Abstract
VII. Bipolar Disorder Genome‐Scans and Overlap With Schizophrenia
O55 DIFFERENT INHERITANCE MODELS BY AGE OF ONSET IN BIPOLAR I DISORDER
Grigoroiu‐Serbanescu M^1^, Martinez M^2^, Nöthen MM^3^, Grinberg M^4^, Sima D^4^, and Propping P^5^
^1^Biometric Psychiatric Genetics Research Unit, Alexandru Obregia Psychiatric Hospital, Sos. Berceni, 10, O.P. 8 R‐75622, Bucharest, Romania, Phone: 40‐1‐332.39.29; 40‐1‐683.57.62; Fax: 40‐1‐334.71.64; E‐mail: [email protected]
^2^I.N.S.E.R.M., Unité 358, EPI 06, Paris, France
^3^Department of Medical Genetics, University of Antwerp, Belgium
^4^Biometric Psychiatric Genetics Research Unit, Alexandru Obregia Psychiatric Hospital, Bucharest, Romania
^5^Institute of Human Genetics, University of Bonn, Germany
In bipolar affective disorder, where the majority of linkage studies have produced conflicting results, studies reporting clinical characteristics and familial occurrence of disease have suggested that age of onset might serve as an indicator for identifying more homogenous subgroups of disease. Our study was the first to examine this hypothesis by the means of segregation analysis. We investigated a sample of 177 bipolar I probands recruited from consecutive admissions and their first‐ and second‐degree relatives (2,407 subjects). Probands were subdivided into an early‐onset (N=107) and a late‐onset group (N=70) using an age of onset of 25 as a cut‐off point. This age was chosen because the observed age of onset distribution was bimodal with a cut‐off of 25 years. Morbid risks for affective disorder were found significantly higher (P=.01) in relatives of probands with an early‐onset than in probands with late‐onset of disease. The segregation analysis showed that the disease is transmitted differently in early‐ and late‐onset groups. In the early‐onset group a non‐Mendelian major gene with a polygenic component was favored while the data in the late‐onset group were compatible with a multifactorial model. This result may have important implications for molecular studies.
O56 THE RISK FOR SCHIZOPHRENIA AND BIPOLAR DISORDER IN SIBLINGS TO PROBANDS WITH SCHIZOPHRENIA AND BIPOLAR DISORDER
Ösby U, Brandt L, and Terenius L
Department of Clinical Neuroscience Karolinska Institutet 171 75 Stockholm, Sweden, Phone: 46 70 772 70 93; Fax: 46 8 27 70 76; E‐mail: [email protected]
All patients in Sweden with an inpatient diagnosis of schizophrenia or bipolar disorder from 1973 to 1995 were identified from the Swedish patient register. All siblings were identified by the second‐generation register and their inpatient diagnoses were determined from the patient register. Standardized incidence ratios (SIR) for full and half siblings were calculated in 5‐year age and calendar time classes. There were 13,870 schizophrenia probands with 23,223 full and 8,369 half siblings, and 5,400 bipolar disorder probands with 8,846 full and 2,758 half siblings. In siblings to schizophrenia probands, SIR for schizophrenia was 7.4 for full and 4.4 for half siblings, and 3.6 for full and 2.8 for half siblings for bipolar disorder. In siblings to bipolar probands, SIR for bipolar disorder was 12.8 for full and 8.1 for half siblings, and 4.4 for full and 2.2 for half siblings for schizophrenia. If both parents were affected, the risk increased for full siblings in both schizophrenia and bipolar disorder. One affected parent increased the risk in bipolar disorder only. When the first admission for the proband was before age 25, the risk increased for schizophrenia in full siblings to schizophrenia probands but not for bipolar disorder in full siblings to bipolar probands.
O57 A SEARCH FOR SPECIFIC AND COMMON SUSCEPTIBILITY LOCI FOR SCHIZOPHRENIA AND BIPOLAR DISORDER
Mérette C, Phaneuf D, Fournier A, Roy MA, Cliche D, Dion C, and Maziade M
Centre de recherche Université Laval Robert‐Giffard, 2601, de la Canardière Beauport, PQ G1J 2G3 Canada, Phone: 418‐663‐5741; Fax: 418‐663‐9540; E‐mail: [email protected]
Schizophrenia (SZ) and bipolar disorder (BP) are prevalent major psychoses underlain by complex genetic components. To identify the susceptibility loci contributing to these disorders, we have undertaken a two‐stage genome wide scan on 480 individuals from 21 multigenerational pedigrees of Eastern Québec. Here we report the second stage based on 220 microsatellite markers. In addition to testing susceptibility loci specific to each disorder, we also tested the hypothesis that some susceptibility loci might be common to both SZ and BP using an affection status that included both disorders. Two‐point and multipoint model‐based linkage analyses were performed and the resulting mod scores will be reported. In the first stage of the genome scan targetting 13 candidate chromosomes, the strongest linkage signals were detected at D18S1145 (in 18q12; Lod=4.03) for BP, and at D6S334 (net Lod=3.47; theta=0.66) for SZ. The 18q12 result met the Lander & Krugliak (1995) criterion for a genome wide significant linkage and, moreover, provided support for a susceptibility region that may overlap SZ and BP. Three other chromosomal areas (3q, 10p, and 21q) yielded positive linkage signals. Chromosomes 4p, 5q, 6q, 8p, 11q, and 22q showed no evidence of linkage.
O58 ASSOCIATION OF CAG REPEAT LOCI ON CHROMOSOME 22 WITH SCHIZOPHRENIA AND BIPOLAR DISORDER
Jain S, Saleem QP, Dash D, Gandhi C, Benegal V, Mukherjee O, and Brahmachari SK
Department of Psychiatry, Molecular Genetics Laboratory, National Institute of Mental Health and Neuro‐Sciences, Hosur Road, Bangalore, Karnataka 560029 India, Centre for Biochemical Technology, Delhi University Campus, Mall Road, Delhi 110007
Chromosome 22 has been implicated in schizophrenia and bipolar disorder in a number of studies. CAG repeat expansion may also be involved in these diseases. To explore the involvement of CAG repeats on Chr.22, we created an integrated map of all CAG repeats >5 on this chromosome together with microsatellite markers associated with these diseases. Of the 52 CAG repeat loci identified, four repeat stretches in regions previously implicated by linkage analyses were chosen for further study. Three of the four repeat containing loci were found in the coding region with the CAG repeats coding for glutamine, and were expressed in the brain. All the loci studied showed varying degrees of polymorphism, and one locus had two alleles of 7 and 8 CAG repeats. The 8 repeat allele was significantly over represented in patient groups when compared to ethnically matched controls, while alleles at the other three loci did not show any difference. The repeat lies within a gene that shows homology to an androgen receptor related apoptosis protein in rat. We also identified other candidate genes in the vicinity of this locus. Our results suggest that the repeats within this gene or other genes in the vicinity of this locus are likely to be implicated in bipolar disorder and schizophrenia.
O59 LINKAGE ANALYSIS USING QUANTITATIVE PHENOTYPES IN BIPOLAR DISORDER: A GENOME SCAN OF A SIB‐PAIR SAMPLE
O'Mahony E, Corvin A, Craddock N, and Gill M
Dept of Psychiatry, Trinity Centre for Health Sciences, St James Hospital Dublin 8, Ireland, Phone: 353 1 608 2465; Fax: 353 1 608 3405; E‐mail: [email protected]
In a previous sibling‐pair study of bipolar illness the authors investigated the degree of familial aggregation of a number of demographic and clinical features: age at onset; frequency of manic and depressive episodes; proportion of manic to depressive episodes; dimension scores for mania, depression, psychosis and incongruence of psychotic symptoms with mood. Of these, intra‐pair Spearman correlations were most significant for dimension scores for psychosis (r=0.332, P<0.001) and age at onset (r=0.293, P<0.001). On the basis of the hypothesis that different aspects of the bipolar phenotype may be primarily influenced by different genes we have sought to apply a quantitative scale to phenotype assignment in our study of familial bipolar illness.
We used 398 highly polymorphic microsatellite markers with an average inter‐marker distance of 9.6cM to genotype all individuals and GENEHUNTER 2.0 was used for non‐parametric analysis of the quantitative phenotype data.
We identified 8 regions, suggestive of linkage for the ‘age at onset’ phenotype; These were on chromosomes 1q, 2p, 3p, 4q, 7p, 10p, 16p and 20p. With regard to the ‘psychosis dimension’ phenotype, we identified 6 regions suggestive of linkage; on 1p, 2p, 5p, 10p, 13q and 18p.
