Src kinase as a mediator of convergent molecular abnormalities leading to NMDAR hypoactivity in schizophrenia

ORIGINAL ARTICLE

Src kinase as a mediator of convergent molecularabnormalities leading to NMDAR hypoactivity inschizophreniaA Banerjee1,5, H-Y Wang2,5, KE Borgmann-Winter1, ML MacDonald1, H Kaprielian1, A Stucky2, J Kvasic2, C Egbujo1, R Ray1, K Talbot1,SE Hemby3, SJ Siegel1, SE Arnold1, P Sleiman4, X Chang4, H Hakonarson4, RE Gur1 and C-G Hahn1

Numerous investigations support decreased glutamatergic signaling as a pathogenic mechanism of schizophrenia, yet themolecular underpinnings for such dysregulation are largely unknown. In the post-mortem dorsolateral prefrontal cortex (DLPFC),we found striking decreases in tyrosine phosphorylation of N-methyl-D aspartate (NMDA) receptor subunit 2 (GluN2) that is criticalfor neuroplasticity. The decreased GluN2 activity in schizophrenia may not be because of downregulation of NMDA receptors asMK-801 binding and NMDA receptor complexes in postsynaptic density (PSD) were in fact increased in schizophrenia cases. At thepostreceptor level, however, we found striking reductions in the protein kinase C, Pyk 2 and Src kinase activity that in tandem candecrease GluN2 activation. Given that Src serves as a hub of various signaling mechanisms affecting GluN2 phosphorylation, wepostulated that Src hypoactivity may result from convergent alterations of various schizophrenia susceptibility pathways and thusmediate their effects on NMDA receptor signaling. Indeed, the DLPFC of schizophrenia cases exhibit increased PSD-95 and erbB4and decreased receptor-type tyrosine-protein phosphatase-α (RPTPα) and dysbindin-1, each of which reduces Src activity viaprotein interaction with Src. To test genomic underpinnings for Src hypoactivity, we examined genome-wide association studyresults, incorporating 13 394 cases and 34 676 controls. We found no significant association of individual variants of Src and itsdirect regulators with schizophrenia. However, a protein–protein interaction-based network centered on Src showed significantenrichment of gene-level associations with schizophrenia compared with other psychiatric illnesses. Our results togetherdemonstrate striking decreases in NMDA receptor signaling at the postreceptor level and propose Src as a nodal point ofconvergent dysregulations affecting NMDA receptor pathway via protein–protein associations.

Molecular Psychiatry advance online publication, 21 October 2014; doi:10.1038/mp.2014.115

INTRODUCTIONMultiple lines of evidence support glutamatergic dysregulation asa key pathogenic mechanism of schizophrenia (SCZ), widelyknown as the N-methyl-D aspartate (NMDA) receptor (NMDAR)hypofunction hypothesis.1–3 This postulate originated from clinicalobservations in which NMDAR blockade induced psychosis inhealthy human subjects and worsened symptoms in patients withSCZ.4–6 In addition, rodent studies offered further supportdemonstrating that NMDAR antagonists or mutation of NMDARgenes induced neurophysiological and behavioral endopheno-types of psychosis.3,7,8 Furthermore, treatment studies demon-strated that clinical manifestations of SCZ in patients and SCZ-likephenotypes in animal models are reduced by enhancement ofglutamatergic neurotransmission.9,10

Despite the noted evidence for NMDAR hypofunction in SCZ,mechanistic underpinnings for this dysfunction are presentlyunknown. Presynaptically, decreased D-serine and glutamatecarboxypeptidase II, and increased kynurenic acid, were observedin the CSF or post-mortem brains of patients with SCZ.11,12 Inaddition, brain imaging studies in particular have shown

alterations in gluthathione.13 Although some of the results arenot fully in concert, these together suggest dysregulations inpresynaptic glutamate signaling in SCZ.1,13

In search of postsynaptic mechanisms, multiple groups haveinvestigated mRNA or protein levels of NMDAR subunits and theirsignaling partners in the post-mortem brains of SCZ cases, andhave reported various interesting findings, albeit somewhatinconsistent.14–16 Given that NMDAR activity arises from thereceptor complexes that are targeted to specific subcellularmicrodomains and involves extensive intracellular signalingmechanisms,17,18 here we examined NMDAR activity specificallyin synaptic regions and its underlying signaling mechanisms inpost-mortem brains.Glutamate binding to NMDARs activates a cascade of kinases at

the postreceptor level (called ‘postreceptor kinases’ henceforth). Itenhances Ca2+ influx that activates Ca2+/calmodulin-dependentkinase II and protein kinase C (PKC).17,19 The latter phosphorylatesproline-rich tyrosine kinase 2 (Pyk2), leading to activation of thesarcoma tyrosine kinase known as Src.20,21 Conversely, postrecep-tor kinases in turn regulate NMDAR channel activity via tyrosinephosphorylation of GluN2 subunits, critical for sustained

1Department of Psychiatry, University of Pennsylvania, Philadelphia, PA, USA; 2Department of Physiology, Pharmacology and Neuroscience, City University of New York MedicalSchool, New York, NY, USA; 3Department of Physiology and Pharmacology, Wake Forest University School of Medicine, Winston-Salem, NC, USA and 4The Center for AppliedGenomics, The Children’s Hospital of Philadelphia, and Department of Pediatrics, University of Pennsylvania School of Medicine, Philadelphia, PA, USA. Correspondence: Dr C-GHahn, Center for Neurobiology and Behavior, Department of Psychiatry, University of Pennsylvania, 2216 TRL, 125 South 31st Street, Philadelphia, PA 19104-3403, USA.E-mail: [email protected] authors contributed equally to this work.Received 18 February 2014; revised 19 August 2014; accepted 21 August 2014

Molecular Psychiatry (2014), 1–10© 2014 Macmillan Publishers Limited All rights reserved 1359-4184/14

www.nature.com/mp

http://dx.doi.org/10.1038/mp.2014.115

mailto:[email protected]

http://www.nature.com/mp

enhancement of channel activity in synaptic plasticity.22 Thus,NMDARs and their downstream postreceptor kinases are inter-dependent for their activation, forming a loop of cascade:NMDAR→ PKC→ Pyk2 → Src→pYGluN2. Of the postreceptorkinases in this cascade, Src plays a primary role as it directlyphosphorylates GluN2 subunits as a molecular hub of varioussignaling pathways, including those outside the NMDAR pathway,affecting tyrosine phosphorylation of GluN2. As such, Src serves asa conduit for various non-NMDAR signaling mechanisms thatregulate tyrosine phosphorylation of GluN2 and synapticplasticity-related NMDAR function.18

Residing in the postsynaptic density (PSD), synaptic NMDARsare associated with scaffolding proteins and many signalingmolecules and comprise a macromolecular receptor complex.23,24

Hence, the NMDAR complex can reflect alterations in diversesignaling pathways, including those implicated for SCZ. Increasinggenetic evidence indeed supports this possibility. Among geneticvariants found to be associated with SCZ are those of DTNBP1,25

