American Journal of Medical Genetics Part B (Neuropsychiatric Genetics) 147B:654–657 (2008)

Brief Research CommunicationPRODH Gene is Associated With Executive Functionin Schizophrenic FamiliesTao Li,1,2,3,4* Xiaohong Ma,1,3 Xun Hu,2,4 Yingcheng Wang,1,3 Chengying Yan,1,3

Huaqing Meng,5 Xiehe Liu,1 Timothea Toulopoulou,2 Robin M. Murray,2 and David A. Collier2,4

1Psychiatric Laboratory, Department of Psychiatry, West China Hospital, Sichuan University, Sichuan, China2King’s College London, Division of Psychological Medicine, Institute of Psychiatry, London, UK3State Key Laboratory of Biotherapy, West China Hospital, Sichuan University, Sichuan, China4King’s College London, SGDP Centre, Institute of Psychiatry, London, UK5Department of Psychiatry, Chongqing University of Medical Science, Chongqing, China

The aim of this study was to investigate therelationship between polymorphisms in thePRODH and COMT genes and selected neurocog-nitive functions. Six SNPs in PRODH and twoSNPs in COMT were genotyped in 167 first-episodeschizophrenic families who had been assessed bya set of 14 neuropsychological tests. Neuropsycho-logical measures were selected as quantitativetraits for association analysis. The haplotype ofSNPs PRODH 1945T/C and PRODH 1852G/A wasassociated with impaired performance on theTower of Hanoi, a problem-solving task mainlyreflecting planning capacity. There was no sig-nificant evidence for association with any otherneuropsychological traits for other SNPs or hap-lotypes of paired SNPs in the two genes. Thisstudy takes previous findings of associationbetween PRODH and schizophrenia further byassociating variation within the gene with per-formance on a neurocognitive trait characteristicof the illness. It fails to confirm previous reportsof an association between COMT and cognitivefunction. � 2007 Wiley-Liss, Inc.

KEY WORDS: schizophrenia; executive func-tion; PRODH; COMT; quantita-tive traits

Please cite this article as follows: Li T, Ma X, Hu X, WangY, Yan C, Meng H, Liu X, Toulopoulou T, Murray RM,Collier DA. 2008. PRODH Gene is Associated WithExecutive Function in Schizophrenic Families. Am JMed Genet Part B 147B:654–657.

INTRODUCTION

Both linkage analysis and linkage disequilibrium studieshave suggested that chromosome 22 contains a locus forschizophrenia [Riley and McGuffin, 2000; Lewis et al., 2003].Moreover, the 22q11 deletion syndrome (22q11DS) is charac-terized by a variable phenotype which includes cognitive andbehavioral problems; 10–30% of those affected developpsychosis in adulthood, making it one of the highest riskfactors for schizophrenia [Bassett et al., 2003]. The study ofcandidate genes within the 3 Mb deletion, especially thenarrower 1.5 Mb ‘psychosis critical region’ at 22q11 offers aunique opportunity to further understanding of the patho-genesis of schizophrenia. Indeed, several genes in 22pter-q12.3, including catechol-O-methyltransferase (COMT) [Liet al., 2000; Shifman et al., 2002] and proline dehydrogenase(PRODH) [Liu et al., 2002; Li et al., 2004] have been associatedwith schizophrenia. However several studies have also failed toreplicate these associations: a recent meta-analysis of associ-ation studies of COMT with schizophrenia did not find evidencefor a main effect, and other studies have failed to replicateassociation with PRODH and COMT [Munafo et al., 2005;Williams et al., 2003; Glaser et al., 2006]. On the other handthere is evidence for a gene-by-environment effect betweencannabis use and COMT [Caspi et al., 2005]. Thus theassociations seen with COMT and PRODH in schizophreniacould be weak genetic effects partially dependent on environ-mental factors for their full expression as illness susceptibilitygenes, making main effects hard to detect.

