Psilocybin-induced default mode network hypoconnectivity is blunted in alcohol-dependent rats

Alcohol Use Disorder (AUD) adversely affects the lives of millions of people, but still lacks effective treatment options. Recent advancements in psychedelic research suggest psilocybin to be potentially efficacious for AUD. However, major knowledge gaps remain regarding (1) psilocybin’s general mode of action and (2) AUD-specific alterations of responsivity to psilocybin treatment in the brain that are crucial for treatment development. Here, we conducted a randomized, placebo-controlled crossover pharmaco-fMRI study on psilocybin effects using a translational approach with healthy rats and a rat model of alcohol relapse. Psilocybin effects were quantified with resting-state functional connectivity using data-driven whole-brain global brain connectivity, network-based statistics, graph theory, hypothesis-driven Default Mode Network (DMN)-specific connectivity, and entropy analyses. Results demonstrate that psilocybin induced an acute wide-spread decrease in different functional connectivity domains together with a distinct increase of connectivity between serotonergic core regions and cortical areas. We could further provide translational evidence for psilocybin-induced DMN hypoconnectivity reported in humans. Psilocybin showed an AUD-specific blunting of DMN hypoconnectivity, which strongly correlated to the alcohol relapse intensity and was mainly driven by medial prefrontal regions. In conclusion, our results provide translational validity for acute psilocybin-induced neural effects in the rodent brain. Furthermore, alcohol relapse severity was negatively correlated with neural responsivity to psilocybin treatment. Our data suggest that a clinical standard dose of psilocybin may not be sufficient to treat severe AUD cases; a finding that should be considered for future clinical trials.


