Interactions of hallucinogens with the glutamatergic system: permissive network effects mediated through cortical layer V pyramidal neurons

Marek frames this chapter as an integration of decades of electrophysiological, pharmacological, lesion, behavioural and clinical evidence linking serotonergic hallucinogens to glutamatergic signalling in the prefrontal cortex (PFC) and neocortex. Earlier work established that LSD, mescaline and psilocybin share potent agonism at 5-HT2A receptors and that activation of these receptors can produce psychotomimetic effects. Over time, however, it became clear that simple 5-HT2A blockade does not fully account for the clinical picture of psychosis and that interactions with glutamate transmission, especially onto layer V (L5) pyramidal neurons, offer a richer mechanistic account relevant to perception, cognition, mood and potential therapeutics. This review sets out to synthesise findings showing that 5-HT2A receptor activation evokes glutamate (Glu) release onto the apical dendritic field of L5 pyramidal cells, producing spontaneous and late excitatory postsynaptic currents (EPSCs) and recurrent network activity. Marek aims to connect cellular and circuit-level mechanisms (including the likely involvement of midline/intralaminar thalamic nuclei, claustrum and deep PFC neurons) to behavioural readouts in rodents and to clinical observations, with attention to the modulatory roles of many G-protein-coupled receptors (GPCRs) that either enhance or suppress this Glu release.

This paper is a narrative review assembling evidence from multiple experimental modalities rather than reporting a new empirical study. The material marshalled includes in vitro electrophysiology from rodent brain slices (intracellular and whole-cell recordings from L5 pyramidal neurons), focal electrical stimulation paradigms, pharmacological manipulations (agonists, antagonists, PAMs and NAMs at a range of GPCRs), fibre-sparing and chemical lesions (notably of midline and intralaminar thalamic nuclei), anterograde labelling and two-photon calcium imaging, and a variety of in vivo rodent behavioural assays (head-twitch response, 5-choice serial reaction time test, differential-reinforcement-of-low rate 72-s operant schedules, locomotor and antidepressant-like screens). Clinical data integrated into the review comprise results from trials and case series testing selective 5-HT2A antagonists/inverse agonists, mGlu2 ligands, opioid-modulating agents, orexin antagonists, and rapidly acting antidepressants such as ketamine and scopolamine. Where available, receptor autoradiography, immunohistochemistry and mRNA localisation studies are used to infer receptor distributions and potential presynaptic versus postsynaptic localisation. The extracted text does not report systematic search methods, inclusion/exclusion criteria, databases searched or formal meta-analytic procedures; the review therefore appears to be a focused synthesis of selected primary studies and mechanistic evidence rather than a systematic review with explicit reproducible methodology.

At the cellular level, multiple slice studies demonstrate that bath application of serotonin (5-HT) or serotonergic hallucinogens (examples: LSD 10 nM, DOI 3 µM) increases the frequency of spontaneous EPSCs recorded in the apical dendritic field of PFC and neocortical L5 pyramidal neurons. These 5-HT-induced synaptic currents are: potently blocked by 5-HT2A antagonists; largely dependent on AMPA receptors (blocked by LY293558); abolished by TTX and by zero-calcium/high-magnesium perfusion; minimally affected by GABAA antagonism (<15% suppression by bicuculline); and mimicked by focal iontophoretic 5-HT application to layers I and Va. When combined with focal low-frequency electrical stimulation (≈0.1–1 Hz), serotonergic hallucinogens induce late EPSCs or prolonged recurrent ‘‘Glu spillover’’ and UP-state-like network activity that can persist long after washout. Several lines of evidence implicate subcortical thalamic afferents as a major source of the glutamate release onto L5 apical dendrites: midline and intralaminar thalamic lesions reduce 5-HT-induced EPSC frequency by ~61–67% and reduce mGlu2/3 and μ-opioid receptor binding in layers I and Va (reported reductions ≈20% for mGlu2/3 and ≈40% for μ-opioid sites). Complementary experiments using anterograde labelling and two-photon calcium imaging show hypocretin-2 (OX2 receptor agonist) induced calcium transients in apical spines apposed to thalamic terminals, and thalamic lesions suppress hypocretin-induced EPSCs. Nonetheless, lesion data and some gene-rescue experiments leave open additional potential sources including the claustrum and intrinsic cortical projections. Pharmacologically, a consistent theme is reciprocal modulation of the 5-HT2A–driven Glu release by multiple GPCRs. Agonists or PAMs at predominantly Gi/o-coupled receptors suppress the 5-HT/DOI-induced EPSCs; these include mGlu2 (orthosteric agonists LY354740, LY379268 and selective mGlu2 PAMs), mGlu4, mGlu8, adenosine A1, and μ-opioid receptors. Gs-coupled receptor activation (dopamine D1/D5 agonists like SKF38393, adenylyl cyclase activation by forskolin, or 8-Br-cAMP) also suppresses DOI-induced recurrent activity, mechanistically linked to upregulation of EAAT3 and downstream cAMP signalling. Conversely, activation of several Gq/11-coupled receptors (α1-adrenergic, mGlu5, OX2, NK3) or α4β2 nicotinic receptors similarly induces feedforward Glu release onto L5 cells, with characteristic dependence on TTX and AMPA receptors and suppression by μ-opioid agonists. Behavioural correlates in rodents align with the slice physiology: the DOI-induced head-twitch response (HTR) is attenuated by mGlu2/3 agonists and mGlu2 PAMs, adenosine A1 agonists, μ-opioid agonists, mGlu4 PAMs, some NK3 antagonists, α1-adrenergic antagonists, OX2 antagonists and β2-adrenergic agonists. The 5-CSRTT and DRL-72s paradigms link 5-HT2A activation and Glu modulation to motoric impulsivity and antidepressant-like effects respectively: intra-mPFC DOI increases impulsivity (suppressed by mGlu2/3 agonists and α1 antagonists), while 5-HT2A antagonists and agents that suppress Glu release (mGlu2 PAMs, adenosine A1 agonists, OX2 blockade) produce antidepressant-like rightward shifts in DRL behaviour. Not all modulators produce consistent behavioural translations—for example, nicotine suppresses HTR despite increasing spontaneous EPSCs and dopamine D1/5 effects show mixed behavioural findings. At the receptor interaction level, in vitro BRET/FRET studies support co-localisation of 5-HT2A and mGlu2 receptors and raise the possibility of heterocomplexes; functional evidence is mixed. Some mutational data implicate direct 5-HT2A–mGlu2 interactions, yet autoradiography, lesion and distribution studies emphasise a largely presynaptic role for mGlu2 and a predominantly postsynaptic localisation for 5-HT2A, leaving open three hypotheses: purely functional interactions, heterocomplex-mediated interactions, or a combination. Clinical translations are heterogeneous. Selective 5-HT2A antagonists showed limited efficacy as monotherapy in schizophrenia, but pimavanserin (a 5-HT2A inverse agonist) produced clinically meaningful benefit in Parkinson's disease psychosis and has encouraging signals in Alzheimer's disease psychosis. A clinical trial of an mGlu2 PAM added to SSRI/SNRI treatment showed only a non-significant trend in major depressive disorder; buprenorphine (μ-opioid partial agonist/κ antagonist) has open-label antidepressant signals. Rapid-acting antidepressants (ketamine, scopolamine) produce synaptogenic and electrophysiological changes in L5 pyramidal neurons—ketamine increases spontaneous EPSCs and induces new spine formation within 24 h, effects tied to mTOR signalling and potentiated by lithium—supporting the circuit-level relevance of these cellular phenomena to mood outcomes.