There may be an advantage in improving the definition of thephenotype and/or reducing the phenotypic complexity ofschizophrenia for genetic studies by using endophenotypicmarkers [Gottesman and Gould, 2003]. Previous studiessuggested that both patients with schizophrenia and theirrelatives have neuropsychological deficits in several cognitivedomains, particularly memory, and executive function [Bilderet al., 2000; Toulopoulou et al., 2003]. Several recent studiesshowed that certain of these features are relatively specific toschizophrenia, with relatives showing deficits intermediatebetween those of their affected relatives and normal subjects[Egan et al., 2001; Toulopoulou et al., 2003], though not allstudies are consistent with this [Faraone et al., 1996]. Werecently examined neurocognitive functions in 235 first-episode schizophrenic patients, 322 of their first-degreerelatives and 133 unrelated normal controls. Of the threegroups, schizophrenic patients performed poorest on allneuropsychological measures, suggestive of a broad range ofneurocognitive deficits. Their first-degree relatives showeda narrower pattern of impaired performance in tests ofmemory and attention (Arithmetic, Digit Symbol, DigitSpan, Delayed Logical Memory), as well as Trail Making(attention, visual searching, mental processing speed andcontrol), Verbal Fluency Test (speed and flexibility of verbal

Grant sponsor: The National Natural Science Foundation ofChina; Grant numbers: 30530300, 30125014; Grant sponsor:NARSAD Independent Investigator Award; Grant sponsor: Well-come Trust for an International Collaborative award; Grantsponsor: Schizophrenia Research Fund; Grant sponsor: NationalBasic Research Program of China (973 Program); Grant number:2007CB512301.

*Correspondence to: Tao Li, M.D., Ph.D., Division of Psycholo-gical Medicine, PO82, Institute of Psychiatry, King’s CollegeLondon, London SE5 8AF, UK. E-mail: sphatal@iop.kcl.ac.uk

Received 27 June 2007; Accepted 18 September 2007

DOI 10.1002/ajmg.b.30648

� 2007 Wiley-Liss, Inc.

thought processes), Tower of Hanoi (executive functions andstrategy planning) and the Wisconsin Card Sorting Test(WCST-M) (frontal lobe functioning). These patterns of neuro-cognitive deficits may represent ‘‘endophenotypes’’ denotingvarying degrees of vulnerability to schizophrenia [Ma et al.,2007].

COMT has previously been associated with neurocognitivefunctions, especially frontal lobe functions such as workingmemory, prefrontal blood flow and prefrontal Frontal P300ERPs and the Wisconsin card sorting test [reviewed inWinterer and Goldman, 2003]. PRODH has not been testedat the time of writing for a role in influencing cognitive functionin humans but in mice chronic hyperprolinemia provokes amemory deficit [Bavaresco et al., 2005].

In the present study, we performed a quantitative linkagedisequilibrium analysis to examine the effect of PRODH andCOMT genes on those neurocognitive trait measures whichshowed family aggregation in our sample.

MATERIALS AND METHODS

Subjects

We recruited 167 Han Chinese families which included 167first-episode schizophrenic patients (80 Male, 87 female), 136of their fathers (3 affected), 135 mothers (1 affected), and 32siblings (2 affected) from Southwest China, Sichuan Province.Detailed information regarding the samples has beendescribed in previous publication [Ma et al., 2007]. In brief,all patients were at the initial stage of the illness when theneuropsychological assessments were implemented, and themajority of them were neuroleptic naı̈ve at the time ofneuropsychological assessment. The age and age at onsetrange of probands was 16–49 years (25.89� 8.08) and 14–44 years (24.78� 7.86), respectively. The years of educationranged from 2 to 18 years (11.57� 3.12). All subjects wereinterviewed by a psychiatrist using the SCID. Diagnosis ofschizophrenia was assigned on the basis of the interview andmedical records according to DSM-IV criteria. Patients wereexcluded if they had a history of neurological disorders or headinjury, or reported intellectual disability. Informed consentswere obtained from all participants.