Fig. S1: Mean sample entropy of the default mode network (DMN).
Psilocybin administration did not significantly affect mean DMN sample entropy compared to placebo (F1,19= 0.58, p>0.05).For mean entropy calculation, all voxels of the DMN as defined in the main part of the manuscript were used.Subgroup-by-treatment interaction effect 18 minutes after psilocybin administration, comparing psilocybin effects in ADE to control animals.Of note, no clusters survived a cluster-correction (pFWE cluster-corrected>0.05) with a clusterdefining threshold of t>2.54 (corresponding to p<0.01) and t>3.58 (corresponding to p<0.001).Psilocybin-induced GBC decrease was slightly more pronounced in the ADE group compared to the control animals in blueish regions.(B) NBS, (C) global and local graph metrics, (D) within-DMN strength and entropy of the DMN regions did not reveal any significant subgroup-by-treatment interaction effect (p>0.05).ADE, alcohol deprivation effect rats (n=15); ANOVA, analysis of variance; Con, control rats (n=6); ΔC, deviation of the clustering coefficient from both lattice and random networks constructed with the same number of nodes and the same degree distribution; ΔL, deviation of the network's characteristic path length from both lattice and random networks constructed with the same number of nodes and the same degree distribution; Psi, psilocybin; Sal, saline; # p<0.10;For abbreviation of brain regions, see Fig. 2. All psilocybin and saline values were first compared to its respective baseline, before feeding the difference into the repeated measures ANOVA.
Fig. S3: Subgroup-specific psilocybin effects on GBC and FC compared to placebo.(A) Unthresholded t-value map illustrating post-hoc voxel-wise GBC analysis of the within-subject treatment condition comparing psilocybin to placebo in ADE rats, depicted on the left panel.Regions exhibiting reduced GBC under psilocybin are indicated in blue to light blue, while red to yellow indicates regions with increased GBC.Correspondingly, the right panel demonstrates an unthresholded t-value map of GBC comparison using a post-hoc analysis of the within-subject treatment condition comparing psilocybin to the placebo condition in control rats.(B) areas surviving clustercorrection (pFWE cluster-corrected<0.05) of the respective unthresholded t-value map of panel A, with a cluster-defining threshold of t>2.62 (dark blue and red, cluster-size ≥ 6991 voxels, ADE) for the left panel and of t>3.36 (dark blue, cluster-size ≥ 82 voxels, Con) for the right panel.(C) NBS results (lower panels) comparing FC alterations during psilocybin treatment and placebo demonstrate a comparable but slightly stronger pattern of psilocybin-induced cortical hypoconnectivity (blue) in control animals and ADE rats (left panel for ADE: pNBS<0.05;primary threshold tpt>2.47,corresponding to ppt<0.01;right panel for control animals: pNBS<0.05;primary threshold tpt>2.76,corresponding to ppt<0.01).Data is illustrated in clockwise manner in a connectivity matrix as t-values (psilocybin vs saline) with black boxes marking the connections of the cluster surviving the NBS.The upper panels depict significant t-test results of a test for directionality of the main effect in control animals (upper right panel) and in ADE rats (upper left panel).Significant correlations are marked in bold (puncorrected<0.05)with green background.Importantly, while several associations between relapse intensity and within DMN strength/mean DMN FC reached significance (puncorrected<0.05for PL, Cing2, Aud2, RS, CA1, mean DMN FC), no significant associations to baseline drinking rate were found.For (C), voxel-wise correlation coefficients (Spearman's rsp) were calculated between the GBC scores of the 15 ADE rats and their baseline drinking rates before being cluster corrected (same methodological approach as in Fig. 5B of the main part of the manuscript).No cluster survived the correction (pcluster-corrected<0.05;cluster-defining threshold t>2.65 (corresponding to p<0.01), dark blue; cluster-defining threshold t>3.85 (corresponding to p<0.001), light blue).ADE, alcohol deprivation effect rats (n=15); Con, control rats (n=6); GBC, global brain connectivity Fig. S5: Baseline comparison between ADE and control rats at first measurement.We performed GBC and NBS analysis of the baseline condition (= pre-injection of the first session, 8.5 minutes) to evaluate baseline differences between ADE and controls without the pharmacological challenge.(A) The unthresholded t-value map comparing GBC between ADE and control animals at the time point before injection shows only very weak differences between the two groups (mostly green color) with no cluster surviving cluster-correction even at a lenient cluster defining threshold of t>2.54 (corresponding to pCDT<0.01).Similarly, (B) NBS analysis between ADE and control rats demonstrated very weak differences between the two groups at baseline (light blue to light yellow color) with no cluster surviving NBS at a primary cluster defining threshold of ppt<0.01.ADE, alcohol deprivation effect rats (n=15); Con, control rats (n=6); CDT, cluster-defining threshold; GBC, global brain connectivity, NBS, network-based statistics; pt, primary threshold Fig. S6: HTR2A transcript levels show no expression differences between alcohol and control rats.Expression patterns of HTR2A mRNA levels were measured by illumina microarray in the nucleus accumbens and are presented as logarithmic values of the normalized expression intensities.Two-sample t-test between the two groups did not reveal significant differences (n=6 animals per group, p>0.05, unpublished data).ADE, alcohol deprivation effect rats, HTR2A, 5-hydroxytryptamine receptor 2A No significant psilocybin treatment effects could be found in the ADE and control group, respectively (post-hoc paired t-test p>0.05 between psilocybin and saline condition).Further, no significant treatment-by-subgroup interaction effect could be demonstrated (unpaired t-test comparing FDPsi-FDSal between ADE and control condition, p>0.05).(B) Exemplary illustration of the region-specific correction for physiological noise (cardiac and respiratory signal) applied by Aztec software 78 .Of note, this correction method particularly takes into account that the brain is nonhomogeneously affected by cardiorespiratory activity.Bright red and blue colors depict areas with relatively strong correction for physiological noise.(C) Comparison of heart rate differences (Δpost-pre-injection, left panel) between psilocybin and saline treatment demonstrated a significant drug effect (within effect: F(1,19)=8.082,p=0.010, pperm=0.001):While heart rate increased slightly under psilocybin, saline led to a decrease of heart rate.Comparison of respiration rate differences (Δpost-pre-injection, right panel) between psilocybin and saline treatment also demonstrated a significant drug effect (within effect: F(1,19)=6.611,p=0.019, pperm=0.019):Respiration rate decreased significantly stronger under psilocybin treatment then under saline treatment.(D) Both, mean heart (left panel) and respiration rate (right panel), calculated as the mean value of the baseline in session 1 and session 2, were significantly lower in ADE animals compared to control animals (heart rate: pperm=0.0187;respiration rate: pperm=0.0014).ADE, alcohol deprivation effect rats (n=15); Con, control rats (n=6); Psi, psilocybin; Sal, saline; * p<0.05; ** p<0.01.

