NeuroImage 56 (2011) 2283–2291

Contents lists available at ScienceDirect

NeuroImage

j ourna l homepage: www.e lsev ie r.com/ locate /yn img

Differential effects of DAAO on regional activation and functional connectivity inschizophrenia, bipolar disorder and controls

Sergio Alessandro Papagni a,b,⁎, Andrea Mechelli a, Diana P. Prata a,c, Joseph Kambeitz a, Cynthia H.Y. Fu a,Marco Picchioni a,d, Muriel Walshe a, Timothea Toulopoulou a, Elvira Bramon a, Robin M. Murray a,David A. Collier c, Antonello Bellomo b, Philip McGuire a

a Department of Psychosis Studies, Institute of Psychiatry, King's College London, De Crespigny Park, SE5 8AF, UKb Section of Psychiatry and Clinical Psychology, Department of Medical Sciences, University of Foggia, Italyc Social, Genetic and Developmental Psychiatry Centre, Institute of Psychiatry, King's College London, De Crespigny Park, SE5 8AF, UKd St Andrews Academic Centre, Kings College London, Institute of Psychiatry, Northampton, NN1, UK

⁎ Corresponding author at: PO BOX 67, Department oPsychiatry, King's College London, De Crespigny Park, Lo

E-mail address: papagni.sergio@libero.it (S.A. Papag

1053-8119/$ – see front matter © 2011 Elsevier Inc. Aldoi:10.1016/j.neuroimage.2011.03.037

a b s t r a c t

a r t i c l e i n f o

Article history:Received 24 November 2010Revised 10 February 2011Accepted 14 March 2011Available online 21 March 2011

Keywords:Imaging geneticsDAAOfMRIVerbal fluencySchizophreniaBipolar disorder

Recent studies have identified DAAO as a probable susceptibility gene for schizophrenia and bipolar disorder.However, little is known about how this gene affects brain function to increase vulnerability to thesedisorders. We examined the impact of DAAO genotype (rs3918346) on brain function in patients withschizophrenia, patients with bipolar I disorder and healthy controls. We tested the hypothesis that a variationin DAAO genotype would be associated with altered prefrontal function and altered functional connectivity inschizophrenia and bipolar disorder. We used functional magnetic resonance imaging to measure brainresponses during a verbal fluency task in a total of 121 subjects comprising 40 patients with schizophrenia, 33patients with bipolar disorder and 48 healthy volunteers. We then used statistical parametric mapping (SPM)and psycho-physiological interaction (PPI) analyses to estimate the main effects of diagnostic group, the maineffect of genotype, and their interaction on brain activation and on functional connectivity. Inferences weremade at pb0.05, after correction for multiple comparisons across the whole brain. In the schizophrenia grouprelative to the control group, patients with one or two copies of the T allele showed lower deactivation in theleft precuneus and greater activation in the right posterior cingulate gyrus than patients with two copies ofthe C allele. This diagnosis×genotype interaction was associated with differences in the functionalconnectivity of these two regions with other cortical and subcortical areas. In contrast, there were nosignificant effects of diagnosis or of genotype in comparisons involving bipolar patients. Our results suggestthat genetic variation in DAAO has a significant impact on both regional activation and functionalconnectivity, and provide evidence for a diagnosis-dependent pattern of gene action.

f Psychosis Studies, Institute ofndon, SE5 8AF, UK.

ni).

l rights reserved.

© 2011 Elsevier Inc. All rights reserved.

Introduction

Schizophrenia and bipolar disorder are severe psychiatric disorderswith a significant genetic component (Owen et al., 2007; Craddock et al.,2006; Berrettini, 2003; Cardno et al., 2002). The identification ofsusceptibility genes has been problematic because of phenotypiccomplexity and a mode of transmission compatible with multiple loci.However, association studies have provided converging evidence infavour of a number of positional genes mapped to both schizophreniaand bipolar disorder linkage regions (Park et al., 2004; Lewis et al., 2003;Segurado et al., 2003); one such candidate is the D-Amino Acid Oxidase(DAAO) gene (D. Prata et al., 2008; D.P. Prata et al., 2008).

The DAAO gene is located at 12q23, a region linked to schizophreniaand bipolar disorder (Sklar, 2002). It was originally associated withschizophrenia by Chumakov et al. (2002) using Canadian samples.Further evidence of association comes from several replication studies(Schumacher et al., 2004; Liu et al., 2004;Kapoor et al., 2006; Corvin et al.,2007; Madeira et al., 2008; Bass et al., 2009). Recently, two geneticstudies (D. Prata et al., 2008;D.P. Prata et al., 2008; Fallin et al., 2005)havealso implicated the DAAO gene in bipolar disorder. First, Fallin et al.reported a significant association of this gene with bipolar I disorder butnot with schizophrenia (Fallin et al., 2005). Prata et al. subsequentlyfound significant association of a two-SNP risk haplotype for DAAOwithbipolar disorder (D. Prata et al., 2008; D.P. Prata et al., 2008). Takencollectively, these studies suggest that DAAO may confer biologicalsusceptibility to psychosis across the traditional Kraepelian dichotomy ofschizophrenia and bipolar disorder (Craddock et al., 2007).

The DAAO gene encodes the DAAO enzyme that oxidises D-serine, aco-agonist for the NMDA glutamate receptor (Boks et al., 2007). The

2284 S.A. Papagni et al. / NeuroImage 56 (2011) 2283–2291

NMDAreceptor is known toplay an important role in synaptic plasticity,neurodevelopment and excitotoxicity (Tuominen et al., 2005). It ischaracterised by twodistinct sub-units known asNR1 andNR2; NR1 is abinding site for co-agonists glycine and D-serinewhileNR2 is the agonistbinding site for glutamate. NR1 must be occupied for glutamate to beable toopen the channel; this requires theavailability of glycineand, to agreater extent, D-serine. Thus, selective degradation of D-serine by theDAAO enzyme results in reduced NMDA neurotransmission. There isgrowing evidence that NMDA neurotransmission is altered in psychosis(Verrall et al., 2010; Stoneet al., 2009; Boks et al., 2007;Hashimoto et al.,2003; Coyle et al., 2003; Tanii et al., 1994; Javitt and Zukin, 1991). Forexample, D-serine levels are decreased in the cerebrospinal fluid andserum of patients with schizophrenia (Hashimoto et al., 2005) andadministration of D-serine may reduce negative, positive and cognitivesymptoms in schizophrenia (Heresco-Levy, 2005). It has therefore beenproposed that increased activity of the DAAO enzyme may lead toincreased degradation of D-serine in schizophrenia, resulting in relativeNMDA hypofunction (Boks et al., 2007). Dysfunction of NMDAneurotransmission has also been suggested in bipolar disorder, giventhe glutamatergic action of mood stabilisers (Li et al., 2002) andincreased glutamine/glutamate concentration in the dorsolateral pre-frontal cortex and cingulate gyrus in bipolar disorder (Michael et al.,2003; Dager et al., 2004).

While the effects of DAAO at molecular and cellular levels areemerging, little is known about this gene's effects on the brain atmacroscopic or systems levels. There have been only two previousstudies investigating the association between DAAO genotype andcognitive function (Stefanis et al., 2007; Goldberg et al., 2006). Stefaniset al. (2007) found an effect of DAAO haplotype on spatial workingmemory in a population of 2243malemilitary conscripts. In contrast, ina sample of over 600 healthy controls, patients with schizophrenia andtheir nonpsychotic siblings, Goldberg et al. (2006) found no genotype orgenotype×diagnosis interactions for any of the three DAAO SNPs used(MDAAO 5–7, using the nomenclature of Chumakov). To date, therehave been no neuroimaging studies that investigated the impact ofDAAO genotype on brain function in healthy controls or psychoticpatients. Thus, little is known about how this gene affects brain functionat a macroscopic or system level. Furthermore, it is unclear whether ornot the impact of DAAO on brain function is expressed consistentlyacross different diagnostic categories, given that recent studies on othercandidate genes for schizophrenia and bipolar disorder have providedevidence for a disease-dependent pattern of gene action (Mechelli et al.,2008; D. Prata et al., 2008; D.P. Prata et al., 2008).