DISC1,26 ERBB4,27 neuregulin 1 (NRG1)28,29 and PTPRA,30,31 whosegene products can affect NMDAR complexes.26,28,30 Moreover,genome-wide investigation of de novo copy number variations aswell as de novo mutations point to multiple genes within NMDAR/activity-regulated cytoskeleton-associated (ARC) complexes,32,33

suggesting that NMDAR complexes may serve as molecularsubstrates upon which various SCZ susceptibility factors converge.A critical next step then will be to delineate molecular under-pinnings for NMDAR hypofunction in SCZ and to identify themechanisms of convergence in the brain of patients with SCZ.Previously, we reported that NMDA/glycine-induced tyrosine

phosphorylation of GluN2 subunits, critical for sustained enhance-ment of NMDAR activity, was decreased in the dorsolateralprefrontal cortex (DLPFC) of SCZ subjects as compared withcontrols.28 Tyrosine phosphorylation of GluN2 subunits is gov-erned by Src kinase22 that can serve as a molecular hub of alte-rations in various signaling pathways including the SCZ suscept-ibility pathways.22 Indeed, NRG1, ErbB4, PSD-95, dysbindin-1 orRPTPα, each of which has been implicated for SCZ, can modulateSrc kinase.22,31,34,35 These together led us to hypothesize that Srckinase may mediate convergent dysregulations of various SCZsusceptibility pathways to reduce tyrosine phosphorylation ofGluN2 and thereby precipitate NMDAR hypoactivity.Examination of receptor activity and intracellular signaling is a

methodological challenge in post-mortem studies. We developedseveral novel methodologies to quantify NMDAR signaling activity,protein–protein interactions (PPIs) in NMDAR complexes and theenzyme activity of Src in post-mortem brain tissue.28,34,36 Amatched-pairs design was used to minimize the effects of age andsex in observed differences between SCZ and control cases. Wefocused on the DLPFC because disruption of this brain area iscommon in SCZ,37,38 contributes to deficits in selective attention,working memory and executive function in this disorder37,39,40

and has been shown to display NMDAR hypofunction in SCZ. Here,we present evidence pointing to the postreceptor segment of theNMDAR pathway as a locus of dysregulation in the DLPFC of SCZcases and demonstrate that Src hypoactivity, resulting fromconvergent alterations of susceptibility pathways, mediatesreduction of tyrosine phosphorylation of GluN2 in the DLPFC ofSCZ cases.

MATERIALS AND METHODSPost-mortem brainsFlash-frozen DLPFC tissues from 17 pairs of SCZ subjects and of controls,matched for age, sex and post-mortem interval, obtained from the PennBrain Bank at the University of Pennsylvania, were used for the study.Approved by the Institutional Review Board at the University ofPennsylvania, subjects were prospectively diagnosed by DSM-IV (Diag-nostic and Statistical Manual of Mental Disorders, 4th Edition) criteria and

consents for autopsy were obtained from the next of kin or a legalguardian. Subjects with a history of substance abuse, neurological illnessesor the need for ventilator support near death were excluded.Neuropathological assessments were performed on the PFC of all subjectsto exclude cases with gross neurodegenerative process and abnormallevels of senile plaques or Lewy bodies (additional details are inSupplementary Data). Detailed demographic data are shown inSupplementary Table S1 and S2.

Animal studiesRodent tissues examined in this study included PFC tissues fromhomozygous sdy mice (n= 6),41 from Ptpra− /− -null mice (n= 7)31

(provided by Jan Sap (University of Paris, Paris, France) and CatherinePallen (University of British Columbia, Vancouver, BC, Canada) respectively)and from antipsychotic-treated rhesus monkeys (n= 8/9) (provided byScott E Hemby, Wake Forest University, Winston-Salem, NC, USA; seetreatment details in Supplementary Figure S2). All protocols wereapproved by the Institutional Animal Care and Use Committee at theUniversity of Pennsylvania and all experiments on mice were conducted inaccordance with the University Laboratory Animal Resources guidelines.

MK-801 assayAs previously described,42 10 μg of P2 membrane extracts were incubatedwith varying concentrations of [3H]MK-801 (Perkin Elmer, Waltham, MA, USA),ranging from 0 to 60 nM, in the presence of 1 μM gycine, 10 μM glutamate and10 μM spermidine in 5mM Tris-HCl (pH 7.4). The samples were incubated at30 °C for 2 h and the reactions were precipitated on GF/C (GE, Pittsburgh, PA,USA) glass filters and radioactivity was measured using Scintiverse cocktail(Fisher Scientific, Fair Lawn, NJ, USA) in a liquid scintillation counter(Pharmacia, Peapack, NJ, USA).

Post-mortem brain receptor activationTissue slices (100 μm) from DLPFC of frozen post-mortem brains wereincubated with 1 μM glycine and 10 μM NMDA for 30min at 37 °C and synapto-somal fractions (S) were prepared as previously described in detail.28,43

Src kinase assayP2 protein extracts (10 μg) were incubated with 1 mM of EPQ(pY)EEIPIA,which activates src, or EPQYEEIPIA (Anaspec, Fremont, CA, USA), controlpeptide, overnight at 4 °C. Src kinase activity was determined bymeasuring incorporation of γATP in the peptide substrate provided bythe manufacturer (Millipore, Billerica, MA, USA) in liquid scintillationcounter (Pharmacia, Peapack, NJ, USA).

Tissue fractionationPSD fractions were isolated from flash frozen coronal slabs of the DLPFC,including Brodmann areas 9 and/or 46, by the method previouslydescribed in detail36 (see Supplementary Text).

Immunoprecipitation, antibodies and western blottingSee Supplementary Text.

Src binding assay with biotinylated peptideFirst, Src proteins were affinity purified from synaptic membranes of theDLPFC of SCZ and CTRL cases using Src antibodies. The EPQ(pY)EEIPIA andEPQYEEIPIA peptides (Anaspec) were biotinylated following the protocol byThermo Scientific (Fremont, CA, USA). Then, 1 μg of purified Src proteins wasincubated with biotinylated peptides at 30 °C for 30min. The reaction wasterminated using ice‐cold immunoprecipitation buffer and incubated with20 μl of Nutravidin beads at 4 °C for 60min. Pelleted beads were analyzed bywestern blotting with streptavidin–horseradish peroxidase conjugate.

GWAS data setsThe SCZ data set consisted of the results of a genome-wide associationstudy (GWAS) meta-analysis of 15 cohorts that included 13 394 cases, ofwhich 11 985 were classified as schizophrenia, 377 as schizoaffective and1032 as bipolar, and 34 676 controls (described in Sleiman et al.44). Single-nucleotide polymorphisms (SNPs) with missing data in more than twocohorts were excluded from further analysis. The major depressive disorder

Src kinase hypoactivity in schizophreniaA Banerjee et al

2

Molecular Psychiatry (2014), 1 – 10 © 2014 Macmillan Publishers Limited

(MDD), bipolar disorder (BP) and attention deficit hyperactivity disorder(ADHD) data sets were downloaded from the Psychiatric GenomicsConsortium (PGC) website (http://www.med.unc.edu/pgc) and consistedof 9240 MDD cases and 9519 controls;45 7481 BP cases and 9250controls;46 2064 ADHD trios, 896 ADHD cases and 2455 controls.47 TheCrohn’s disease (CD) data set comprised 3685 cases and 5968 controls.48

Gene-level associationsGene-level associations were calculated using VEGAS.49 VEGAS performsMonte Carlo simulations from the multivariate normal distribution basedon the patterns of linkage disequilibrium in a reference population andassigns an estimated P-value to each gene. SNPs located within 50 kbupstream and 50 kb downstream of gene boundaries were included in theanalysis in order to capture regulatory regions and SNPs in linkagedisequilibrium. Gene-level significance was set at P≤ 0.05.

