Abstracts S96 229. ASSOCIATION BETWEEN ALLELIC VARIATION the U.K. blood transfusion service. Genotyping was done IN THE SEROTONIN TRANSPORTER GENE AND PER- on an ABI 373 DNA sequencer using the ABI GENESCAN VASIVE DEVELOPMENTAL DISORDER. 1Zwaigenbaum and GENOTYPER software. Patient and control data were L, 2Mejia J, 1Szatmari P, 2Palmour R, 3Jones MB, 2Paliouras analysed by CLUMP vs 1.1 and 2 × 2 contingency tables G, 1Bryson SE, 1Bartolucci G, 1Mahoney W.. 1Centre for were analysed by Pearson chi square test. Studies of Children at Risk, Department of Psychiatry, Alleles containing 3 and 4 repeats were most fre- McMaster University, Hamilton, Ontario; 2Department of quent, making up 35% and 62% of the sample respect- Human Genetics and Psychiatry, McGill University, Mon- ively. No significant differences were observed in allele treal, Quebec; 3Department of Behavioural Science, Penn frequency between patients and controls, p=0.188. When State, Hershey, Pennsylvania. the alleles were grouped into high and low expression, Serotonin has long been suggested to play a role in again no significant differences were observed, ␹2=1.81, the pathophysiology of autism/pervasive developmental p=0.25. Male and female data were also analysed separ- disorder (PDD). Cook and colleagues (1997) first reported ately (male, n=108; female, n=46), with no significant dif- an preliminary evidence of linkage and association ferences. between autism and the serotonin transporter (5HTT) gene (short variant of the second intron of the promoter region). Subsequent efforts have failed to replicate this finding, 1 Sabol et al. (1998): Hum Genet 103: 273–279 although one group has found preferential transmission of 2 Williams et al. (1998): Am J Med Genet (Neuropsychiatric Genet) the opposite (long) variant in a PDD sample (Klauck et al., 81: 453 1997). The purpose of this study is to further assess 5HTT 3 Norton et al. (1998): Am J Med Genet (Neuropsychiatric Genet) as a candidate gene in trios consisting of probands with 81: 529 PDD and both parents. We have collected DNA samples from 116 families for TDT analysis. Probands were diag- nosed with PDD by consensus best estimate. Preliminary analysis considreted on 76 families shows a non-signifi- 231. EXCESS OF LONG MONOAMINE OXIDASE A cant trend towards preferential transmission of the long GENE PROMOTER ALLELES IN FEMALE PATIENTS variant of the 5HTT promoter to PDD probands (TDT chi- = = = WITH PANIC DISORDER: FUNCTIONAL CHARAC- squared 2.13, d.f. 1, p 0.14). We will complete the TERIZATION AND REPLICATION. Deckert J, Catalano M, analysis on the entire sample, and assess for genetic het- Syagailo YV, Bosi M, Okladnova O, DiBella D, No¯then erogeneity on the basis on gender, I.Q., and PDD subtype MM, Maffei P, Franke P, Fritze J, Maier W, Propping P, (autism vs. non-autism). Beckmann H, Bellodi L and Lesch KP.. Departments of Psychiatry, Universities of Wu¨ rzburg and Bonn, Institute Cook EH Jr, Courchesne R, Lord C et al. (1997). Evidence of linkage between the serotonin transporter and autistic disorder. Molecular of Human Genetics Bonn, Germany and IRCCS H San Raf- Psychiatry, 2, p247–250. faele Milan, Italy. Klauk SM, Poustka F, Benner A et al. (1997). Serotonin transporter Increased frequency of the longer of several monoam- gene variants associated with autism? Human Molecular Genetics, ine oxidase A gene promoter alleles has been observed 6, p2233–8. among female patients in a German sample (Am J Med Genet 81/6, 514, 1998). To probe the relevance of this observation we performed a functional characterization of the different promoter alleles and complemented the orig- 3. Monomine Oxidase inal finding by a replication study in an independent Ital- ian sample. Panic disorder was diagnosed according to DSMIII-R 230. NO EVIDENCE FOR ASSOCIATION BETWEEN using as interviews SADS-LA, CIDI and DIS-R (total n=209, 80 German and 129 Italian). Controls were of the SCHIZOPHRENIA AND MAO-A PROMOTER POLYMOR- = PHISM. Norton N,. Bray NJ, Williams NM, Cardno AC, same ethnic background (total n 190, 80 German and 110 Murphy KC, Jones LA, Sanders RD, McCarthy G, McGuffin Italian). Genotyping was performed as previously P, and Owen MJ.. Department of Psychological Medicine, described. In addition, functional characterization of the University of