Neuropsychological Assessment

All subjects were assessed by a comprehensive set of 14neuropsychological tests. In brief, 19 measures derived from 14tests, which can be broken down as follows: (1) Number ofCorrect Responses in items completed within 120 sec for StroopColor and Stroop Color-Word Interference test respectively; (2)Illegal moves, Errors and Time to complete task for TrailMaking Part A and Trail Making B-M respectively; (3) Correctwords and repeated words (animal) within 60 sec for VerbalFluency Test; (4) Planning Time, Executive Time, IllegalMoves, Total Score of Tower of Hanoi; (5) Correct Number,Total Errors, Perseverative Errors, Nonperseverative errors,and Catagories of WCST-M; (6) Five subtests (Information,Arithmetic, Digit Symbol, Digit Span, and Block Design) fromthe Wechsler Adult Intelligence Scale-Revised, Chinese Ver-sion, scores were adjusted for the age of the subject; and (7)Subtest scores from WMS-RC, including Immediate andDelayed Logical Memory, Immediate and Delayed VisualReproduction, Scores of delayed Visual Reproduction.

Genotyping

Six SNPs within the PRODH gene, PRODH� 1195C/T,PRODH� 1482C/T, PRODH� 1483A/G, PRODH� 1766A/G,PRODH� 1852G/A, and PRODH� 1945T/C, were also geno-typed as described elsewhere [Liu et al., 2002; Li et al., 2004].

SNPs 472A/G (ss6312623) and �1217A/G (ss6312621) weregenotyped in the COMT gene. SNPs 472A/G result in an aminoacid change (Val to Met), at codon 158 of MB-COMT and codon108 of S-COMT, that decreases the activity level of the COMTenzyme three- to fourfold [Lachman et al., 1996]. Polymor-phism �1217A/G, which we previously named as �287C/T, islocated at 50 promoter region of the longer MB-COMT mRNAspecies with an A/G SNP altering a HindIII restriction site [Liet al., 2000]. SNPs 472A/G was genotyped using the Amplifi-cation Refractory Mutation System (ARMS) assay [Hoda et al.,1996], and �1217A/G was genotyped as described elsewhere[Li et al., 2000].

Statistical Analysis

Fifteen measures from 10 neuropsychological tests/subtestswere selected as quantitative traits for quantitative trait LDanalysis: Arithmetic, Digit symbol, Digit span, Immediatelyand Delayed logical memory, Correct words of and Repeatedwords of Verbal Fluency, Illegal Moves, Time of Trail Making Aand Trail Making B, Scores of Hanoi Tower, Correct Number,Perseverative Error and Categories of WCST. These measureswere selected because our previous analysis showed they wereimpaired in the families of our schizophrenic patients, suggest-ing they may be a specific familial indicator of vulnerability toschizophrenia [Ma et al., 2007].

To adjust for the effects of age, gender and education level onneuropsychological performance, linear regression was per-formed by using neuropsychological measures as dependentvariables while age, sex and educational years were used asindependent variables. Only sex and education years wereused as independent variables for scores on Arithmetic, DigitSymbol, and Digit Span because these measures had alreadybeen adjusted for age effects. Standardized residuals weresaved as quantitative phenotypic measures. They werenormally or approximately normally distributed variables.

The program QPDTPHASE was used for analysis (http://www.rfcgr.mrc.ac .uk/registered/option/unphased.html).QPDTPHASE implements the quantitative trait pedigreedisequilibrium test, which can be used with multiple familytypes in addition to trios and which has been extended to dealwith haplotypes and missing data [Dudbridge, 2003]. Inaddition, the permutation procedure allows a reasonablecorrection for tests of multiple haplotypes and multiple loci.This permutation corrected the P-value for the amounts ofSNPs and haplotypes tested but not for the multiple measuresfrom the neuropsychological tests.