PPI network analysisThe PPI network was constructed based on indices of protein interactionsderived from primary interaction databases including BIND, BioGRID,CORUM, DIP, HPRD, InnateDB, IntAct, MatrixDB, MINT, MPact, MPIDB, MPPIand OPHID as compiled by iRefindex.50 To control the false positive rate,only human–human protein interactions supported by at least twopublications listed in PubMed were considered for this study. Based onthe PPIs we constructed a network of the Src pathway genes and theirinteractors (Figure 5b). The resulting Src network consisted of 30 genes.

Enrichment analysisTo test whether the src PPI network was enriched for genic associationwith SCZ, we generated gene-level P-values for each of the 30 genes usingVEGAS from 5 separate GWAS data sets including SCZ, MDD, BP, ADHD anda non-psychiatric disease, CD. To determine enrichment of significance, weidentified all SNPs within each of the 30 genes, including all SNPs 50 kbupstream and downstream of the genes, as was done to generate gene-level significance for each of the data sets separately. Total SNP numbersvaried between each of the studies depending on the genotypingplatform. SNP-level significance was set at P≤ 0.05. Enrichment ofassociation between SCZ and each of the four other phenotypes wastested using two-tailed Fisher’s exact test as well as between CD and thefour psychiatric disorders. Because of the low cell counts in the gene-levelassociations, enrichment statistics were generated on individual single-nucleotide variations within the genes.

RESULTSThis study employs multiple methods that were devised speci-fically to assess NMDAR function in post-mortem brains.28,36,43,51 Weconducted extensive preliminary work28,36,43,51 to verify the stabilityof these measures in post-mortem tissues, with special attention tothe range of ages, antipsychotic doses and post-mortem interval ofour cases in human (Supplementary Figure S1), rodent and monkeybrains (Supplementary Figures S2–S4). The post mortem study wasconducted on 17 SCZ cases (Supplementary Tables S1–S2), eachmatched to a normal control of the same sex, similar age(80.12±1.40 years (mean± s.d.) for SCZ cases vs 79.53±2.31 yearsfor controls), equivalent pH (6.56±0.039 for SCZ cases vs6.59±0.089) and comparable, low post-mortem interval(12.12±0.98 h vs 8.76±0.08 for controls).

NMDAR complexes are increased in the PSD of SCZ casesThe abundance of NMDAR complexes was assessed in synapticmembranes of the DLPFC using a binding assay with thenoncompetitive NMDAR antagonist MK-801 in the presence ofglutamate and glycine.52,53 The Bmax for MK-801 binding showedthat the NMDAR density in SCZ cases was ∼ 30% higher than that incontrols (Figures 1a and b, t(13) = 2.837, P=0.014). To test whethersuch increases in NMDAR complexes occur in the PSD, the PSD-enriched fractions were isolated from the DLPFC and proteinassociations in NMDAR complexes were assessed using immuno-precipitation experiments for GluN1 or PSD-95. The results showed

a higher level of PSD-95 in GluN1 immunoprecipitates (t(13) = 2.279,P=0.0389, Figures 1c and d), and a higher level of GluN1 in PSD-95immunoprecipitates in SCZ brains (t(13) = 2.405 P=0.0318, Figures1c and d). These results indicate increased number of NMDARcomplexes in the PSD of SCZ cases, which may not be a likelyexplanation for decreased tyrosine phosphorylation of GluN2. Thus,we turned our attention to molecular events downstream to thereceptors, that is, the postreceptor level.

NMDAR signaling is hypoactive at the postreceptor level inschizophreniaWe measured NMDAR signaling responses to 10 μM NMDA+1 μMglycine applied to 50 μm thick slices from the DLPFC of patientsand healthy matched controls. SCZ cases showed decreases inNMDA/glycine-induced tyrosine phosphorylation of GluN2 that iscritical for NMDAR channel activity (P= 0.001, t(11) = 6.066, Figures2a and b). Next, we examined PKCγ, phospholipase C-γ (PLCγ) andSrc that in tandem govern tyrosine phosphorylation of GluN2. Insynaptic membranes derived from the tissues that were incubatedwith NMDA/glycine, we found decreases in GluN1 association withPLCγ and PKCγ54,55 (Figures 2c and d, P= 0.027, t(13) = 2.406 forPLCγ and P≤ 0.026, t(13) = 2.512 for PKCγ), indicative of decreasedNMDAR activation. In addition, pY402Pyk2 and pY419Src, reflectiveof the activity of these kinases, were also decreased in GluN1immunoprecipitates in SCZ cases compared with their matchedcontrols (Figures 2c and d, P= 0.029, t(13) = 2.831 for pY402Pyk2and Po0.043, t(13) = 2.701 for pY419Src).To further test Src phosphorylation, we examined GluN2A

complexes in which Src directly phosphorylates tyrosine residuesof this subunit. GluN2A immunoprecipitates from the DLPFC of theSCZ group showed a decrease in pY419Src, which activates Srckinase, and an increase in pY530Src, which suppresses this kinase(Figures 2a and b; P= 0.042, t(11) = 2.303 for pY419Src andP= 0.002, t(11) = 4.171 for pY530 Src). These results togetherdemonstrate that the activity of postreceptor kinases PKC, Pyk2and Src is attenuated in the DLPFC of SCZ cases.All parameters of NMDAR signaling described above were

derived from membrane fractions that incorporate NMDARcomplexes in synaptic as well as extrasynaptic regions. To testwhether the altered NMDAR signaling occurs in the synapticregion, we examined PSD fractions of the DLPFC. In PSD fractions,the activity of postreceptor kinases is difficult to assess because ofthe presence of detergents, but protein associations resultingfrom the activity of kinases can be measured. We examined PSDfractions for NMDAR association with PLCγ and PKCγ54,56 assurrogate measures of GluN2 activation that was significantlyreduced in the SCZ cases compared with controls (Figure 2e and f;Po0.0104, t(13) = 2.991 for PLCγ and Po0.004, t(13) = 3.434, forPKCγ). This indicates that synaptic NMDAR complexes arehypoactive in the DLPFC of SCZ cases.

Src hypoactivity occurs independently of its upstream kinases inthe NMDAR pathwayAs noted above, Src serves as a molecular hub for various non-NMDAR signaling mechanisms affecting tyrosine phosphorylationof GluN2, including NRG1, erbB4 and RPTPα, SCZ susceptibilitypathways. Based on this, we hypothesized that Src hypoactivity isnot merely a result of hypofunction of upstream kinases within theNMDAR pathway but integrates alterations in signaling pathwaysthat are not canonical components of the NMDAR pathway. Wefirst tested whether Src hypoactivity is present in SCZ cases whenSrc activity is assessed independently of activation of NMDAR. Instudies represented in Figure 2, postreceptor NMDAR signalingwas assayed in response to NMDAR activation, and thus the resultscannot resolve whether Src hypoactivity results from alterationsat the receptor level or from the postreceptor segment. Todistinguish between these two possibilities, we assessed Src


3

© 2014 Macmillan Publishers Limited Molecular Psychiatry (2014), 1 – 10

http://www.med.unc.edu/pgc

activity with and without NMDAR activation in postmortemtissues. First, DLPFC slices were incubated with NMDA+glycineso that NMDAR complexes were activated and Src kinase activitywas measured (Figure 3a). Second, we isolated synaptic mem-branes from the DLPFC tissues, that were then incubated with thepeptide that directly activates Src, EPQ(pY)EEIPIA, instead ofNMDA+glycine34,35 (Figure 3b). NMDA+glycine and EPQ(pY)EEIPIAboth enhanced Src activity. EPQ(pY)EEIPIA increased Src enzymeactivity on average 3.02 ± 0.37-fold in the control group, but onlyby 1.75 ± 0.30-fold on average in the SCZ group (Figure 3b;P= 0.022, t(16) = 2.859). As predicted, we found a correspondingreduction in basal levels of activated Src (pY419) in the patientgroup compared with controls (Figures 3c and d, P= 0.015, t(13) = 3.230), whereas Src protein levels remained unaltered. Ofparticular note, DLPFC tissue from the SCZ patients showed levelsof Pyk2 and its activated form (pY402 Pyk2) that were comparableto those of the control group (Figures 3c and d). These findingsindicate that Src hypoactivity in SCZ cannot be explained fully byupstream abnormalities in the NMDAR→ PKC→ Pyk2→ Src cas-cade. As Src hypoactivity is independent of upstream kinases, wenext determined the binding capacity of Src for its activators.

