5-HT2CR Is as Important as 5-HT2AR in Inducing Hallucinogenic Effects in Serotonergic Compounds

Hallucinogens such as mescaline act on serotonergic systems and have long been associated with activation of serotonin (5-HT) 2A receptors (5-HT2AR), with frontal cortex-dependent head‑twitch response (HTR) in rodents commonly used as a preclinical proxy for hallucinogenic activity. Although genetic or pharmacological inactivation of 5-HT2AR blocks HTR and other psychedelic behaviours, not all 5-HT2AR agonists produce hallucinogenic effects, and other 5-HT receptor subtypes have been implicated, including 5-HT1A and 5-HT2C receptors (5-HT2CR). Recent anatomical work shows co‑localisation of 5-HT2AR and 5-HT2CR in apical dendrites of medial prefrontal cortex pyramidal neurons, and prior genetic deletion work suggests 5-HT2CR contributes to the behavioural effects of some phenethylamines, leaving open the question whether 5-HT2AR activation alone is sufficient to explain psychedelic effects. Custodio and colleagues set out to reassess the relative roles of 5-HT2AR and 5-HT2CR by combining receptor binding and transporter assays with in vivo pharmacology and neurochemical measures. Using two mescaline derivatives, methallylescaline (MAL; a 3,4,5-trisubstituted analogue) and BOD (a 2,4,5-trisubstituted analogue), the study examines receptor affinities, SERT uptake inhibition, HTR with pharmacological antagonism, downstream signalling, neurochemical endpoints (TPH2 and 5-HT levels) and behavioural indices of neurotoxicity in mice. The paper is a preprint and has not been peer reviewed.

This is a preclinical, mechanistic study combining in vitro receptor/transporter pharmacology with in vivo behavioural, biochemical and histological assays in mice. Receptor binding assays used HEK-293 cells transiently transfected with human 5-HT2A or 5-HT2C constructs and radioligand displacement to derive IC50 and K i values. Transporter uptake studies employed HEK-293 cells stably expressing serotonin, dopamine or norepinephrine transporters; uptake was measured with tritiated substrates after incubation with test compounds (typically 1 μM). Behavioural testing in mice included the head‑twitch response (HTR) as the primary index of hallucinogenic-like activity, with groups of n=10 mice per treatment. HTR was recorded after acute dosing (0.3, 1 and 3 mg•kg−1 for MAL and BOD; 1 mg•kg−1 for DOI and saline controls) on day 1 and day 7. Separate cohorts were pre-treated with the 5-HT2CR antagonist SB-242084 (SB) and the preferential 5-HT2AR antagonist ketanserin (KS) to probe receptor involvement; the Methods section reports a pre-treatment dose of 0.1 mg•kg−1, whereas some Results text cites 0.2 mg•kg−1, an inconsistency in the extracted text. To assess neurotoxic-like effects, mice (n=10 per group) received a single high dose (30 mg•kg−1) and were evaluated using open-field locomotion, rotarod (motor coordination), novel-object recognition (attention) and Y‑maze (spontaneous alternation) tests. Behavioural endpoints were recorded with automated video tracking where applicable. Neurochemical and molecular assays included western blotting of prelimbic cortex (PLc) lysates (30 μg per lane), immunofluorescence on 40 μm brain sections targeting 5-HT2AR, 5-HT2CR, TPH2, GFAP and IBA‑1, and ELISA measurement of 5-HT in dorsal raphe nucleus (DRN), hippocampus and PLc. Immunohistochemistry counted TPH2-positive cells in DRN and quantified corrected fluorescence in target regions. Statistical analyses used GraphPad Prism v9; normality was assessed with D'Agostino–Pearson omnibus test and groups compared with one‑ or two‑way ANOVA (with or without repeated measures) followed by Tukey's test. A significance threshold of P≤0.05 was applied and experiments were replicated at least three times. Drugs (MAL, BOD) were synthesised in-house; DOI, KS and SB were commercial hydrochloride salts dissolved in saline and administered intraperitoneally.