Marek interprets the assembled data to support a working model in which activation of cortical 5-HT2A receptors leads to glutamate release onto the apical dendrites of L5 pyramidal neurons, promoting recurrent cortical activity that can influence downstream limbic and subcortical circuits. This ‘‘permissive network’’ effect, mediated at least in part by thalamocortical and possibly claustral or intracortical afferents, provides a unifying mechanism linking hallucinogen pharmacology to observable behaviours in rodents and to certain clinical phenomena in humans. The review emphasises that a broad array of GPCRs can modulate this 5-HT2A-driven Glu release in predictable ways: Gi/o-coupled receptors tend to suppress Glu spillover, whereas multiple Gq/11-coupled receptors can themselves induce feedforward Glu release. These interactions offer mechanistic explanations for why manipulations at mGlu2, adenosine A1, opioid, orexin, adrenergic and dopaminergic systems alter both electrophysiological signatures and behavioural outcomes such as head twitches, impulsivity and DRL performance. Marek positions these mechanistic insights as a rationale for exploring non-schizophrenia therapeutic indications—examples being pimavanserin in Parkinson's and Alzheimer's psychosis, and adjunctive opioid or orexin strategies in mood disorders. The authors acknowledge important uncertainties and limitations. The precise source(s) of the glutamatergic afferents remain incompletely resolved: midline/intralaminar thalamic nuclei are strongly implicated but do not account for the entirety of effects, leaving room for claustral or intrinsic cortical contributions. Technical limitations—such as antibody sensitivity for presynaptic 5-HT2A detection, promoter-driven expression patterns in gene-rescue models, and the interpretative limits of in vitro heteromer assays—are discussed. Translational gaps are also noted: robust preclinical modulation of 5-HT2A–Glu interactions has not consistently produced clear clinical efficacy in schizophrenia, though successes in neurodegenerative psychoses and with rapid-acting antidepressants demonstrate that circuit-informed targets can translate under some conditions. Marek calls for further mechanistic work using modern approaches (optogenetics, gene-based targeting, refined receptor pharmacology) to resolve receptor localisation, heterocomplex existence and circuit-level causality, and to better guide clinical development.

Marek concludes that serotonergic hallucinogens exert prominent effects on PFC and neocortical circuitry by engaging 5-HT2A receptors and provoking glutamate release onto L5 pyramidal neurons, with downstream consequences for network dynamics and behaviour. A wide range of receptor systems can modulate this Glu release and the associated behaviours, and some of these targets have produced promising clinical effects (notably pimavanserin in Parkinson's disease psychosis and synaptogenic effects of ketamine and scopolamine relevant to antidepressant action). Nonetheless, important questions about afferent sources, receptor–receptor interactions, post-receptor signalling and circuit-level roles remain, and resolving these will be necessary to translate mechanistic insights into new treatments for psychosis, depression and related conditions.