The aim of the present studywas therefore to investigate the impactof genetic variation in DAAO (the rs3918346 polymorphism) on brainfunction in 3 diagnostic groups: healthy volunteers, patients withschizophrenia and patients with bipolar I disorder. We used functionalmagnetic resonance imaging (fMRI) to measure brain responses duringa verbal fluency task that required subjects to generate and articulatewords in response to letter cues. This task was chosen for two mainreasons: first, it engages a distributed network of fronto-temporalcortical and sub-cortical brain regions that have been implicated inschizophrenia and bipolar disorder (Fu et al., 2005; Fu et al., 2002; Curtiset al., 2001); second, it taps into executive cognitive processes that arecompromised in both schizophrenia and bipolar disorder (Daban et al.,2006; Kravariti et al., 2005; Krabbendam et al., 2005; Curtis et al., 2001).A “clustered” image acquisition sequence was used in order to recordovert vocal responses in the absence of scanner noise (Fu et al., 2005,2002) and then model correct and incorrect trials separately in thestatistical analysis.

Our first hypothesis was that DAAO genotype would have ameasurable impact on activation during the verbal fluency task. Whilethere is evidence for DAAO's action in the prefrontal cortex (Verrall et al.,2010; Hashimoto et al., 2009; Verrall et al., 2007), this gene is thought tobe expressed throughout the cortex; we therefore tested for its effects inthe whole brain using an appropriate correction for multiple compari-

sons. Our second hypothesis was that the impact of DAAO on theprefrontal function would vary across the three diagnostic groups asrevealed by significant genotype×diagnosis interactions.

Materials and methods

Some of the functional neuroimaging data examined in the presentinvestigation have been included in previous studies which investigatedbrain dysfunction in psychosis (Fu et al., 2005) or the impact of othercandidate genes (Mechelli et al., 2008;D. Prata et al., 2008;D.P. Prata et al.,2008; Prata et al., 2009a, 2009b).

Subjects

A total of 121 subjects were investigated, including 48 healthyvolunteers, 40 patients with schizophrenia and 33 patients withbipolar I disorder. All participants were native English speakers andgave written informed consent in accordance with protocolsapproved by the Local and Multicentre Research Ethics Committee(LREC, MREC). Healthy volunteers were recruited through localadvertisement and had no family history of psychiatric illness asassessed using the FIGS (Family Interview for Genetic Studies).Patients with schizophrenia and bipolar disorder were recruitedthrough the South London and Maudsley NHS Trust or other NHSMental Health Trusts and met relevant DSM-IV criteria, determinedafter a detailed clinical interview and systematic review of themedical records. Full-scale IQ was assessed using the WAIS-III(Wechsler Adult Intelligence Scale-III; Wechsler, 1997), the WAIS-R(Wechsler Adult Intelligence Scale-R; Wechsler, 1981), the WASI-FSIQ-4 (Wechsler Abbreviated Scale of Intelligence; Wechsler, 1999)or the Quick Test (Ammons and Ammons, 1962). The WAIS-IIIcorrelates highly both with the WAIS-R (93.9%; Wechsler, 1997) andwith the WASI-FSIQ-4 (92%; Wechsler, 1999). The Quick test has alsobeen shown to yield comparable results to WAIS (Quick Test, 78±7and WAIS, 83±6 in schizophrenia; Frith et al., 1991). The proportionof subjects assessed with each method was matched betweengenotype groups.

All participants were genotyped for the DAAO SNP rs3918346 (seeMaterials and methods below) with C and T being the two alleles.Demographic data (including age, full-scale IQ, years of education,handedness, gender and ethnicity) and clinical data (includingpositive and negative symptoms, duration of illness and anti-psychotic medication) for each diagnostic group and for eachgenotypic group are reported in Table 1 and described in supple-mentary materials (S1).

Genotyping

Genomic DNA was extracted from blood or cheek swabs by SGDP(Social Genetic and Developmental Psychiatry) laboratory techniciansfollowing standard methodology (Freeman et al., 2003). Genotypingof the DAAO rs3918346 single nucleotide polymorphism (SNP), wasperformed blind to status under contract by KBioscience (Herts, UK;http://www.kbioscience.co.uk/) using a competitive allele specificPCR system (CASP). This single nucleotide polymorphism (SNP) waschosen because it was the single marker most frequently associatedwith schizophrenia and bipolar disorder in previous studies eitherindividually or in haplotype form (D. Prata et al., 2008; D.P. Prata et al.,2008; Corvin et al., 2007; Wood et al., 2007; Schumacher et al., 2004).The genotype counts retrieved were: 29 CC, 16 CT and 3 TT in thecontrol group; 18 CC, 18 CT and 4 TT in the schizophrenia group; and19 CC, 12 CT and 2 TT in the bipolar disorder group. The genotypingresults of our sample were in HardyWeinberg equilibrium (X2=0.34;p=0.6291).

Table 1Demographics and behavioural performance in control (C), schizophrenic (S) and bipolar (BD) samples with CC genotype and CT or TT genotype respectively. Degrees of freedom(df), F or χ2 test value and p-value are reported for comparisons between diagnostic groups that reached significance at pb0.05. The Scale for the Assessment of Positive Symptoms(SAPS) and the Scale for the Assessment of Negative Symptoms (SANS) measure symptoms experienced within the month prior to the interview. The asterisk (*) indicates asignificant difference between the CC genotype and the CT and TT genotypes combinedwithin schizophrenic sample for CPZ rate (F=8.229; df=1; p=0.007) andwithin the bipolarsample for scores on the Beck Depression Inventory (BDI; F=5.371; df=1; p=0.027) and on Altman Self-Rated Mania Scale (ASRM; F=4.275; df=1; p=0.47).CPZ=chlorpromazine; N.=number; sub.=subjects; n.s.=not significant.

Diagnosis C S BD Group comparisons

Genotype CC CT/TT CC CT/TT CC CT/TT C vs S C vs BD S vs BD

N. of subjects 29 19 18 22 19 14Age: mean (sd) 34.9 (10.5) 33.5 (10.9) 34.5 (12.6) 35.8 (10.8) 40.1 (12.2) 37.0 (11.9) n.s. n.s. n.s.IQ: mean (sd) 115.2

(12.7)115.8(8.9)

98.4(15.0)

97.6(15.6)

104.1(15.7)

110.6(14.1)

F=38.462df=1pb0.001

F=8.598df=1p=0.004

n.s.

Gender: male/female 15/14 9/10 15/3 18/4 8/11 5/9 χ2=10.001df=1p=0.001

n.s. χ2=14.416df=1pb0.001

Ethnicity: cauc/black/carib/black-afric/mixed

28/0/1/0/0 18/0/0/0/1 16/1/0/0/1 17/4/1/0/0 19/0/0/0/0 12/1/1/0/0 n.s. n.s. n.s.