Binding capacity of Src for activated upstream kinases isdecreased in SCZWe isolated Src from DLPFC tissues and then assessed binding ofthese kinases to biotinylated EPQ(pY)EEIPIA. The SCZ group

showed a significantly lower association of Src with EPQ(pY)EEIPIAat 0.1, 1.0 and 10 nM of the substrate (Figures 4a and b, P= 0.013,t(11) = 2.945 for 0.1 nM; P= 0.024, t(11) = 2.620 for 1 nM; P= 0.049,t(11) = 2.213 for 10 nM). Thus, the binding capacity of the SH-2domain in Src is decreased in SCZ, decreasing Src activityindependently of the NMDAR→ PKC→ Pyk2 cascade. The nextstep was to identify molecular alterations that may affect thebinding capacity of Src and/or its association with its activators.

Convergent molecular alterations reduce Src binding withupstream kinases in SCZWe found a number of molecular alterations in the DLPFC of SCZcases that can affect Src association with its activators. RPTPαincreases the activity of Src via dephosphorylation of SrcY530.30

RPTPα association with Src, assessed in PSD-95 immunoprecipi-tates, was significantly decreased in SCZ cases (Figures 4c and d,P= 0.006, t(11) = 3.390). Given that the N-terminal of PSD-95 caninterfere with Src binding with its activators,34 increased PSD-95 inNMDAR complexes as shown in the DLPFC of SCZ cases (Figures1c and d) may be another molecular underpinning for Srchypoactivity activity. In addition, increased erbB4 signaling, eithervia altered expression57 or cleavage of NRG158 or erbB4 activity, isexpected to decrease Src activity. Indeed, Src -NMDAR complexescaptured by PSD-95 exhibited increased erbB4 in the SCZ group(P=0.0413, t(13)=2.246) (Figure 4c and d).

Figure 1. The abundance of N-methyl-D aspartate receptor (NMDAR) complexes is increased in the postsynaptic density (PSD) of schizophrenia(SCZ) subjects. (a, b) Synaptic membranes obtained from 14 pairs of control (CTL) and SCZ subjects were incubated with 0 to 60 nM of 3H-MK-801 in the presence of 10 μM of glutamate and 1 μM of glycine. (a) Bmax of MK-801 binding is enhanced in the SCZ group by ∼ 30%(P= 0.014). (b) Representative binding kinetics of MK-801 assay for a pair of CTL (-●-) and SCZ (-○-) subjects. (c, d) GluN1–PSD-95 complexesare increased in SCZ. PSD fractions from the prefrontal cortex (PFC) of 14 matched pairs were immunoprecipitated with antibodies for GluN1or PSD-95. GluN1 association with PSD-95 was increased in the GluN1 complexes (black bar; P≤ 0.038) and PSD-95 association with GluN1(black bar; P¼ 0.031) was increased in PSD-95 complexes in the SCZ group. All data are means± s.e.m., Student’s paired t-test, two tailed.*P≤ 0.05, **P≤ 0.01, ***P≤ 0.001.


4


We next further examined whether these dysregulations canindeed lead to Src hypoactivity. ErbB4 suppresses Src activity28,35

and is increased in PSD-95-Src complexes derived from the DLPFCof SCZ subjects (Figures 4c and d). We predicted that activation oferbB4 via NRG1 would lead to a greater reduction in Src activity ofSCZ cases than of controls. Indeed, NRG1-induced suppression ofSrc activation was more pronounced in SCZ cases compared withcontrols (Figure 4e, P= 0.005, t(11) = 3.433). Hence, the decreasedRPTPα in PSD-95-Src complexes as observed in SCZ (Figures 4cand d) should decrease Src activity. Indeed, mice lacking RPTPα,which were found to display several SCZ phenotypes,31 alsoexhibit Src hypoactivity (Figure 4e, P= 0.015, t(8) = 2.3114).Dysbindin-1, encoded by DTNBP1, is another molecule that isdecreased in the DLPFC of SCZ cases.59 Mice lacking dysbindin-1due to deletion mutation in Dtnbp1 (sdy − /− ), which showseveral SCZ phenotypes59 as well as NMDAR hypofunction,60 alsoexhibited Src hypoactivity (Figure 4e, P= 0.0021, t(7) = 4.728).

A PPI-based Src network is associated with SCZWe next asked whether the molecular alterations in PSD-95, erbB4,RPTPα or dysbindin in the DLPFC of SCZ cases could be geneticallydetermined. We conducted a meta-analysis of SCZ GWAS studies,consisting of 15 cohorts that included 13 394 cases and 34 676controls. Src, PKCγ, Pyk2, PSD-95 and RPTPα (Src, PRKCG, PTK2B,DLG4 and PTPRA) as well as other direct interactors of Src showedno evidence of SNP or gene-level associations (Figure 5a). Givenour results suggesting that Src hypoactivity results from alteredprotein association of Src with its interactors, we hypothesizedthat a wider PPI-based Src network may be associated with SCZ.We generated a PPI network centered around Src that consisted of30 genes (Figure 5, Supplementary Table S3 in SupplementaryInformation). Of these, 14 showed gene-level significant associa-tion with SCZ (Figure 5b). To determine whether the observedgene-level associations in the Src network is specific to SCZ, we

Figure 2. Postreceptor N-methyl-D aspartate receptor (NMDAR) signaling is attenuated in the post-mortem dorsolateral prefrontal cortex(DLPFC) of schizophrenia (SCZ) patients. The DLPFC from 14 pairs of SCZ (black bar) and control subjects (white bar) was incubated with 10 μMof NMDA and 1 μM of glycine for 15min and solubilized membrane extracts were immunoprecipitated with antibodies for (a, b) GluN2 (2A/2B)or (c, d) GluN1 and immunoblotted as indicated. (a, b) GluN2A/B tyrosine phosphorylation (pYGluN2) is attenuated in SCZ compared withcontrol (P= 0.001). pY419 Src is decreased and pY530 Src is increased in GluN2 complex in SCZ (P= 0.042 and P= 0.016, respectively). (c, d)Ligand-induced enhancement of GluN1 association with its signaling partners, phospholipase C-γ (PLCγ) and protein kinase C-γ (PKCγ), isreduced in SCZ subjects compared with their controls (P= 0.027 and P= 0.026, respectively). Tyrosine phosphorylation of 402Pyk2 (P= 0.029)and 419Src (P= 0.0435) are also decreased in SCZ subjects in GluN1 complexes. (e, f) Postsynaptic density (PSD) fractions from the DLPFC of 14matched pairs were immunoprecipitated with antibody for GluN1. Association of GluN1 with PLCγ (P= 0.010) and PKCγ (P= 0.004) isdecreased in the SCZ group (black bar) compared with that of the control (white bar). All data are means± s.e.m., Student’s paired t-test, two-tailed. *P≤ 0.05, **P≤ 0.01, ***P≤ 0.001.