O60 GENOME‐WIDE GENETIC LINKAGE STUDIES IN BIPOLAR DISORDER: A REVIEW
Segurado R and Gill M
Trinity College, University of Dublin, Department of Genetics, Dublin, IE Dublin, 2 Ireland, Phone: 353 1 608 2444; Fax: 353 1 679 8558; E‐mail: [email protected]
Genetic linkage studies are prone to publication bias, as are genome scans which have been published in incomplete form, sometimes before the completion of genotyping and analysis across the entire genome. In order to overview genetic linkage to Bipolar Affective Disorder in an unbiased and objective manner we have reviewed all genome scans, and on the basis of pre‐determined criteria conducted an informal meta‐analysis on data from the eleven complete and independent scans published. Results indicated areas exceeding our set thresholds (Lod>1 or P<0.01) on chromosome 5q35 overlapped in four scans out of six, and taking top three ranked regions, areas on 18p11 which overlapped in three scans out of ten. Simulation of randomly positioned regions indicated that these results do not overlap more than expected by chance. However the methods used are expected to be conservative. Our study provides little justification for fine mapping of positional candidate genes in any chromosomal region, however further more formal meta‐analysis of genome scan data is required. Full publication, perhaps on the Web, of raw data, would greatly assist future studies.
O61 GENOME‐WIDE SCAN FOR PREDISPOSING LOCI FOR BIPOLAR DISORDER IN FINNISH FAMILIES; EVIDENCE FOR A LOCUS ON 4q28.3
Ekholm JM, Kieseppa T, Partonen T, Paunio T, Perola M, Lonnqvist J, and Peltonen L
UCLA Department of Human Genetics, Gonda Neuroscience and Genetics Research Center, 695 Charles E. Young Drive South, Room 6309 Los Angeles, CA 90095‐7088, Phone: 310‐794‐9802; Fax: 310‐794‐5446; E‐mail: [email protected]
We have completed the first stage of a two‐stage genome wide scan in a Finnish bipolar disorder family set. In the first stage all the affected subjects from the 41 families were screened with 384 microsatellite markers covering the genome with a ∼10cM resolution. All the families contained at least two siblings that were defined by DSM‐IV as bipolar disorder I or schizoaffective disorder, manic type. We identified one distinct locus on 4q28.3 that gave significant evidence of linkage (Zmax=3.3). Furthermore, one other locus at 12q23.2 gave a LOD score > 3.0 and three loci with a LOD score>2.0 were observed on 1q31.1, 16p11.1 and at Xq25, a locus already earlier identified in one Finnish extended pedigree. Total of thirteen loci; 1q43, 2pq, 3p12.2, 3q13.31, 3q26.31, 5p12, 7p14, 8p22, 8q24.3, 9p21, 11pq, 14q21, 14pq the two‐point LOD score exceeded 1.0. Interestingly, regions of 1q31 (Zmax=2.3; D1S1660), 4q28.3 (Zmax=3.3; D4S1629), 5p13.3 (Zmax=1.6; D5S1470), 12q23.2 (Zmax=3.0 PAH), 16p11.1 (Zmax= 2.9; D16S769) have provided evidence for linkage also in previous studies from other populations. We are currently fine mapping these regions to establish conclusive evidence for their involvements in the genetic background of bipolar disease in Finland.
O62 THE WELLCOME TRUST UK‐IRISH BIPOLAR SIB‐PAIR GENOME SCREEN: SECOND STAGE PROGRESS REPORT: CHROMOSOMES 4, 7 AND X
Middle F, Bennett PP, Jones II, Heron JJ, Gill MM, and Craddock NN
University of Birmingham, Department of Psychiatry, Queen Elizabeth Psychiatric , Birmingham, 0 B15 2QZ UK Phone: 0121 414 3838; Fax: 0121 414 8025; E‐mail: [email protected]
The Wellcome Trust UK‐Irish Bipolar Sib‐pair Study is a two‐stage collaborative genome screen funded by the Wellcome Trust involving the University of Birmingham, UK, and Trinity College, Dublin, Eire. The first stage screening sample included 509 subjects, consisting of 154 narrowly defined affected pairs (DSMIV BPI) and 258 broadly defined affected pairs (DSMIV BPI, BPII, SA BP, BP NOS or MDD(R)). Analysis of genotyping data from ∼ 200 markers (∼ 10 cM spacing) covering the 11 chromosomes under investigation in Birmingham confirmed the presence of 7 chromosomal regions exceeding our second stage inclusion criterion, namely a MLS>0.74 under the narrow diagnostic model. In particular, regions on chromosomes 4 (D4S419‐D4S1597; MLS 0.79 at D4S405), 7 (D7S516‐D7S515; MLS 1.40 at D7S630), and X (DXS1060‐DXS1001; MLS 1.18 at DXS990) have now undergone further detailed study. This has involved both increasing marker density (to ∼ 5 cM) and the addition of a second stage sample set containing 360 further individuals (giving a total of 220 narrowly defined affected pairs and 396 broadly defined affected pairs). An interim analysis, involving all new markers for the first stage sample set as well as a number of markers for the second stage sample set, supports results originally obtained for chromosomes 4 and 7, while those for chromosome X are less consistent. A detailed analysis including all markers and those individuals in the second stage sample set will be presented at the 2001 World Congress.
VIII. PHARMACOGENETIcS
O63 GENETIC PREDICTORS OF OLANZAPINE RESPONSE
Arranz MJ, Staddon S, Mata I, Beperet M, Munro J, Osborne S, and Kerwin RW
Clinical Neuropharmacology, Institute of Psychiatry, London, UK SE5 8AF Phone: 44 (0)20 7848 0343; E‐mail: [email protected]
In a previous study we showed that clinical response to the potent antipsychotic clozapine could be predicted using a combination of genotypes from neurotransmitterreceptors (Arranz et al., 2000). In this study we try to predict response to the drug olanzapine using a similar strategy. Olanzapine is an atypical antipsychotic which resembles clozapine in its affinity for 5‐HT2A, 5‐HT2C, 5‐HT6, Histamine‐1 and Muscarine‐1 receptors. However, olanzapine has a relatively higher affinity for D2 and D3 receptors than clozapine and a lower affinity for Histamine‐2 receptors. We have investigated known genetic variants in these receptors and their contribution to treatment outcome. Preliminary results in a sample of 70 schizophrenic patients of British and Spanish origin and treated with olanzapine show that a combination of polymorphisms in 5‐HT2A, 5‐HT2C, 5‐HT6, and D3 receptor genes and in the serotonin transporter 5‐HTT could result in a level of prediction of 78% (P=ns, sensitivity 75%, specificity=81%). This level of prediction was higher (83%, P= 0.04) on a subset of Spanish patients of Basque origin (Sensitivity=90%, specificity=71%). The validity of these results is being investigated in a larger sample.
O64 PHARMACOGENETIC DISSECTION OF CENTRAL AND PERIPHERAL CANDIDATE GENES IN ATYPICAL ANTIPSYCHOTIC‐INDUCED WEIGHT‐GAIN
Vincenzo S Basile,^*^ Mario Masellis, Vincenzo DeLuca, Herbert Y Meltzer, Jeffrey A Lieberman, and James L Kennedy
Neurogenetics Section, Clarke Division, Centre for Addiction and Mental Health (CAMH), University of Toronto, 250 College Street, Toronto (ON) M5T 1R8, Canada ([email protected])
A drawback to the treatment of schizophrenia with atypical antispsychotics (AAs) is the occurrence of weight gain. Variability exists among individuals regarding this weight gain and genetic predisposition has been suggested. AAs may potentially disrupt both central mechanisms regulating appetite and peripheral mechanisms governing energy expenditure to cause weight gain. Evidence supports a role for the serotonin 2C and histamine H1 receptors in central hypothalamic regulation of food intake and consequently for AA induced weight gain. There is also support for peripheral AA disruption of thermogenesis and metabolic rate via TNFα, β~3~ and α~1~adrenergic receptors. We investigated these candidate genes in 77 patients with DSM‐IIIR diagnoses of schizophrenia that were prospectively assessed for clozapine‐induced weight gain. ANCOVA analyses correcting for covariates were utilized to detect differences among genotypes. Although results were predominantly negative, trends were noted for TNFα (F[2,72]=2.58, P=0.12; means per genotype of 7.4 ± 3.7, 2.3 ± 4.1 and 3.9 ± 4.5) and the α~1~adrenergic receptor (F[2,57]=1.45, P=0.24; Arg/Arg=3.24 ± 3.1; Arg/Cys =4.89 ± 4.9; Cys/Cys=0.85 ± 3.4). These candidate genes may be involved in clozapine induced weight gain, although replication is necessary.
