The psychoplastogen tabernanthalog induces neuroplasticity without proximate immediate early gene activation

Cortical atrophy and dysfunction are central features of many neuropsychiatric disorders, and small molecules called psychoplastogens can rapidly promote structural and functional neuronal growth in the prefrontal cortex (PFC). Classic serotonergic psychedelics (for example, psilocybin, LSD, 5-MeO) are potent psychoplastogens but their profound subjective effects limit clinical scalability. Nonhallucinogenic psychoplastogens structurally related to psychedelics, such as tabernanthalog (TBG), have been developed as potential alternatives that might stimulate similar plasticity without producing hallucinatory effects. Aarrestad and colleagues set out to test whether TBG promotes cortical neuroplasticity via the same intracellular signalling pathway attributed to classic psychedelics (involving 5-HT2A receptors, TrkB, mTOR and AMPA receptors), and whether immediate early gene (IEG) induction and acute glutamate bursts are required for psychoplastogen-induced spinogenesis and antidepressant-like behaviour. The study uses complementary pharmacological and genetic tools in rodents, in vitro assays and in vivo imaging to compare TBG with hallucinogenic congeners and to map acute neuronal activity and gene-expression responses associated with each compound.

The investigators used a multi-level preclinical approach combining in vitro receptor pharmacology, cultured neuron assays, and in vivo rodent and porcine experiments. Key methods included radioligand binding, aequorin calcium assays, IP1 accumulation and BRET assays to characterise TBG’s activity at 5-HT2 receptor subtypes and downstream Gq and β-arrestin pathways. TrkB functional assays and screens for MAO-A, NET and SERT interactions were also performed. In vivo and ex vivo work employed mice (wild-type and 5-HT2A receptor knockout), rats and Danish domestic pigs. Structural plasticity was assessed by Golgi-Cox staining of layer 5 pyramidal neurons in medial PFC and by spinogenesis/dendritogenesis assays in cultured cortical neurons. Functional plasticity measures included whole-cell patch-clamp recordings of spontaneous excitatory postsynaptic currents (sEPSCs). Behavioural endpoints comprised the tail suspension test (TST) in mice, the forced swim test (FST) in rats (including an IFNα-induced depressive-like model), chronic social defeat (CSD) social interaction testing, SmartCube high-throughput phenotyping and the mouse head-twitch response as a proxy for hallucinogenic potential. Causal tests of spine necessity used a viral construct expressing an activated synapse-targeted photoactivatable Rac1 (AS-PaRac1) in the PFC; TBG-induced spines were selectively ablated with blue light and subsequent behaviour measured. Neurochemical measurements included microdialysis in rat PFC for glutamate, GABA, monoamines and acetylcholine. In vivo receptor engagement was measured by PET ([11C]Cimbi-36) in pigs. Neuronal activity was probed with fiber photometry using iGluSnFR (glutamate) and GCaMP6f (calcium) in the mPFC, complemented by two-photon imaging through GRIN lenses to track single-cell responses across sessions. Brain-wide immediate early gene (IEG) protein mapping used tissue clearing and light-sheet fluorescence microscopy (c-Fos and NPAS4), and single-nucleus RNA sequencing (snRNA-seq) of PFC tissue was used to map cell-type-specific transcriptional responses. Treatments were randomised and experimenters were blinded to conditions; statistical analyses used standard parametric tests unless otherwise noted and are reported in supplementary materials.