5


compared the association results of a SCZ GWAS meta-analysisagainst the results of four other data sets: MDD, BP, ADHD and CD.The enrichment of SNP associations within the Src networkof 30 genes was significant in schizophrenia compared with allother phenotypes (SCZ vs MDD P= 6.921 × 10− 11; SCZ vs ADHDP= 3.429 × 10− 12; SCZ vs BP P= 9.338 × 10− 8; SCZ vs CDP= 7.47 × 10− 7); individual SNP counts for each of the fivediseases are listed in Supplementary Table S4 (See Supplemen-tary Information).

DISCUSSIONThe goal of this study was to delineate the alterations inpostsynaptic NMDAR signaling in SCZ and to investigate itsunderlying mechanisms. Our results point to the postreceptorkinases as a locus of dysregulation in SCZ and suggest that Srchypoactivity may mediate convergent abnormalities of suscept-ibility pathways to decrease tyrosine phosphorylation of GluN2subunits.NMDAR hypofunction can result from decreased number of

NMDAR complexes, which has been extensively tested in previousstudies.14–16 Our approach was to measure the number of NMDARassemblies in the synaptic region by MK-801 binding assays ofsynaptic membranes and by immunoprecipitation of the corescaffold of NMDAR complexes, the GluN1–PSD-95 associations.Our findings of increased MK-801 binding and NMDAR complexesin the PSD suggest that NMDAR hypofunction in SCZ may notresult from decreases in NMDAR complexes in synaptic regions.

At the postreceptor level, however, we found robust reductionsin multiple parameters indicative of hypoactive NMDAR signaling.These include decreases in (1) the activity of PKCγ, Pyk2 and Src,(2) tyrosine phosphorylation of GluN2 subunits and (3) NMDARassociation with its signaling partners, PLCγ and PKCγ. These threeparameters are interconnected since the postreceptor kinasesenhance tyrosine phosphorylation, which in turn recruits PLCγ intoNMDAR complexes (Figure 5c). Consistent reductions in thesethree interconnected parameters therefore point to this specificsegment of the NMDAR pathway, namely the postreceptorkinases, as a locus of dysregulation in SCZ.The reduced postreceptor kinase activities as observed in SCZ

cases could potentially result from presynaptic alterations,1 suchas decreased glutamate release or increased kynurenic acidobserved in SCZ cases.11,12 Our assessment of postreceptor kinasefunction, however, is based on applying relatively high concentra-tions of NMDA and glycine, conditions under which assay resultsmay not capture presynaptic alterations. Thus, dysregulations inpostreceptor kinases as assessed in this study are unlikely toreflect presynaptic dysregulations.It is possible that these postreceptor kinase dysregulations are

secondary to ante-mortem medication effects. In particular,previous studies have shown that chronic clozapine treatmentdecreased the activity and/or expression of PKC and PLCγ.57 Wetherefore examined these signaling molecules in rhesus macaques(age range: 5.9–7.9 years) that were treated with a low dose ofhaloperidol (0.07 mg kg− 1 b.i.d.; n= 8), high dose of haloperidol(n= 8; 2.0 mg kg− 1 b.i.d.), clozapine (n= 9; 2.6 mg kg− 1; b.i.d.) orvehicle (n= 9) for 6 months. GLUN1 association with PKCγ or PLCγ

Figure 3. Src activity is attenuated in schizophrenia (SCZ). (a) Src kinase activity was measured by P32γ-ATP incorporation into Src substrates insynaptic membranes, extracted from the tissues that were incubated with 10 μM N-methyl-D aspartate (NMDA) and 1 μM glycine. Ligand-induced Src activation is decreased in SCZ (black bars; P= 0.003). (b) Synaptic membranes from the tissues that were not incubated withNMDA/glycine were reacted with the EPQ(pY)EEIPIA peptide that binds and activates the SH-2 domain of Src. EPQ(pY)EEIPIA-induced Srcactivity is attenuated (black bar; P= 0.022) in the SCZ group compared with the control (white bar). (c, d) EPQ(pY)EEIPIA peptide-inducedenhancement of pY419 of Src was significantly reduced (P= 0.015) but the peptide-mediated pY402-Pyk2 was unaltered in SCZ subjects (blackbars) compared with controls (white bars) unlike NMDA/glycine-induced condition. All data are means± s.e.m., Student’s paired t-test, twotailed. *P≤ 0.05, **P≤ 0.01.


6


showed trends for increase or no changes in the three treatmentconditions (Supplementary Figure S2 and SupplementaryInformation). Thus, decreases in GluN1's association with PLCγand PKCγ in SCZ are unlikely to be the result of antipsychotictreatment.Among the postreceptor kinases, Src is of particular interest as it

directly phosphorylates tyrosine residues of GluN2 and thereforeserves as a point of convergence for various pathways affectingNMDAR signaling.22,61 Our results offer multiple lines of evidence

supporting the notion that Src hypoactivity is a mere result ofhypoactivity of upstream kinases in the NMDAR pathway.Specifically, Src activity was decreased in SCZ cases even whenmeasured under the condition that NMDARs were not activatedand Src binding capacity was also reduced, independent of theNMDAR→ PKC→ Pyk2 cascade. These findings strongly suggestthat Src hypoactivity may result from changes in molecular eventsthat are outside of the canonical components of the NMDARpathway.

Figure 4. Underlying mechanisms for Src hypoactivity in schizophrenia (SCZ). (a, b) The Src capacity for binding with SH-2 binding domain isdecreased in SCZ. Purified Src (1 μg) from prefrontal cortex (PFC) tissues was incubated with 10 nM of biotinylated EPQYEEIPIA (1) or EPQ(pY)EEIPIA peptide in 0.1 (2), 1 (3) and 10 (4) nM concentrations in healthy control (white bars) and patients with SCZ (black bars) subjects. Bindingof Src to pY peptide was significantly reduced in SCZ subjects at all concentrations (P= 0.013 for 0.1 nM, P= 0.024 at 1 nM and P= 0.049 at10 nM). (c, d) Src hypoactivity is accompanied by alterations in Src associations with receptor-type tyrosine-protein phosphatase-α (RPTPα),postsynaptic density protein-95 (PSD-95) and erbB4 in SCZ. Synaptic membranes of the post-mortem dorsolateral prefrontal cortex (DLPFC)were immunoprecipitated with antibodies for PSD-95. RPTPα is decreased but erbB4 is increased in the PSD-95 complexes derived from SCZ(black bars) subjects (P= 0.006 for RPTPα and P=0.041 for ErbB4), whereas csk or Src are not altered. In addition, the ratios of PSD-95 withrespect to GluN1 were increased in the N-methyl-D aspartate receptor (NMDAR) complexes derived from PSD fractions (Figure 1c and d). (e)Increased NRG1–erbB4 signaling and decreased expression of RPTPα and dysbindin decrease Src activity. NRG1-mediated Src activitymeasured by 32P-γATP incorporation in the presence of NMDA/glycine was decreased in the DLPFC of human post-mortem brains (n= 12;P≤ 0.005). Src activity was also measured in presence of Src-SH2 binding pY peptide, EPQ(pY)EEIPIA in dysbindin (sdy− /− ) and RPTPα(Ptpra− /− ) knockout mice. In both sdy− /− and Ptpra− /− mice, Src activity was attenuated compared with the wild type. In sdy− /− mice,src activity was decreased by ~ 37%, (n= 6; P≤ 0.002), whereas it was decreased in Ptpra− /− mice by ~ 47%, (n= 7; P≤ 0.015). All data aremeans± s.e.m., Student’s paired t-test, two tailed.