Fig. S2 :
Fig. S2: Subgroup-by-treatment interaction effects for global brain connectivity (GBC), network-based statistic (NBS), graph analysis, within-default-mode-network (DMN) strength and entropy within the DMN.(A)Subgroup-by-treatment interaction effect 18 minutes after psilocybin administration, comparing psilocybin effects in ADE to control animals.Of note, no clusters survived a cluster-correction (pFWE cluster-corrected>0.05) with a clusterdefining threshold of t>2.54 (corresponding to p<0.01) and t>3.58 (corresponding to p<0.001).Psilocybin-induced GBC decrease was slightly more pronounced in the ADE group compared to the control animals in blueish regions.(B) NBS, (C) global and local graph metrics, (D) within-DMN strength and entropy of the DMN regions did not reveal any significant subgroup-by-treatment interaction effect (p>0.05).ADE, alcohol deprivation effect rats (n=15); ANOVA, analysis of variance; Con, control rats (n=6); ΔC, deviation of the clustering coefficient from both lattice and random networks constructed with the same number of nodes and the same degree distribution; ΔL, deviation of the network's characteristic path length from both lattice and random networks constructed with the same number of nodes and the same degree distribution; Psi, psilocybin; Sal, saline; # p<0.10;For abbreviation of brain regions, see Fig.2.All psilocybin and saline values were first compared to its respective baseline, before feeding the difference into the repeated measures ANOVA.

Fig. S4 :
Fig. S4: Correlation between different drinking measures and rsfMRI metrics.The association between baseline drinking rate and the mean FC decrease in the default mode network is illustrated in (A).The table in (B) demonstrates Spearman's correlation coefficients and p-values for the association between alcohol consumption during relapse (% of baseline) and baseline (right panel) and the within DMN strength and mean DMN functional connectivity.

Fig. S7 :
Fig. S7: Assessment and correction for motion, heart and respiration rate.(A) Analysis of average framewise displacement (FD) indicates a generally very low level of motion during the fMRI experiment (mean FD<<0.05mm).No significant psilocybin treatment effects could be found in the ADE and control group, respectively (post-hoc paired t-test p>0.05 between psilocybin and saline condition).Further, no significant treatment-by-subgroup interaction effect could be demonstrated (unpaired t-test comparing FDPsi-FDSal between ADE and control condition, p>0.05).(B) Exemplary illustration of the region-specific correction for physiological noise (cardiac and respiratory signal) applied by Aztec software 78 .Of note, this correction method particularly takes into account that the brain is nonhomogeneously affected by cardiorespiratory activity.Bright red and blue colors depict areas with relatively strong correction for physiological noise.(C) Comparison of heart rate differences (Δpost-pre-injection, left panel) between psilocybin and saline treatment demonstrated a significant drug effect (within effect: F(1,19)=8.082,p=0.010, pperm=0.001):While heart rate increased slightly under psilocybin, saline led to a decrease of heart rate.Comparison of respiration rate differences (Δpost-pre-injection, right panel) between psilocybin and saline treatment also demonstrated a significant drug effect (within effect: F(1,19)=6.611,p=0.019, pperm=0.019):Respiration rate decreased

Fig. S8 :
Fig. S8: Correlation of mean physiological parameters (mean respiration and mean heart rate) with region-ofinterested (ROI)-based functional connectivity (FC) and global brain connectivity (GBC).Spearman's correlation coefficients between mean respiration rate (A) and mean heart rate (B) at baseline and ROI-based FC are illustrated in a matrix.Correlation coefficients were calculated with n=42 baseline sessions, associating the respective FC values between two ROIs with the mean respiration and heart rates at baseline.Of note, correlation values were relatively low for both comparisons and did not survive FDR correction (pFDR>0.05,corrected for the number of connections n=946), arguing against a strong influence of physiological data on ROI-based FC.Spearman's correlation coefficients between physiological metrics at baseline and whole-brain voxel-wise GBC values are illustrated in (C) for mean respiration rate and (D) for mean heart rate.Correlation coefficients were calculated with n=42 baseline sessions, associating the respective GBC values with the baseline physiological metrics.Of note,