The postsynaptic density (PSD) contains a collection of scaffold proteins used for assembling synaptic signaling complexes. However, it is not known how the core-scaffold machinery associates in protein-interaction networks or how proteins encoded by genes involved in complex brain disorders are distributed through spatiotemporal protein complexes. Here using immunopurification, proteomics and bioinformatics, we isolated 2,876 proteins across 41 in vivo interactomes and determined their protein domain composition, correlation to gene expression levels and developmental integration to the PSD. We defined clusters for enrichment of schizophrenia, autism spectrum disorders, developmental delay and intellectual disability risk factors at embryonic day 14 and adult PSD in mice. Mutations in highly connected nodes alter protein–protein interactions modulating macromolecular complexes enriched in disease risk candidates. These results were integrated into a software platform, Synaptic Protein/Pathways Resource (SyPPRes), enabling the prioritization of disease risk factors and their placement within synaptic protein interaction networks.
This work was supported by grants from the National Institute of Child Health and Human Development (MH104603-01A1 to M.P.C.), NIH grant MH108728 (to K.W. and H.Y.), Simons Foundation Autism Research Initiative (SFARI) grants 248429 and 345034, and DOD CDMRP AR110189 (to T.A.). B.N. acknowledges support from the Stanley Center for Psychiatric Research.
Integrated supplementary information
Interactomes dataset Tab 1 (PSD Scaffolds) Protein interactions identified in MS analysis of protein complexes from PSD scaffolds (Dlg4, Dlgap1, Shank3) All protein interactors have no less than 2 peptides identified together in at least triplicate samples by MASCOT and SEQUEST, and are not present in negative control purifications. Column A: Developmental stage/Cellular localization/Cell type Column B: Target Protein Column C Protein interaction (Gene name) Supplementary table contains interactomes of the PSD scaffolding proteins (sheet 1) and PSD scaffold-interactors (sheet 2) described in the main text. Developmental stage/localization column describes at what developmental stage the immunopurification was performed and whether it was with PSD or non-PSD fractions. Tab 2 (Scaffold interactors) Protein interactions identified in MS analysis of protein complexes from PSD proteins that are scaffold interactors All protein interactors have no less than 2 peptides identified together in at least triplicate samples by MASCOT and SEQUEST, and are not present in negative control purifications. Column A: Developmental stage/Cellular localization/Cell type Column B: Target Protein Column C Protein interaction (Gene name) The protein column contains the target protein used for immunopurification while the Interactor columns contain the interactors of that particular protein identified via mass spectrometry with the corresponding mouse, human, or rat annotation corresponding gene symbol according to the Entrez Gene Database.The Mentha and Biogrid columns contain a yes if that particular interaction was reported in the protein-protein interaction databases for the corresponding species Tab 3 Annotations for all protein interactions identified in MS assays A: Interactor gene name B: Cytogenetic location C: Genomic coordinates (GRCh37)D: HUGO Gene Nomenclature Committee (HGNC) id E: Mouse Genome Informatics (MGI) id F-H: Online Mendelian Inheritance in Man (OMIM) annotation
Reported interactions dataset Supplementary table contains previously reported protein-protein interactions overlapping with the current dataset. With a majority of previously reported interactions being derived from BioGRID (release 3.4.142) the table contains the BioGRID-specifc identifiers such as BioGRID interaction and organism ID. Organism ID corresponds to: Homo sapiens = 9606, Mus Musculus = 10090, Rattus norvegicus = 10116. Table also contains Entrez Gene Database identifiers, official gene symbols, reference pubmed ID, and the experimental system for each previously reported interaction.
Protein domain analysis of interactomes Proteins isolated by HPLC-MS/MS from mouse PFC Column A: Gene ID Column B: MGI ID Column C: Description Protein domain composition of PSD Proteins (SMART/Pfam databases) Column A: Database Column B: Domain ID Column C: Number of PSD proteins with domain Column D: p-value Column E: Gene IDs with domains Column F: Total number of domains Column G: Total-Database Column H: Enrichment Column I: Correction (Bonferroni) Column J: False Discovery Rate
Protein domain annotation and enrichment for developmental scaffold protein complexes Individual protein complexes are shown in separate spreadsheets Column A: Domain ID (SMART/Pfam) Column B: Domain name Column C: Number of proteins containing the domain in the protein complex Column D: p-value Column E: Corrected p-value (Bonferroni)
Clustering of developmental Dlg4/Dlgap1/Shank3 protein complexes Node name and network parameters for Dlg4, Dlgap1 and Shank3 (e14, p7, p14 adult) protein complexes. Network parameters include: number of nodes, average degree (average connectivity of a node in the network), network density (normalize average connectivity of each node), clustering coefficient (ratio of N / M, where N is the number of edges between the neighbors of this node, and M is the maximum number of edges that could possibly exist between the neighbors of the node), clustering coefficient of a node is always a number between 0 and 1.The network clustering coefficient is the average of the clustering coefficients for all nodes in the network. Here, nodes with less than two neighbors are assumed to have a clustering coefficient of 0.
Protein domain annotation and enrichment for scaffold interactors protein complexes Individual protein complexes are shown in separate spreadsheets Column A: Domain ID (SMART/Pfam) Column B: Domain name Column C: Number of proteins containing the domain in the protein complex Column D: p-value Column E: Corrected p-value (Bonferroni)
Testing for gene set enrichment among de novo mutations in brain disorders Individual spreadsheets indicate datasets, description and analysis: SCZ, ASD, DD, ID, SIB_CONTROL, CHD-NS, CHD-S.
Clustering of Adult PSD protein complexes Dataset includes: Tnik, Dlg4, Dlgap1, Shank3, Tsc1, Homer1, Nckap1, Cyfip1, Cyfip2, Syngap1, Fmr1, Cnksr2 nodes Columns A-N: Node name and network parameters Columns O-Q: Disrupted interactions in MAGUK protein complexes with Tnik−/− mutation. Columns R-T: Disrupted interactions in Dlgap1 protein complexes with Tnik−/− mutation. Columns U-W: Disrupted interactions in Shank3 protein complexes with Tnik−/− mutation. Columns X-Z: Disrupted interactions in Dlgap1 protein complexes with Shank3ΔC−/+ mutation Columns AA-AC: Disrupted interactions in Shank3 protein complexes with Shank3ΔC−/+ mutation All protein interactions were determined in triplicate samples. Normalized spectral abundance factor (NSAF) was calculated for each protein34, 54 and used for comparison between wt/mutant samples. Table shows disruption of highly connected nodes. Tnik and Shank3 spreadsheets, shows quantitation of total protein levels by WB analysis of PSD proteins with impaired associations to PSD protein complexes in wt and Tnik−/−, Shank3ΔC/+ mice. Columns D and E shows ratios wt/mutant mice rations and standard deviation for triplicate assays. Protein ratios did not present statistical significance (0.66/1.5 p<0.05). Bottom tables shows exact p values for quantitation of protein interactions in scaffold complexes by western blot assays
Non-specific proteins determined by MS. Column A: Uniprot id, Column B: Gene name
Non-specific RNA binding proteins determined by MS. Table shows RNA binding proteins determined by MS in samples treated with RNAse T1/A. Column A: Uniprot id, Column B: Gene name
Antibodies used in IP, and WB assays