7


Figure 5. The Src pathway is associated with schizophrenia (SCZ) not via common variants of their own but by upstream or downstreameffectors connected through protein–protein interactions. (a) Gene-level associations with SCZ for core members of the Src pathway. Gene-level significance was calculated based on meta-analysis of genome-wide association studies (GWAS) that included a total of 13394 cases and34 676 controls. Src, PKC (PRKCG), Pyk2 (PTK2B), PSD-95 (DLG4), CSK and RPTPα (PTPRA) failed to show gene-level significance of the P-value˂0.05. (b) The protein interaction-derived Src subnetwork in the SCZ and Crohn’s disease cohorts.48 Based on protein–protein interactionspreviously reported, a subnetwork of Src was constructed that consists of 30 genes. In this illustration, nodes represent genes in thesubnetwork and connecting lines (or edges) show previously reported direct protein interactions. Derived from brain expression data fromthe Allen Brain Atlas, the information on the degree of coexpression between the proteins is incorporated into the connecting lines. The widthof connecting lines corresponds to the strength of the degree of coexpression between the proteins, with solid lines signifying positive, anddotted lines negative correlations. Of the 30 genes in the subnetwork, 14 showed gene-wise significance in the schizophrenia cohort, whereasonly 3 showed gene-wise significance in the Crohn’s disease cohort. Gene-wise P-value significance is denoted on a green to red color scale.Circled nodes are genes with single-nucleotide polymorphisms (SNPs) that had P-values o1× 10− 3 for genetic association with the illnessand all others are boxed. (c) Summary diagram: Src serves as a nodal point for various molecular alterations in the dorsolateral prefrontalcortex (DLPFC) of SCZ cases, leading to reduced postreceptor N-methyl-D aspartate receptor (NMDAR) activity. (1) NRG1–erbB4 signaling canreduce Src activity,35 and therefore the increased NRG1–erbB4 signaling as shown in SCZ cases28 will decrease Src activity (Figure 4e). (2)Postsynaptic density protein-95 (PSD-95) decreases Src activity34 and thus the increased PSD-95 in NMDAR complexes in SCZ cases (Figure 1c)will reduce Src activity. (3) Receptor-type tyrosine-protein phosphatase-α (RPTPα) increases Src activity and thus decreased RPTPα in SCZ cases(Figure 4d) will reduce Src activity, (4) Sdy− /− mice, mutated for dysbindin, exhibit reduced Src activity. Decreased dysbindin 1, as previouslyseen in SCZ cases,59 can be another underlying mechanism for Src hypoactivity. Src hypoactivity precipitated by these convergentdysregulations (arrows in red) will reduce phosphorylation of its substrate, the GluN2 subunits, that is critical for the channel activity ofNMDARs, as well as its downstream signaling (arrows in blue). Symbols → and ⊥ indicate positive or negative regulation of Src activity by themolecule or the pathway. Downward or upward arrows (blue) indicate alterations of the function of the molecules in the post-mortem DLPFCin SCZ cases.


8


Indeed, the DLPFC of SCZ cases exhibit a number of molecularalterations that can precipitate Src hypoactivity (Figure 5c). First,we found decreased RPTPα in Src-NMDAR complexes in SCZ cases.This will lead to Src hypoactivity, as RPTPα enhances Src activity bydecreasing phosphorylation of pY530 of Src.30 Second, as PSD-95can inhibit Src activity,34 increased PSD-95 in NMDAR–Srccomplexes in SCZ cases will also precipitate Src hypoactivity.Third, NRG1–erbB4 signaling can reduce Src activity,35 and wefound increased erbB4–PSD-95 association in SCZ cases. Finally,decreased dysbindin 1 shown in DLPFC of SCZ cases59 maydecrease Src activity as observed in the case of sdy− /− mice.Based on these findings, we propose that Src serves as a nodalpoint for these molecular alterations in susceptibility path-ways leading to decreased GluN2 tyrosine phosphorylation(Figure 5c).It will be important to ask whether Src hypoactivity arises from

genomic effects. A meta-analysis of 15 cohorts demonstrate thatcommon variants of the core members of the Src pathway, Src,PRKCG, PTK2B, DLG4, CSK, and PTPRA, are not associated with SCZ.This is not unexpected as the expression of these genes in theDLPFC of SCZ cases was comparable to that of control subjects(data not shown). The results from the PPI-based gene associationstudy, however, suggest that a wider PPI-based network centeredon Src consisting of 30 genes showed significant enrichment ofgene-level associations with SCZ compared with other diseases(Figure 5). Together, these genetic findings are consistent with thenotion that the Src pathway as a whole is associated with SCZ, notvia common variants of Src or its direct interactors but via protein–protein associations with a wider network.Src-mediated tyrosine phosphorylation of GluN2 is critical for

the synaptic plasticity (long-term potentiation),62–65 the substratesfor learning and memory. Thus, Src hypoactivity and its down-stream effects observed in the DLPFC can be a basis for deficits inthe neurocognitive functions modulated by the DLPFC, such asthe working memory and executive function. We propose Srchypoactivity as a nodal mechanism for integrating convergentdysregulation in SCZ susceptibility pathways to reduce NMDARactivity (Figure 5c) and suggest that the Src pathway, particularlyprotein interactions therein, be considered as a novel therapeutictarget for SCZ.

CONFLICT OF INTERESTThe authors declare no conflict of interest.

ACKNOWLEDGMENTSWe express most heartfelt gratitude to the donors of post-mortem brain tissues andtheir family members. We also thank Dr Wade Berrettini for reading this manuscriptand making valuable suggestions. This project was supported by the RO1-MH075916and R01MH059852 (C-GH), grants from NARSAD and Stanley Medical ResearchInstitute (to C-GH), RO1MH07462 (to SJS), RO1MH074313 and a grant from theStanley Medical Research Institute (to SEH).

AUTHOR CONTRIBUTIONSAB performed biochemical fractionations. AB and H-YW performed all otherexperiments including immunoprecipitation, receptor and kinase activationassays with help from KEB-W, MLM, AS, HK, JK CE, RR and KT. PS, XC and HHconducted network analysis of GWAS data; SEH and SJS conductedantipsychotic treatment of rhesus monkeys and mice. The post-mortemtissue and brains were collected by SEA, KT and REG, who also ensured theintegrity of post-mortem brain tissues. PS, XC and HH conducted analyses ofgenetic data. AB and C-GH performed all other data analyses. C-GH conceivedof and directed this study. The manuscript was written by AB, KBW, KT andC-GH and edited by all authors.

REFERENCES1 Coyle JT. Glutamate and schizophrenia: beyond the dopamine hypothesis. Cell

Mol Neurobiol 2006; 26: 365–384.2 Javitt DC, Schoepp D, Kalivas PW, Volkow ND, Zarate C, Merchant K et al. Trans-

lating glutamate: from pathophysiology to treatment. Sci Transl Med 2011; 3:102mr102.

3 Moghaddam B, Javitt D. From revolution to evolution: the glutamate hypothesisof schizophrenia and its implication for treatment. Neuropsychopharmacology2012; 37: 4–15.