O65 INVESTIGATION OF AN ASSOCIATION BETWEEN A CYP1A2 5′ FLANKING SNP (T‐3591G) AND RESPONSE TO CLOZAPINE
Aitchison KJ, Zhao JH, Munro J, Collier DA, Makoff AJ, and Kerwin RW
Institute of Psychiatry, Section Clinical Neuropharmacology, 1 Windsor Walk, Denmark Hill, London, SE5 8AF UK, Phone: 44 (0)207 848 0034/0793; Fax: 44 (0)207 848 0051/0059; E‐mail: [email protected]
The cytochrome P450 enzyme CYP1A2 is involved in the metabolism of several psychotropic agents, including clozapine (1,2). There is wide interindividual variation in CYP1A2 activity (1), and several polymorphisms in CYP1A2 have been identified, including SNPs in the 5′ flanking region (3).
We have determined the frequency of one of these SNPs (T‐3591G) by PCR‐RFLP analysis (3), in 194 patients treated with clozapine. Prescribing consultants provided information regarding the patients’ response to clozapine. Analysing the genotyping results under a recessive model, Fisher's exact test gave a 1‐tailed P value of 0.16; while analysing the results by allele, Fisher's exact test gave a 1‐tailed P value of 0.10. The results show a trend towards an excess of the mutant (G‐3591) allele in those with a relatively poor response to clozapine. Although this SNP has been shown to be nonfunctional, it is possible that it is in linkage disequilibrium with a functional CYP1A2 polymorphism, and genotyping for this SNP, in addition to other genetic markers, could be useful in predicting response to clozapine.
References:
1 Aitchison et al. Drug Metab Drug Interactions 2000;16: 15–38.
2 Aitchison et al. J Psychopharmacol 2000; 14: 353– 359.
3 Aitchison et al. Pharmacogenetics 2000; 10: 695– 704.
O66 QUANTITATIVE GENETICS OF MONOAMINE METABOLITES IN PEDIGREED BABOONS, PAPIO HAMADRYAS
Rogers J, Comuzzie AG, Martin L, Mann JJ, and Kaplan JR
Southwest Reg. Primate Research Center, 7620 N.W. Loop, 410 San Antonio, TX 78245, Phone: 210‐258‐9532; Fax: 210‐670‐3344; E‐mail: [email protected]
The monoamine neurotransmitters serotonin, dopamine and norepinephrine are related to a number of fundamental neurophysiological processes, as well as to psychiatric diseases such as depression, anxiety disorders, bipolar disorder and risk of suicide. While it is clear that genetic variation among people influences levels of neurotransmitters found in the cerebrospinal fluid, the details of this genetic control are not understood. In an effort to learn more about the genetics of monoamine levels, we measured levels of metabolites for these monoamines (5‐HIAA, HVA and MHPG) in 270 pedigreed baboons. All study animals can be linked into a single large pedigree, providing unique opportunities to examine quantitative genetic parameters. All three compounds exhibit significant heritability (5‐HIAA h2=0.30, HVA h2=0.50, MHPG h2=0.36). Bivariate genetic correlations calculated using variance components methods reveal that the genetic correlation between MHPG and HVA is rho=0.91, indicating that 83% of the genetic variance is shared between these two traits. In addition, it has long been known that HVA and 5‐HIAA are phenotypically correlated in humans. Our results show that in baboons the genetic correlation is 0.50 and environmental correlation 0.71, both statistically significant. This means the common phenotypic correlation of HVA and 5‐HIAA is due both to shared genes and shared environmental effects. A genome scan is planned to locate specific loci controlling monoamine levels in the baboons.
O67 MICROARRAY STUDIES OF CHANGES IN GENE EXPRESSION IN MOUSE BRAIN INDUCED BY ANTI‐MANIC DRUGS
Adams LJ and Schofield PR
Garvan Institute, 384 Victoria Street Darlinghurst, Sydney, Phone: 61 2 9295 8287; Fax: 61 2 9295 8281; E‐mail: [email protected]
Genes found to be regulated by one or more anti‐manic drugs are candidates that may have a role in the etiology of bipolar disorder. We are using the mouse as a model to investigate the biology underlying susceptibility to this disorder through the use of microarray studies in which we can monitor changes in gene expression. We have validated treatment doses in mice for both lithium and valproate, and are currently expanding this to include other anti‐manic drugs such as carbamazepine and lamotrigine. We have obtained brains from mice treated for seven days with either valproate (400 mg/kg/day, n=10), lithium (340 mg/kg/day) which have high (0.8–1.2 mM; n=10) or medium (0.4–0.7 mM; n=10) serum levels, or saline controls (n=10). Our initial experiments were undertaken using Affymetrix Murine U74A GeneChips, each containing approximately 12,000 genes and ESTs, and mice (n=5) with high lithium serum levels (mean ± SD =0.94 ± 0.03 mM) or saline controls (n=5). Analysis of these GeneChips identified 20 transcripts that are expressed at least two‐fold more, and two genes that are expressed at least two‐fold less in the treated animals. These represent members of different gene families, including protein kinases, transcription proteins, and integral membrane proteins.
O68 IDENTIFICATION OF POTENTIAL GLUCOCORTICOID RESPONSE ELEMENTS IN REGULATORY REGIONS OF CANDIDATE GENES FOR SUSCEPTIBILITY TO BIPOLAR DISORDER
Tremblay S, Tremblay M, Harvey M, and Barden N
CHUL Research centre, 2705 Blvd. Laurier Sainte‐Foy, Qc G1V 4G2 Canada, Phone: 418 654 2152; Fax: 418 654 2753, E‐mail: [email protected]
We have identified the chromosome 12q24.11–24.31 region as a site for bipolar disorder susceptibility gene(s). Since stress often precipitates manic or depressive episodes and the hypothalamic‐pituitary‐adrenal axis has been shown to be functionally disturbed in depression, we have developed a method to identify glucocorticoid sensitive sites in genes. A fusion‐protein, GST‐DBD‐hGRa (glutathione‐S‐transferase/DNA‐binding‐domain of the human glucocorticoid receptor a), was produced, coupled to sepharose 4B‐glutathione and used to enrich genomic DNA fragments containing GRE sequences. Experimental conditions were developed using the MMTV GRE (5′‐GATTACAaacTGTTCT‐3′). DBD‐hGRa showed high affinity for GRE sequences and competition assays against other hormone response elements (PRE/GRE, ERE, RXR, etc) confirmed the DBD‐hGRa/GRE binding specificity. Method validation was performed using human genes known to contain GRE sequence and cloned in BACs. The DBD‐hGRa/GRE interaction permitted enrichment of these gene fragments and we are currently using this strategy to identified regulatory GREs in the chromosome 12 linkage region as potential sites for mutations.
O69 CHRONIC ANTIDEPRESSANT TREATMENT EFFECTS ON KINASE GENE EXPRESSION PATTERNS
Rausch JL, Fei Y, Johnson ME, Jackson B, Ganapathy V, Hobby HM, and Leibach FH
Veterans Administration, The Medical College of Georgia, 1515 Pope Ave. Augusts, GA 30912, Phone: 706 721‐7793; Fax: 706 721‐7796; E‐mail: [email protected]
The serotonin transporter, SERT, is a phosphoprotein whose function is determined by its phosphorylation state via multiple kinase pathways. At least two fundamental phosphorylation mechanisms may regulate SERT activity. Phosphorylation through protein kinases may serve both to regulate sequestration of SERT from the cell membrane, and also down‐regulate SERT expression. Ligand occupancy has been recently shown to affect this system, with SSRIs blocking 5‐HT's ability to inhibit phosphorylation‐mediated sequestration of SERT. Consequently, the kinases may serve to regulate extracellular 5‐HT concentrations with SERT upregulation in response to increased 5‐HT, and down regulation of SERT in response to SSRIs. If this were true, then we might expect to see differences in protein kinase expression in response to SSRI treatment. To investigate this possibility, we studied 5 groups of rats 5 rats in each group (N=25, males). Rats were assigned randomly to osmotic mini‐pump treatment with placebo 3 days, placebo 21 days, fluoxetine 3 days, fluoxetine 21 days, or citalopram 21 days. Total RNA was isolated and labeled as cRNA, and incubated with Affymetrix gene chip and stained with Streptavidin‐phycoerythrin conjugate, and read for changes in the kinase expression system as a result of antidepressant treatment. The results indicated that expression of several protein kinases increased with acute and decreased with chronic antidepressant treatment. The results are consistent with homeostasis of SERT function through a decrease in PK manufacture, in response to antidepressant treatment. The results suggest that gene variation in this system may underlie differences in response to antidepressant treatment since kinase down‐regulation would counterbalance the SSRI effect, by lessening inhibition of SERT function, in response to treatment.