Handedness: right/left/mixed 27/2/0 18/1/0 15/3/0 21/1/0 16/3/0 14/0/0 n.s n.s. n.s.Years of education: mean (sd) 14.5 (2.3) 15.9 (3.1) 14.2 (2.3) 14.0 (2.2) 14.6 (2.9) 15.9 (3.3) n.s. n.s. n.s.Years of illness: mean (sd) 14.56 (11.13) 10.72 (7.98) 16.16 (10.08) 12.71 (9.91) n.s. n.s.CPZ rate: mean (sd) 783.3

(410.9)*397.7(432.5)*

136.8 (260.9)N. of sub.=7

50.0 (116.0)N. of sub.=4

Years of antipsychotic medication: mean (sd) 14.1 (11.1) 10.7 (8.0) 13.6 (10.7) 9.1 (5.5)SAPS: mean (sd) 9.61 (8.87) 5.86 (4.90)SANS: mean (sd) 9.72 (5.37) 7.05 (3.86)BDI: mean (sd) 6.21 (5.58)* 12.79 (10.56)*ASRM [mean (sd)] 2.58 (2.46)* 4.57 (3.08)*N. of errors: mean (sd) 10.62 (7.64) 7.52 (4.98) 12.61 (5.30) 15.63 (7.75) 11.05 (8.59) 11.79 (11.08) F=11.108 n.s. n.s.N. of easy errors: mean (sd) 4.28 (3.56) 1.78 (2.30) 4.67 (3.09) 6.59 (4.76) 3.79 (3.24) 5.00 (6.16) df=1N. of hard errors: mean (sd) 6.34 (4.54) 5.73 (3.28) 7.94 (3.32) 9.05 (4.59) 7.26 (6.17) 6.78 (6.20) p=0.001

2285S.A. Papagni et al. / NeuroImage 56 (2011) 2283–2291

Verbal fluency task and image acquisition

The task and image acquisition was performed as described before(Fu et al., 2002), see S2 for further details. In brief, during a“generation” condition, subjects were visually presented with a seriesof letters and required to overtly articulate a word beginning with thepresented letter. This condition was contrasted with a “repetition”condition, in which subjects were presented with the word “rest” andwere required to say rest out loud. Task difficulty was manipulated bypresenting separate sets of “easy” and “hard” letters. Verbal responseswere recorded allowing the identification of “incorrect” trials, inwhich the subject did not generate any response or generatedrepetitions, derivatives, or grammatical variations of a previous word.

fMRI data analysis

The analysis of the fMRI data was performed using StatisticalParametric Mapping (SPM5) software (Friston, 2003), running underMatlab 6.5. A 3×2×2 full-factorial ANOVA was used in whichdiagnostic category and genotype were modelled as between-subjectfactors and task difficulty was modelled as a within subject factor. SeeS3 for details. Statistical inferences were made in SPM5 using astatistical threshold of pb0.05 after family-wise error (FWE)correction for multiple comparisons across the whole brain and anextent threshold of 5 voxels.

Functional connectivity analysis

To establish putative alterations in functional integration that wereunderlying genotype-related differences in regional activation, we usedthe psycho-physiological interaction (PPI)method (Friston et al., 1997).PPI is a well established analytical technique based on multipleregression that allows the investigation of the functional couplingbetween regions (“functional connectivity”) as a functionof oneormoreexperimental manipulations of interest. In brief, for each region ofinterest (defined as spheres with a 6 mm radius) the time series was

extracted from each participant using the “VOI extraction” tool in SPMsoftware; the choice of regions of interest were based on voxels whichshowed a significant interaction between genotype and diagnosticcategory in the standard analysis of regional responses. For each regionof interest, the time series from each subject wasmultiplied by a vectorwhich contrasted verbal fluency against baseline. This resulted in a “PPIvector” which encoded the interaction between the activation of aregion of interest and the experimental context (verbal fluency vsbaseline). A correlation analysis was then performed between this PPIvector and the rest of the brain in a subject specific fashion; finally, thesubject-specific contrast images were entered into an ANOVA to makeinference at the second level. This procedure allowed the identificationof alterations in functional integration underlying the interactionsbetween genotype and diagnostic category detected in the standardanalysis of regional responses. Statistical inferencesweremade in SPM5using a statistical threshold of pb0.05 after family-wise error (FWE)correction for multiple comparisons across the whole brain and anextent threshold of 5 voxels.

Results

Performance

The effects of diagnostic group, genotype, task difficulty and theirinteraction on the accuracy of verbal responses during scanning(Table 1) were tested using a repeated measures ANOVA in SPSS(Statistical Package for Social Sciences— version 15.0). The number oferrors significantly differed as a function of diagnostic group(F=4.621; df=2; p=0.012). Post hoc t-tests revealed that patientswith schizophrenia made significantly more errors than healthyvolunteers (F=11.108; df=1; p=0.001). Patients with bipolardisorder made an intermediate number of errors and did notsignificantly differ (pN0.05) from either healthy volunteers orpatients with schizophrenia. The number of errors did not differ(pN0.05) as a function of genotype, irrespective of whether the 3diagnostic groups were considered separately or in combination. The

2286 S.A. Papagni et al. / NeuroImage 56 (2011) 2283–2291

main effect of task difficulty was significant, with the “hard” conditionassociated with a greater number of errors than the “easy” condition(F=54.174; df=1 pb0.001). Finally, there were no 2-way or 3-wayinteractions between diagnostic group, genotype and task difficulty(pN0.05).

Functional MRI

Effects of generation vs. baselineA distributed network of regions including the left inferior frontal

gyrus, left insula and left superior temporal gyrusweremore activatedduring generation than baseline (Table 2). Conversely, the leftprecuneus, bilateral angular gyrus, bilateral anterior cingulate, rightmedial frontal gyrus, bilateral insula and right postcentral gyrusexpressed increased activation during baseline relative to generation(Table 2).

Main effect of diagnostic groupControls expressed greater activation (i.e. generationNbaseline)

relative to patients with schizophrenia in the left posterior cingulategyrus (x=30 y=−48 z=14 Z-score=5.08 p=0.001 after FWEcorrection for multiple comparisons). Conversely, no regionsexpressed greater activation in patients with schizophrenia relativeto controls. There were no significant differences between bipolarpatients and controls and between bipolar and schizophrenic patients.

Main effects of genotypeThere was a significant effect of genotype in the precuneus bilaterally

(right: x=14 y=−50 z=12 Z-score=5.10; p-value=0.001 after FWEcorrection; left: x=−14 y=−54 z=16Z-score=4.52; p-value=0.014after FWE correction). Exploration of the parameter estimates revealedthat this region expressed greater deactivation during generation relativeto baseline in CC homozygotes than thosewith one or two copies of the Tallele (Fig. 1).

Effects of interaction of diagnostic group×genotypeThe interaction between diagnostic group and genotype was

examined by contrasting the effect of genotype in one diagnosticgroup against the same effect in another diagnostic group within ourfull-factorial ANOVA. This revealed two regions which expressed a

Table 2Coordinates and Z-scores (in brackets) of the brain regions that expressed significant(pb0.05 after FWE correction) decreased activation during baseline relative togeneration and increased activation during generation relative to baseline consistentlyacross all diagnostic groups and genotypes.

Brain regions Coordinates (Z-score)

BaselineNgenerationLeft precuneus 4 −52 34 (8.89)

−10 −60 24 (8.24)Left angular gyrus −50 −64 26 (7.64)

−44 −66 38 (6.92)−52 −60 35 (6.29)

Right angular gyrus 46 −54 24 (7.28)Left anterior cingulate −2 36 −1 (7.49)Right anterior cingulated 2 50 0 (7.32)Right medial frontal gyrus 4 46 −8 (7.82)Right insula 38 −16 14 (5.77)

36 −12 −6 (4.65)Left insula −38 −16 0 (5.35)

−38 −20 18 (4.96)Right postcentral gyrus 62 −28 38 (5.23)

56 −8 28 (4.51)

GenerationNbaselineLeft inferior frontal gyrus −46 6 28 (8.61)Left insula −38 18 14 (8.01)

−34 22 2 (8.00)Left superior temporal gyrus −50 −42 10 (4.42)

significant diagnosis×genotype interaction at pb0.05 after FWEcorrection for multiple comparisons. In these regions, the significantinteraction contrast was “CCNCT & TT in healthy volunteersNpatientswith schizophrenia”. One region to show a significant diagnosis×-genotype interaction was the left precuneus (x=−14, y=−56,z=26, Z-score=4.79, p=0.004 after FWE correction; see Fig. 2),which also expressed deactivation during generation relative tobaseline across different diagnostic and genotypic groups. Explorationof the parameter estimates revealed that, in this region, deactivationduring generation relative to baseline was less pronounced inschizophrenic patients with one or two copies of the T allele thanthose without it (x=−16, y=−58, z=28, Z-score=4.55, p=0.012after FWE correction); in contrast the opposite trend was expressedamongst healthy volunteers (x=−14, y=−52, z=26, Z-score=2.77, pb0.01 uncorrected). The other region to show asignificant diagnosis×genotype interaction were the right posteriorcingulate gyrus (x=20, y=−54, z=12, Z-score=4.42, p=0.021after FWE). In this region, there was activation during generation inpatients with schizophrenia who had at least one copy of the T allelewhile there was deactivation in those without it (x=16, y=−52,z=12, Z-score=5.11, p=0.001 after FWE); in contrast no such effectof DAAO genotype was expressed in healthy volunteers. There wereno diagnostic group×genotype interactionswhich involved the groupof bipolar patients.