4 Farber NB. The NMDA receptor hypofunction model of psychosis. Ann NY Acad Sci2003; 1003: 119–130.

5 Javitt DC, Zukin SR. Recent advances in the phencyclidine model of schizophrenia.Am J Psychiatry 1991; 148: 1301–1308.

6 Krystal JH, Karper LP, Seibyl JP, Freeman GK, Delaney R, Bremner JD et al. Sub-anesthetic effects of the noncompetitive NMDA antagonist, ketamine, in humans.Psychotomimetic, perceptual, cognitive, and neuroendocrine responses. Arch GenPsychiatry 1994; 51: 199–214.

7 Moghaddam B, Krystal JH. Capturing the angel in ‘angel dust’: twenty years oftranslational neuroscience studies of NMDA receptor antagonists in animalsand humans. Schizophr Bull 2012; 38: 942–949.

8 Mohn AR, Gainetdinov RR, Caron MG, Koller BH. Mice with reduced NMDAreceptor expression display behaviors related to schizophrenia. Cell 1999; 98:427–436.

9 Carty NC, Xu J, Kurup P, Brouillette J, Goebel-Goody SM, Austin DR et al. Thetyrosine phosphatase STEP: implications in schizophrenia and the molecularmechanism underlying antipsychotic medications. Transl Psychiatry 2012; 2: e137.

10 Noetzel MJ, Jones CK, Conn PJ. Emerging approaches for treatment of schizo-phrenia: modulation of glutamatergic signaling. Discov Med 2012; 14: 335–343.

11 Erhardt S, Schwieler L, Nilsson L, Linderholm K, Engberg G. The kynurenic acidhypothesis of schizophrenia. Physiol Behav 2007; 92: 203–209.

12 Schwarcz R, Rassoulpour A, Wu HQ, Medoff D, Tamminga CA, Roberts RC.Increased cortical kynurenate content in schizophrenia. Biol Psychiatry 2001; 50:521–530.

13 Poels EM, Kegeles LS, Kantrowitz JT, Slifstein M, Javitt DC, Lieberman JA et al.Imaging glutamate in schizophrenia: review of findings and implications for drugdiscovery. Mol Psychiatry 2014; 19: 20–29.

14 Akbarian S, Sucher NJ, Bradley D, Tafazzoli A, Trinh D, Hetrick WP et al. Selectivealterations in gene expression for NMDA receptor subunits in prefrontal cortex ofschizophrenics. J Neurosci 1996; 16: 19–30.

15 Kristiansen LV, Patel SA, Haroutunian V, Meador-Woodruff JH. Expression of theNR2B-NMDA receptor subunit and its Tbr-1/CINAP regulatory proteins in post-mortem brain suggest altered receptor processing in schizophrenia. Synapse2010; 64: 495–502.

16 Weickert CS, Fung SJ, Catts VS, Schofield PR, Allen KM, Moore LT et al. Molecularevidence of N-methyl-D-aspartate receptor hypofunction in schizophrenia. MolPsychiatry 2013; 18: 1185–1192.

17 McBain CJ, Mayer ML. N-methyl-D-aspartic acid receptor structure and function.Physiol Rev 1994; 74: 723–760.

18 Salter MW. Cellular neuroplasticity mechanisms mediating pain persistence. JOrofac Pain 2004; 18: 318–324.

19 Coultrap SJ, Bayer KU. CaMKII regulation in information processing and storage.Trends Neurosci 2012; 35: 607–618.

20 Hsin H, Kim MJ, Wang CF, Sheng M. Proline-rich tyrosine kinase 2 regulateshippocampal long-term depression. J Neurosci 2010; 30: 11983–11993.

21 MacDonald JF, Kotecha SA, Lu WY, Jackson MF. Convergence of PKC-dependentkinase signal cascades on NMDA receptors. Curr Drug Targets 2001; 2: 299–312.

22 Salter MW, Kalia LV. Src kinases: a hub for NMDA receptor regulation. Nat RevNeurosci 2004; 5: 317–328.

23 Husi H, Ward MA, Choudhary JS, Blackstock WP, Grant SG. Proteomic analysis ofNMDA receptor-adhesion protein signaling complexes. Nat Neurosci 2000; 3:661–669.

24 Kennedy MB. The postsynaptic density at glutamatergic synapses. Trends Neurosci1997; 20: 264–268.

25 Tang TT, Yang F, Chen BS, Lu Y, Ji Y, Roche KW et al. Dysbindin regulates hip-pocampal LTP by controlling NMDA receptor surface expression. Proc Natl AcadSci USA 2009; 106: 21395–21400.

26 Ma TM, Abazyan S, Abazyan B, Nomura J, Yang C, Seshadri S et al. Pathogenicdisruption of DISC1-serine racemase binding elicits schizophrenia-like behaviorvia D-serine depletion. Mol Psychiatry 2013; 18: 557–567.

27 Law AJ, Kleinman JE, Weinberger DR, Weickert CS. Disease-associated intronicvariants in the ErbB4 gene are related to altered ErbB4 splice-variant expression inthe brain in schizophrenia. Hum Mol Genet 2007; 16: 129–141.

28 Hahn CG, Wang HY, Cho DS, Talbot K, Gur RE, Berrettini WH et al. Altered neur-egulin 1-erbB4 signaling contributes to NMDA receptor hypofunction in schizo-phrenia. Nat Med 2006; 12: 824–828.


9


29 Agim ZS, Esendal M, Briollais L, Uyan O, Meschian M, Martinez LA et al. Discovery,validation and characterization of Erbb4 and Nrg1 haplotypes using data fromthree genome-wide association studies of schizophrenia. PLoS One 2013; 8:e53042.

30 Bjorge JD, Pang A, Fujita DJ. Identification of protein-tyrosine phosphatase 1B asthe major tyrosine phosphatase activity capable of dephosphorylating and acti-vating c-Src in several human breast cancer cell lines. J Biol Chem 2000; 275:41439–41446.

31 Takahashi N, Nielsen KS, Aleksic B, Petersen S, Ikeda M, Kushima I et al. Loss offunction studies in mice and genetic association link receptor protein tyrosinephosphatase alpha to schizophrenia. Biol Psychiatry 2011; 70: 626–635.

32 Fromer M, Pocklington AJ, Kavanagh DH, Williams HJ, Dwyer S, Gormley P et al. Denovo mutations in schizophrenia implicate synaptic networks. Nature 2014; 506:179–184.

33 Kirov G, Pocklington AJ, Holmans P, Ivanov D, Ikeda M, Ruderfer D et al. De novoCNV analysis implicates specific abnormalities of postsynaptic signalling com-plexes in the pathogenesis of schizophrenia. Mol Psychiatry 2012; 17: 142–153.

34 Kalia LV, Pitcher GM, Pelkey KA, Salter MW. PSD-95 is a negative regulator of thetyrosine kinase Src in the NMDA receptor complex. EMBO J 2006; 25: 4971–4982.

35 Pitcher GM, Kalia LV, Ng D, Goodfellow NM, Yee KT, Lambe EK et al. Schizophreniasusceptibility pathway neuregulin 1-ErbB4 suppresses Src upregulation of NMDAreceptors. Nat Med 2011; 17: 470–478.

36 Hahn CG, Banerjee A, Macdonald ML, Cho DS, Kamins J, Nie Z et al. The post-synaptic density of human postmortem brain tissues: an experimental studyparadigm for neuropsychiatric illnesses. PLoS One 2009; 4: e5251.

