Models of psychedelic drug action: modulation of cortical-subcortical circuits

This theory-building paper (2021) presents a new model of how psychedelic drugs may act in the brain. The new model, the cortico-claustro-cortical model (CCC model), proposes that psychedelics disrupt 5-HT_2A-mediated network coupling between the claustrum (a region of the brain where 5-HT_2A receptors are densely expressed) and the cortex, leading to attenuation of canonical cortical networks. This model is discussed in relation to two previously described models, the CSTC and REBUS.

Abstract

“Classic psychedelic drugs such as psilocybin and lysergic acid diethylamide (LSD) have recaptured the imagination of both science and popular culture, and may have efficacy in treating a wide range of psychiatric disorders. Human and animal studies of psychedelic drug action in the brain have demonstrated the involvement of the serotonin 2A (5-HT_2A) receptor and the cerebral cortex in acute psychedelic drug action, but different models have evolved to try to explain the impact of 5-HT_2A activation on neural systems. Two prominent models of psychedelic drug action (the cortico-striatal thalamo-cortical, or CSTC, model and relaxed beliefs under psychedelics, or REBUS, model) have emphasized the role of different subcortical structures as crucial in mediating psychedelic drug effects. We describe these models and discuss gaps in knowledge, inconsistencies in the literature, and extensions of both models. We then introduce a third circuit-level model involving the claustrum, a thin strip of grey matter between the insula and the external capsule that densely expresses 5-HT_2A receptors (the cortico-claustro-cortical, or CCC, model). In this model, we propose that the claustrum entrains canonical cortical network states, and that psychedelic drugs disrupt 5-HT_2A-mediated network coupling between the claustrum and the cortex, leading to attenuation of canonical cortical networks during psychedelic drug effects. Together, these three models may explain many phenomena of the psychedelic experience, and using this framework, future research may help to delineate the functional specificity of each circuit to the action of both serotonergic and non-serotonergic hallucinogens.”

Authors: Manoj K. Doss, Maxwell B. Madden, Andrew Gaddis, Mary Beth Nebel, Roland R. Griffiths, Brian M. Mathur & Frederick S. Barrett

Summary

Abstract

Psychedelic drugs affect the serotonin 2A (5-HT2A) receptor and the cerebral cortex, and may have efficacy in treating a wide range of psychiatric disorders. Three circuit-level models have been proposed to explain psychedelic drug effects, including the cortico-claustro-cortical (CCC) model.

Abbreviations: 5-HT1B receptor, 5-HT2A receptor, 5-HT2C receptor, CCC model, CSTC model, DMN model, DOI model, FPN model, IOR model, PPI model, REBUS model, SN model.

Introduction

Psychedelic drugs, such as psilocybin and LSD, have recaptured the imagination of both science and popular culture. They may have efficacy in treating a wide range of medical indications, including mood, substance use, and headache-related disorders.

The cerebral cortex is thought to be central to the neural mechanisms underlying the psychedelic experience due to its high expression of 5-HT2A receptors. 5-HT2A receptor activation via the psychedelic DOI induces a substantial increase in the frequency and amplitude of excitatory postsynaptic current events in rodent prefrontal layer V pyramidal neurons.

Two models of the neural basis of psychedelic drug action differ on the importance of different subcortical structures and extracortical 5-HT2A receptors in psychedelic effects. One model asserts that psychedelic drugs modulate the cortex and subcortical structures and the other model asserts that the reverse occurs.

Here, we review evidence for the CSTC and REBUS models of psychedelic effects and introduce a third circuit-level model centered on the claustrum. This model emphasizes the function of the claustrum in the control of cortical network states.

The thalamus is a collection of subcortical nuclei that relay sensory information from sensory organs to the cortex. These thalamic relay nuclei act as gates, restricting the flow of information to permit only a subset of sensory information at any given time.

Cortico-Striatal-Thalamic-Cortical Loops

In the original conception of psychedelic disruption of CSTC loops, 5-HT2A-containing pyramidal neurons of the medial prefrontal layer V were thought to inhibit GABAergic pallido-thalamic neurons, which in turn inhibited thalamic activity, resulting in inundation of the cortex with sensory information.