Effects of functional connectivityThe aim of the PPI analysis was to examine the task-dependent

changes in functional connectivity which mediated the diagnosis×-genotype interactions reported above. We therefore performed twoPPI analyses which focussed on the left precuneus and the rightposterior cingulate respectively.

For the first region of interest, the left precuneus, the PPI analysisrevealed that the diagnosis×genotype interaction on regionalactivation was mediated by significant changes in functional connec-tivity (pb0.05 after FWE correction) between this region and adistributed network including left and right precuneus, left putamen,right posterior cingulate gyrus, left caudate and right angular gyrus(Fig. 3A and Table 3). Here the significant interaction contrast was “CTand TTNCC in healthy volunteersNpatients with schizophrenia”;exploration of the parameter estimates indicated that less pro-nounced deactivation during generation relative to baseline amongstschizophrenic patients with one or two copies of the T allele wasassociated with weaker functional connectivity.

Finally for the second region of interest, the right posterior cingulategyrus, we found that the diagnosis×genotype interaction on regionalactivation was mediated by significant changes in functional connec-tivity (pb0.05 after FWE correction) between this region and twoareas: the right precuneus and the left insula (Fig. 3B and Table 3). Herethe significant interaction contrast was “CCNCT and TT in healthyvolunteersNpatients with schizophrenia”; exploration of the parameterestimates indicated that greater activation during generation relative tobaseline amongst schizophrenic patients with one or two copies of theT allele was associated with stronger functional connectivity.

Possible confounding variablesWe considered the possibility that the above diagnosis×genotype

interactions could be explained by inter-subject variability in age,gender, handedness, IQ, years of education and ethnicity. Wetherefore repeated the statistical analysis modelling of these factorsas covariates of no interest; however this did not alter any of theresults. We also considered the possibility that our results could bedriven by the significant difference in anti-psychotic medicationbetween schizophrenic patients with and without the T allele. Whenmedication variables including dose, type (first vs. second generation)and duration of antipsychotic treatment were entered into thestatistical analysis as covariates of no interest, they did not change

Fig. 1. Effect of genotype in the right precuneus. Plot shows that this region expressed deactivation during task performance in CC homozygotes but not in those with one or twocopies of the T allele. C=control; S=schizophrenic; B=bipolar; easy=easy version of task; hard=hard version of task.

2287S.A. Papagni et al. / NeuroImage 56 (2011) 2283–2291

the foci of maximal significance or reduce the associated Z-scores. Wealso performed a correlation analysis to test for a possible associationbetween medication dose and brain activation in the three regionswhich expressed a significant diagnosis×genotype interaction. Wefound no evidence for a significant correlation between dose ofantipsychotic medication and brain activation in these regions(pN0.05), suggesting that the diagnosis×genotype interactionswere unlikely to be explained by medication effects. Likewise, theinclusion of dose of lithium medication as covariate of no interest didnot affect the results and the amount of activation in the leftprecuneus and the right posterior cingulate gyrus was not related tothis variable (pb0.05 uncorrected).

Fig. 2. A. Results of the interaction effect between diagnosis and genotype in the left precuneusless pronounced in schizophrenic patients with one or two copies of the T allele than those winteraction effect betweendiagnosis and genotype in the right posterior cingulate gyrus; plot shwho had at least one copy of the T allele while there was deactivation in those without it; in c

Discussion

We examined the effect of DAAO genotype on brain functionduring a verbal fluency task which engages brain regions andcognitive processes that are impaired in both schizophrenia andbipolar disorder (Daban et al., 2006; Krabbendam et al., 2005; Fu et al.,2002, 2005; Curtis et al., 2001). Independent of diagnostic group, theleft precuneus showed greater deactivation during task performancerelative to baseline in CC homozygotes than those with one or twocopies of the T allele (pb0.05 after FWE correction). In addition, asignificant genotype×diagnostic group interaction was found in theleft precuneus and the right posterior cingulate gyrus. These findings

; plot shows that this region expressed deactivation during generation relative to baselineithout it; in contrast the opposite was expressed in healthy volunteers. B. Results of theows that this region expressed activation during generation in patients with schizophreniaontrast no such effect of DAAO genotype was expressed in healthy volunteers.

Fig. 3. A. Results of the PPI analysis with the left precuneus as region of interest; the plot indicates that less pronounced deactivation during generation relative to baseline inschizophrenic patients with one or two copies of the T allele was associated with weaker functional connectivity. B. Results of the PPI analysis with the right posterior cingulate gyrusas region of interest; the plot indicates that greater activation during generation relative to baseline in schizophrenic patients with one or two copies of the T allele was associatedwith stronger functional connectivity.

2288 S.A. Papagni et al. / NeuroImage 56 (2011) 2283–2291

are unlikely to be attributable to differences in age, gender,handedness, ethnicity or years of education, since these factors weremodelled in the statistical analysis as covariates of no interest.

The precuneus and posterior cingulate gyrus are part of the “restingstate” or “default” network that often shows a decrease in activationduring task performance relative to baseline (Vanhaudenhuyse et al.,2010; Mechelli et al., 2008; McKiernan et al., 2006; 2003; Binder et al.,1999; Shulman et al., 1997; McGuire et al., 1996). This network isthought to be implicated in ongoing internally generated processes thatoccur during any low level baseline condition. The latter are disrupted

Table 3Results of the psycho-physiological interaction (PPI) analyses. Coordinates and Z-scores(in brackets) refer to regions which showed task-dependent changes in functionalconnectivity (FC) which mediated the diagnosis×genotype interactions in regionalactivation (pb0.05 after FWE correction).

Brain regions Coordinates (Z-score)

FC with left precuneusLeft precuneus −22 −70 30 (5.40)Right precuneus 10 −64 30 (4.75)Left putamen −28 −6 4 (4.88)Right cingulate gyrus 8 −40 42 (4.62)Left caudate −6 6 2 (4.62)

−18 22 −6 (4.29)−12 22 2 (4.24)

Right angular gyrus 32 −62 24 (4.51)44 −56 22 (4.27)

FC with right posterior cingulate gyrusRight precuneus 6 −46 10 (4.63)Left insula −34 0 −12 (4.55)

when the subject has to then engage in amoredemandingexperimentaltask (McKiernan et al., 2006; 2003; Binder et al., 1999), and task-relateddeactivation within the default network may reflect a reallocation ofcognitive resources (McKiernan et al., 2006; 2003; Binder et al., 1999).Furthermore, in previous studies, this network was functionally alteredin patients with schizophrenia relative to healthy controls (Kim et al.,2009;Whitfield-Gabrieli et al., 2009). One possible interpretation of ourfinding is that the active task might have produced differentialinterference with ongoing internally generated processes in schizo-phrenic patients as a function of genotype.

In order to characterise the task-dependent changes in functionalintegration which underlay the interactions between DAAO genotypeand diagnosis, we performed two PPI analyses. These revealed that thediagnosis-dependent effect of genotype in the left precuneus wasmediated by differential functional connectivity with a distributedcortical and sub-cortical network, while the diagnosis-dependent effectof genotype in the right posterior cingulate gyrus was mediated bydifferential functional connectivitywith the right precuneus and the leftinsula. Thus, the present study indicates that the impact of DAAOgenotype on brain function is not limited to regional responses but isalso evident in the functional integration between regions; this isconsistent with previous reports that genetic vulnerability to psychosisis associated with altered functional integration (Whalley et al., 2005;Whitfield-Gabrieli et al., 2009).