O70 A COMMON P‐GLYCOPROTEIN POLYMORPHISM IS ASSOCIATED WITH NORTRIPTYLINE‐INDUCED POSTURAL HYPOTENSION IN PATIENTS TREATED FOR MAJOR DEPRESSION
Kennedy MA, Roberts RL, Joyce PR, and Mulder RT
University of Otago, Christchurch School of Medicine, University of Otago, PO Box 4345, Christchurch, New Zealand Department of Pathology, Christchurch School of Medicine, P.O. Box 4345 Christchurch, SI 8015 New Zealand, Phone: 640‐3641222; Fax: 640‐3640525; E‐mail: martin.kennedy@AEA‐chmeds.ac.nz
The human multi‐drug resistance gene ABCB1 encodes the P‐glycoprotein (P‐gp) that regulates movement of many substances across the blood‐brain barrier. Evidence from a knockout mouse lacking P‐gp suggests that the TCA amitriptyline and its metabolites are substrates for P‐gp. In these mice penetration of amitriptyline, but not the SSRI fluoxetine, into the brain is enhanced. We reasoned that polymorphisms of P‐gp may affect responses of patients to antidepressant drugs. A polymorphism of ABCB1 (3435C+AD4‐T) was recently correlated with expression levels and in vivo function of P‐gp. We developed a DNA test for this SNP, and genotyped 165 Caucasian patients with major depression enrolled in a randomized antidepressant treatment trial of nortriptyline and fluoxetine. We observed a significant association between nortriptyline‐induced postural hypotension (NIPH) and 3435C+AD4‐T (chi‐square +AD0‐ 6.75, df +AD0‐ 2.73, p+AD0‐0.034). Of nortriptyline‐treated patients, 25+ACU‐ (4 of 16) who were TT and 7+ACU‐ (3 of 43) who were heterozygous suffered symptomatic NIPH. None of the 17 patients who were CC and no fluoxetine‐treated patients experienced postural hypotension. Our results suggest that presence of one or more T alleles at the 3435C+AD4‐T polymorphism of ABCB1 is a risk factor for occurrence of NIPH.
O71 EFFECTS OF THE ‐1438‐A/G SEROTONIN 2A RECEPTOR PROMOTOR POLYMORPHISM ON EXPRESSION
Parsons MJ, D'Souza U, Makoff A, Arranz MJ, and Kerwin R
Institute of Psychiatry De Crespigny Park Denmark Hill, London, United Kingdom SE5 8AF, Phone: 011442087610953; E‐mail: [email protected]
The 5‐HT2A‐receptor (5‐HT2A‐R) promoter polymorphism ‐1438‐A/G is associated with psychiatric disorders and plays a potential role in clozapine response. The ‐1438‐A/G polymorphism was found to affect the degree of radioligand affinity for the 5‐HT2A receptor, suggesting that the polymorphism may have functional consequences on 5‐HT2A‐R gene expression. Conversely, the two ‐1438‐A/G variants had equal levels of basal expression. In order to clarify whether the polymorphism has functional effects, either at the promoter or enhancer level, we cloned part of the 5‐HT2A‐R promoter (‐1536 to ‐536) for both ‐1438‐A/G variants into the following vectors: pCAT‐basic, negative control; pCAT‐enhancer, to test for promoter activity; and p‐CAT‐promoter, to test for enhancer affects. The pCAT‐control vector was used as a positive control. These vectors were transformed into SHSY‐5Y cells (neuroblastoma cells) and CAT activity was determined using a CAT ELISA kit. Our preliminary findings suggest that the expression levels were lower for the ‐1438‐G variant within pCAT enhancer. There appears to be no other differences. Though further replication is required, these results suggest that the ‐1438‐G variant decreases basal promoter activity, but may not disrupt an enhancer element.
IX. Manifestations of Known Genetic Disorders/Anorexia
O72 AN ANALYSIS OF HUNTINGTON'S DISEASE MUTATIONS IN INDIA SUGGESTS PREVALENCE RATES CLOSER TO THE WEST AND MULTIPLE ORIGINS FOR THE DISEASE MUTATION
Padiath QS, Roy S, Murgood U, Muthane U, Verma IC, Saxena R, Jain S, Anand A, and Brahmachari SK
Centre for Biochemical Technology CBT (CSIR), Mall Rd., Delhi University Campus, Delhi 110 007 NCBS, TIFR Centre Bangalore, 560 065 India, Phone: 91 80 3636421; Fax: 91 11 766 7471; E‐mail: [email protected]
Huntington's disease (HD) is a progressive neurodegenerative disorder caused by the expansion of CAG repeats in the coding region of the Huntingtin gene,. There is a wide variation in the prevalence of the disease with West Europeans having the highest rates and non‐Caucasian populations exhibiting an extremely low prevalence. We analysed 30 unrelated HD families from geographically distinct parts of India for CAG repeat expansions and polymorphisms at loci previously shown to be tightly linked to the HD mutation. Forty nine individuals exhibited expanded repeats with sizes ranging from 36 to 86, the largest size being a case of juvenile HD. We observed a variation in intergenerational instability, with male transmissions being more unstable. Even within male transmissions large variations were observed, suggesting the role of other cis or trans factors in the modulation of instability. The HD mutation in our families did not show any significant over representation of either the (CCG)7 or (CCG)10 allele. The CAG repeats in the normal population and the D2642 poly‐morphism showed a greater similarity to the West European populations rather than non‐Caucasians suggesting that the disease prevalence of HD may be intermediate between these two groups or even closer to that seen in the West. The analysis of the D4S127 locus suggests the presence of a founder mutation in a subset of South Indian families which is different from what is seen in the North, providing evidence for multiple and geographically distinct origins for the HD mutation in India.
O73 GENOTYPE‐PHENOTYPE CORRELATIONS IN PEOPLE WITH TUBEROUS SCLEROSIS COMPLEX
Lewis JC, Murphy KC, and Sampson JR
Institute of Medical Genetics, University of Wales College of Medicine, Heath Park Cardiff, South Wales CF14 4XN UK, Phone: 44 2920 744730; Fax: 44 2920 746551; E‐mail: [email protected]
Tuberous sclerosis complex (TSC) is an autosomal dominant disorder associated with mutations affecting one of two tumour‐suppressor genes, TSC1and TSC2. In this study, we attempted to determine whether TSC1 or TSC2 associated tuberous sclerosis are disorders of different clinical severity and whether the class of mutation affecting each gene is also a determinant of disease severity.
Methods: Comprehensive mutational analysis of TSC1 and TSC2 genes was performed on a group of 150 subjects with TSC. In addition, psychiatric and neuropsychological assessments were performed using standardised neuropsychological testing, semi‐structured clinical interviews and a review of all available medical notes.
Results: To date, mutations of TSC1 have been identified in 23 subjects and mutations of TSC2 have been identified in 106 subjects. Clinical assessments have been performed in 71 TSC subjects; 34 males and 37 females, with an age range of 6‐70 years (mean 26 years). In the adults able to complete self‐report questionnaires (n = 26), results from the Hospital Anxiety and Depression Scale showed that, for the depression subscale, 81% were ‘normal,’ 11% ‘mild depression,’ and 8% ‘severe depression’. For the anxiety subscale, 42% were ‘normal,’ 27% ‘mild anxiety,’ 23% ‘moderate anxiety’ and 8% ‘severe anxiety’. Analysis of other psychological measures is still to be completed.
Conclusions: In this study, we have identified specific patterns of psychological morbidity that occur in paople with TSC and are now investigating genotype‐phenotype correlations within this group.
O74 PREMUTATION EXPANSION OF CGG TRIPLET REPEATS AFFECTS BRAIN; A STUDY OF MALE CARRIERS OF FRAGILE X SYNDROME
Daly E, Moore CJ, Schmitz N, Jacobs PA, Davis KE, Murphy KC, and Murphy DGM
Dept Psychological Medicine, Institute of Psychiatry, De Crespigny Park, Denmark Hill London, UK SE5 8AF England, Phone: 44 0207848 0349; E‐mail: [email protected]
Expansion of trinucleotide repeats (Trs) is associated with several neuropsychiatric disorders. It is currently thought that a ‘threshold’ of Tr expansion needs to be crossed before biological effects are manifested. Nonetheless few studies have directly related Tr expansion to brain anatomy and function. People with Fragile X syndrome (FraX) allow us to investigate this question because FraX is associated with an expansion in Trs. People most affected by FraX have > 200 Trs and a behavioural and cognitive phenotype typically caused by methylation of the Fragile X Mental Retardation gene (FMR‐1) and subsequent loss of FMR‐1 protein (FMRP) production. Clinically unaffected ‘premutation’ FraX carriers have 50–200 CGG Trs. Thus we investigated neuroanatomical differences between 20 premutation male carriers of FraX (NTMs) and 17 age and IQ matched controls using MRI. Data were analysed with SPM99. Grey matter volume was significantly (P=>0.001) smaller in NTMs relative to control subjects in a number of brain areas (including cerebellum, inferior temporal cortices, hippocampus, cuneus, lingual gyrus, postcentral gyri and insula). This is the first study to demonstrate that premutation expansion of CGG triplet repeats in clinically unaffected NTMs significantly affects brain regions crucial to higher cognitive function and implicated in neuropsychiatric disorders. Thus the currently accepted genetic explanation for FraX may need to be modified.
