The serotonin 5-HT2C receptor and the non-addictive nature of classic hallucinogens

Canal and colleagues review the pharmacology and addiction-related properties of classic hallucinogens (CH), a class that includes tryptamines (for example psilocybin, DMT), ergolines (LSD), and phenylalkylamines (for example mescaline, DOB). They emphasise that CH share high-affinity agonism at serotonin 5-HT2 GPCR subtypes (5-HT2A, 5-HT2B, 5-HT2C) and that blockade of 5-HT2 receptors attenuates many of their psychoactive effects. Despite these shared pharmacological features, CH are generally reported by experienced users and by controlled preclinical studies to have low addiction liability, a contrast to many other Schedule I substances; the mechanistic basis for this apparent non-addictive profile remains uncertain. This paper sets out to synthesise behavioural, neuroanatomical and neurochemical preclinical findings and to propose a testable neurobiological framework that could explain both the non-addictive nature of CH and their potential to treat substance dependence. Specifically, the authors develop and defend a hypothesis in which activation of 5-HT2C receptors—particularly on medium spiny GABAergic neurons in the nucleus accumbens (NAc) shell—modulates intrinsic neuronal excitability (via inhibition of Kv1.x potassium channels), thereby suppressing mesolimbic dopamine signalling that underlies addictive behaviours.

This article is a narrative, integrative review rather than a systematic meta-analysis: Canal and colleagues draw on a broad array of preclinical data (behavioural pharmacology, electrophysiology, autoradiography, genetic knockout studies, neurochemistry, and some optogenetic evidence) alongside historical clinical reports to build a mechanistic hypothesis. The extracted text does not report a formal literature search strategy, inclusion/exclusion criteria, databases searched, or dates covered, so it should be regarded as a synthesis of selected preclinical and clinical findings rather than a reproducible systematic review. The authors place emphasis on evidence types that bear on addiction liability and mechanism: intravenous self-administration and reinstatement paradigms in nonhuman primates and rodents, conditioned place preference and intracranial self-stimulation (ICSS) assays, receptor localisation (autoradiography, immunohistochemistry), receptor pharmacology including studies with selective agonists/antagonists and knockout mice, slice electrophysiology, and biochemical measures of dopamine release in nucleus accumbens (NAc) and dorsal striatum. They identify and integrate findings across these modalities to support a focal mechanistic proposition linking 5-HT2C receptor activation to modulation of Kv1.x potassium channels on NAc medium spiny neurons (MSN).