The genotype×diagnostic group interactions in the left precuneusand the right posterior cingulate gyrus were driven by effects of DAAOgenotype which were evident in patients with schizophrenia but notin healthy controls. This pattern of results is consistent with previousreports of disease-specific effects of candidate genes for psychosis onbrain structure and function (Addington et al., 2007; Frodl et al., 2004;

2289S.A. Papagni et al. / NeuroImage 56 (2011) 2283–2291

Mechelli et al., 2008). The mechanisms that lead to effects of DAAOgenotype being evident in patients with schizophrenia but not inhealthy controls are unclear. One possibility is that the effects ofvariation in a given gene depend on the genetic context (Weinberger,2010). Patients with schizophrenia are likely to carry a larger numberof risk genes than healthy controls; these risk genes may interact withthe DAAO gene such that its expression is evident in patients but notin controls. An alternative explanation is that the effect of a gene mayvary with differences in environmental exposure (Caspi and Moffitt,2006); which may also differ between patients and controls. A finalpossibility is that the effects of a given gene may interact with theeffects of schizophrenia and bipolar disorder on the function of thebrain.

It should be noted that, although a number of genetic studies haveassociated the T allele for DAAO rs3918346 with increased risk ofschizophrenia and bipolar disorder relative to the C allele, the resultshave not always been consistent as to which of these alleles confersthe higher risk (Liu et al., 2004; Schumacher et al., 2004; Fallin et al.,2005; Yamada et al., 2005; Liu et al., 2006; Shinkai et al., 2007; Li andHe, 2007; Wood et al., 2007; Corvin et al., 2007; Ohnuma et al., 2009;Bass et al., 2009); this could be due to false positive results or reflectallelic heterogeneity (i.e. different alleles in the same marker beingassociated with disease) (Moskvina and O'Donovan, 2007). In thepresent investigation, we have therefore avoided reference to theterms low or high risk, and have referred to the alleles instead.Assuming that DAAO rs3918346 does moderate the risk of schizo-phrenia and bipolar disorder, it remains to be established whether ornot the diagnosis-dependent effects identified in this study lie uponthe pathway between genes and clinical phenotype (Owen, 2010).This is an important question, since imaging endophenotypes maymediate the increased risk conferred by genes but also reflect geneeffects which do not necessarily result in increased risk, consistentwith the notion of pleiotropy (Owen, 2010).

In this study all significant DAAO effects were from the contrastbetween patients with schizophrenia and healthy controls with nosignificant effects in the bipolar patients. However, the directcomparison between schizophrenic and bipolar patients was notsignificant and, furthermore, exploration of the parameter estimatesindicated that the bipolar patients consistently showed intermediatevalues between schizophrenic patients and controls. This suggeststhat there may be effects of DAAO in the bipolar population that werenot detected due to the relatively small sample size. Alternatively, theeffects of DAAO on brain function may be more pronounced inschizophrenia than bipolar disorder due to systematic differences inother candidate genes between the two diagnostic groups. Futurestudies with larger sample size are required in order to clarify thisissue.

The present investigation has a number of limitations whichrequire careful consideration. Firstly, within the DAAO gene, severalSNPs have been identified by genetic association in addition tors3918346, and there is no agreement as to which is the marker withthe most significant effect on disease-risk; at present a functionalpolymorphism with plausible causality is yet to be identified.Secondly, genotypic groups were not balanced for dose of antipsy-chotic medication within the schizophrenic sample. However we didnot detect any significant correlation between regional activation andchlorpromazine equivalences. Thirdly, individuals with two copies ofthe T allele constitute only a small fraction of the population and in thepresent investigation were collapsed with the group of individualswith CT genotype. Thismeans that our data cannot reveal whether theaction of the T allele on brain function is best described by a dominantor an additive model.

In conclusion, we have shown for the first time that DAAOgenotype has a measurable impact on brain function in healthyvolunteers, and in patients with schizophrenia and bipolar disorder.Our findings also suggest that the effects of DAAO genotype differ

between patients with schizophrenia and healthy volunteers. AsDAAO genotype is thought to influence glutamate neurotransmission,this disease-dependent pattern of gene action provides indirectsupport for the hypothesis that glutamate dysfunction contributesto the pathophysiology of schizophrenia (Olney, 1995). A morecomprehensive understanding of how variation in DAAO affects brainfunction will require investigation of interactions with othercandidate genes and with environmental risk factors.

Supplementarymaterials related to this article can be found onlineat doi:10.1016/j.neuroimage.2011.03.037.

References

Addington, A.M., Gornick, M.C., Shaw, P., Seal, J., Gogtay, N., Greenstein, D., et al., 2007.Neuregulin 1 (8p12) and childhood-onset schizophrenia: susceptibility haplotypesfor diagnosis and brain developmental trajectories. Mol. Psychiatry 12 (2),195–205.

Ammons, R.B., Ammons, C.H., 1962. The Quick Test. Psychological Test Specialists,Missoula, MT.

Bass, N.J. ., Datta, S.R., McQuillin, A., Puri, V., Choudhury, K., Thirumalai, S., Lawrence, J.,Quested, D., Pimm, J., Curtis, D., Gurling, H.M., 2009. Evidence for the association ofthe DAOA (G72) gene with schizophrenia and bipolar disorder but not for theassociation of the DAO gene with schizophrenia. Behav. Brain Funct. 5, 28 Jul 8.

Berrettini, W., 2003. Bipolar disorder and schizophrenia: not so distant relatives?WorldPsychiatry 2 (2), 68–72 Jun.

Binder, J.R., Frost, J.A., Hammeke, T.A., Bellgowan, P.S., Rao, S.M., Cox, R.W., 1999.Conceptual processing during the conscious resting state. A functional MRI study. J.Cogn. Neurosci. 11 (1), 80–95.

Boks, M.P., Rietkerk, T., van de Beek, M.H., Sommer, I.E., de Koning, T.J., Kahn, R.S., 2007.Reviewing the role of the genes G72 and DAAO in glutamate neurotransmission inschizophrenia. Eur. Neuropsychopharmacol. 17 (9), 567–572 Review.

Cardno, A.G., Rijsdijk, F.V., Sham, P.C., Murray, R.M., McGuffin, P., 2002. A twin study ofgenetic relationships between psychotic symptoms. Am. J. Psychiatry 159 (4),539–545 Apr.

Caspi, A., Moffitt, T.E., 2006. Gene–environment interactions in psychiatry: joiningforces with neuroscience. Nat. Rev. Neurosci. 7 (7), 583–590 Jul, Review.

Chumakov, I., Blumenfeld, M., Guerassimenko, O., Cavarec, L., Palicio, M., Abderrahim, H.,Bougueleret, L., Barry, C., Tanaka, H., La Rosa, P., Puech, A., Tahri, N., Cohen-Akenine, A.,Delabrosse, S., Lissarrague, S., Picard, F.P., Maurice, K., Essioux, L., Millasseau, P., Grel, P.,Debailleul, V., Simon, A.M., Caterina, D., Dufaure, I., Malekzadeh, K., Belova, M., Luan, J.J.,Bouillot, M., Sambucy, J.L., Primas, G., Saumier, M., Boubkiri, N., Martin-Saumier, S.,Nasroune, M., Peixoto, H., Delaye, A., Pinchot, V., Bastucci, M., Guillou, S., Chevillon, M.,Sainz-Fuertes, R., Meguenni, S., Aurich-Costa, J., Cherif, D., Gimalac, A., Van Duijn, C.,Gauvreau, D., Ouellette, G., Fortier, I., Raelson, J., Sherbatich, T., Riazanskaia, N., Rogaev, E.,Raeymaekers, P., Aerssens, J., Konings, F., Luyten,W.,Macciardi, F., Sham, P.C., Straub, R.E.,Weinberger, D.R., Cohen, N., Cohen, D., 2002. Genetic and physiological data implicatingthe newhuman gene G72 and the gene for D-amino acid oxidase in schizophrenia. Proc.Natl Acad. Sci. U. S. A. 99 (21), 13675–13680 Oct 15, Epub 2002 Oct 3.