O75 NEUROANATOMICAL EFFECT OF FMR1 GENE mRNA IN PREMUTATION CARRIERS OF FRAGILE X SYNDROME
Moore CJ, Daly E, Tassone F, Jacobs PA, Davies KE, Murphy KC, and Murphy KGM
Dept Psychological Medicine, Institute of Psychiatry, De Crespigny Park, Denmark Hill London, UK SE5 8AF England, Phone: 44 2078480349; E‐mail: [email protected]
Fragile X syndrome (FraX) is associated with an expansion of CGG triplet repeats (Trs). People most affected by FraX have>200 CGG Trs with methylation of the Fragile X Mental Retardation gene (FMR‐1) and subsequent loss of FMR‐1 protein (FMRP) production. Clinically unaffected ‘premutation’ FraX carriers have 50–200 CGG Trs, an unmethylated FMR‐1 gene and normal FMRP production. It was previously assumed that Tr expansion of<200 has no biological effect. Nonetheless, we recently reported that male carriers of FraX (NTMs) have structural abnormalities in brain anatomy. Tassone et al. (2000) found that relative levels of leukocyte FMR‐1 mRNA were elevated in a sample of NTMs. Thus, in this study, we related FMR‐I mRNA to brain anatomy in 17 NTMs using MRI. mRNA levels were calculated using previously published methods (Tassone et al., 2000) and neuroimaging data were analysed with SPM99. We found that grey matter volume was significantly (P => 0.001) positively related to mRNA levels in cerebellum, bilateral lingual and temporal occipital cortices, left cuneus and right insula, overlapping with brain regions that we previously reported to differ significantly between NTMs and controls. This is, to our best knowledge, the first study to demonstrate a relationship between mRNA and brain anatomy implicated in higher cognitive function and behaviour in humans.
Reference:
Tassone et al. (2000) Am. J. Hum. Genet. 66:6–15.
O76 EVIDENCE FOR A SUSCEPTIBILITY GENE FOR RESTRICTING ANOREXIA NERVOSA ON CHROMOSOME 1
Grice D, Halmi KA, Fichter MM, Strober M, Woodside DB, Kaplan AS, Treasure AS, Magistretti PJ, Goldman D, Kaye WH, Bulik CM, and Berrettini WH
Univ of Pennsylvania Room 135A, CRB, Psychiatry, 415 Curie Blvd Philadelphia, PA 19104, Phone: 215 573 4582; Fax: 215 573 2041; E‐mail: [email protected]
Eating disorders, such as anorexia nervosa, have been shown to have a significant genetic component. A recent genome wide linkage analysis of 196 affected relative pairs with anorexia nervosa and related eating disorders, including bulimia nervosa, gave only modest evidence for linkage. This may be due to multiple interacting genes of weak to moderate effect or sample heterogeneity. Reducing sample heterogeneity would increase power to detect linkage. In our current study we have carried out linkage analysis in a subset (n=37) of families where at least two affected relative pairs had diagnoses of anorexia nervosa, restricting subtype. Restricting anorexia nervosa (RAN) is a defined subtype of anorexia nervosa characterized by severe limitation of food intake without the presence of binge eating or purging behavior. When we restricted the linkage analysis to this clinically more homogenous subgroup, the highest multipoint NPL score of 3.03 was observed on chromosome 1p. Genotyping additional markers in this region led to a peak multipoint NPL score of 3.45. The data are consistent with the presence of a susceptibility locus for the restricting type of anorexia nervosa on chromosome 1p.
O77 LINKAGE ANALYSIS OF ANOREXIA NERVOSA IN A LARGE FAMILY WITH MULTIPLE AFFECTED INDIVIDUALS
Waller DA, Neville M, Barnes R, and Hobbs HH
Department of Psychiatry and McDermott Center for Human Growth and Development, UT Southwestern Medical Center, 5323 Harry Hines Blvd. Dallas, TX 75390‐8589, Phone: 214‐648‐4412; Fax: 214‐648‐4330; E‐mail: [email protected]
Twin and family studies suggest that genetic factors contribute to anorexia nervosa. We have performed a genome‐wide scan in two generations of a family in which multiple family members have had anorexia nervosa. Blood was collected and systematic semi‐structured interviews were performed on 11 family members, including four with a history of anorexia nervosa, one with a history of bulimia nervosa, one with obsessive compulsive disorder, and two with severe eating disorder symptoms. A total of eight family members were classified as affected. A whole genome scan was performed using 516 markers. Parametric linkage analysis was performed using the following diagnostic criteria and assumptions: 1) Affected status was given to all family members with a lifetime history of the diagnosis of anorexia nervosa, bulimia nervosa, or obsessive compulsive disorder using DSM‐IV criteria, or serious symptoms within the spectrum of eating disorders but not reaching DSM‐IV criteria; 2) autosomal dominant inheritance; and 3) 100% penetrance. Parametric linkage analysis revealed one area of linkage with LOD score >2. This region includes markers D11S1981, D11S902, and D11S1397 on chromosome 11p14.3. The lod score was 2.4. Additional families are being collected in an attempt to further delineate gene sequence differences that contribute to this disorder.
X. Cognitive/Neurodevelopment
O78 A GENOME‐WIDE ALLELIC ASSOCIATION SCAN OF 1847 DNA MARKERS FOR GENERAL COGNITIVE ABILITY: A FIVE‐STAGE DESIGN USING DNA POOLING
Hill L, Craig IW, McGuffin P, Lubinksi D, Thompson LA, Owen MJ, and Plomin R
SGDP Research Centre, Institute of Psychiatry, SE5 8AF UK, London
Our goal is to identify quantitative trait loci (QTLs) associated with high g versus average g. As a first step towards a systematic genome scan for allelic association, we used DNA pooling to screen 1847 simple‐sequence repeat (SSR) markers throughout the genome in a five‐stage design: (1) case‐control DNA pooling (101 cases with mean IQ of 136 and 101 controls with mean IQ of 100), (2) case‐control DNA pooling (96 cases with IQ > 160 and 100 controls with mean IQ of 100), (3) individual genotyping of Stage 1 sample, (4) individual genotyping of Stage 2 sample, (5) transmission disequilibrium test (196 parent‐child trios for offspring with IQ > 160). The numbers of markers surviving each stage using a conservative allele‐specific directional test were 108, 6, 4, 2, and 0, respectively, for the five stages. Several markers that were close to significance at all stages are being investigated further. At least 100,000 markers are needed to exclude QTL associations relying on linkage disequilibrium, but we are not planning to genotype additional SSR markers. Instead we are using the same design to screen markers such as cSNPs and SNPs in regulatory regions that are likely to include functional polymorphisms in which the marker can be presumed to be the QTL.
O79 A CANDIDATE GENE ANALYSIS OF TWO PHOSPHOLIPASE GENES THAT MAP TO THE CHROMOSOME 15Q15.1–15.3 REGION ASSOCIATED WITH READING DISABILITY
Morris DW, Robinson L, Turic D, Duke M, Owen MJ, O'Donovan MC, and Willia J University of Wales College of Medicine Heath Pk., Dept. of Psychological Medicine, Cardiff, CF144XN UK Phone: 00442920743244;
E‐mail: [email protected]
We have identified a region associated with dyslexia/reading disability (RD) on chromosome 15q by linkage disequilibrium mapping using microsatellite markers (Morris et. al. (2000) Hum Mol Genet 9 (5): 843–848). Two phospholipase genes map to our associated region. They are phospholipase C beta‐2 (PLCB2) and phospholipase A2, group IVB (cytosolic; PLA2G4B). Evidence supports an association between increased levels of cytosolic phospholipase A2 and dyslexia in adults. We have completed mutation detection analysis of PLCB2 and PLA2G4B and genotyped 14 SNPs in pooled samples of 143 RD cases and 171 controls. Two SNPs in PLA2G4B show evidence of association with RD (P<0.05) by case control analysis. We are currently analysing these SNPs in a family‐based association sample (178 RD trios).