RESULTS

PRODH Gene

Quantitative pedigree disequilibrium test was carried outusing six SNPs in the PRODH gene using the 15 cognitivemeasures mentioned above as quantitative traits. SNPsPRODH 1945T/C showed significant association with Scoresof the Tower of Hanoi (w2¼ 4.94, P¼ 0.03) and categories ofWCST (w2¼ 5.34, P¼ 0.02). SNPs PRODH 1195C/T showedsignificant LD with Immediately logical memory (w2¼ 4.81,P¼ 0.03), and SNPs PRODH 1852G/A showed significant LDwith Illegal times of trail making A (w2¼ 4.43, P¼ 0.04).However, after we applied the permutation procedure with1,000 permutations to allow correction for tests of multiple loci,none of these P values remained significant (>0.05).

We also performed moving window pairwise transmissiondisequilibrium analysis of haplotypes with each neurocogni-tive trait. The haplotype of PRODH 1945/1852 showedsignificant LD with Tower of Hanoi scores (Global w2¼ 9.07,P¼ 0.03; haplotype PRODH 1945T-1852G: Z¼ 2.62, P¼ 0.009;haplotype PRODH 1945T-1852A: Z¼�2.22, P¼ 0.03) and

PRODH Gene and Executive Function 655

categories of WCST (Global w2¼ 7.19, P¼ 0.07; haplotypePRODH 1945T-1852G: Z¼ 2.11, P¼ 0.04; haplotype PRODH1945T-1852A: Z¼�2.15, P¼ 0.03) before correction for multi-ple testing. The permutation procedure with 1,000 permuta-tions was carried out to allow correction for tests of multiple lociand multiple haplotypes. No P values remained significantafter correction (P> 0.05) except for individual haplotypePRODH 1945T-1852A (P¼ 0.01).

COMT

QPDT was carried out with Val158Met, �1217A/G, orhaplotypes of both SNPs using 15 measures as derived fromthe 10 neuropsychological tests/subtests mentioned above asquantitative traits. There was no significant evidence forassociation with any quantitative trait for either SNPs orhaplotypes of the two SNPs.

DISCUSSION

PRODH

This gene is located in a region (22q11) previously linked toschizophrenia, and has recently being reported to be in linkagedisequilibrium (LD) with schizophrenia [Liu et al., 2002]. Liuet al. [2002] found association between the SNP PRODH 1945or a two-marker haplotype, PRODH 1945/1766 and schizo-phrenia by both TDT and case-control analysis. Our recentfindings [Li et al., 2004] are consistent with the earlier study ofLiu et al. [2002], which shows a different haplotype PRODH1945/1852 associated with schizophrenia in the Han Chinese.The haplotype PRODH 1945/1852 is in fact a sub-haplotype ofthe risk haplotype identified by Liu et al. [2002], supporting thenotion that the causative risk alleles are towards the 30 end ofthe gene. However, three other studies have failed to supportthe association between PRODH and schizophrenia [Fan et al.,2003; Williams et al., 2003; Glaser et al., 2006].

Previous studies [Egan et al., 2001; Toulopoulou et al., 2003]have shown that performance on neuropsychological tests,particularly those involving executive function and memory, isimpaired in schizophrenic patients and some of their biologicalrelatives. In this study, we found that the haplotype of pairedSNPs PRODH 1945/1852 was associated with impairedperformance on the Tower of Hanoi, a complex problem-solvingtask mainly reflecting the planning capacity of executivefunction. This result takes previous findings of associationbetween PRODH and schizophrenia [Liu et al., 2002] furtherby associating it with a neurocognitive trait characteristic ofthe illness.

