The postsynaptic proteome of excitatory synapses comprises ~1,000 highly conserved proteins that control the behavioral repertoire, and mutations disrupting their function cause >130 brain diseases. Here, we document the composition of postsynaptic proteomes in human neocortical regions and integrate it with genetic, functional and structural magnetic resonance imaging, positron emission tomography imaging, and behavioral data. Neocortical regions show signatures of expression of individual proteins, protein complexes, biochemical and metabolic pathways. We characterized the compositional signatures in brain regions involved with language, emotion and memory functions. Integrating large-scale GWAS with regional proteome data identifies the same cortical region for smoking behavior as found with fMRI data. The neocortical postsynaptic proteome data resource can be used to link genetics to brain imaging and behavior, and to study the role of postsynaptic proteins in localization of brain functions.
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Support came from the Medical Research Council (Brain Bank MR/L016400/1) and European Union Seventh Framework Programme (FP7 grant agreement no. 604102) and Horizon 2020 (agreement no. 720270). We thank K. Elsegood for laboratory management, J. DeFelipe for comments on the manuscript, D. Maizels for artwork, and T. Le Bihan and L. Imrie at SynthSys, University of Edinburgh for mass spectrometry sample analysis. The LC-MS QExactive equipment was purchased by a Wellcome Trust Institutional Strategic Support Fund and a strategic award from the Wellcome Trust for the Centre for Immunity, Infection and Evolution (095831/Z/11/Z). Data were extracted from NIFTI (Neuroimaging Informatics Technology Initiative) files using a custom automated script written by J. J. Roy, MEMEX, Inc., Burlington, Ontario, Canada. MRI data were provided by the Human Connectome Project, WU-Minn Consortium (Principal Investigators: D. Van Essen and K. Ugurbil; 1U54MH091657) funded by the 16 NIH Institutes and Centers that support the NIH Blueprint for Neuroscience Research; and by the McDonnell Center for Systems Neuroscience at Washington University.
Integrated Supplementary Information
The magnitude of Anova P-value (x-axis) plotted against the maximum fold change (y-axis) for differentially expressed proteins. Dots correspond to specific proteins (total 1215). Green, 149 differentially expressed proteins with level of significance p < 0.05 and fold change >1.5.
Supplementary Figure 2 Overlap between the 1,213 PSD proteins identified in this study and other published synaptic proteomes
Overlap between the 1,213 PSD proteins identified in this study and other published synaptic proteomes. A) Venn diagram showing the overlap between this study and the human PSD proteomes published in Bayes et al., 2011 and Bayes et al., 2014. B) Venn diagram showing the overlap between this study and the PSD proteomes published in rat and mouse models from Distler et al., 2014, Collins et al., 2006, Fernandez et al., 2009 and Trinidad et al., 2008.
Similarity of the PSD proteome between individuals. Individual clustering dendrograms displaying the similarities between the 12 neocortical regions based upon PSD proteome abundances observed for each individual: A) SD025/13, B) SD042/13, C) SD032/13 and D) SD023/13.
(A) Confidence intervals for the mean PSD protein abundance (Y-axis) plotted for each individual brain A) SD025/13, B) SD042/13, C) SD032/13 and D) SD023/13 (y-axis). (B) Pairwise comparison of the PSD proteome abundances in individuals A-D.
Effect of individual variation on PPM robustness. (A) Bootstrap analysis of PPM cluster robustness with k-means and PAM methods shows all modules were found to be stable (≥0.5 membership robustness). (B) Correlation of the PPMs/clusters across three combined samples (omitting the longest postmortem interval sample) (y-axis) and all four samples (x-axis). (C) Correlation of the PPMs/clusters between three combined samples (omitting the female sample) (y-axis) and all four samples (x-axis). Cr, cluster number; Mr, PPM number.
Supplementary Figure 6 Circos plots showing circular hierarchical clustering of the distribution of PPMs in 12 Bas
Circos plots showing circular hierarchical clustering of the distribution of PPMs in 12 BAs. Shown left, an overlay of all seven protein modules for all 12 cortical regions and individual plots of PPM1 (middle) and PPM2 (right) illustrate differential distributions. PPM 1 (black), PPM 2 (red), PPM 3 (green), PPM 4 (blue), PPM 5 (cyan) and PPM 6 (magenta) and PPM 7 (yellow). Width of the link is proportional to the fraction of the regional proteins contributed while its color corresponds to the respective PPM.
Supplementary Figure 7 Alignment of BAs in our study with the multimodal parcellated human cortex from the Human Connectome Project (HCP)
Alignment of Brodmann areas (BAs) in our study with the multi-modal parcellated human cortex from the Human Connectome Project (HCP). A) Line plot showing the trends in abundance for the four myelin proteins (Mog, dark blue; Plp2, red; Omg, green; Pmp2, purple), the average abundance of all four proteins (Avgenes, shown in cyan) and the average myelin abundance detected by MRI reported by Glasser et al., 2016 from all parcels corresponding to BAs (Avmyelin, shown in orange) and from selected parcels, which give the best correlations score (av1, shown in light blue). B) Atlas of 40 HCP regions (annotated) corresponding to BA areas used for synapse proteomics.
Supplementary Figure 8 Correlations between the Postsynaptic Proteome Module (PPM) abundance and task-fMRI scores across 12 BAs
Significant correlations between the Postsynaptic Proteome Module (PPM) abundance (averaged protein abundance for respective module) values and structural (Thickness, myelin) and task-fMRI scores across 12 Brodmann areas. The coefficients with an absolute value above 0.53 were significant at the FDR level 5%. Scale bar indicates positive and negative correlations. See Methods and supplementary Table 12 for definitions of tfMRI nomenclature.
The distribution of correlation scores obtained for 10,000 reshuffled samples is shown for module2/GI and module3/GI pairs. Note that that the observed correlation values (vertical black line) lie far from the random distribution, indicating that the observed correlation with PET data is extremely unlikely to have occurred by chance (resampling-based q-value < 10−4).
Supplementary Figures 1–9.
Human brain tissue used in this study.
LC-MS/MS quantitation and comparison of 1,213 PSD proteins throughout the human neocortex one-way ANOVA.
Summary of 149 proteins differentially abundant across the 12 regions of the neocortex.
New PSD proteins detected only in this study.
Differential stability analysis: correlation of PSD proteome abundances between individual brains.
Differential stability analysis: functional enrichment in PSD proteins that correlate between individual brains.
LC-MS/MS quantitation and comparison of the human PSD proteome throughout the human neocortex.
Ranked abundance of PSD proteins in 12 neocortical regions.
TOST-test analysis: proteins that do not change between brain regions.
Functional enrichment analysis: proteins that do not change between brain regions.
Differential abundance across the human neocortex of key molecules involved in cognition and memory.
Composition of protein modules (PMMs).
Nomenclature used for the task-fMRI (tfMRI) data. A full list of the tfMRI terms as well as their corresponding full descriptions as per Glasser et al., 2016, Supplementary Information file 3, Table 3 (https://doi.org/doi:10.1038/nature18933).
Correlation between myelin values obtained by LC-MS/MS in this study and the parcellated cortex Glasser et al., 2016 study.
Correlation values between functional MRI scores from Glasser et al., 2016 and abundance values of the 1,213 PSD proteins identified in this study.
Significant correlation values (P < 0.05, corrected for multiple testing by 10,000 permutations) between functional MRI scores from Glasser et al., 2016 and abundance values of the 1,213 PSD proteins identified in this study. Protein, protein name; funct, functional term from Glasser et al., 2016; pval, P value; corr.val, R2.