O80 A NOVEL TRANSCRIPTION FACTOR IS MUTATED IN A SEVERE SPEECH AND LANGUAGE DISORDER
Lai CS, Fisher SE, Hurst JA, Vargha‐Khadem F, and Monaco AP
The Wellcome Trust Centre for Human Genetics, XRoosevelt Drive, Headington Oxford, OX3 7BN United Kingdom, Phone: (44)1865 287517; E‐mail: [email protected]
Individuals with Specific Language Impairment (SLI) experience significant difficulties in acquiring language despite having normal hearing and intelligence. Although twin studies have consistently suggested the involvement of genetic factors in SLI, simple Mendelian patterns of inheritance are seldom observed, and the neural basis of SLI remains elusive. We have studied a rare 3‐generation family, KE, in which a severe speech and language disorder is transmitted as an autosomal‐ dominant monogenic trait. In previous work, we mapped the locus responsible for the disorder (SPCH1) to a 5.6cM interval of chromosome 7q31 and used genomic sequence to assemble a comprehensive transcript map of this region. We now show that a misssense mutation in a gene encoding a novel transcription factor is present in all affected individuals of the KE family. This changes an invariant amino acid in the DNA‐binding domain that is likely to be critical for the function of the protein. We have also demonstrated that this gene is directly disrupted by a translocation in an unrelated patient who has a very similar phenotype to that of the KE family. We propose that haploinsufficiency of this gene leads to abnormal development of neural structures that are important for speech and language.
O81 THE PREVALENCE AND ROLE OF GENETIC ABNORMALITIES IN A GENERAL POPULATION SAMPLE WITH MILD MENTAL RETARDATION
Simonoff E, Wood N, Gringras P, Chadwick O, Maney JA, and Higgins S
GKT Medical School and Institute of Psychiatry, Guy's Campus, Department of Child and Adoelscent Psychiatry, Munro Centre, Snowsfields London, SE1 3SS UK, Phone: 00442073783225; Fax: 00442073783243;
E‐mail: [email protected]
We report on a general population sample of 13 to 15 year old children with mild mental retardation (IQ<70) ascertained through individual cognitive screening in schools. Over 2,000 children participated in the screening. All those scoring in the lowest 5%, along with a stratified random sample of low normal scorers (6th to 25th centile) and normal scorers (>25th centile) were selected for in‐depth assessments. A full medical examination, including a standardized examination for dysmorphic features and other stigmata of genetic disorders associated with mental retardation was conducted. Cytogenetics, fragile x, subtelomeric deletions and organic acidemias were evaluated in all cases. We report here on the prevalence and type of known and suspected genetic abnormalities found. Final results are still pending at the time of submission; currently of the 62 on whom results of medical examinations, cytogenetics and fragile X are available, only 5 (8.1%) showed definite or strong evidence of a genetic disorder (excluding autism, N=3). Medical abnormalities raised the possibility of genetic disorders in a further 5, raising the rate of definite/suspected genetic causes to 16% excluding autism and 21% including autism. The rates are lower than those reported for Scandinavian studies but similar to those from US and UK samples.
O82 EVIDENCE OF GENOMIC IMPRINTING IN FAMILIES WITH GILLES DE LA TOURETTE SYNDROME IN CHINESE SUBJECTS
Huang Y^1^, Li T^1,2^, Liu X^1^, Guo L^1^, Zhao J^2^, Sham PC^2^, and Collier DA^2^
^1^Institute of Mental Health, The First University Hospital, West China University of Medical Sciences, Chengdu 610041, P R China,
^2^Department of psychological Medicine, The Institute of Psychiatry, De Crespigny Park, Denmark Hill, London SE5 8AF, UK, Tel: 0044‐20‐78480343, Fax: 0044‐20‐78480051; E‐mail: [email protected]
Previous attempts using large multigenerational families to localize susceptibility loci of Gilles de la Tourette Syndrome (GTS) have been unsuccessful ,which primarily due to the complex mode of its inheritance. Recently, several studies reported that genomic imprinting may be involved in the transmission of GTS in caucasion population.In this study, a semi‐structured schedule for the genetic research of Tourette syndrome and related disorders was used in the family study of genomic imprinting in 171 probands with GTS. The family data include information from 342 first degree relatives,1283 second degree relatives and 2310 third degree relatives in addition to probands. Our data suggest that maternal transmission was associated with the symptom of complex motor tics in the proband; P=0.01. Maternal transmission was more likely to present earlier‐onset of the disease than paternal transmission 5.56+0.85 Yr; 6.07+1.10 Yr; t=2.34; P=0.02. However, paternal transmitted GTS was characterized by the increased attention problem score in CBCL behavioral scale of the proband t=2.78; P=0.01. This result indicated that parental specific expression exists in the transmission of GTS, which gives evidence that genomic imprinting may be involved in the inheritance of GTS in Chinese people.
O83 PROTOCADHERIN XY—A BRAIN‐EXPRESSED CELL SURFACE ADHESION MOLECULE AS A CANDIDATE GENE FOR LANGUAGE, CEREBRAL ASYMMETRY AND PREDISPOSITION TO PSYCHOSIS
Crow TJ, Williams NA, Giouzeli M, Ross NA, Priddle T, Groome N, DeLisi LE, Sargent CA, Affara NA, and Blanco P POWIC, University Department of Psychiatry, Warneford Hospital, Oxford OX3 7JX UK
Background: It is proposed that psychosis represents a component of variation associated with the asymmetry that characterises the human brain and underlies language, and that this variation is determined by a gene that is present on X and Y chromosomes (Crow, 1993). The Xq21.3/Yp11.2 region of homology that was created by a translocation from X to Y that occurred after the separation of the chimpanzee and hominid lineages has been subject to a subsequent paracentric inversion and other changes (Sargent et al, 2001). Within this region a gene has been identified that codes for a protocadherin, a member of a class of cell surface adhesion proteins that act as axonal guidance molecules (Blanco et al, 2000). This gene differs in its structure on the X and the Y chromosome in a way that could account for sex differences such as are seen in age of onset of psychosis, verbal ability and handedness (Crow et al, 1998).
Methods: We have studied the gross structure of the Xq21.3/Yp11.2 region of homology for variation that may reflect its recent evolutionary history and selective pressures. We have investigated the sequence structure of protocadherin XY in Homo sapiens, the chimpanzee, orang‐outang and gorilla and are investigating sequence variations in relation to schizophrenia and schizo‐affective disorder with DHPLC. We are developing monoclonal antibodies.
Findings: Rearrangements have been identified around the marker DXS214 that may be relevant to the expression of genes in this region including protocadherin X. The coding sequence of protocadherin Y has diverged more than that on the human X from that of the nonhuman primate X. It appears that some forms of protocadherin XY are expressed on the surface of pyramidal cells on the human cerebral cortex. No sequence variation relating to psychosis has so far been detected.
Conclusions: From its evolutionary history and structure it is clear that protocadherin XY has played a specific role in the evolution of the human cerebral cortex. Its putative role in cerebral asymmetry and relationship to psychosis remain to be demonstrated.
O84 HERITABILITY OF NEUROCOGNITIVE FUNCTIONS AND NUMBER OF QUANTITATIVE TRAIT LOCI CONTRIBUTING TO THEM IN FAMILIES WITH SCHIZOPHRENIA
Annamari Tuulio‐Henriksson, MSc,^1^ Jari Haukka, PhD,^1^ Teppo Varilo, MD PhD,^2^ Timo Partonen, MD, PhD,^1^ Tyrone D. Cannon, PhD,^1,3^ Tiina Paunio, MD, PhD,^2^ Jesper Ekelund, MD,^2^ Joanne M. Meyer, PhD,^4^ and Jouko Lönnqvist, MD, PhD^1^
^1^Department of Mental Health and Alcohol Research, and ^2^Department of Human Genetics, National Public Health Institute of Finland, Helsinki, and ^3^Departments of Psychology, Psychiatry and Human Genetics, UCLA, Los Angeles, and ^4^Millennium Pharmaceuticals Inc, Boston
Despite evidence for several chromosomal loci linked to schizophrenia, no susceptibility genes have been identified for the disorder. Using quantitative measures of phenotypic affection in place of clinical diagnostic categories or dichotomous classifications may be more effective when susceptibility genes are searched. Neurocognitive traits have been suggested as putative quantitative endophenotypes of the disorder, but their heritabilities are only sparsely known. We investigated the heritability of working memory, verbal declarative memory and its different components, and both verbal and visual ability functions. We also estimated the number of quantitative trait loci (QTL) contributing to these neurocognitive functions.