Executive function is characteristically impaired in schizo-phrenia. Many studies have used the WCST to measureexecutive function [Lezak, 1995; Egan et al., 2001], but wedid not find significant association between PRODH gene andWCST in this study. This may be because the WCST is acomplex task assessing various cognitive processes includingconcept formation and cognitive flexibility. Thus, a subject mayperform poorly on WCST for various reasons including deficitsin concept formation, or dysfunction in response shifting.Tower of Hanoi is a complex task assessing the development ofstrategies, and the issue of goal/subgoal conflict. It also tapsinto aspects of working memory since one has to conceptualizeand execute a plan of action which one has to hold on line untilthe execution of the plan is complete. Within this, there is asensorymotor component, since one needs to scan the initialdisplay, form and maintain a representation of the moves to bemade and then physically change the display according to themental representation of the plan. Gogos et al. [1999] showedthat PRODH is involved in sensorimotor gating, a neuro-cognitive neural filtering process that allows attention to befocused on a given stimulus, and which is often affected in

patients with neuropsychiatric disorders or head injury. Thusthe association between impaired performance on the Tower ofHanoi and PRODH gene may reflect the effect of the gene onsensorimotor gating in schizophrenic families. It should benoted that deficits in the sensorimotor gating seen in humanscan be modeled in rats by administration of MK-801, a non-competitive antagonist of NMDA receptors. Thus our currentfindings may suggest that the performance of Tower of Hanoimay reflect not only the known neuropsychological domain,that is, problem solving, working memory or executivefunctions, but also with more complex components such asattention modulation. We performed an association analysisbetween the PRODH SNPs and the neurocognitive measuresin 465 healthy controls. There was no significant evidence forassociation with any quantitative trait for individual SNPs orhaplotypes of the SNPs in PRODH gene (unpublished data).This finding, in conjunction with the chromosomal location ofPRODH, suggests a potential involvement of this gene in the22q11-associated cognitive, psychiatric and behavioral pheno-types.

COMT

Egan et al. [2001] found that the common functionalpolymorphism Val158Met in the COMT gene was related toperformance on the WCST, and explained 4% of variance infrequency of perseverative errors. Bilder et al. [2002] andMalhotra et al. [2002] also provided evidence that the poly-morphism influences neurocognitive functions in schizophre-nia, especially measures of Processing Speed and Attention.However, in a recent meta-analysis aimed at exploring theeffects of the COMT Val158Met polymorphism on executivefunction in schizophrenics and health controls, small butsignificant effects of Val158Met genotype on executive functionin healthy individuals but not in schizophrenia were found[Barnett et al., 2007]. Our results are in line with the findings ofthe meta-analysis and showed no significant evidence forlinkage disequilibrium with any neurocognitive measures forthe individual SNPs COMT 158Val/Met and COMT�1217G/A,or haplotypes of these two SNPs. Thus, our results do notsupport the direct involvement of these two SNPs in the COMTgene with neurocognitive functions in schizophrenia. Theearlier findings by Egan et al. [2001] and others might be falsepositives, or the association sufficiently weak or heterogeneousbetween populations that detection is difficult.

ACKNOWLEDGMENTS

This work was partly funded by National Nature ScienceFoundation of China (30530300 and 30125014, TL), NARSADIndependent Investigator Award (TL), the Wellcome Trust foran International Collaborative award (TL, DAC and XL), theSchizophrenia Research Fund (TL and DAC), and NationalBasic Research Program of China (973 Program:2007CB512301, TL).

REFERENCES

Barnett JH, Jones PB, Robbins TW, Muller U. 2007. Effects of the catechol-O-methyltransferase Val158Met polymorphism on executive function: Ameta-analysis of the Wisconsin Card Sort Test in schizophrenia andhealthy controls. Mol Psychiatry 12:502–509.

Bassett AS, Chow EW, AbdelMalik P, Gheorghiu M, Husted J, Weksberg R.2003. The schizophrenia phenotype in 22q11 deletion syndrome. Am JPsychiatry 160:1580–1586.