Wales College of Medicine, Heath Park, Car- different promoter alleles (3, 3a, 4 and 5 repeats) was per- diff, CF4 4XN. formed in SH-SY5Y cells by means of a luciferase assay. Monoamine Oxidase, (MAO-A), is a mitochondrial As in the German sample higher frequency of the longer enzyme which preferentially oxidizes serotonin and nore- alleles (3a, 4 and 5) and of genotypes containing only piniphrine. The gene encoding MAO-A maps to longer alleles was observed among females in the Italian = = Xp11.23–11.4. sample (p 0.01 and p 0.05 respectively). Combining the Sabol et al, 19981, reported a VNTR polymorphism two female subsamples increased significance to p=0.001 within the promoter of MAO-A, whereby alleles with 3.5 for the allele distribution and p=0.01 for the genotype dis- and 4 repeat elements produced a significantly higher tribution. Functional characterization demonstrated transcription rate than alleles containing 3 and 5 repeat approximately 50% higher activity of the longer alleles in elements as determined by gene fusion and transfection comparison with the short allele. experiments. MAO-A inhibitors are an effective treatment for Following the highest lod score of 1.6 in the region patients with panic disorder, in particular females. This Xp11 from our schizophrenia affected sib pair genome clinical observation is consistent with our finding that scan2,3, we tested for allelic association of the MAO-A homozygosity for monoamine oxidase A gene promoter VNTR polymorphism and schizophrenia. 154 caucasian, alleles with high expression of monoamine oxidase A may DMS-IV schizophrenia patients from the U.K. were age constitute a vulnerability factor in the pathogenesis of and sex matched with 154 control subjects collected from panic disorder. Abstracts S97 232. GENOTYPE OF THE MAO A PROMOTER-ASSO- with 4 copies of the allele, (Fisher’s PSLD-test; p = 0.026, CIATED VNTR AND CELLULAR MAO A ACTIVITY IN p = 0.018, p = 0.010, respectively). HUMAN MALE SKIN FIBROBLASTS. Denney RM, Koch The present results suggest that the structure of the HG and Craig IW.. MAOA promoter might be one factor underlying the SGDP Research Centre, Institute of Psychiatry, De expression of certain personality characteristics in Cauca- Crespigny Park, Denmark Hill, London, UK and Depart- sian males, possibly by influencing transcription of the ment of HBC and G, University of Texas Medical Branch, MAOA gene, and thereby the amount of active enzyme Galveston, TX. AVNTR located 1.2 kb upstream from molecules. MAO A exon 1 (uVNTR) is reportedly associated with panic disorder and compulsive eating disorder in females, and antisocial behavior in male alcoholics. Two labora- 234. ASSOCIATION ANALYSIS BETWEEN A FUNC- tories have shown that the three repeat allele at this locus TIONAL VNTR POLYMORPHISM IN THE PROMOTER confers lower promoter activity than the four repeat allele REGION OF MONOAMINE OXIDASE-A GENE AND when tested by transient expression in cultured cell lines. MOOD DISORDERS. Ishida S1, Kunugi H1, Kato T2, Tats- We report that uVNTR genotype has a quantitatively simi- umi M3, Sakai T3, Hattori M1, Hirose T1, Nanko S1. lar effect on cellular MAO A activity in cultured human 1Department of Psychiatry, Teikyo University School of skin fibroblasts. MAO A specific activity was significantly Medicine, Tokyo, Japan 2Department of Neuropsychiatry, lower in cultures with three repeats than that in cultures Faculty of Medicine, University of Tokyo, Tokyo, Japan with four (1.6 ± 1.1 and 13 ± 12 nmol/hr/mg, respectively; 3Department of Psychiatry, Showa University School of P = 0.032). Immunofluorescence microscopy and flow Medicine, Tokyo, Japan. cytometry showed that MAO A expression was confined Sabol and colleagues (Hum Genet 103: 273–9, 1998) to a cell sub population varying in size from 0.5–85% of identified a common polymorphism of a variable number cells in different cultures. This variability was not associa- tandem repeat (VNTR) in the promoter region of the ted with genotype, and generated additional variation in monoami oxidase-A (MAOA) gene, which was shown to activity within genotypes and overlap between genotypes. be associated with MAOA transcriptional activity. We Mean specific activity in MAO A+ cells (whole culture examined whether this functional polymorphism is