Thalamic Nuclei of Interest

Human studies with acute psychedelic manipulations have not parcelled the thalamus into its constituent nuclei, despite differential 5-HT2A expression, functional specificity, and anatomical connectivity of thalamic nuclei. The reticular nucleus expresses 5-HT2A receptors and is thought to be involved in regulating the frequency of cortical rhythms, especially gamma waves.

The MD is a higher-order nucleus that receives inputs from the cortex, medial temporal lobe, pallidum, and other thalamic nuclei. It projects to the medial prefrontal cortex, and expresses 5-HT2A receptors, suggesting multiple points within an MD to medial prefrontal cortex loop that could be targeted by 5-HT2A receptor activation.

The ventrobasal thalamus is a first-order relay nucleus in the thalamus that carries tactile information to the somatosensory cortex. LSD increases functional connectivity between the ventrobasal thalamus and somatosensory cortex in humans.

Compared to somatosensation, psychedelics appear to have more reliable and complex effects on vision, possibly due to modulation of the pulvinar, a nucleus that receives higher-order visual inputs from visual cortex.

Behavioral Evidence, Limitations, and Extensions

Thalamic nuclei within thalamocortical circuits, including MD and sensory nuclei, facilitate and control pre-attentive sensory gating processes, which are thought to be impacted by psychedelics. Animal models demonstrate that LSD and DOI reduce PPI, but human evidence is mixed. DMT has been found to blunt neural responses in primary sensory cortices during sustained attention tasks, but this may represent increased noise and decreased sensitivity to experimentally controlled stimuli.

Mixed evidence for the CSTC model comes from the prediction that psychedelics should produce increased activity within prefrontal cortex. However, both decreases and increases in overall prefrontal and thalamic activity have been reported in humans.

The CTSC model focuses on psychedelic drugs disrupting thalamo-cortical striatal-thalamic circuits, but a revised view of thalamic anatomy suggests that thalamic nuclei receiving first-order thalamic relays also project to higher-order thalamic nuclei and also branch to synapse on subcortical motor centers. Psychedelics may drive aberrant predictions via 5-HT2A receptors in thalamic nuclei and layer V pyramidal neurons, potentially explaining sensory distortions and a sense of surprise that are often encountered during psychedelic experiences.

Relaxed Beliefs Under Psychedelics (REBUS) and the Disruption of Cortico-Medial Temporal Constraints

The REBUS model proposes that psychedelics disrupt the hierarchical nature of the brain by allowing greater influence of incoming stimuli in prediction circuits. This increases cortical “entropy” and putatively allows updates or changes to prior beliefs.

The REBUS model focuses on higher-level cortical networks, such as the default mode, frontoparietal, and salience networks, and is rather vague regarding the subcortical structures of greatest importance to psychedelic drug effects.

DMN Centricity, Egocentricity, and Medial Temporal Entropy Release

The dorsolateral prefrontal network (DMN) is one of the most studied networks in the brain and is implicated in a number of high-level cognitive functions including episodic memory, theory of mind, semantic processing, and self-referential processing.

Psychedelic drugs have been shown to impact the activity of regions within and outside of the default mode network, but whether these effects are increases or decreases or even selective to the default mode network has varied by study. Psychedelics increase bottom-up information flow from the hippocampus and parahippocampal gyrus to higher-level cortical areas, especially the DMN42. The entorhinal cortex of the parahippocampal gyrus is a particularly important site for serotonergic transmission. Although speculative, psychedelics may attenuate entorhinal gating of information to the hippocampus and drive erroneous predictions, but empirical evidence suggests that psychedelics increase top-down information flow from the parahippocampal gyrus to visual cortex.

Psychedelics increase cortical entropy and complexity, and desynchronize scalp electrophysiological oscillations, especially in lower frequencies. These increases are more widespread and greater in magnitude in the SN, FPN, and even sensory and motor networks.

Behavioral Evidence, Limitations, and Extensions

A behavioral prediction put forth by the REBUS model is that higher-level cognitive functions (e.g., attention and memory) should be more impaired than lower-level functions (e.g., perception). However, psychedelics impair pre-attentive processes (PPI and IOR) that are arguably even lower levels of sensory memory than extinction learning.