Corvin, A., McGhee, K.A., Murphy, K., Donohoe, G., Nangle, J.M., Schwaiger, S., Kenny, N.,Clarke, S., Meagher, D., Quinn, J., Scully, P., Baldwin, P., Browne, D., Walsh, C.,Waddington, J.L., Morris, D.W., Gill, M., 2007. Evidence for association and epistasisat the DAOA/G30 and D-amino acid oxidase loci in an Irish schizophrenia sample.Am. J. Med. Genet. B Neuropsychiatr. Genet. 144B (7), 949–953 Oct 5.

Coyle, J.T., Tsai, G., Goff, D., 2003. Converging evidence of NMDA receptor hypofunctionin the pathophysiology of schizophrenia. Ann. N. Y. Acad. Sci. 1003, 318–327 Nov.

Craddock, N., O'Donovan, M.C., Owen, M.J., 2006. Genes for schizophrenia and bipolardisorder? Implications for psychiatric nosology. Schizophr. Bull. 32 (1), 9–16 Jan,Epub 2005 Nov 30, Review.

Craddock, N., O'Donovan, M.C., Owen, M.J., 2007. Phenotypic and genetic complexity ofpsychosis. Invited commentary on… Schizophrenia: a common disease caused bymultiple rare alleles. Br. J. Psychiatry 190, 200–203 Mar.

Curtis, V.A., Dixon, T.A., Morris, R.G., Bullmore, E.T., Brammer, M.J., Williams, S.C.,Sharma, T., Murray, R.M., McGuire, P.K., 2001. Differential frontal activation inschizophrenia and bipolar illness during verbal fluency. J. Affect. Disord. 66 (2–3),111–121 Oct.

Daban, C., Martinez-Aran, A., Torrent, C., Tabarés-Seisdedos, R., Balanzá-Martínez, V.,Salazar-Fraile, J., Selva-Vera, G., Vieta, E., 2006. Specificity of cognitive deficits inbipolar disorder versus schizophrenia. A systematic review. Psychother. Psycho-som. 75 (2), 72–84.

Dager, S.R., Friedman, S.D., Parow, A., Demopulos, C., Stoll, A.L., Lyoo, I.K., Dunner, D.L.,Renshaw, P.F., 2004. Brain metabolic alterations in medication-free patients withbipolar disorder. Arch. Gen. Psychiatry 61 (5), 450–458 May.

Fallin,M.D., Lasseter, V.K., Avramopoulos, D., Nicodemus, K.K.,Wolyniec, P.S.,McGrath, J.A.,Steel, G., Nestadt, G., Liang, K.Y., Huganir, R.L., Valle, D., Pulver, A.E., 2005. Bipolar Idisorder and schizophrenia: a 440-single-nucleotide polymorphism screen of 64candidate genes amongAshkenazi Jewish case–parent trios. Am. J. Hum.Genet. 77 (6),918–936 Dec, Epub 2005 Oct 28.

Freeman, B., Smith, N., Curtis, C., Huckett, L., Mill, J., Craig, I.W., 2003. DNA from buccalswabs recruited by mail: evaluation of storage effects on long-term stability andsuitability for multiplex polymerase chain reaction genotyping. Behav. Genet. 33(1), 67–72.

2290 S.A. Papagni et al. / NeuroImage 56 (2011) 2283–2291

Friston, K.J., 2003. Introduction: experimental design and statistical parametricmapping. In: Frackowiak, R.S.J., Friston, K.J., Frith, C., Dolan, R., Friston, K.J., Price,C.J., Zeki, S., Ashburner, J., Penny, W.D. (Eds.), Human Brain Function. AcademicPress, pp. 599–632.

Friston, K.J., Buechel, C., Fink, G.R., Morris, J., Rolls, E., Dolan, R.J., 1997. Psychophysiologicaland modulatory interactions in neuroimaging. Neuroimage 6 (3), 218–229 Oct.

Frith, C.D., Leary, J., Cahill, C., Johnstone, E.C., 1991. Performance on psychological tests.Demographic and clinical correlates of the results of these tests. Br. J. PsychiatrySuppl. (13), 26–29 Oct, 44-6.

Frodl, T., Meisenzahl, E.M., Zill, P., Baghai, T., Rujescu, D., Leinsinger, G., Bottlender, R.,Schüle, C., Zwanzger, P., Engel, R.R., Rupprecht, R., Bondy, B., Reiser, M., Möller, H.J.,2004. Reduced hippocampal volumes associated with the long variant of theserotonin transporter polymorphism in major depression. Arch. Gen. Psychiatry 61(2), 177–183 Feb.

Fu, C.H., Morgan, K., Suckling, J., Williams, S.C., Andrew, C., Vythelingum, G.N., 2002. Afunctional magnetic resonance imaging study of overt letter verbal fluency using aclustered acquisition sequence: greater anterior cingulate activation withincreased task demand. NeuroImage 17 (2), 871–879.

Fu, C.H., Suckling, J., Williams, S.C., Andrew, C.M., Vythelingum, G.N., McGuire, P.K.,2005. Effects of psychotic state and task demand on prefrontal function inschizophrenia: an fMRI study of overt verbal fluency. Am. J. Psychiatry 162 (3),485–494 Mar.

Goldberg, T.E., Straub, R.E., Callicott, J.H., Hariri, A., Mattay, V.S., Bigelow, L., Coppola, R.,Egan, M.F., Weinberger, D.R., 2006. The G72/G30 gene complex and cognitiveabnormalities in schizophrenia. Neuropsychopharmacology 31 (9), 2022–2032Sep, Epub 2006 Mar 22.

Hashimoto, T., Volk, D.W., Eggan, S.M., Mirnics, K., Pierri, J.N., Sun, Z., Sampson, A.R.,Lewis, D.A., 2003. Gene expression deficits in a subclass of GABA neurons in theprefrontal cortex of subjects with schizophrenia. J. Neurosci. 23 (15),6315–6326.

Hashimoto, K., Engberg, G., Shimizu, E., Nordin, C., Lindström, L.H., Iyo, M., 2005. ReducedD-serine to total serine ratio in the cerebrospinal fluid of drug naive schizophrenicpatients. Prog. Neuropsychopharmacol. Biol. Psychiatry 29 (5), 767–769.

Hashimoto, K., Fujita, Y., Horio, M., Kunitachi, S., Iyo, M., Ferraris, D., Tsukamoto, T.,2009. Co-administration of a D-amino acid oxidase inhibitor potentiates the efficacyof D-serine in attenuating prepulse inhibition deficits after administration ofdizocilpine. Biol. Psychiatry 65 (12), 1103–1106 Jun 15, Epub 2009 Feb 12.

Heresco-Levy, U., 2005. Glutamatergic neurotransmission modulators as emergingnew drugs for schizophrenia. Expert Opin. Emerg. Drugs 10 (4), 827–844Review.

Javitt, D.C., Zukin, S.R., 1991. Recent advances in the phencyclidine model ofschizophrenia. Am. J. Psychiatry 148 (10), 1301–1308 Oct, Review.

Kapoor, R., Lim, K.S., Cheng, A., Garrick, T., Kapoor, V., 2006. Preliminary evidence for alink between schizophrenia and NMDA-glycine site receptor ligand metabolicenzymes, D-amino acid oxidase (DAAO) and kynurenine aminotransferase-1 (KAT-1). Brain Res. 1106 (1), 205–210 Aug 23, Epub 2006 Jul 7.

Kim, D.I., Manoach, D.S., Mathalon, D.H., Turner, J.A., Mannell, M., Brown, G.G., Ford, J.M.,Gollub, R.L., White, T., Wible, C., Belger, A., Bockholt, H.J., Clark, V.P., Lauriello, J.,O'Leary, D., Mueller, B.A., Lim, K.O., Andreasen, N., Potkin, S.G., Calhoun, V.D., 2009.Dysregulation of working memory and default-mode networks in schizophreniausing independent component analysis, an fBIRN and MCIC study. Hum. BrainMapp. 30 (11), 3795–3811 Nov.