asso- specific activity divided by the proportion of immunopos- ciated with the risk of developing mood disorders in a itive cells) again was lower for cells with three repeats Japanese sample of 161 patients with bipolar disorder, 98 than for cells with four (7.2 ± 3.1 and 23.9 ± with unipolar depression, and 258 controls. For the par- 9.5 nmol/hr/mg protein, respectively; P = 0.0013), with ticipation in the study, written informed consent was much less variation within and no overlap in activity obtained from each subject. We found a novel allele which between genotypes. Finding low fibroblast MAO A was not described by Sabol et al (1998), although its fre- activity in cells with three repeats is consistent with tran- quency was rare. Excluding this allele from the analyses, sient expression data, and suggests that uVNTR genotype there was no significant genotypic or allelic association, may be a major source of individual differences in sero- suggesting that the functional VNTR polymorphism in the tonin and norepinephrine oxidizing capacity. MAOA gene is unlikely to play a major role in the patho- genesis of bipolar disorder or unipolar depression in the Japanese sample. In addition, we found no significant 233. MAOA PROMOTER GENOTYPE IS RELATED TO association between the polymorphism and a history of PERSONALITY TRAITS. Garpenstrand H1, Damberg M1, suicide attempt. Hallman J1, Forslund K2, Rylander G2, sberg M2, and Ore- land L1.. 1Department of Neuroscience, Section of Pharma- cology, Uppsala University, Sweden 2Department of Clini- 235. MONOAMINE OXIDASE A PROMOTER POLY- cal Neuroscience, Karolinska Hospital, Stockholm, MORPHISM AND RELATIONSHIPS TO MONOAMINE Sweden. METABOLITE CONCENTRATIONS IN CSF. Jo¯nsson EGa, Abnormalities in monoamine oxidase type A (MAO- Norton Nb, Gustavsson JPa, Oreland Lc, Owen MJb, Sedvall A) have been connected with deviating behavior and psy- GCa. aDepartment of Clinical Neuroscience, Psychiatry chiatric disorders. The gene encoding MAO-A is located Section, Karolinska Institute, Stockholm, Sweden; bInsti- on the X chromosome and contains a polymorphic region tute of Medical Genetics, University of Wales College of with a variable number of a 30 bp repeat sequence which Medicine, Heath Park, Cardiff, UK; cDepartment of Neuro- is located approximately 1.2 kb upstream of the coding science, Section of Pharmacology, University of Uppsala, sequence. It has been reported that differences in this Uppsala, Sweden. polymorphism affects the transcriptional activity of the Concentrations of monoamine metabolites (MM) in MAOA gene. Thus, this variable region of the MAOA pro- lumbar cerebrospinal fluid (CSF) have been used exten- moter could be involved in the underlying mechanisms sively as indirect estimates of monoamine turnover in the for inter-individual differences in MAO-A catalytic brain. We investigated possible relationships between a activity, and may be involved in the expression of person- putative functional promoter polymorphism in the ality traits. monoamine oxidase A (MAOA) gene and CSF concen- In the present study, we investigated if MAOA pro- trations of homovanillic acid (HVA), 5-hydroxyindolea- moter genotype is associated with personality character- cetic acid (5-HIAA), and 3-methoxy-4-hydoxyphenylgly- istics as estimated by the Karolinska Scales of Personality col (MHPG) in healthy volunteers (n=86). Women carrying (KSP). The MAOA promoter allele status and KSP scores at least one of the more efficiently transcribed alleles dis- were determined for 55 healthy, non-smoking, Cauca- played higher concentrations of HVA (p=0.01) and 5- sian males. HIAA (p=0.01). However, men with the more efficiently Individuals with 3 copies of the 30-repeat allele dis- transcribed alleles tended to have lower HVA-levels played significantly lower KSP scores for muscular ten- (p=0.07). The results suggest that MAOA genotypes may sion, guilt, and monotony avoidance than individuals participate differentially in the regulation of dopamine Abstracts S98 and serotonin turnover rate under presumed steady state wanted to analyze if this variant is also involved in the in the central nervous system. etiology of major depressive disorder. Method: We perfor- med a genetic association study including 150 patients who fulfilled DSM-IV criteria for major depressive dis- 236. NO ASSOCIATION OF THE MAOA GENE WITH order (43 single episode and 107 recurrent major ANTISOCIAL PERSONALITY DISORDER. Lu RB1*, Lee depression) and 229 controls. Genetic analyses were car- JF2,KoHC3, Chen K4, Shih JC4. 