Changes in global and thalamic brain connectivity in LSD-induced altered states of consciousness are attributable to the 5-HT2A receptor

This double-blind, randomized, counterbalanced, cross-over study (n=24) investigated the effects of LSD (100 μg) on global brain connectivity during resting-state and observed the synchronization of sensory and somatomotor functional networks and the dis-integration of associative networks.

Abstract

“Background: Lysergic acid diethylamide (LSD) has agonist activity at various serotonin (5-HT) and dopamine receptors. Despite the therapeutic and scientific interest in LSD, specific receptor contributions to its neurobiological effects remain unknown.

Methods: We therefore conducted a double-blind, randomized, counterbalanced, cross-over study (ClinicalTrials.gov, NCT02451072) during which 24 healthy human participants received either (i) placebo+placebo, (ii) placebo+LSD (100 mg po), or (iii) Ketanserin, a selective 5-HT2A receptor antagonist,+LSD. We quantified resting-state functional connectivity via a data-driven global brain connectivity method and compared it to cortical gene expression maps.

Findings: LSD reduced associative, but concurrently increased sensory-somatomotor brain-wide and thalamic connectivity. Ketanserin fully blocked the subjective and neural LSD effects. Whole-brain spatial patterns of LSD effects matched 5-HT2A receptor cortical gene expression in humans.

Conclusion: Together, these results strongly implicate the 5-HT2A receptor in LSD’s neuropharmacology. This study therefore pinpoints the critical role of 5-HT2A in LSD’s mechanism, which informs its neurobiology and guides rational development of psychedelic-based therapeutics.”

Authors: Katrin H. Preller, Joshua B. Burt, Jie Lisa Ji, Charles H. Schleifer, Brendan D. Adkinson, Philipp Stämpfli, Erich Seifritz, Grega Repovs, John H. Krystal, John D. Murray, Franz X. Vollenweider & Alan Anticevic

Summary

Introduction

The current study aims to map the time-dependent pharmacological effects of LSD on neural functional connectivity in healthy human adults using resting-state functional connectivity (rs-fcMRI). This technique has been proven sensitive to the effects of neuropharmacological agents.

Focused analyses of specific regions revealed effects of intravenously administered LSD on functional connectivity between V1 and distributed cortical and subcortical regions, but such an approach has limited ability to detect pharmacologically-induced dysconnectivity not predicted a priori.

Preller et al. have used brain imaging to explore the effects of LSD on the brains of healthy volunteers. They found that LSD reduced communication among brain areas involved in planning and decision-making, but increased communication between areas involved in sensation and movement.

Research into how LSD affects the brain may help us to better understand psychiatric disorders and develop more effective treatments.

A data-driven approach called Global Brain Connectivity (GBC) was used to examine the connectivity of the brain after intravenously administered LSD. The study did not take into account the influence of global signal artefacts (GS), which can induce elevated relationships across the brain.

The current study aimed to determine the extent to which the neural and behavioral effects of LSD are mediated by 5-HT2A receptors. It was found that the effects of LSD were blocked by ketanserin, a selective antagonist at 5-HT2A and a-adreno receptors.

Here we leverage recent advances in human cortical gene expression mapping to predict the spatial topography of neuropharmacologically-induced changes in data-driven connectivity. This prediction validates the contribution of the 5-HT2A receptor to LSD neuropharmacology.

This study demonstrates that LSD effects are mediated by the 5-HT2A receptor and that LSD effects can be mapped onto the spatial expression profile of the 5-HT2A receptor.

LSD modulates global brain connectivity and induces marked subjective drug effects

The main effect of drug on GBC was found to be hyper-connectivity in sensory and somatomotor areas, and hypo-connectivity in areas associated with associative networks.

Figure 1A shows mean connectivity strength (Fz) for each drug condition and Figure 1B shows significant hyper- and hypo-connectivity for LSD compared to (Ket+LSD)+Pla conditions. We observed significant hyper-connectivity in predominantly sensory areas and significant hypo-connectivity in associative networks in the LSD group compared to the control group. This result indicates that LSD-induced alterations in information flow across these networks are predominantly attributable to its agonistic activity onto the 5-HT2A receptor. A repeated-measures ANOVA revealed significant main effects for drug condition and scale, and a significant interaction of drug condition and scale. Bonferroni corrected simple main effect analyses showed increased ratings on all 5D-ASC scales in the LSD condition.