Receptor pharmacology established that TBG is a relatively balanced partial agonist at 5-HT2A receptors while acting as a full agonist at 5-HT2C receptors; additional assays showed 5-HT2B antagonism in some readouts. Radioligand and functional assays supported partial 5-HT2A activity across multiple platforms. In vivo, a single intraperitoneal dose of TBG (commonly 50 mg kg-1 in mice) increased dendritic spine density in layer 5 pyramidal neurons of the medial PFC and produced sustained increases in sEPSC frequency and amplitude 24 h after administration. These structural and functional effects were present in wild-type animals but absent in 5-HT2A receptor knockout mice, indicating that 5-HT2A activation is necessary for TBG-induced cortical plasticity. Behaviourally, TBG reduced immobility in the mouse TST 24 h after dosing in wild-type but not 5-HT2A KO mice, linking receptor engagement to antidepressant-like effects. Using AS-PaRac1-mediated photoablation to remove TBG-induced spines, the researchers demonstrated that selective ablation of those newly formed spines abolished the sustained antidepressant-like effect of TBG in the TST, whereas photoablation of a random set of spines did not, supporting a causal role for cortical spinogenesis in the behavioural effect. Comparative in vitro work showed that nonhallucinogenic congeners from multiple chemical scaffolds promoted dendritogenesis in rat cortical cultures to a similar extent as hallucinogenic counterparts, and these effects required 5-HT2A, TrkB, mTOR and AMPA receptor activity. TBG did not act as a TrkB positive allosteric modulator and had only very low-potency interactions with MAO-A, NET and SERT, arguing against classical antidepressant-like monoamine transporter mechanisms at behaviourally relevant doses. Pharmacokinetics and microdialysis indicated that TBG is brain-penetrant, reaching micromolar brain concentrations at low doses, but produces minimal to no increases in extracellular glutamate, GABA, acetylcholine or dopamine at doses ≤10 mg kg-1 in rat PFC; slight elevations of norepinephrine and serotonin were seen at 30 mg kg-1. Behaviourally, TBG produced rapid and sustained antidepressant-like effects in the rat FST comparable to ketamine and rescued IFNα- and CSD-induced depressive-like/social deficits; SmartCube phenotyping clustered TBG closer to psilocybin/psilocin than to SSRIs. PET imaging in pigs with [11C]Cimbi-36 showed dose-dependent 5-HT2A receptor occupancy by TBG, with doses ≥3 mg kg-1 producing >46% occupancy. Despite appreciable occupancy, TBG produced only mild startle/head-shake behaviours in pigs and, when given to mice before 5-MeO, fully blocked 5-MeO-induced head-twitch responses. Neuronal-activity measures revealed a dissociation between calcium and glutamate signalling: both TBG and 5-MeO increased cytosolic Ca2+ in the same population of mPFC excitatory neurons (two-photon imaging showed ≈76% concordance in single-cell responses between drugs), but 5-MeO produced an acute increase in glutamate transients while TBG decreased the frequency of glutamate transients and did not increase amplitude. Two-photon and photometry data suggested similar numbers of cells were activated by either compound, but patterns and magnitudes of activity differed. IEG mapping showed that 5-MeO robustly increased c-Fos and NPAS4 protein expression across multiple brain regions and specifically induced an IEG transcriptional programme in layer 4/5 intratelencephalic (IT) glutamatergic neurons in the PFC. In contrast, TBG (and other nonhallucinogenic psychoplastogens tested, BOL-148 and AAZ-A-154) produced minimal proximate increases in c-Fos or NPAS4 protein 1–1.5 h after administration, and snRNA-seq revealed no significant IEG induction by TBG in any analysed cell cluster. At the gene level, 5-MeO upregulated classical IEGs such as Fos and Arc and other genes (for example, Tiparp showed average log2 fold change = 3.08, adjusted P = 9.47 × 10-12) in an Htr2a+ cell-enriched population, whereas TBG produced no comparable transcriptional response.

Aarresta and colleagues interpret their data to mean that TBG, although nonhallucinogenic, promotes cortical structural and functional plasticity through the same intracellular signalling cascade implicated for psychedelics and ketamine, namely 5-HT2A receptor engagement leading to TrkB, mTOR and AMPA receptor-dependent neuron growth. Crucially, their genetic and pharmacological loss-of-function experiments indicate that 5-HT2A receptor activation is necessary for TBG's plasticity and antidepressant-like effects, and selective photoablation experiments demonstrate that TBG-induced spinogenesis in the PFC is required for its sustained behavioural action. However, the study separates proximate physiological signatures of classical psychedelics from the mechanisms necessary for lasting plasticity: unlike the hallucinogenic 5-MeO, TBG does not induce a prominent glutamate burst or proximate immediate early gene (IEG) expression, despite occupying 5-HT2A receptors in vivo and increasing intracellular calcium in the same neuronal populations. The authors suggest that excessive glutamate signalling and consequent IEG induction may be more closely tied to the acute hallucinogenic or seizure-like activity of full 5-HT2A agonists, rather than being essential for the induction of long-term structural plasticity. The investigators acknowledge uncertainties and limitations reported in the paper. They note that the precise factors that allow some compounds to stimulate cortical neuron growth without an acute IEG response remain unclear, and that increases in cytosolic calcium—though implicated—are not proven here to be sufficient on their own to drive spinogenesis. They also point out that TBG has not been administered to humans, so its hallucinogenic potential in humans can only be inferred from preclinical measures. Finally, the authors position partial 5-HT2A agonism as a promising strategy for developing psychoplastogens with low hallucinogenic potential and argue that identifying compounds that induce minimal proximate IEG responses may aid clinical scalability, while recognising that further work is required to define the determinants of plasticity versus acute gene-activation signatures.