1Department of Psychiatry, ried out according to the protocol of Deckert et al2. Results National Defense Medical Center, Tri-Service General and Discussion: Neither in the overall sample nor in the Hospital, Taipei, Taiwan, R.O.C. 2Armed Forces Pei-Tou sub-samples of patients with recurrent major depression Hospital, Military Psychiatry Center, Taipei, Taiwan, and a single episode of major depression, was the distri- R.O.C. 3Institute of Behavioral Medicine, National Cheng bution of alleles different from the control group. This Kong University Medical College, Tainan, Taiwan, R.O.C. held true when females and males were analyzed separ- 4Department of Molecular Pharmacology and Toxicology, ately. School of Pharmacy, University of Southern California, Thus, our results do not support the hypothesis that CA, U.S.A. the MAO-A gene is involved in the etiology of depress- There is evidence that MAOA gene might be related ive disorders. to aggressive behavior in mice and humans (Cases et al. 1995, Brunner et al. 1993). Aggressive behavior is one of 1 Sabol SZ, Hu S, Hamer D (1998) A functional polymorphism in the monoamine oxidase A gene promoter. Hum Genet; 103; 273–279 the major symptoms of antisocial personality disorder. 2 Deckert J, Catalano M, Syagailo YV, Bosi M, Okladnova O, Di Bella This study examines whether there is an association D, No¨then MM, Maffei Piermario, Franke P, Fritze J, Maier W, between antisocial personality disorder and an allele of Propping P, Beckmann H, Bellodi L, Lesch KP (1999) Excess of high variable number of tandem repeat located upstream activity monoamine oxidase A gene promoter alleles in female (uVNTR), and EcoRV polymorphisms, both individually patients with panic disorder. Hum Mol Genet; 8(4); 621–624 and as haplotypes, at the MAOA gene. From Taipei Jail, 164 criminals, including 122 cases meeting DSM-IV cri- FUNCTIONAL MONOAMINE OXIDASE A GENE teria for antisocial personality disorder and 42 cases not 238. PROMOTER ALLELES: A CASE-CONTROL STUDY IN meeting for antisocial personality disorder, were sampled MANIC-DEPRESSIVE DISEASE, MONOPOLAR for this study. In addition, 74 individuals with no crimi- DEPRESSION AND CYCLOID PSYCHOSIS. Gerald Stober, nal history were recruited as controls from the community ¨ Yana V. Syagailo, Olga Okladnova, Jobst Meyer, Helmut of Taipei. All of the subject are Chinese Han males. Sig- Beckmann, Klaus-Peter Lesch. Department of Psychiatry, nificant linkage disequilibrium occurs for the two poly- University of Wurzburg, Fuchsleinstr, 15, 97080 Wurz- morphic sites. No significant intergroup differences were ¨ ¨ ¨ burg, Germany. observed among the three subject groups on the allele fre- A possible dysregulation of monoamine oxidase quencies of the two polymorphisms at the MAO locus, (MAO) genes has been implicated in the etiology of affect- tested individually and as haplotypes. This suggests that ive disorders as well as acute schizophreniform no association exists between genetic variation at the psychoses. The MAO-A gene is localized on chromosome MAOA locus and antisocial personality disorder in Chi- Xp. Four alleles of the MAO-A gene promoter polymor- nese Han criminals. phism were observed (3, 3a, 4 and 5 30-bp repeats). Using luciferase assays, transcriptional activity of smaller MAO- A promoter alleles (containing three repeat units) has pre- 237. NO ASSOCIATION BETWEEN A FUNCTIONAL viously been shown to be significantly reduced. In the POLYMORPHISM IN THE MONOAMINE OXIDASE PRO- present study we investigated allelic and genotypic vari- MOTER REGION AND MAJOR DEPRESSIVE DISORDER ations of the functional repeat polymorphism within the Schulze TG1,. Muller DJ1, Scherk H1, Syagailo YV2, Wind- ¨ MAO-A gene promoter in a case-control association study: emuth Csp 3, Neidt H4, Stober G2,Nothen MM4, Maier W1, ¨ ¨ 229 controls (134 males), 155 patients with affective dis- Lesch KP2, Rietschel M1. 