Zalsupindole is a Nondissociative, Nonhallucinogenic Neuroplastogen with Therapeutic Effects Comparable to Ketamine and Psychedelics

Cortical atrophy and loss of synapses in the prefrontal cortex (PFC) are implicated in depression and related disorders such as PTSD and substance use disorder. Earlier research has shown that diverse interventions that promote structural neuroplasticity in the PFC — ranging from electroconvulsive therapy and exercise to fast-acting agents such as ketamine and classical psychedelics — can produce rapid and sustained therapeutic benefits. However, most of the advanced psychoplastogens produce hallucinogenic or dissociative effects, and some carry cardiotoxicity or abuse potential, limiting their scalability and suitability for many patients. This safety profile has driven interest in so-called neuroplastogens: compounds that promote structural and functional plasticity without causing hallucinations or dissociation. Agrawal and colleagues set out to characterise zalsupindole (ZAL, AAZ-A-154), an isotryptamine-derived neuroplastogen, across pharmacology, neuroplasticity, electrophysiology, and behaviour. The study aimed to compare ZAL's in vitro and in vivo effects with those of clinically relevant psychoplastogens (ketamine, psilocybin, and DMT), to profile its receptor pharmacology and pharmacokinetics, and to test whether ZAL could produce rapid and sustained antidepressant-like effects without hallucinogenic or dissociative properties. The work spans multiple preclinical assays, including neurite outgrowth assays, radioligand and functional receptor profiling, Golgi-Cox spine analysis, electrophysiology, microdialysis, and behavioural testing in rodents.

This was a preclinical, multi-laboratory characterisation using in vitro and in vivo assays. Neurite outgrowth (neuritogenesis) assays were performed at three independent laboratories (Olson lab, Neurofit, Cellectricon) using primary cortical neurons from rodent embryos; concentration–response curves and blocker experiments used ketanserin (a 5-HT2 antagonist) and rapamycin (mTOR inhibitor) to probe mechanism. Radioligand binding and functional assays were carried out across a 55-target CNS panel (single concentration 10 μM screening) and by concentration–response for selected targets (5-HT2A/2C/2B, sigma-1, SERT). Functional IPOne, GTPγS, and cAMP HTRF assays assessed agonist/antagonist activity at serotonergic receptors, and MAO-A and transporter assays were performed by contract labs. Pharmacokinetics were measured in male Sprague‑Dawley rats following intraperitoneal (IP) dosing of ZAL at 3, 10, and 30 mg/kg; plasma and whole brain concentrations were quantified by LC–MS/MS (LLOQ 1 ng/mL or 1 ng/g). Microdialysis in the PFC measured extracellular neurotransmitters (DA, NE, 5‑HT, Glu, GABA, ACh) after ZAL 3, 10, or 30 mg/kg IP (group sizes N = 8–9 per dose), with analyses by HPLC–MS/MS and expression as percent change from basal output. Golgi‑Cox staining assessed dendritic morphology and spine density in layer 5 pyramidal neurons of the medial PFC 24 h after single IP dosing of ketamine (10 mg/kg), psilocybin (1–3 mg/kg), or ZAL (3–30 mg/kg); N = 10 per treatment group. Electrophysiological recordings quantified spontaneous excitatory postsynaptic currents (sEPSCs) in mPFC slices 24 h after dosing (groups of N = 8–10) to evaluate functional plasticity. Behavioural testing included the forced swim test (FST) in male rats (N = 10 per group) using a preswim paradigm with single IP dosing of saline, ketamine (10 mg/kg), or ZAL (10 mg/kg) and assessments at 24 h and 7 days postdose. Hallucinogenic potential was assessed with the mouse head‑twitch response (HTR) in C57BL/6J males using psilocybin, DOI, and two doses of ZAL (10 and 30 mg/kg). Statistical analysis used GraphPad Prism (means ± SEM), concentration–response fitting, and conventional IC50/Ki calculations; specific inferential tests are described within individual experiments in the text.