Methods: Polygenic heritability of the neurocognitive functions was estimated in a sample of schizophrenia patients and their first‐degree relatives (N=264) from an isolated geographical subregion in Finland. The number of QTLs was analysed using Markov Chain Monte Carlo segregation analysis.
Results: Significant heritabilities were found in working memory and ability functions. Furthermore, the working memory functions revealed the most restricted number of QTLs. The mean number of loci for verbal and visual working memory was 1.2 and 1.0, respectively, with corresponding posterior probabilities of 73% and 70% for at least one locus. In declarative memory variables the number of loci was more dispersed.
Conclusions: Our results suggest that neurocognitive measures, particularly working memory, provide valid quantitative phenotypic traits for linkage analyses searching predisposing genes for schizophrenia.
O85 SYNAPSIN III: A CANDIDATE SCHIZOPHRENIA SUSCEPTIBILITY GENE THAT REGULATES NEUROGENESIS
Hung‐Teh Kao,^1,2^ Barbara Porton,^2^ Vincent Pieribone,^3^ Lynn E. DeLisi,^1^ and Paul Greengard^2^
^1^Department of Psychiatry, New York University School of Medicine
^2^Laboratory of Molecular and Cellular Neuroscience, The Rockefeller University
^3^Department of Molecular and Cellular Physiology, Yale University School of Medicine
Synapsin III is the most recently identified member of a family of neuronal phosphoproteins that are involved in neurodevelopment and synaptic transmission. The synapsin III gene is located on chromosome 22q12–13, a region previously identified as a potential schizophrenia susceptibility locus. We have begun to search for polymorphisms in the synapsin III gene in schizophrenic subjects whose families display linkage to this region. Preliminary studies indicate that a polymorphism exists in exon 12, and affects a site known to be phosphorylated in the synapsin III protein. This polymorphism is likely to be rare, and further work is underway to determine if it segregates with the disease. In the adult brain, synapsin III is expressed in the hippocampus and olfactory bulb, regions known to contain high levels of neurogenesis. In mice bearing a null mutation in the synapsin III gene, neurogenesis is markedly decreased in the hippocampus. These findings could explain certain features reported in schizophrenia, such as reduced hippocampal and olfactory bulb volumes, and abnormalities in learning, memory, and smell. Synapsin III is one of the first genes found to play a role in adult neurogenesis, and therefore provides us with a molecular tool for understanding the mechanism by which new neurons are added to the mature brain.
O86 SCREENING OF CANDIDATE GENES RELATED TO MYELINATION FOR MUTATIONS ASSOCIATED WITH SCHIZOPHRENIA
Williams NM,^*^ Spurlock G,^*^ Williams H,^*^ Norton N,^*^ Davis KL,^#^ Buxbaum JD,^#^ Haroutunian V,^#^ Saunders R,^*^ Cardno AG,^*^ McCarthy G,^*^ O'Donovan MC,^*^ and Owen MJ^*^
^*^Department of Psychological Medicine, University of Wales College of Medicine, Heath Park, Cardiff, CF14 4XN UK
^#^Department of Psychiatry, Mount Sinai School of Medicine, New York, E‐mail: [email protected]
Hakak and colleagues recently used DNA microarray analysis to assay the expression levels of over 6000 genes in the postmortem dorsolateral prefrontal cortex of 12 schizophrenics and matched controls. A total of 89 genes that are involved in a range of biological processess, including synaptic plasticity and neuronal development, were reported to show an altered expression profile in schizophrenics. However, their most notable finding was the differential expression of a number of genes related to myelination (MAL, CNP, MAG, transferin gelsolin and ErbB3), suggesting a disruption in oligodendrite function in schizophrenia. We have now screened each of these 6 genes for sequence variants using DHPLC and sequencing. All identified SNP's have been genotyped by primer extension and their allele frequencies estimated in an association sample of 174 DSMIV schizophrenic patients and 174 matched controls by a method of DNA pooling based on the ABI SnaPshot^2^ assay. SNP's with a significant difference in allele frequencies were then typed in a second sample of similar size by the same pooling method. Only SNP's that yielded significant results in both samples were subsequently genotyped individually in the same sample to confirm the allelic association. A full report of the SNP's identified and the results of the association analysis will be presented.
O87 DIFFERENTIAL PROMOTER ACTIVITY OF THE MOUSE APOLIPOPROTEIN (APOE) GENE IN PRIMARY NEURONS, ASTROCYTOMA AND IN PC12 CELLS
Lahiri D, Ge YW, Chen X, Nurnberger JI Jr, Farlow MR, and Du Y
Indiana Univ Sch Medicine, Institute of Psychiatric Res, 791 Union Drive, Indianapolis, IN 46202, Phone: 317‐274‐2706; Fax: 317‐274‐1365; E‐mail: [email protected]
The APOE gene, which constitutes a major susceptibility factor for the development of the familial and sporadic forms of late‐onset Alzheimer's disease (AD), encodes a 34 kDa protein. It plays a critical role in mobilization and redistribution of cholesterol and phospholipid during membrane remodeling and synaptic plasticity. The gene is located on 19q13 of the human chromosome. The proximal 5′‐flanking region of the APOE gene is highly conserved in the mouse, rat and human; the relative position of the ‘TATA box’ and the two copies of ‘GC box’ are identical. To study the transcription control of the mouse (m) APOE gene, we assayed in different cell types the promoter activity of a 725 nucleotide (nt) 5′‐flanking region, which is located 772 nt upstream from the translation initiation codon. We cloned the 725 nt region into a promoterless vector upstream of the reporter chloramphenicol acetyl transferase (CAT) gene. The mAPOE promoter and vector DNAs were independently transfected in primary rat cortical neurons, human astrocytoma (U138) and PC12 cell lines. In mAPOE‐transfected U‐138 cells, we observed a 5‐fold increase in CAT reporter activity from the promoterless vector. As compared to U138 cells, we detected a reduced CAT activity in rat cortical neurons and PC12 cell lines. In both these cells, the mAPOE promoter displayed significantly higher levels of activity than the vector. Our results suggest that mAPOE can also be expressed in neuronal cells in addition to the astrocytic cells. Characterization of mAPOE promoter is important for the APOE transgenic mice studies, which are used for the AD drug development discovery.
Acknowledgment: Supported by grants from the Alzheimer's Association and NIH.
XI. Bipolar Candidate Genes and Chromosome Regions
O88 FURTHER EVIDENCE FOR A BIPOLAR RISK GENE ON CHROMOSOME 12q24 SUGGESTED BY INVESTIGATION OF HAPLOTYPE SHARING AND ALLELIC ASSOCIATION IN PATIENTS FROM THE FAROE ISLANDS
Degn B, Lundorf MD, Wang AG, Vang M, Mors O, Kruse TA, and Ewald H
Institute for Biological Psychiatry, Psychiatric Hospital in Aarhus, Skovagervej 2 Risskov, na 8240 Denmark, Phone: 45 77 89 28 23; E‐mail: [email protected]
A number of studies have strongly suggested a susceptibility locus for bipolar affective disorder on chromosome 12q24. The present study investigates for a shared chromosomal segment among distantly related patients with bipolar affective disorder from the Faroe Islands, using 17 microsatellite markers covering 24 cM in the previously suggested region on chromosome 12q24. Possible allelic association to bipolar affective disorder (P‐value using CLUMP below 0.01) and increased sharing among cases of two‐marker haplotypes (P‐values using CLUMP around or below 0.001) were suggested in a 6 cM region bounded by markers D12S1614 and D12S1675. This area contains the minimum interesting region between suggested by the previously reported haplotypes in two Danish families with bipolar affective disorder which have yielded significant linkage to this region. (Molecular Psychiatry, in press)
O89 POSITIONAL CLONING OF BIPOLAR SUSCEPTIBILITY GENE IN THE DARIER REGION OF CHROMOSOME 12q23–q24
Craddock N, Glaser B, Green E, O'Donovan MC, Jones I, Owen MJ, and Jones I
University of Birmingham, Queen Elizabeth Psychiatric Hospital, Division of Neuroscience, Birmingham, B75 5QL UK, Phone: 44 121 678 2358; Fax: 44 121 678 2351; E‐mail: [email protected]
We have described two pedigrees (324: max. lod 2.1; 5501: max. lod 3.6) in which Bipolar Disorder segregates with markers in the region of the Darier's disease gene on 12q23–q4.1. We and other groups have reported independent evidence for linkage of markers in the 12q23–q24 region with susceptibility to Bipolar Disorder in pedigrees unselected for Darier's disease. As is usual in complex disorders, the signals span a broad region of interest. Haplotype studies using a dense map of microsatellite markers in pedigrees 324 and 5501 have allowed us to refine the most likely location of the Bipolar susceptibility gene and we are using a combination of positional and candidate approaches for identification of the pathogenically relevant gene. Using direct mutation/polymorphism analysis of known and predicted genes within this region of interest we have already excluded the coding and known promoter sequences of 30 genes. Evidence supportive of fine localization has been provided by systematic linkage disequilibrium mapping studies using microsatellite and SNP markers across the region in outbred case‐control samples. In this presentation we will provide a progress report of our search.