We repeated the analyses presented above without GSR, and found significant predominantly left-hemispheric widespread differences in GBC between drug conditions. The results also showed significant hyper- and hypo-connectivity for LSD compared to (Ket+LSD) +Pla conditions. The comparison between LSD and Pla conditions shows that LSD induces hypo-connectivity in the right insula and hyper-connectivity in the cerebellum. LSD induces hyper-connectivity predominantly in the right hemisphere and a significant positive correlation between hyper- and hypo-connectivity was found when GSR was not performed. We tested whether the hyper/hypo relationship changed with GSR by examining the areas that survived type I error correction following TFCE. The hyper/hypo relationship remained highly stable even without GSR.

Characterizing the directionality of LSD-induced effects on association versus sensory-somatomotor areas

We reported robust and widespread differences between GBC analyses results with and without GSR following LSD administration, which call into question GBC interpretations.

We calculated the mean variance of the GS across all grey matter grayordinates, and found that the variance does not differ significantly between conditions when computing the mean across all gray matter grayordinates.

LSD administration does not change the average grayscale signal significantly. However, the GS signal itself may have a distinct spatial configuration. We computed a beta map of the GS for each subject to investigate the possibility of a shifted topography of the GS under LSD. The beta map revealed a bi-directional spatial shift of the GS under LSD, which correlated highly with the spatial organization of the LSD-induced changes on GBC. To quantify this transformation, we calculated the relationship between the LSD-Pla contrast GBC map and the LSD-Pla contrast GS beta map. The results show that the transformation is negative after GSR.

This analysis of GBC effects does not inform the ‘ground truth’ effect of LSD on baseline connectivity, because the GBC metric is affected by the shared variance across all grayordinates (i.e. the map of the GS).

To address this, we designed a complementary analysis that yields a map that is interpretationally consistent irrespective of GSR-related shifts. We computed thalamic maps by extracting average time-series across all grayordinates in each subject’s anatomically defined bilateral thalamus and conjuncted correlation and covariance as methods of statistical association. This ensured that the resulting regions exhibit thalamic coupling irrespective of processing or statistical method. LSD would decrease connections in the top 10% of the thalamus, which represent thalamo-associative coupling, and elevate connections in the bottom 10%, which represent thalamo-sensory coupling.

Figure 4B and C show that LSD decreases coupling in associative areas and increases FC in sensory-somatomotor regions. This effect was preserved for the LSD-(Ket+LSD) analysis, but inconsistent results emerged without GSR. LSD decreased thalamic coupling in association areas and increased thalamic coupling in sensory-somatomotor areas, irrespective of analysis method, after GSR. However, LSD did not consistently decrease connections that were in the top 10% or elevate connections that were in the bottom 10% without GSR.

Session impacts global brain connectivity in Ketanserin+LSD condition

The authors conducted two resting-state scans to investigate the potentially distinct temporal phases of LSD pharmacology. They found no significant differences in GBC between session 1 and 2 within the Pla and the LSD condition, but significant increases in GBC in the Ket+LSD condition.

We investigated the effects of drug on seven functionally-defined networks using parcellations derived by Yeo et al. (2011), Buckner et al. (2011) and Choi et al. (2012). We found that there were significant main effects for drug condition in all networks except for the dorsal attention network.

Global brain connectivity in somatomotor network correlates with subjective effects

To evaluate the relationship between LSD-induced changes in GBC in functional networks and subjective LSD-induced effects, we correlated the mean 5D-ASC short version score with the change in Fz mean connectivity in the somatomotor network. We investigated the relationship between somatomotor network Fz connectivity and subjective effects using the 5D-ASC short version scale, and found that all five scale scores were significantly correlated with Fz mean connectivity change in the somatomotor network.

cortical gene expression maps

The expression of the main gene of interest (HTR2A) is negatively correlated with the expression of HTR7. Figure 9C illustrates the cortical distribution of HTR2A gene expression, and Figure 9F illustrates that the correlation between the HTR2A cortical gene expression map and the unthresholded GBC Z-score map with GSR is significantly stronger than the correlation between the HTR2A cortical gene expression map and the GBC Z-score map without GSR.

Discussion

The current study closes major knowledge gaps in the area of psychedelic effects in humans by showing that LSD increases GBC across sensory and somatomotor functional networks and reduces GBC in associative networks, which is sensitive to GS removal.