The assembled literature indicates a consistent pattern: CH show low abuse liability in humans and are rarely self-administered by laboratory animals. Historical intravenous self-administration studies in rhesus monkeys failed to demonstrate reliable self-injection of mescaline or DOM, and DMT produced persistent self-administration only under extreme conditions (sensory deprivation) in a minority of monkeys. Human reports and preclinical data converge on the absence of drug craving and limited withdrawal syndromes following CH exposure, despite the ability of CH to elevate mood and induce strong subjective effects. Behavioural pharmacology implicates the 5-HT2C receptor subtype as an important anti-addictive mediator. Selective 5-HT2C agonists attenuate a range of addiction-relevant behaviours: they reduce cocaine-induced facilitation of ICSS, suppress cocaine self-administration and reinstatement in rodents and primates, blunt locomotor- and discriminative-stimulus effects of cocaine, and reduce cue- and drug-induced relapse in several models. Conversely, 5-HT2C antagonists and genetic deletion of 5-HT2C enhance psychostimulant effects, increase self-administration or reinstatement, and in some cases maintain self-administration when substituted for cocaine. Lorcaserin, a relatively selective 5-HT2C agonist, is reported to suppress cocaine discrimination and self-administration in primates, and clinical trials of lorcaserin for cocaine use disorder were noted as ongoing at the time of writing. Neurochemical and circuit-level data show that 5-HT2 receptor subtypes diverge in their effects on dopamine: 5-HT2A activation generally facilitates dopamine neurotransmission (for example increasing VTA firing and prefrontal dopamine), whereas 5-HT2C activation tends to suppress mesolimbic dopamine signalling. Evidence supports multiple loci for 5-HT2C action: activation of 5-HT2C receptors on GABA neurons in the VTA can inhibit VTA dopamine neurons, and 5-HT2C receptors in the NAc exert inhibitory regulation over dopamine signalling there. Autoradiography and immunohistochemistry reveal dense 5-HT2C expression in the NAc—especially the shell—relative to the dorsal striatum, consistent with observed regional specificity in how 5-HT2C manipulations alter psychostimulant-elicited dopamine release (effects are prominent in the NAc but minimal in the dorsal striatum). Building on these data, Canal and colleagues propose a cellular mechanism involving Kv1.x voltage-gated potassium channels (Kv1.1, Kv1.2, Kv1.3). Several experimental lines indicate that 5-HT2C activation suppresses Kv1.x currents in diverse preparations, whereas psychostimulant exposure (for example repeated cocaine) increases Kv1.x conductance—notably Kv1.2—in NAc shell MSN, reducing intrinsic excitability and firing rate. The authors synthesise these findings into a hypothesis: 5-HT2C receptor agonism on NAc shell MSN inhibits Kv1.x channels, increasing MSN intrinsic excitability and GABAergic output to the VTA, thereby tempering mesolimbic dopamine release and opposing the pro-addictive adaptations induced by psychostimulants. Supporting observations cited include cocaine-induced increases in Kv1.2 expression and persistent reductions in NAc shell MSN firing after cocaine, as well as manipulations of potassium channels (Kir2.1 overexpression, Kv1.1 knockdown) that alter psychostimulant behavioural responses.

Canal and colleagues interpret the convergent behavioural, anatomical and neurochemical evidence as consistent with an important anti-addictive role for 5-HT2C receptor activation, and they propose the 5-HT2C–Kv1.x–NAc shell circuit as a parsimonious mechanistic account for the low addiction liability of CH and for their capacity to alleviate substance dependence. They emphasise that 5-HT2C-mediated increases in NAc shell MSN intrinsic excitability would enhance GABAergic inhibition of the VTA and thereby reduce phasic dopamine signals that drive reinforcement and relapse. The authors also acknowledge alternative or complementary mechanisms: CH may produce long-lasting downregulation of 5-HT2A receptors, and the psychedelic subjective effects themselves (mystical or 'peak' experiences) may contribute to therapeutic outcomes in addiction by altering entrenched behavioural patterns. Several limitations and uncertainties are discussed. The distribution and cell-type specificity of 5-HT2 receptor subtypes across discrete circuits remain incompletely defined, and some evidence indicates 5-HT2A receptors can also modulate Kv1.x channels in particular cell types, complicating simple subtype–function mappings. The pharmacological tools available are imperfect: many so-called selective 5-HT2C agonists have appreciable activity at 5-HT2A or other receptors, which clouds interpretation of systemic administration studies. The central role of 5-HT2B is under-studied owing to low CNS expression and safety concerns (cardiac valvulopathy with peripheral 5-HT2B agonism), so its contribution to addiction liability is unclear. Importantly, the authors caution that repeated CH exposure could alter 5-HT2C receptor expression or function and might, under some circumstances, increase susceptibility to addiction; epidemiological signals such as higher rates of addictive drug use among psilocybin users are noted. To advance the field, Canal and colleagues propose a set of concrete, testable experiments: compare abuse liability of highly 5-HT2A-selective agonists (for example 25CN-NBOH) with CH; test whether 5-HT2C knockout animals self-administer CH; assess whether selective 5-HT2C antagonists unmask CH abuse liability; map precisely whether 5-HT2C receptors are expressed on D1-expressing MSN of the NAc shell; and dissect 5-HT2C–Kv1.x interactions in forebrain regions implicated in addiction. They also call for improved pharmacological probes and cell-type specific approaches to clarify receptor contributions. Overall, the authors present their model as empirically tractable and as a useful framework that could inform development of selective 5-HT2C-based therapeutics for addiction, while urging caution and further rigorous testing.