37 Glausier JR, Lewis DA. Dendritic spine pathology in schizophrenia. Neuroscience2012; 251: 90–107.

38 Lewis DA. Neuroplasticity of excitatory and inhibitory cortical circuits in schizo-phrenia. Dialogues Clin Neurosci 2009; 11: 269–280.

39 Goghari VM, Sponheim SR, MacDonald AW 3rd. The functional neuroanatomy ofsymptom dimensions in schizophrenia: a qualitative and quantitative review of apersistent question. Neurosci Biobehav Rev 2010; 34: 468–486.

40 Lesh TA, Niendam TA, Minzenberg MJ, Carter CS. Cognitive control deficits inschizophrenia: mechanisms and meaning. Neuropsychopharmacology 2011; 36:316–338.

41 Cox MM, Tucker AM, Tang J, Talbot K, Richer DC, Yeh L et al. Neurobehavioralabnormalities in the dysbindin-1 mutant, sandy, on a C57BL/6J genetic back-ground. Genes Brain Behav 2009; 8: 390–397.

42 Steele JE, Bowen DM, Francis PT, Green AR, Cross AJ. Spermidine enhancement of[3H]MK-801 binding to the NMDA receptor complex in human cortical mem-branes. Eur J Pharmacol 1990; 189: 195–200.

43 Wang HY, Stucky A, Liu J, Shen C, Trocme-Thibierge C, Morain P. Dissociatingbeta-amyloid from alpha 7 nicotinic acetylcholine receptor by a novel therapeuticagent, S 24795, normalizes alpha 7 nicotinic acetylcholine and NMDA receptorfunction in Alzheimer’s disease brain. J Neurosci 2009; 29: 10961–10973.

44 Sleiman P, Wang D, Glessner J, Hadley D, Gur RE, Cohen N et al. GWAS metaanalysis identifies TSNARE1 as a novel Schizophrenia/Bipolar susceptibility locus.Sci Rep 2013; 3: 3075.

45 Major Depressive Disorder Working Group of the Psychiatric GWAS Consortium,Ripke S, Wray NR, Lewis CM, Hamilton SP, Weissman MM et al. A mega-analysis ofgenome-wide association studies for major depressive disorder. Mol Psychiatry2013; 18: 497–511.

46 Psychiatric GWAS Consortium Bipolar Disorder Working Group. Large-scale gen-ome-wide association analysis of bipolar disorder identifies a new susceptibilitylocus near ODZ4. Nat Genet 2011; 43: 977–983.

47 Neale BM, Medland SE, Ripke S, Asherson P, Franke B, Lesch KP et al. Meta-analysisof genome-wide association studies of attention-deficit/hyperactivity disorder. JAm Acad Child Adolesc Psychiatry 2010; 49: 884–897.

48 Franke A, McGovern DP, Barrett JC, Wang K, Radford-Smith GL, Ahmad T et al.Genome-wide meta-analysis increases to 71 the number of confirmed Crohn’sdisease susceptibility loci. Nat Genet 2010; 42: 1118–1125.

49 Liu JZ, McRae AF, Nyholt DR, Medland SE, Wray NR, Brown KM et al. A versatilegene-based test for genome-wide association studies. Am J Hum Genet 2010; 87:139–145.

50 Razick S, Magklaras G, Donaldson IM. iRefIndex: a consolidated protein interactiondatabase with provenance. BMC Bioinformatics 2008; 9: 405.

51 MacDonald ML, Ciccimaro E, Prakash A, Banerjee A, Seeholzer SH, Blair IA et al.Biochemical fractionation and stable isotope dilution liquid chromatography-mass spectrometry for targeted and microdomain-specific protein quantificationin human postmortem brain tissue. Mol Cell Proteomics 2012; 11: 1670–1681.

52 Ebert B, Andersen S, Hjeds H, Dickenson AH. Dextropropoxyphene acts as anoncompetitive N-methyl-D-aspartate antagonist. J Pain Symptom Manage 1998;15: 269–274.

53 Kornhuber J, Mack-Burkhardt F, Kornhuber ME, Riederer P. [3H]MK-801 bindingsites in post-mortem human frontal cortex. Eur J Pharmacol 1989; 162: 483–490.

54 Codazzi F, Di Cesare A, Chiulli N, Albanese A, Meyer T, Zacchetti D et al. Synergisticcontrol of protein kinase Cgamma activity by ionotropic and metabotropic glu-tamate receptor inputs in hippocampal neurons. J Neurosci 2006; 26: 3404–3411.

55 Takagi N, Besshoh S, Morita H, Terao M, Takeo S, Tanonaka K. Metabotropicglutamate mGlu5 receptor-mediated serine phosphorylation of NMDA receptorsubunit NR1 in hippocampal CA1 region after transient global ischemia in rats. EurJ Pharmacol 2010; 644: 96–100.

56 Gurd JW, Bissoon N. The N-methyl-D-aspartate receptor subunits NR2A and NR2Bbind to the SH2 domains of phospholipase C-gamma. J Neurochem 1997; 69:623–630.

57 Dwivedi Y, Pandey GN. Effects of treatment with haloperidol, chlorpromazine, andclozapine on protein kinase C (PKC) and phosphoinositide-specific phospholipaseC (PI-PLC) activity and on mRNA and protein expression of PKC and PLC isozymesin rat brain. J Pharmacol Exp Ther 1999; 291: 688–704.

58 Marballi K, Cruz D, Thompson P, Walss-Bass C. Differential neuregulin 1 cleavagein the prefrontal cortex and hippocampus in schizophrenia and bipolar disorder:preliminary findings. PLoS One 2012; 7: e36431.

59 Tang J, LeGros RP, Louneva N, Yeh L, Cohen JW, Hahn CG et al. Dysbindin-1 indorsolateral prefrontal cortex of schizophrenia cases is reduced in an isoform-specific manner unrelated to dysbindin-1 mRNA expression. Hum Mol Genet 2009;18: 3851–3863.

60 Talbot K, Louneva N, Cohen JW, Kazi H, Blake DJ, Arnold SE. Synaptic dysbindin-1reductions in schizophrenia occur in an isoform-specific manner indicating theirsubsynaptic location. PLoS One 2011; 6: e16886.

61 Ohnishi H, Murata Y, Okazawa H, Matozaki T. Src family kinases: modulators ofneurotransmitter receptor function and behavior. Trends Neurosci 2011; 34:629–637.

62 Citri A, Malenka RC. Synaptic plasticity: multiple forms, functions, and mechan-isms. Neuropsychopharmacology 2008; 33: 18–41.

63 Hunt DL, Castillo PE. Synaptic plasticity of NMDA receptors: mechanisms andfunctional implications. Curr Opin Neurobiol 2012; 22: 496–508.

64 Li HB, Jackson MF, Yang K, Trepanier C, Salter MW, Orser BA et al. Plasticity ofsynaptic GluN receptors is required for the Src-dependent induction of long-termpotentiation at CA3-CA1 synapses. Hippocampus 2011; 21: 1053–1061.

65 Perez-Otano I, Ehlers MD. Homeostatic plasticity and NMDA receptor trafficking.Trends Neurosci 2005; 28: 229–238.

Supplementary Information accompanies the paper on the Molecular Psychiatry website (http://www.nature.com/mp)


10


Documents

Src kinase as a mediator of convergent molecular abnormalities leading to NMDAR hypoactivity in schizophrenia