LSD increases GBC across sensory and somatomotor functional networks and reduces GBC in associative networks

We show that LSD induces hyper-connectivity in sensory and somatomotor areas and hypo-connectivity in areas associated with associative networks.

LSD administration increases V1 resting-state connectivity with the rest of the brain, whereas sensory and somatomotor areas exhibit decreased brain-wide shared signal. Results showed that the amygdala exhibited brain-wide hyper-connectivity under LSD, and that participants with the highest LSD-induced coupling within sensory and somatomotor networks also showed the strongest LSD-induced de-coupling in associative networks. This suggests that linked systems-level perturbations may underlie the psychedelic state.

Brain-wide integration of sensory networks and dis-integration of associative networks are observed in LSD-induced altered state of consciousness. This effect is blocked by pre-administration of Ketanserin, indicating that the 5-HT2A receptor is highly dependent on stimulation of LSD-induced alterations in neural and behavioral effects.

LSD’s effect on GBC is sensitive to GS removal

One study investigated the effects of intravenously administered LSD on the default-mode, salience, and frontoparietal attention networks, but the results were contradictory. The results presented in Figure 1 were studied as a function of GS regression, and the results were consistent with the previous study.

GSR statistically attenuates non-neural arts and provides a method to better isolate functional networks in pharmacological resting-state connectivity studies. This dataset is not well-suited for drawing conclusions about GSR suitability for pharmacological neuroimaging, as results are more replicable across session 1 and 2 without GSR. However, experiments that manipulate variables such as breathing rate and vigilance will be key to fully characterize the effects of GSR. There are open knowledge gaps regarding LSD’s effects on neurovascular coupling and hemodynamic response function properties, which need to be addressed in experiments.

Considering the directionality of LSD’s neural effects

We performed additional analyses based on seed-based thalamic functional connectivity, which revealed that LSD-induced changes were consistent after GSR and comparable to GBC effects. However, GS removal attenuated signal components that are neuronal in origin and may be relevant to important LSD-induced properties.

Time-dependent effects of LSD

Animal studies suggest distinct temporal phases of LSD pharmacology, and Ket blocked subjective LSD-induced effects over the whole time course, indicating that subjective effects are most likely attributable to 5-HT2A receptor stimulation. LSD-induced effects on GBC were investigated using two resting-state scans, conducted 75 min (session 1) and 300 min (session 2) after the second drug administration. Ketanersin blocked LSD effects in session one across all networks. The effect of Ket alone on GBC is preferentially inhibited by 5-HT2A receptor activation. Studies using pre-treatment of LSD by antagonists on receptors other than 5-HT2A are needed to determine if the second phase is indeed modulated by another receptor system.

LSD-induced alterations in GBC in the somatomotor network are associated with subjective effects

The somatomotor network was associated with the subjective effects of LSD, including blissful state, disembodiment, changed meaning of percepts, elementary imagery, spiritual experience. The somatomotor network was also associated with the perception of presence and agency, and therefore a sense of self. Somatomotor system activity and connectivity has been implicated in the pathophysiology of schizophrenia, and the current results show that this activity is also closely associated with an LSD-induced psychedelic state.

We used a threshold-free Z-score map of LSD effects relative to Ket blockade and Pla to investigate LSD’s receptor pharmacology. The results showed that LSD-induced changes in functional connectivity after GSR quantitatively match the HTR2A expression.

The results strongly implicate the involvement of the 5-HT2A receptor in LSD’s neuronal and subjective effects, and the expression of the 5-HT7 receptor was highly negatively correlated with LSD-induced changes in functional connectivity. However, it is also possible that the 5-HT7 receptor functionally contributes to LSD-induced effects. We show that GSR improves the spatial match between gene expression maps and GBC maps, and that this approach can be used to relate spatial gene expression profiles to neuropharmacological manipulations in humans.

Conclusion

The current results show that LSD induces widespread alterations of GBC in cortical and subcortical brain areas, which are sensitive to GSR. The 5-HT2A receptor plays a critical role in subjective and neuronal LSD-induced effects, but at a later phase, modulation by receptors other than the 5-HT2A receptor is involved.

Participants

Participants were recruited through advertisements placed in local universities and underwent a screening visit before inclusion in the larger study protocol. They were required to abstain from drug and alcohol use and from smoking for at least 60 min before MRI assessment.