Bavaresco CS, Streck EL, Netto CA, Wyse AT. 2005. Chronic hyper-prolinemia provokes a memory deficit in the Morris water maze task.Metab Brain Dis 20:73–80.

656 Li et al.

Bilder RM, Goldman RS, Robinson D, Reiter G, Bell L, Bates JA, et al. 2000.Neuropsychology of first-episode schizophrenia: Initial characterizationand clinical correlates. Am J Psychiatry 157:549–559.

Bilder RM, Volavka J, Czobor P, Malhotra AK, Kennedy JL, Ni X, et al. 2002.Neurocognitive correlates of the COMT Val(158)Met polymorphism inchronic schizophrenia. Biol Psychiatry 52:701–707.

Caspi A, Moffitt TE, Cannon M, McClay J, Murray R, Harrington H, TaylorA, Arseneault L, Williams B, Braithwaite A, Poulton R, Craig IW. 2005.Moderation of the effect of adolescent-onset cannabis use on adultpsychosis by a functional polymorphism in the catechol-O-methyltrans-ferase gene: Longitudinal evidence of a gene X environment interaction.Biol Psychiatry 57:1117–1127.

Dudbridge F. 2003. Pedigree disequilibrium tests for multilocus haplotypes.Genet Epidemiol 25:115–221.

Egan MF, Goldberg TE, Kolachana BS, Callicott JH, Mazzanti CM, StraubRE, Goldman D, Weinberger DR. 2001. Effect of COMT Val108/158 Metgenotype on frontal lobe function and risk for schizophrenia. Proc NatlAcad Sci USA 98:6917–6922.

Fan JB, Ma J, Zhang CS, Tang JX, Gu NF, Feng GY, St Clair D, He L. 2003. Afamily-based association study of T1945C polymorphism in the prolinedehydrogenase gene and schizophrenia in the Chinese population.Neurosci Lett 338:252–254.

Faraone SV, Seidman LJ, Kremen WS, Toomey R, Lyons MJ, Tsuang MT.1996. Neuropsychological functioning among the elderly nonpsychoticrelatives of schizophrenic patients. Schizophr Res 21:27–31.

Glaser B, Moskvina V, Kirov G, Murphy KC, Williams H, Williams N, OwenMJ, O’Donovan MC. 2006. Analysis of ProDH, COMT and ZDHHC8 riskvariants does not support individual or interactive effects on schizo-phrenia susceptibility. Schizophr Res 87:21–27.

Gogos JA, Santha M, Takacs Z, Beck KD, Luine V, Lucas LR, Nadler JV,Karayiorgou M. 1999. The gene encoding proline dehydrogenasemodulates sensorimotor gating in mice. Nat Genet 21:434–439.

Gottesman II, Gould TD. 2003. The endophenotype concept in psychiatry:Etymology and strategic intentions. Am J Psychiatry 160:636–645.

Hoda F, Nicholl D, Bennett P, Arranz M, Aitchison KJ, al-Chalabi A, KunugiH, Vallada H, Leigh PN, Chaudhuri KR, Collier DA. 1996. No associationbetween Parkinson’s disease and low-activity alleles of catechol O-methyltransferase. Biochem Biophys Res Commun 228:780–784.

Lachman HM, Papolos DF, Saito T, Yu YM, Szumlanski CL, WeinshilboumRM. 1996. Human catechol-O-methyltransferase pharmacogenetics:Description of a functional polymorphism and its potential application toneuropsychiatric disorders. Pharmacogenetics 6:243–250.