1Department of Psychiatry, Uni- order (manic-depressive disease or monopolar depression versity of Bonn, Sigmund-Freud-Str. 25, D-53105 Bonn, according to Leonhard’s classification; 66 males) and 82 Germany 2Department of Psychiatry, University of Wurz- ¨ patients with cycloid psychosis (39 males). Allele fre- burg, Fuchsleinstr. 15, D-97080 Wurzburg, Germany ¨ ¨ quencies between smaller, low activity (three repeat units) 3Institute of Medical Statistics, University of Bonn, Sig- and longer, high activity variants did not differ between mund-Freud-Str. 25, D-53105 Bonn, Germany 4Institute of controls and affective psychosis (␹2= 1.22; p= 0.27) and Human Genetics, University of Bonn, Wilhelmstr. 31, D- cycloid psychosis (␹2= 1.75; p= 0.18). However, patients 53111 Bonn, Germany. with cycloid psychosis showed a higher frequency of Monoamine oxidase A (MAO-A) is one of the key smaller alleles (42%) compared to those with affective enzymes in the metabolism of monoaminergic neuro- psychosis (30%; ␹2= 4.49; p= 0.03). These findings suggest transmitters which play an important role in the etiology that reduced MAO-A activity may be a possible risk factor of affective disorders. Moreover, MAO-A inhibitors are for cycloid psychosis. effective in the pharmacotherapy of depressive syn- dromes. Thus, the X-chromosomal MAO-A gene can be considered a potential candidate gene for affective dis- 239. GENETIC ASSOCIATION STUDY BETWEEN THE orders. Several variants of the MAO-A gene have been pre- MONOAMINE OXIDASE GENE AND BIPOLAR AFFECT- viously described. IVE DISORDERS IN SINGAPOREAN CHINESE. Tut TG, Recently, a novel functional repeat polymorphism in Wang JL, Perera I, Lim CCL. Department of Psychological the promoter of the MAO-A gene has been reported, with Medicine National University Hospital, National Univer- the longer alleles (3a, 4, 5) being more active than the sity of Singapore. shorter one (3)1,2. Deckert et al.2 reported the longer alleles Monoamine oxidase (MAO) is an important enzyme to be associated with panic disorder in females. We in human physiology and behavior and also a critical Abstracts S99 enzyme in degradative deamination of biogenic amines. lation of some aspects of central Executive Function, MAO may be involved in the pathophysiology of major particularly dual attentional processes. psychoses, including mood disorders and schizophrenia. The genetic association between the MAO gene and bipolar affective disorder was reported to be present in 241. OBSESSIVE COMPULSIVE DISORDER IN THE Caucasians and Japanese patients. We investigated a gen- AFRIKANER POPULATION: AN ASSOCIATION STUDY etic association between bipolar affective disorder and the WITH AN NlaIII POLYMORPHISM IN THE COMT GENE alleles of monoamine oxidase A and B (MAOA and Kinnear CJ1,. Weyers JB1, Niehaus DJH, Van Kradenberg J, MAOB) using the case-control method. Replication is Moolman-Smook JC1, Corfield VA1, Brink PA1, du Toit W, essential in case-control method. It is necessary to confirm Els C, Emsley RA, Knowles JA2, Stein DJ.. MRC Unit for the suggestion of alleles at the MAOA locus contribute to Anxiety and Stress disorders, University of Stellenbosch, bipolar affective disorder because the real genetic associ- 1MRC/US Centre for Molecular and Cellular Biology, Uni- ation might be a chance finding, as has frequently versity of Stellenbosch, 2Department of Psychiatry, Uni- occurred with genetic case-control studies. Seventy-seven versity of Columbia. Singaporean Chinese bipolar affective disorder patients Obsessive-compulsive disorder (OCD) is a common and seventy-four healthy normal population controls were psychiatric disorder associated with repetitive intrusive genotyped with two markers on the MAOA gene and one thoughts (images) and repetitive behaviours designed to marker on the MAOB gene. The Fnu4HI polymorphism at neutralize the egodystonic thought content. The suffering the monoamine oxidase A (MAOA) gene (MAOA locus at and significant impairment in work and social situations