Twenty-five participants took part in the study, and 24 were included in the final analysis. No substantial side effects were recorded during the study, and participants were contacted again three months after the last drug administration.

Study design

The study employed a fully double-blind, randomized, cross-over design and included participants who had previously taken placebo, LSD, or Ketanserin. The resting-state scan was conducted 75 and 300 min after treatment administration and a short version of the 5D-ASC was administered 180, 250, and 360 min after drug intake.

Preprocessing

Structural and functional MRI data were first preprocessed according to the methods provided by the Human Connectome Project (HCP), outlined below, and described in detail by the WU-Minn HCP consortium.

The T1w/T2w images were corrected for bias-field distortions and warped to the standard Montreal Neurological Institute-152 (MNI-152) brain template, then converted to the Connectivity Informatics Technology Initiative (CIFTI) volume/surface ‘grayordinate’ space.

BOLD images were corrected for field inhomogeneity distortion, phase encoding direction distortions and susceptibility arts using FSL’s TOPUP algorithm, and then registered to the structural images via FLIRT/FNIRT.

After minimal HCP preprocessing steps, a high-pass filter was applied to the BOLD time series, and a nuisance variable was computed for ventricles and deep white matter. Finally, the mean gray matter time series was regressed to address spatially pervasive artefacts. GSR was performed using standard procedures, excluding ventricles and white matter, and used as a nuisance predictor within a multiple linear regression model. All data were motion-scrubbed as recommended by Power et al. (2013), and subjects with more than 50% frames flagged were completely excluded from all analyses.

Global brain connectivity calculation

Most connectivity studies focus on pre-defined areas (i.e. seed-based approaches). Global brain connectivity (GBC) measures connectivity from a given grayordinate to all other voxelgrayordinates simultaneously by computing average connectivity strength, thereby producing an unbiased approach as to the location of dysconnectivity. The GBC approach was applied to the data using in-house Matlab tools, and the results were transformed to Fisher z-values and then computed as a mean. The GBC results were verified using a non-normalized covariance measure, which did not alter effects.

Thalamic seed functional connectivity

We computed a thalamus correlation and covariation map by extracting average time-series across all grayordinates in each subject’s bilateral thalamus.

Global signal regression

We separately examined results without GSR implemented, because emerging findings suggest that clinical populations exhibit elevated GS variability, and that GS may contain major elements of respiratory artefacts.

Global gray matter signal beta map calculation

We used multiple regression analysis to obtain global signal beta values for each subject by calculating mean raw BOLD signal averaged over all grayordinates for each time point.

The BOLD signal is modeled as a function of time, with the nuisance signal being represented by the Xi term. The residual signal is the preprocessed BOLD signal.

BOLDrawk ðÞ¼t b0 þ bGS GS tðÞþ BOLDpreprocessedk ðÞt ;

The ‘mean GS beta weight’ computation is done by fitting a generalized linear model (GLM) to each grayordinate’s BOLD time series to obtain the GS beta weight (bGS ). This ‘GS beta weight’ map is not a functional connectivity measure.

Quality assurance analyses

We computed signal-to-noise ratio and percentage of ‘scrubbed’ images, and correlated these measures with mean Fz-connectivity with and without GSR. All correlations were non-significant.

Statistical analysis of behavioral data

The 5D-ASC comprises 94 items that are answered on visual analogue scales, and 11 recently validated scales were calculated. The relationship between Fz values within the seven functionally defined networks at session two and subjective drug effects was investigated using Bonferroni-corrected Pearson correlations.

Gene expression preprocessing

We used the Allen Human Brain Atlas (AHBA) to create a cortical expression map of candidate receptors in the left hemisphere, and analyzed the results using the Human Connectome Project’s Multi-Modal Parcellation (MMP1.0).

Author contributions

Katrin H Preller, Joshua B Burt, Jie Lisa Ji, Charles H Schleifer, Brendan D Adkinson, Philipp Sta mpfli, Erich Seifritz, Erich Repovs, Software, Methodology, Writing – review and editing; John H Krystal, John D Murray, Franz X Vollenweider, Alan Anticevic.

Ethics

The study was registered at ClinicalTrials.gov and all participants provided written informed consent statements. The study was approved by the Cantonal Ethics Committee of Zurich.