Lewis CM, Levinson DF, Wise LH, DeLisi LE, Straub RE, Hovatta I,Williams NM, Schwab SG, Pulver AE, Faraone SV, Brzustowicz LM,Kaufmann CA, Garver DL, Gurling HM, Lindholm E, Coon H, MoisesHW, Byerley W, Shaw SH, Mesen A, Sherrington R, O’Neill FA, Walsh D,Kendler KS, Ekelund J, Paunio T, Lonnqvist J, Peltonen L, O’DonovanMC, Owen MJ, Wildenauer DB, Maier W, Nestadt G, Blouin JL,

Antonarakis SE, Mowry BJ, Silverman JM, Crowe RR, Cloninger CR,Tsuang MT, Malaspina D, Harkavy-Friedman JM, Svrakic DM, BassettAS, Holcomb J, Kalsi G, McQuillin A, Brynjolfson J, Sigmundsson T,Petursson H, Jazin E, Zoega T, Helgason T. 2003. Genome scan meta-analysis of schizophrenia and bipolar disorder. Part II. Schizophrenia.Am J Hum Genet 73:34–48.

Lezak MD. 1995. Neuropsychological assessment, 3rd edition. New York:Oxford University Press, Inc.

Li T, Ball D, Zhao J, Murray RM, Liu X, Sham PC, Collier DA. 2000. Family-based linkage disequilibrium mapping using SNP marker haplotypes:Application to a potential locus for schizophrenia at chromosome 22q11.Mol Psychiatry 5:77–84.

Li T, Ma X, Sham PC, Sun X, Hu X, Wang Q, Meng H, Deng W, Liu X, MurrayRM, Collier DA. 2004. Evidence for association between novel poly-morphisms in the PRODH gene and schizophrenia in a Chinesepopulation. Am J Med Genet Part B 129B:13–15.

Liu H, Heath SC, Sobin C, Roos JL, Galke BL, Blundell ML, Lenane M,Robertson B, Wijsman EM, Rapoport JL, Gogos JA, Karayiorgou M.2002. Genetic variation at the 22q11 PRODH/DGCR6 locus presents anunusual pattern and increases susceptibility to schizophrenia. Proc NatlAcad Sci USA 99:3717–3722.

Ma X, Wang Q, Sham PC, Liu X, Rabe-Hesketh S, Sun X, Hu J, Meng H,Chen W, Chen EYH, Deng W, Chan RCK, Murray RM, Collier DA, Li T.2007. Neurocognitive deficits in first-episode schizophrenic patientsand their first-degree relatives. Am J Med Genet Part B 144B:407–416.

Malhotra AK, Kestler LJ, Mazzanti C, Bates JA, Goldberg T, Goldman D.2002. A functional polymorphism in the COMT gene and performance ona test of prefrontal cognition. Am J Psychiatry 159:652–654.

Munafo MR, Bowes L, Clark TG, Flint J. 2005. Lack of association of theCOMT (Val158/108 Met) gene and schizophrenia: A meta-analysis ofcase-control studies. Mol Psychiatry 10:765–770.

Riley BP, McGuffin P. 2000. Linkage and associated studies of schizophre-nia. Am J Med Genet 97:23–44.

Shifman S, Bronstein M, Sternfeld M, Pisante-Shalom A, Lev-Lehman E,Weizman A, Reznik I, Spivak B, Grisaru N, Karp L, Schiffer R, Kotler M,Strous RD, Swartz-Vanetik M, Knobler HY, Shinar E, Beckmann JS,Yakir B, Risch N, Zak NB, Darvasi A. 2002. A highly significantassociation between a COMT haplotype and schizophrenia. Am J HumGenet 716:1296–1302.

Toulopoulou T, Morris RG, Rabe-Hesketh S, Murray RM. 2003. Selectivity ofverbal memory deficit in schizophrenic patients and their relatives. Am JMed Genet Part B 116B:1–7.

Williams HJ, Williams N, Spurlock G, Norton N, Ivanov D, McCreadie RG.2003. Association between PRODH and schizophrenia is not confirmed.Mol Psychiatry 8:644–645.

Winterer G, Goldman D. 2003. Genetics of human prefrontal function. BrainRes Rev 43:134–163.

PRODH Gene and Executive Function 657