Comparison of Functional Connectivity Changes in Humans and Primates in Psychedelic States
Poster No:
2017
Submission Type:
Abstract Submission
Authors:
Frederick Bagdasarian1, Hanne Hansen1,2, Hsiao-Ying Wey1
Institutions:
1Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Boston, MA, 2Neurobiology Research Unit, Copenhagen University Hospital, Copenhagen, Denmark
First Author:
Frederick Bagdasarian, Ph.D.
Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital
Boston, MA
Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital
Boston, MA
Co-Author(s):
Hanne Hansen
Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital|Neurobiology Research Unit, Copenhagen University Hospital
Boston, MA|Copenhagen, Denmark
Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital|Neurobiology Research Unit, Copenhagen University Hospital
Boston, MA|Copenhagen, Denmark
Hsiao-Ying Wey, Assistant Professor
Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital
Boston, MA
Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital
Boston, MA
Introduction:
Psychedelic substances such as psilocybin and LSD have received enormous attention for their positive effects on disorders like depression1 and alcoholism.2 Functional imaging studies on these drugs often report reduced within-network functional connectivity (FC) and an increased between-network FC.3-5 Though animal models are used to characterize and evaluate psychedelics, understanding the translational validity of such assessments between animals and humans is essential for psychedelic research. This work seeks to (a) identify if psychedelic modulation of FC in anesthetized nonhuman primates (NHPs) with psilocybin may have similar findings as a human cohort with LSD previously reported, thereby (b) providing support for the use of a NHP model in psychedelic neuroimaging.
Methods:
Anesthetized NHPs were maintained at 1.0-1.2% isofluorane and scanned on a 3T Siemens TIM-Trio MRI. An anatomical T1-weighted MPRAGE with 1mm isotropic resolution was acquired. For fMRI, NHPs were injected with Ferumoxytol (10mg/kg) to improve SNR. An EPI sequence (TE/TR = 22/3000ms, 1.3mm isotropic resolution) was acquired during drug administration. NHP received an i.v. solution of psilocybin (N=7: 30µg/kg [N=2]; 60µg/kg [N=3]; 90µg/kg [N=2]) mid-scan. Preprocessing included slice-timing correction, motion-correction, brain extraction, template registration, bias-field correction, 4mm spatial smoothing, segmenting 15-min of resting state pre, post-drug data, grand-mean scaling, band-pass filtering, linear/quadratic trend removal, nuisance regression of motion, CSF and white matter. For human analysis, 15 human fMRI datasets with full pre-processing were obtained from OpenNeuro6 from a study by Carhart-Harris et al.7 These data were under conditions (i.v.) of placebo and 75µg LSD. Independent Component Analysis (ICA) derived seeds were used instead of a-priori selected seeds to reduce user biases in the analysis. FSL's MELODIC was used for ICA. Seed selection and analysis was performed via: (1) Assuming 25 ICs for initial ICA model fitting; (2) Extracting non-noisy ICs from the initial 25 ICs; (3) Of those selected, identifying ICs with similar spatial distribution across species (Figure 1) as seeds. Voxel-wise Pearson's correlation coefficients were calculated and converted to Z-scores. Statistical analysis was a paired t-test of psychedelic vs non-psychedelic states and thresholded at Z>2.3 with cluster-corrected at p<0.05.
Results:
Five IC networks across humans and NHP (Figure 1) with similar spatial distributions were identified for seed-based analysis. Figure 2 shows FC differences between non-psychedelic and psychedelic states in both species. Potential overlapping similarities in both NHP and human FC-MRI results are: 1) reduced FC within portions of the auditory network; 2) increased striatal FC to thalamic/hypothalamic areas; 3) reduced FC from sensorimotor areas to occipital, temporal cortices. Human data, in general, had more widespread FC changes globally across the brain, in contrast to NHP, which were isolated to the areas indicated in Figure 2.
·Figure 1: ICA-derived seed regions used for functional connectivity analysis overlayed NHP (left) and Human (right) templates.
·Figure 2: Functional connectivity (FC) from Auditory, Subcortical, Sensorimotor seeds. Blue indicates reduced FC. Red indicates increased FC. Green are portions of seed ROI in slices of interest.
Conclusions:
This work evaluated FC changes by psychedelics in awake humans (LSD) and anesthetized NHPs (psilocybin). Trends may support the use of anesthetized NHPs as a model for evaluating psychedelic drug states. Common characteristics across species were underscored by auditory, sensory, striatal/thalamic, and occipital cortical domains. Psychedelic mediation of cortical-striatal-thalamic circuits was also reported in a work on humans with LSD,8 which our findings provide support in a NHP model. Additional significance in humans may be due to larger sample size and/or use of anesthesia in NHPs. Consistency of human results with the original work by Carhart-Harris et al.7 is evident. Future analyses will incorporate comparisons to human data with psilocybin. We also plan to conduct FC-MRI studies on LSD in NHPs, in addition to larger NHP sample sizes.
Modeling and Analysis Methods:
fMRI Connectivity and Network Modeling
Task-Independent and Resting-State Analysis 1
Perception, Attention and Motor Behavior:
Consciousness and Awareness 2
Physiology, Metabolism and Neurotransmission :
Pharmacology and Neurotransmission
Keywords:
Consciousness
FUNCTIONAL MRI
MRI
Other - Psychedelics; Drug Analysis
1|2Indicates the priority used for review