Receptor pharmacology and transporter activity: MAL showed negligible affinity for 5-HT2AR (reported K i ≈100 nM) but high affinity for 5-HT2CR (K i =5.08 nM), a higher 5-HT2CR affinity than DOI (K i =16.01 nM). BOD displayed high 5-HT2AR affinity (K i =1.38 nM, exceeding DOI's 5-HT2AR K i =3.39 nM) and weaker 5-HT2CR affinity (K i =19.74 nM). At 1 μM, MAL and BOD inhibited serotonin transporter (SERT) uptake; DOI did not inhibit SERT under the same conditions. HTR and receptor blockade: HTR testing showed an inverted U-shaped dose–response for MAL, with the peak effect at 1 mg•kg−1. Pharmacological antagonism indicated that MAL-induced HTR was blocked by the 5-HT2CR antagonist SB-242084 but not by ketanserin, whereas BOD-induced HTR was blocked by both SB and ketanserin. These patterns indicate that MAL produces HTR primarily via 5-HT2CR activation, whereas BOD (and DOI) engage both 5-HT2AR and 5-HT2CR. Protein expression and serotonergic markers: Western blot and immunofluorescence showed that MAL increased 5-HT2CR expression in the prelimbic cortex (PLc), while BOD increased both 5-HT2AR and 5-HT2CR expression. TPH2 immunoreactivity and tissue 5-HT levels were elevated in the DRN, hippocampal CA1a region and PLc following hallucinogen treatment, consistent with increased serotonergic activity. Downstream signalling: Biochemical assays indicated divergent G q α11-mediated signalling downstream of 5-HT2 receptors across compounds. MAL activated G q α11-linked phospholipase C (PLC) via 5-HT2CR but did not elicit phosphorylation of downstream PKC, ERK1/2 or CREB; nevertheless, MAL upregulated c‑FOS and brain‑derived neurotrophic factor (BDNF) and increased NF‑κB signalling, suggesting engagement of a PLA2–NF‑κB axis. By contrast, BOD and DOI activated canonical PLC signalling with phosphorylation of PKC and ERK1/2 and induction of CREB and NF‑κB, culminating in c‑FOS activation; however, BOD and DOI did not increase BDNF and were associated with downregulation of ΔFOSB. Neurotoxicity and inflammation after high-dose exposure: Repeated exposure to a single high dose (30 mg•kg−1) produced behavioural impairments indicative of neurotoxic effects. BOD and DOI reduced open-field locomotion and increased rotarod falling frequency after 7 days of repeated treatment; MAL showed less consistent motor impairment. Across compounds, reduced novel‑object preference (MAL and BOD reported), fewer total Y‑maze arm entries and impaired spontaneous alternation were observed, alongside episodes of hypothermia at selected time points. Neurochemical measures after repeated high-dose treatment showed reduced TPH2 immunoreactivity and lower 5-HT levels in the DRN, consistent with serotonergic damage. Immunohistochemistry revealed increased IBA‑1 (microglial marker) across brain regions, and ELISA/western blot detected elevated pro‑inflammatory cytokines IL‑6 and TNF‑α in the DRN. GFAP (an astrocyte marker) changes were compound- and region-specific: DOI failed to enhance GFAP in the DRN and hippocampal dentate gyrus and decreased PLc GFAP in some measures, while other compounds had variable GFAP effects. Overall, the data indicate that these serotonergic hallucinogens can produce both acute hallucinogenic-like responses at 1 mg•kg−1 and neuroinflammatory/neurotoxic alterations at a single high dose of 30 mg•kg−1. Experiments were reported to be replicated at least three times; sample sizes for behavioural assays were typically n=10 per group and immunofluorescence used n=6 per group.

Custodio and colleagues interpret their findings as evidence that 5-HT2CR activation is at least as important as 5-HT2AR activation for producing hallucinogenic-like effects in mice. The central observation is that MAL, which lacks appreciable 5-HT2AR affinity but binds strongly to 5-HT2CR, elicits an HTR comparable to BOD and DOI and that this response is selectively blocked by a 5-HT2CR antagonist. In contrast, BOD and DOI, which bind both receptors, show HTRs blocked by antagonists at both 5-HT2AR and 5-HT2CR, supporting a model in which either receptor alone or their concerted activation can produce hallucinogenic-like behaviour. The authors highlight divergent intracellular signalling as a notable mechanistic finding: MAL engages PLC and PLA2-linked pathways downstream of 5-HT2CR and increases c‑FOS, BDNF and NF‑κB without canonical PKC/ERK/CREB phosphorylation, whereas BOD and DOI activate canonical PLC–PKC–ERK–CREB cascades. The paper suggests that differences in substitution patterns (3,4,5 versus 2,4,5 tri-substitution) may underlie distinct receptor binding modes and signalling outcomes, although they acknowledge that structure–activity relationship studies are needed to confirm this. Regarding safety, the researchers report that a single high dose (30 mg•kg−1) produces behavioural impairments (motor, attention, memory), reductions in serotonergic markers and activation of microglia and pro‑inflammatory cytokines, indicating neurotoxic potential under these dosing conditions. They note some compound- and region-specific differences in astrocyte (GFAP) responses. Limitations and next steps noted by the authors include the possibility that other receptors might contribute to the observed effects (mescaline and related compounds have multiple receptor affinities) and the need for further SAR and mechanistic studies to clarify how structural modifications determine receptor selectivity and downstream signalling. The authors also reiterate that the study is a preprint and has not been peer reviewed, and they state that their reporting follows recommended preclinical transparency and rigor guidelines.