Krabbendam, L., Arts, B., van Os, J., Aleman, A., 2005. Cognitive functioning in patientswith schizophrenia and bipolar disorder: a quantitative review. Schizophr. Res. 80(2–3), 137–149.

Kravariti, E., Dixon, T., Frith, C., Murray, R., McGuire, P., 2005. Association of symptomsand executive function in schizophrenia and bipolar disorder. Schizophr. Res. 74(2–3), 221–231 May 1.

Lewis, C.M., Levinson, D.F., Wise, L.H., DeLisi, L.E., Straub, R.E., Hovatta, I., Williams, N.M.,Schwab, S.G., Pulver, A.E., Faraone, S.V., Brzustowicz, L.M., Kaufmann, C.A., Garver, D.L.,Gurling, H.M., Lindholm, E., Coon, H., Moises, H.W., Byerley, W., Shaw, S.H., Mesen, A.,Sherrington, R., O'Neill, F.A., Walsh, D., Kendler, K.S., Ekelund, J., Paunio, T., Lönnqvist, J.,Peltonen, L., O'Donovan, M.C., Owen, M.J., Wildenauer, D.B., Maier, W., Nestadt, G.,Blouin, J.L., Antonarakis, S.E., Mowry, B.J., Silverman, J.M., Crowe, R.R., Cloninger, C.R.,Tsuang, M.T., Malaspina, D., Harkavy-Friedman, J.M., Svrakic, D.M., Bassett, A.S.,Holcomb, J., Kalsi, G., McQuillin, A., Brynjolfson, J., Sigmundsson, T., Petursson, H.,Jazin, E., Zoëga, T., Helgason, T., 2003. Genome scanmeta-analysis of schizophrenia andbipolar disorder, part II: schizophrenia. Am. J. Hum. Genet. 73 (1), 34–48 Jul, Epub 2003Jun 11.

Li, D., He, L., 2007. G72/G30 genes and schizophrenia: a systematic meta-analysis ofassociation studies. Genetics 175 (2), 917–922 Feb, Epub 2006 Dec 18.

Li, X., Ketter, T.A., Frye, M.A., 2002. Synaptic, intracellular and neuroprotectivemechanisms of anticonvulsants: are they relevant for the treatment and courseof bipolar disorders? J. Affect. Disord. 69 (1–3), 1–14 May, Review.

Liu, X., He, G., Wang, X., Chen, Q., Qian, X., Lin, W., Li, D., Gu, N., Feng, G., He, L., 2004.Association of DAAO with schizophrenia in the Chinese population. Neurosci. Lett.369 (3), 228–233 Oct 21.

Liu, Y.L., Fann, C.S., Liu, C.M., Chang, C.C., Wu, J.Y., Hung, S.I., Liu, S.K., Hsieh, M.H.,Hwang, T.J., Chan, H.Y., Chen, J.J., Faraone, S.V., Tsuang, M.T., Chen, W.J., Hwu, H.G.,2006. No association of G72 and D-amino acid oxidase genes with schizophrenia.Schizophr. Res. 87 (1–3), 15–20 Oct, Epub 2006 Jul 13.

Madeira, C., Freitas, M.E., Vargas-Lopes, C., Wolosker, H., Panizzutti, R., 2008. Increasedbrain D-amino acid oxidase (DAAO) activity in schizophrenia. Schizophr. Res. 101(1–3), 76–83 Apr, Epub 2008 Apr 2.

McGuire, P.K., Paulesu, E., Frackowiak, R.S., Frith, C.D., 1996. Brain regions active duringstimulus independent thought. Neuroreport 7 (13), 2095–2099 Sep 2.

McKiernan, K.A., Kaufman, J.N., Kucera-Thompson, J., Binder, J.R., 2003. A parametricmanipulation of factors affecting task-induced deactivation in functional neuro-imaging. J. Cogn. Neurosci. 15 (3), 394–408 Apr 1.

McKiernan, K.A., D'Angelo, B.R., Kaufman, J.N., Binder, J.R., 2006. Interrupting the“stream of consciousness”: an fMRI investigation. Neuroimage 29 (4), 1185–1191Feb 15, Epub 2005 Nov 2.

Mechelli, A., Prata, D.P., Fu, C.H., Picchioni, M., Kane, F., Kalidindi, S., McDonald, C.,Demjaha, A., Kravariti, E., Toulopoulou, T., Murray, R., Collier, D.A., McGuire, P.K.,2008. The effects of neuregulin1 on brain function in controls and patients withschizophrenia and bipolar disorder. Neuroimage 42 (2), 817–826 Aug 15, Epub2008 May 27.

Michael, N., Erfurth, A., Ohrmann, P., Gössling, M., Arolt, V., Heindel, W., Pfleiderer, B.,2003. Acute mania is accompanied by elevated glutamate/glutamine levels withinthe left dorsolateral prefrontal cortex. Psychopharmacology (Berl.) 168 (3),344–346 Jul, Epub 2003 Apr 9.

Moskvina, V., O'Donovan, M.C., 2007. Detailed analysis of the relative power of directand indirect association studies and the implications for their interpretation. Hum.Hered. 64 (1), 63–73 Epub 2007 Apr 27.

Ohnuma, T., Shibata, N., Maeshima, H., Baba, H., Hatano, T., Hanzawa, R., Arai, H., 2009.Association analysis of glycine- and serine-related genes in a Japanese populationof patients with schizophrenia. Prog. Neuropsychopharmacol. Biol. Psychiatry 33(3), 511–518 Apr 30, Epub 2009 Feb 14.

Olney, J.W., Farber, N.B., 1995. Glutamate receptor dysfunction and schizophrenia. Arch.Gen. Psychiatry 52 (12), 998–1007.

Owen, M., 2010. Imaging genetics in psychiatry: disease path or garden path?Schizophr. Res. 117, 123.

Owen, M.J., Craddock, N., Jablensky, A., 2007. The genetic deconstruction of psychosis.Schizophr. Bull. 33 (4), 905–911 Jul, Epub 2007 Jun 5.

Park, N., Juo, S.H., Cheng, R., Liu, J., Loth, J.E., Lilliston, B., Nee, J., Grunn, A., Kanyas, K.,Lerer, B., Endicott, J., Gilliam, T.C., Baron, M., 2004. Linkage analysis of psychosis inbipolar pedigrees suggests novel putative loci for bipolar disorder and sharedsusceptibility with schizophrenia. Mol. Psychiatry 9 (12), 1091–1099 Dec.

Prata, D., Breen, G., Osborne, S., Munro, J., St Clair, D., Collier, D., 2008a. Association ofDAO and G72(DAOA)/G30 genes with bipolar affective disorder. Am. J. Med. Genet.B Neuropsychiatr. Genet. 147B (6), 914–917 Sep 5.

Prata, D.P., Mechelli, A., Fu, C.H., Picchioni, M., Kane, F., Kalidindi, S., McDonald, C.,Kravariti, E., Toulopoulou, T., Miorelli, A., Murray, R., Collier, D.A., McGuire, P.K.,2008b. Effect of disrupted-in-schizophrenia-1 on pre-frontal cortical function. Mol.Psychiatry 13 (10), 915–917 Oct, 909.

Prata, D.P., Mechelli, A., Fu, C.H., Picchioni, M., Toulopoulou, T., Bramon, E., Walshe, M.,Murray, R.M., Collier, D.A., McGuire, P., 2009a. Epistasis between the DAT 3′ UTRVNTR and the COMT Val158Met SNP on cortical function in healthy subjects andpatients with schizophrenia. Proc. Natl Acad. Sci. U. S. A. 106 (32), 13600–13605Aug 11, Epub 2009 Jul 28.

Prata, D.P., Mechelli, A., Fu, C.H., Picchioni, M., Kane, F., Kalidindi, S., McDonald, C.,Howes, O., Kravariti, E., Demjaha, A., Toulopoulou, T., Diforti, M., Murray, R.M.,Collier, D.A., McGuire, P.K., 2009b. Opposite effects of catechol-O-methyltransferaseVal158Met on cortical function in healthy subjects and patients with schizophre-nia. Biol. Psychiatry 65 (6), 473–480 Mar 15, Epub 2008 Dec 3.