Ketamine (10 mg/kg IP) and psilocybin (3 mg/kg oral, measured as psilocin) produced rapid spinogenesis in layer 5 pyramidal neurons of the rat PFC 24 h after a single dose. Golgi‑Cox analysis of 60 dendritic segments across 10 animals per group showed no gross alterations in dendritic arbors but significant increases in dendritic spine density across apical and basal compartments. The increase was not restricted to one morphological spine class; thin, stubby, mushroom, filopodia-like, and branched spines were all represented among the increases. ZAL increased neuritogenesis in primary cortical cultures in a concentration-dependent manner at three independent sites, and its in vitro effect was blocked by ketanserin and by rapamycin, implicating 5-HT2 receptor signalling and mTOR. Pharmacokinetic studies in rats indicated very rapid plasma and brain clearance with first-order kinetics and an estimated half-life of ~0.35 h; brain-to-plasma ratios of ~10–25 showed high brain penetration. Primary metabolic pathways included demethylation and oxidative metabolism followed by glucuronidation, hydroxylation, and sulfation; the three major metabolites (M1–M3) were inactive in neuritogenesis assays. In vivo, a single IP dose of ZAL increased dendritic spine density in the mPFC 24 h after dosing in a dose-dependent fashion (tested at 3, 10, 30 mg/kg), again without gross changes to dendritic arborisation. ZAL exerted its strongest effects on mushroom and filopodia-like spines. Comparisons of calculated effect sizes found no statistical differences in spine density between ZAL, ketamine, and psilocybin treatment groups. Electrophysiologically, ZAL, psilocybin, and DMT produced statistically significant increases in sEPSC frequency in mPFC neurons 24 h after dosing; ketamine’s effect on frequency approached significance (p = 0.06). Notably, only ZAL and DMT significantly increased sEPSC amplitude. A comparison of effect sizes indicated that ZAL produced a larger effect on functional plasticity in the mPFC than either ketamine or psilocybin; ZAL’s frequency effect resembled DMT though its amplitude increase was slightly smaller. Receptor profiling showed relative selectivity for serotonergic targets. In the 55-target radioligand panel at 10 μM, ZAL inhibited binding by >50% only at 5‑HT7, 5‑HT2B, and SERT. Follow-up concentration–response assays produced Ki values <20 μM at 5‑HT2A, 5‑HT2B, 5‑HT2C, sigma‑1, and SERT. Functional assays characterised ZAL as a low potency (EC50 = 8.2 μM), low efficacy (span ≈17%) partial agonist at 5‑HT2A, with modest to low potency/efficacy at 5‑HT2C, 5‑HT1D, and 5‑HT7; it did not agonise or antagonise 5‑HT1A at concentrations below 30 μM. ZAL acted as a full antagonist at 5‑HT2B and was a low potency inhibitor of 5‑HT3 and monoamine oxidase A. Behavioural and neurochemical findings supported a nonhallucinogenic profile. ZAL did not activate psychLight, did not induce immediate early gene activation in the PFC, and failed to elicit a head‑twitch response in mice at the tested doses. Unpublished phase I data (reported by the authors) were said to show oral bioavailability and CNS functional effects in humans without hallucination or dissociation. In the rat FST, a single 10 mg/kg IP dose of ZAL produced a rapid decrease in immobility at 24 h that was sustained for at least 7 days, matching ketamine’s temporal profile. Microdialysis from the PFC revealed that at 10 mg/kg ZAL transiently doubled extracellular serotonin for ~2 h but produced no substantial increases in norepinephrine, dopamine, acetylcholine, GABA, or glutamate compared with vehicle; at 30 mg/kg transient larger increases in serotonin and norepinephrine were observed. The authors therefore infer that transient neurotransmitter efflux is unlikely to be the primary mechanism for ZAL’s neuroplastic and behavioural effects and that partial 5‑HT2A activation in the PFC likely contributes to its actions.

Agrawal and colleagues interpret their data to indicate that zalsupindole is a brain‑penetrant neuroplastogen which promotes both structural (spinogenesis) and functional (increased sEPSC frequency and amplitude) neuroplasticity in the rat medial PFC after a single dose. The investigators found that these neural changes dissociate from ZAL’s pharmacokinetics, since spine and electrophysiological effects persisted after the compound had been cleared from brain and plasma, and that a single administration produced a rapid and sustained antidepressant‑like response in the forced swim test analogous to ketamine. In comparing ZAL with ketamine, psilocybin, and DMT, the study team emphasises that ZAL’s magnitude of structural plasticity was comparable to those agents and that ZAL produced a particularly robust effect on functional plasticity. The authors argue that ZAL’s markedly reduced hallucinogenic potential stems from its pharmacology: low‑efficacy partial agonism at 5‑HT2A receptors, absence of a cortical glutamate burst at therapeutic doses, and a selective receptor profile that includes full antagonism at 5‑HT2B (reducing putative cardiotoxicity risk). They note that blockade of ZAL’s in vitro neuritogenic effect by ketanserin and rapamycin implicates 5‑HT2 signalling and mTOR pathways in its mechanism, though other targets engaged (5‑HT7, sigma‑1, SERT, MAO‑A) could contribute. The authors acknowledge limitations and remaining uncertainties. Mechanistic links between receptor engagement, intracellular signalling and enduring behavioural outcomes remain incompletely defined. Although phase I data are referenced as supportive of oral bioavailability and absence of hallucination in humans, those clinical results are described as unpublished and are not detailed in this report. Additionally, the metabolites tested in vitro were inactive, but the full relevance of ZAL metabolism to safety and efficacy requires further study. Finally, while multiple preclinical assays and independent laboratories reproduced neuritogenic effects, translation to clinical efficacy and safety will require controlled human trials. Overall, the investigators conclude that ZAL represents a promising nonhallucinogenic neuroplastogen with a pharmacological and efficacy profile that may offer advantages for scalability and safety relative to first‑generation psychoplastogens. They frame ZAL’s first‑in‑human testing as a milestone in efforts to harness compound‑induced neuroplasticity for disorders characterised by cortical atrophy, and they suggest further clinical and mechanistic work to validate these preclinical findings.