Xp 11.23–11.4), MAOA-CA and MAOB-GT repeat poly- make early diagnosis and treatment essential. Biological morphisms were studied in 77 Chinese Bipolar affective markers for subsets of OCD may allow more specific treat- disorders (27 males and 50 females) and 74 healthy Chi- ment based on the presence or absence of these factors, nese controls (39 males and 35 females). Analyses did not and permit early detection of subclinical OCD. show significant allelic association between MAO markers Previously, an allele of a polymorphism in the cat- and bipolar affective disorder in Singaporean Chinese. echolamine-O-methyltransferase gene (COMT), which has These data do not support genetic association between been associated with a 3–4 fold reduction in enzyme MAO gene and bipolar affective disorder. activity, has been implicated in OCD in a Mixed North American population (1). The Afrikaner population, a genetically homogeneous group, provides a suitable study 4. Catechol-O-Methyl-Transferase population in which to investigate this association. The sample consisted of 49 OCD patients of Afrikaner descent and 47 ethnically matched controls. Patients and 240. GENETIC AND COGNITIVE FACTORS IN HYPNO- controls underwent a structural clinical interview (SCID, TIZABILITY: ASSOCIATION BETWEEN THE LOW Mini) and diagnosis was based on DSM-IV criteria. The ENZYME ACTIVITY CATECHOL O- METHYL TRANS- region containing the COMT polymorphism was PCR- FERASE (COMT) MET ALLELE AND HIGH HYPNOTIZ- amplified from genomic DNA obtained from all individ- ABILITY Ebstein RP, Bachner-Melman R & Lichtenberg P. uals, the amplified samples digested with NlaIII, and Herzog Hospital P.O. Box 35300, Jerusalem 91351, Israel. digestion products electrophoresed on polyacrylamide Only recently have studies of electrocortical activity, gels. Bands were visualized by silver staining. ␹2 event-related potentials, and regional cerebral blood flow analysis of the distribution of COMT genotypes begun to shed light on the anatomical and neurobiological (high-high; high-low and low-low) did not support an underpinnings of hypnosis. Since twin studies show a sig- association between the low activity allele of the COMT = nificant heritable component for hypnotizability we were gene and OCD (p not significant). prompted to examine the role of a common, functional This contradicts previous findings (1). This may indi- polymorphism in contributing to individual differences in cate that the COMT gene is not involved in OCD, or that hypnotizability. A group of 109 male and female subjects OCD is a heterogeneous disorder in which contributions were administered three psychological instruments and from multiple gene, which may differ from population to inventoried for the high/low enzyme activity COMT val→- population, are involved. Investigation of biological mark- met polymorphism. We observed a significant correlation ers in different homogeneous populations may help to between hypnotizability measured by the Stanford Hyp- clear up the underlying mechanism. notic Susceptibility Scale (SHSS:C), ability to partition 1 Maria Karayiogou, Margaret Altemus, Brandi L. Gallie, David Gold- attention (Differential Attentional Processes Inventory or men, Dennis L Murphy, Jurg Ott and Joseph A Gogos (1997). Geno- DAPI) and absorptive capacities (Tellegen Absorption type determining low catechol-O-methyltransferase activity as a risk Scale or TAS). A significant effect of COMT on hypnotiz- factor for obsessive compulsive disorder. ability was observed when SHSS:C total hypnosis scores are parsed by COMT genotype (P=0.002) or by the pres- ence or absence of the low enzyme activity COMT met 242. QTL LINKAGE ANALYSIS BETWEEN POLYMOR- allele (P=0.001). Individuals with the low COMT met PHISMS IN THE COMT GENE AND SCHIZOPHRENIA IN allele are more hypnotizable. Three of the four subscales A CHINESE POPULATION. Li T1,2, Liu XH2, Sham PC1, of the DAP1 that correlate with hypnotizability also show Cai G2, Zhao J1, Murray RM1 and Collier DA1. 1Department a trend for the low COMT met allele to be associated with of Psychological Medicine, Institute of Psychiatry, Lon- higher scores indicating extended ability in these individ- don SE5 8AF, UK. 2Department of Psychiatry, West China uals to divide attention. No significant relationship was University of Medical Sciences, Sichuan 610041, P R observed between the COMT polymorphism and TAS. We China. suggest that the role of the low enzyme activity COMT A number of lines of evidence suggest that COMT met allele in accounting for 10% of the variance in hypno- gene is involved in the causes of psychoses and the regu- tizability scores is likely mediated by this gene’s modu- lation behaviour. In this study, sib-pair analysis were car-