Schumacher, J., Jamra, R.A., Freudenberg, J., Becker, T., Ohlraun, S., Otte, A.C., Tullius, M.,Kovalenko, S., Bogaert, A.V., Maier, W., Rietschel, M., Propping, P., Nöthen, M.M.,Cichon, S., 2004. Examination of G72 and D-amino-acid oxidase as genetic riskfactors for schizophrenia and bipolar affective disorder. Mol. Psychiatry 9 (2),203–207 Feb.

Segurado, R., Detera-Wadleigh, S.D., Levinson, D.F., Lewis, C.M., Gill, M., Nurnberger Jr., J.I.,Craddock, N., DePaulo, J.R., Baron, M., Gershon, E.S., Ekholm, J., Cichon, S., Turecki, G.,Claes, S., Kelsoe, J.R., Schofield, P.R., Badenhop, R.F., Morissette, J., Coon, H., Blackwood,D., McInnes, L.A., Foroud, T., Edenberg, H.J., Reich, T., Rice, J.P., Goate, A., McInnis, M.G.,McMahon, F.J., Badner, J.A., Goldin, L.R., Bennett, P., Willour, V.L., Zandi, P.P., Liu, J.,Gilliam, C., Juo, S.H., Berrettini, W.H., Yoshikawa, T., Peltonen, L., Lönnqvist, J., Nöthen,M.M., Schumacher, J., Windemuth, C., Rietschel, M., Propping, P., Maier, W., Alda, M.,Grof, P., Rouleau,G.A., Del-Favero, J., VanBroeckhoven,C.,Mendlewicz, J., Adolfsson, R.,Spence, M.A., Luebbert, H., Adams, L.J., Donald, J.A., Mitchell, P.B., Barden, N., Shink, E.,Byerley,W., Muir, W., Visscher, P.M., Macgregor, S., Gurling, H., Kalsi, G., McQuillin, A.,Escamilla, M.A., Reus, V.I., Leon, P., Freimer, N.B., Ewald, H., Kruse, T.A., Mors, O.,Radhakrishna, U., Blouin, J.L., Antonarakis, S.E., Akarsu, N., 2003. Genome scan meta-analysis of schizophrenia and bipolar disorder, part III: Bipolar disorder. Am. J. Hum.Genet. 73 (1), 49–62 Jul, Epub 2003 Jun 11.

Shinkai, T., De Luca, V., Hwang, R., Müller, D.J., Lanktree, M., Zai, G., Shaikh, S., Wong, G.,Sicard, T., Potapova, N., Trakalo, J., King, N., Matsumoto, C., Hori, H., Wong, A.H.,Ohmori, O., Macciardi, F., Nakamura, J., Kennedy, J.L., 2007. Association analyses ofthe DAOA/G30 and D-amino-acid oxidase genes in schizophrenia: further evidencefor a role in schizophrenia. Neuromolecular Med. 9 (2), 169–177.

Shulman, G.L., Fiez, J.A., Corbetta, M., Buckner, R.L., Miezin, F.M., Raichle, M.E., Petersen,S.E., 1997. Common blood flow changes across visual tasks. 2. Decreases in cerebralcortex. J. Cogn. Neurosci. 5, 648–663.

Sklar, P., 2002. Linkage analysis in psychiatric disorders: the emerging picture. Annu.Rev. Genomics Hum. Genet. 3, 371–413.

Stefanis, N.C., Trikalinos, T.A., Avramopoulos, D., Smyrnis, N., Evdokimidis, I., Ntzani, E.E.,Ioannidis, J.P., Stefanis, C.N., 2007. Impact of schizophrenia candidate genes onschizotypy and cognitive endophenotypes at the population level. Biol. Psychiatry 62(7), 784–792 Oct 1.

Stone, J.M., Day, F., Tsagaraki, H., Valli, I., McLean, M.A., Lythgoe, D.J., O'Gorman, R.L.,Barker, G.J., McGuire, P.K., 2009. Glutamate dysfunction in people with prodromalsymptoms of psychosis: relationship to gray matter volume. Biol. Psychiatry 66 (6),533–539.

2291S.A. Papagni et al. / NeuroImage 56 (2011) 2283–2291

Tanii, Y., Nishikawa, T., Hashimoto, A., Takahashi, K., 1994. Stereoselective antagonismby enantiomers of alanine and serine of phencyclidine-induced hyperactivity,stereotypy and ataxia in the rat. J. Pharmacol. Exp. Ther. 269 (3), 1040–1048.

Tuominen, H.J., Tiihonen, J., Wahlbeck, K., 2005. Glutamatergic drugs for schizophrenia:a systematic review and meta-analysis. Schizophr. Res. 72 (2–3), 225–234 Review.

Vanhaudenhuyse,A., Noirhomme,Q., Tshibanda, L.J., Bruno,M.A., Boveroux, P., Schnakers, C.,Soddu, A., Perlbarg, V., Ledoux, D., Brichant, J.F., Moonen, G., Maquet, P., Greicius, M.D.,Laureys, S., Boly, M., 2010. Default network connectivity reflects the level ofconsciousness in non-communicative brain-damaged patients. Brain 133 (Pt 1),161–171 Jan.

Verrall, L., Walker, M., Rawlings, N., Benzel, I., Kew, J.N., Harrison, P.J., Burnet, P.W.,2007. D-Amino acid oxidase and serine racemase in human brain: normaldistribution and altered expression in schizophrenia. Eur. J. Neurosci. 26 (6),1657–1669.

Verrall, L., Burnet, P.W., Betts, J.F., Harrison, P.J., 2010. The neurobiology of D-amino acidoxidase and its involvement in schizophrenia. Mol. Psychiatry 15 (2), 122–137.

Wechsler, D., 1981. Manual for theWechsler Intelligence Scale— Revised. PsychologicalCorporation.

Wechsler, D., 1997. Wechsler Adult Intelligence Scale—Third Edition Manual.Psychological Corporation.

Wechsler, D., 1999.Wechsler Abbreviated Scale of Intelligence. Psychological Corporation.Weinberger, D., 2010. On the matter of neuroimaging in the context of schizophrenia

genetics. Schizophr. Res. 117, 109–110.Whalley, H.C., Simonotto, E., Marshall, I., Owens, D.G., Goddard, N.H., Johnstone, E.C.,

Lawrie, S.M., 2005. Functional disconnectivity in subjects at high genetic risk ofschizophrenia. Brain 128 (Pt 9), 2097–2108.

Whitfield-Gabrieli, S., Thermenos, H.W., Milanovic, S., Tsuang, M.T., Faraone, S.V.,McCarley, R.W., Shenton, M.E., Green, A.I., Nieto-Castanon, A., LaViolette, P.,Wojcik, J., Gabrieli, J.D., Seidman, L.J., 2009. Hyperactivity and hyperconnectivityof the default network in schizophrenia and in first-degree relatives of personswith schizophrenia. Proc. Natl Acad. Sci. U. S. A. 106 (4), 1279–1284 Epub 2009Jan 21.

Wood, L.S., Pickering, E.H., Dechairo, B.M., 2007. Significant support for DAO as aschizophrenia susceptibility locus: examination of five genes putatively associatedwith schizophrenia. Biol. Psychiatry 61 (10), 1195–1199.

Yamada, K., Ohnishi, T., Hashimoto, K., Ohba, H., Iwayama-Shigeno, Y., Toyoshima, M.,Okuno, A., Takao, H., Toyota, T., Minabe, Y., Nakamura, K., Shimizu, E., Itokawa, M.,Mori, N., Iyo, M., Yoshikawa, T., 2005. Identification of multiple serine racemase(SRR)mRNA isoforms and genetic analyses of SRR and DAO in schizophrenia and D-serine levels. Biol. Psychiatry 57 (12), 1493–1503.