (mPFC) and electrocorticograms (ECoGs) from primary sensory cortices S1, Au1 and V1 using epidural electrodes. Signals were amplified, filtered (0.1–100 Hz) and after a 5-minute stable baseline a slow injection of 5-MeO-DMT was given; in some experiments antipsychotics or WAY-100635 were administered with defined inter-injection intervals. LFCO was defined as power between 0.15 and 4 Hz. Off-line analysis used Fast Fourier Transform on 18 consecutive ten-second epochs per condition, with LFCO power expressed as percentage of baseline. In vivo microdialysis measured extracellular serotonin (5-HT) in mPFC of freely moving mice 20–24 h after probe implantation. Probes were perfused with aCSF containing 1 μM citalopram to block reuptake, samples were collected every 20 minutes and analysed by HPLC-amperometry. Both systemic (1 mg/kg s.c.) and local (30–300 μM via reverse dialysis) applications of 5-MeO-DMT were tested. Behavioural head twitch response (HTR) was scored during microdialysis sessions in consecutive 5-minute blocks. Statistical analyses comprised Student’s t-test and two-way repeated-measures ANOVA with treatment and genotype (or area) as factors, followed by Newman–Keuls post-hoc tests; significance was set at p<0.05.

Baseline LFCO power did not differ significantly between WT and KO2A mice across mPFC, S1, Au1 and V1. In mPFC systemic 5-MeO-DMT (1 mg/kg) significantly reduced LFCO in WT mice from 0.054±0.004 to 0.030±0.002 μV2, equivalent to 51.1±2.5% of baseline (n=40). KO2A mice also showed a reduction but of smaller magnitude, from 0.064±0.004 to 0.041±0.004 μV2, or 61.4±3.3% of baseline (n=13). Two-way ANOVA indicated significant effects of 5-MeO-DMT treatment and genotype-related interactions. To probe 5-HT1A involvement, KO2A mice were pretreated with the selective antagonist WAY-100635 (0.5 mg/kg). WAY-100635 alone increased LFCO power and fully prevented the 5-MeO-DMT-induced LFCO reduction in KO2A mice. Two-way ANOVA showed significant main effects of 5-MeO-DMT and WAY-100635 and a significant treatment x pre-treatment interaction (for example F(2,34)=14.29, p<0.005 for 5-MeO-DMT; F(2,17)=32.75, p<0.0001 for WAY-100635). Microdialysis revealed comparable basal extracellular 5-HT in mPFC (WT 14.5±1.8 fmol/30 μl, n=15; KO2A 16.2±2.3 fmol/30 μl, n=11). Systemic 5-MeO-DMT decreased extracellular 5-HT similarly in both genotypes, with maximal reductions to 57.0±7.0% (WT) and 43.6±4.9% (KO2A) of baseline; two-way ANOVA showed a significant effect of 5-MeO-DMT (F(9,99)=8.35; p<0.00001) but no genotype interaction. Behaviourally, 5-MeO-DMT increased HTR in WT mice from 0.98±0.29 to 4.09±0.66 occurrences, whereas KO2A mice did not show this increase (0.92±0.34 to 0.39±0.15). Local reverse-dialysis application of 5-MeO-DMT to mPFC produced dose-dependent and genotype-dependent changes in extracellular 5-HT. At 30 μM both genotypes showed similar reductions. Higher concentrations diverged: 300 μM increased extracellular 5-HT to 149.5±22.1% of baseline in WT mice, whereas 100 μM decreased 5-HT to 38.8±8.1% of baseline in KO2A mice. ANOVA on normalized areas under the curve showed significant effects of concentration, genotype and their interaction. Antipsychotic drugs reversed the 5-MeO-DMT-induced LFCO reduction in WT mPFC. Both haloperidol and risperidone restored LFCO towards baseline; two-way ANOVA revealed significant main effects and interactions (for example F(2,26)=109.76, p<0.00001 for 5-MeO-DMT treatment; F(2,13)=4.62, p<0.05 for antipsychotic treatment). In primary sensory cortices, systemic 5-MeO-DMT reduced LFCO in WT mice in S1 to 67.1±4.3% of baseline, in Au1 to 59.3±4.1% and in V1 to 67.1±6.8%. In KO2A mice reductions were seen only in V1 (50.2±5.1% of baseline) but not in S1 or Au1. Statistical analyses indicated significant effects of treatment across areas in WT mice and genotype x treatment and area effects in KO2A mice, consistent with a regionally selective contribution of 5-HT1A and 5-HT2A receptors.

Riga and colleagues interpret their data as confirming that 5-MeO-DMT disrupts low frequency cortical synchrony in PFC and primary sensory areas through actions at both 5-HT1A and 5-HT2A receptors. They highlight that systemic 5-MeO-DMT reduced LFCO in WT mice and produced a smaller but significant reduction in KO2A mice, indicating that receptors other than 5-HT2A contribute to the effect. Prevention of the LFCO reduction in KO2A mice by the 5-HT1A antagonist WAY-100635 supports a compensatory or unmasked role for 5-HT1A receptors when 5-HT2A receptors are absent. The investigators discuss mechanistic distinctions between presynaptic and postsynaptic 5-HT1A activation. Systemic 5-MeO-DMT decreased extracellular 5-HT similarly in WT and KO2A mice, suggesting a predominant presynaptic 5-HT1A autoreceptor-mediated inhibition of dorsal raphe neurons. By contrast, local application in mPFC revealed genotype-dependent concentration–response differences: low concentrations reduced 5-HT in both genotypes, while higher concentrations increased 5-HT in WT but not KO2A mice, implicating cortical 5-HT2A receptor-mediated effects on local serotonergic output. The concomitant increase in HTR in WT but not KO2A mice following systemic 5-MeO-DMT is interpreted as evidence of postsynaptic 5-HT2A activation at the behavioural dose used. Regional differences in receptor contribution are emphasised. The differential sensitivity of sensory cortices to 5-MeO-DMT in KO2A mice—marked in S1 and Au1 but preserved in V1—led the authors to propose a relatively greater role for 5-HT2A receptors in S1/Au1 and for 5-HT1A receptors in V1. They note dense expression and co-localisation of both receptor subtypes in visual cortex and cite prior evidence linking these receptors to modulation of thalamic visual inputs and visual plasticity. The apparent paradox that activation of both excitatory (5-HT2A) and inhibitory (5-HT1A) receptors can reduce LFCO is discussed in terms of complex cellular distribution and interactions, including expression in pyramidal cells and GABAergic interneurons and possible synergistic network effects. Limitations acknowledged by the authors include technical constraints in comparing systemic and local drug application: the dialysis membrane and extracellular clearance reduce effective concentrations reaching tissue, and the microdialysis probe samples only a small neuronal population. Finally, the reversal of 5-MeO-DMT effects by risperidone is attributed to direct competition at 5-HT2A receptors, whereas haloperidol reversal is interpreted at the network level via dopamine D2 receptor blockade affecting excitatory/inhibitory balance. The authors conclude that these findings illuminate receptor- and region-specific mechanisms by which indoleamine hallucinogens perturb cortical circuits related to hallucinations and support the LFCO disruption model as useful for antipsychotic drug development.

The authors conclude that 5-MeO-DMT produces marked disruptions of cortical function in both primary sensory cortices (Au1, S1, V1) and associative cortex (PFC), mediated by 5-HT1A and 5-HT2A receptors with differential regional contributions. They further state that 5-HT1A receptors play a major role in the drug’s effects on visual and prefrontal cortices, and that the ability of antipsychotic drugs to reverse the LFCO reduction supports the model’s relevance for antipsychotic drug development.