The REBUS model shows that psychedelics impair various forms of hippocampally-dependent learning, including trace fear conditioning, learning of a novel spatial location, and encoding of recollection memory.

Although the REBUS model claims that psychedelics impair cognition due to “generalized disengagement”, studies have found that psilocybin and dextromethorphan selectively impair working memory compared to episodic memory, and that perirhinal cortex-dependent mnemonic processes are unaffected or enhanced under psychedelics.

Psychedelics increase entropy and diversity of brain states, but reduce resting state measures of complexity, and impair cognitive flexibility in mice. Furthermore, studies supporting psychedelic-induced increases in the number of brain states are based on task-free conditions and thus are under ambiguous control of cognitive operations.

Canonical Cortical Networks

The claustrum is a thin, ribbon-like telencephalic nucleus located lateral to the putamen, medial to the insula, and in between the external and extreme capsules. It contains 5-HT2A receptors and -opioid receptors, and may be involved in cognitive control functions rather than sensory binding or the generation of consciousness.

The Cortico-claustro-cortical (CCC) Model

The claustrum receives dense projections from primarily contralateral areas of association cortices, and provides expansive ipsilateral projections back to prefrontal cortex, as well as to sensory and parietal association cortices. Claustrum neurons provide a cortical network synchronization broadcast signal necessary for appropriate goal-directed or internally directed mental activity. The claustrum is required for resilient performance on a cognitively demanding five-choice serial reaction time task, and for the deactivation of the default mode network during attention tasks.

The claustrum expresses both 5-HT2A receptor mRNA and the 5-HT2A receptor protein, suggesting it receives 5-HT2A-expressing projections and also contains 5-HT2A receptors. This suggests that disruption of claustrum function by psychedelic drugs may amplify neuronal avalanches that eventually result in the destabilization of network states.

Psilocybin decreases resting state claustrum activity and decreases functional connectivity between the claustrum and cortical networks, but increases connectivity between the right claustrum and FPN. This finding supports the hypothesis that coordination between the claustrum and cortex is critical to establish cortical network states necessary for cognitive control.

Psilocybin may disrupt the prefrontal cortex and claustrum, and may be a hub connecting the three circuit models of psychedelic drug action. This may contribute to cognitive control dysfunction induced by psychedelic drugs.

Opioid Hallucinogen Effects

The claustrum densely expresses 5-HT2A receptors and -opioid receptors, making them the primary target of atypical dissociative hallucinogens such as salvinorin A and enadoline. These hallucinogens produce some subjective effects similar to classic psychedelics such as laughing, visual imagery, and mystical-like experiences.

Salvinorin A has similar effects on human brain function to psychedelics, including decreased power across frequency bands, increased between-network connectivity, and increased entropy. However, the effects are not as substantial as those observed during the acute effects of classic psychedelics.

Data from fMRI studies suggest that serotonergic psychedelics and -opioid receptor agonists disrupt claustrum connectivity and entropy, which may be a polypharmacological mechanism underlying some “psychedelic-like” effects.

-opioid receptors are also densely expressed in prefrontal cortex, and it is possible that -opioid receptor-mediated mPFC disruption or activation of striatal-projecting neurons in the mPFC could result in disinhibition of the thalamus and a flooding of the cortex in line with the CSTC model.

Here, we discussed three circuit models that may explain various effects of classic psychedelic drugs and highlighted potential revisions to each. The CSTC and REBUS models may benefit from more precise definitions in neuroanatomy, and the CCC model is in its infancy.

Although the claustrum is discussed as a possible contributor to disruption of network states, cortical layer V pyramidal neurons and the thalamus could also be directly involved in such disruption. Modulation of one or more of these circuits might explain similarities among and differences between classic and atypical hallucinogenic compounds.

Competing interests

Psychedelics may affect cognition and perception by disrupting the cortico-striatal-thalamic-cortical (CTSC) loops and enhancing bottom-up prediction error signaling to higher-level structures compared to top-down reciprocal signaling.

Figure 3 depicts the cortico-claustro-cortical (CCC) model circuitry, which is disrupted by 5-HT2A activation. This disrupts cortical network states, including synchronization of frontal, associative, and sensory cortices across time.