O90 FURTHER SUPPORT FOR BIPOLAR DISORDER SUSCEPTIBILITY LOCI ON CHROMOSOMES 22q AND 13q IN AN INDEPENDENT SECOND SAMPLE OF FAMILIES
Kelsoe JR, Shaw SH, Mroczkowski‐Parker Z, Remick R, Dessa S, McElroy S, and Keck P
Department of Psychiatry, 0603, UCSD and SDVAHS La Jolla, CA 92093 USA
We have previously reported a genome scan of 20 families with bipolar disorder from the general North American population which indicated evidence of linkage to 22q13 and suggestive evidence of linkage to 13q. We are now conducting a genome scan on a second independent set of 32 families also from the general North American population. This sample includes 194 subjects, 118 of whom are affected under our broad diagnostic model. Chromosomes 22 and 13 were examined in these 32 families using the same markers and parametric models as employed in our first study. On 22q, a maximum lod score of 2.2 was obtained at D22S684, approximately 2 MB from the genome‐wide maximum from the first family sample at D22S278. A lod score of 1.4 was also obtained at D22S419 near the GRK3 gene and a secondary linkage peak from our first study. Similarly, on 13q, our first study identified two markers with lod scores greater than 2.0 which are about 8 cM apart: D13S154 and D13S225. In the second set of 32 families, D13S154 yielded a lod score of 2.3, and D13S225 a lod score of 1.6. Together, these data from a second independent sample of families provide additional support for the evidence for linkage to 22q and 13q that we reported previously.
O91 PROMOTER REGION VARIANTS IN G PROTEIN RECEPTOR KINASE‐3 (GRK3) ARE ASSOCIATED WITH BIPOLAR DISORDER
Barrett TB, Hauger RL, Kennedy JL, Alexander M, Keck P, McElroy S, and Kelsoe JR
University of California at San Diego, Department of Psychiatry, 0603, UCSD 9500 Gilman Dr. La Jolla, CA 92093‐0603 USA
In a genome‐wide linkage survey we previously found evidence indicating chromosome 22q12 contains a susceptibility locus for BPD in the region of GRK3. GRK3 is an excellent candidate risk gene since GRKs play key roles in the homologous desensitization of G protein‐coupled receptors. To identify mutations in GRK3 we sequenced the putative promoter region, exons, and flanking intron in individuals with BPD. We found six variants in the promoter/5′‐UTR region, but no coding or obvious splice variants. TDT analysis of two triad sets indicates two of the promoter/5′‐UTR variants, generally found as a linked haplotype, are associated with BPD in families of Northern European Caucasian ancestry. In 329 triads the transmission to non‐transmission ratio was 25:7, chi‐square=10.1, P=0.003. We have cloned the putative promoter (spanning the locations of all six variants) into a luciferase expression vector. Transfection of this construct into SK‐N‐MC cells demonstrates this region has transcriptional activity. The possibility that the variants associated with BPD are mutations which effect transcription or translation will be studied. These data support the hypothesis that a dysregulation in GRK3 expression which alters signaling desensitization contributes to the development of BPD.
O92 SEARCH FOR A SHARED SEGMENT ON CHROMOSOME 10q26 IN PATIENTS WITH BIPOLAR AFFECTIVE DISORDER OR SCHIZOPHRENIA FROM THE FAROE ISLANDS
Ewald H, Flint TJ, Wang AG, Jorgensen TH, Vang M, Kruse TA, and Mors O
Institute for Basic Psychiatric Research, Psychiatric Hospital in Aarhus Skovagervej 2, 8240 Risskov, Denmark, Phone: 45 77 89 28 23; Fax: 45 77 89 28 99; E‐mail: [email protected]
Due to the relatively few founders, limited population size for centuries, sparse immigration, population bottlenecks and genetic drift, the population on the Faroe Islands could be sufficiently homogenous with respect to disease mutations, risk alleles and related haplotypes to facilitate the genetic mapping of disease genes. Previous linkage studies have suggested a new locus for bipolar affective disorder and possibly also for schizophrenia on chromosome 10q26. We searched for allelic association and chromosome segment and haplotype sharing on chromosome 10q26 among distantly related patients with bipolar affective disorder or schizophrenia and controls from the Faroe Islands by investigating 22 microsatellite markers from a 35 cM region. We used a combined approach with both assumption free tests and tests based on genealogical relationships. An interesting 7.6 cM region between D10S1757 and D10S2322 were supported especially for haplotype sharing among patients with bipolar affective disorder (empirical P‐values around 0.003) and for allelic association to both disorders combined (empirical P‐values around 0.003).The region also received some support as it was estimated that random sharing of a segment without a disease gene inherited from a common ancestor was relatively rare. This region has been implied in previous linkage analyses.
O93 HAPLOTYPE ANALYSIS DEFINES A 4.7Mb PROBABLE DISEASE REGION FOR A BIPOLAR AFFECTIVE DISORDER SUSCEPTIBILITY LOCUS ON CHROMOSOME 4q35
Badenhop RF, Moses MJ, Scimone A, Adams LJ, Donald JA, Mitchell PB, and Schofield PR
The Garvan Institute of Medical Research, 384 Victoria Street, Darlinghurst, Sydney NSW 2010 Australia, Phone: 61 2 9295 8288; Fax: 61 2 9295 8281; E‐mail: [email protected]
We have developed a novel haplotype‐based approach to define a 4q35 bipolar susceptibility region. Linkage analysis in 55 pedigrees gave a maximum two‐point LOD score of 3.01 for D4S1652 and scores between 1.5 and 2.44 for several other markers. 24 linked pedigrees were selected for haplotype analysis based on having LOD scores greater than their maximum expected LOD score for multiple markers. There was no ancestral disease haplotype and no one‐to‐one correspondence between disease and disease haplotype. Therefore we determined a probable disease region based on the percentage of affected individuals within each pedigree sharing the same portion of the disease haplotype and pooled this data across all linked pedigrees. In each pedigree, for each 4q35 marker we calculated the number of affecteds who share the marker allele that forms part of the disease haplotype, identical‐by‐descent (IBD). The number of affected individuals sharing alleles IBD at each marker was pooled to generate a map of percentage of sharing. A probable disease region of 4.7Mb from D4S1540 to the telomere was defined by maximum allele sharing of affected pedigree members. Using this pedigree specific IBD allele sharing approach provides a means for focusing the candidate gene search for this complex trait.
O94 SEQUENCE ANALYSIS OF AN INTRONLESS GENE WITHIN GNAL ON 18Q11 IN INDIVIDUALS AFFECTED WITH BIPOLAR DISORDER
Trofatter JA, Nurnberger KM, Berrettini WH, and Nurnberger JI Jr
Indiana University School of Medicine, 791 Union Dr, B‐03 Indianapolis, IN 46202, Phone: 317‐278‐1716; Fax: 317‐274‐1365; E‐mail: [email protected]
The short and long arms of chromosome 18 have been identified as regions that may contain genes associated with susceptibility to bipolar disease. Extensive work has been carried out to identify candidate genes in both regions. Much of the work in the 18p11.2 region has centered around the twelve exon GNAL/G (olf) gene which encodes a GTP‐binding protein involved in odorant signal transduction. A small intronless gene (Berrettini name 22444; also human EST AW013797) has been identified within the fifth intron of the GNAL gene within 4kbp of the GNAL (CA)n repeat and ending just 5′ to exon 6. This gene encodes a putative 116 amino acid protein of unknown function. Since GNAL has not shown any changes associated with bipolar disease, we hypothesized that bipolar illness may result from genes in or around GNAL. We have sequenced this intronless gene in fourteen individuals with bipolar disease. These individuals were derived from fourteen families in the Bipolar Consortium dataset that had NPL scores over 1 in this region of chromosome 18 based on GeneHunter Plus analysis. These people represented those individuals with the highest likelihood of containing mutations or variants. No variant nucleotides were observed in this analysis.
O95 FINE MAPPING STUDIES OF A BIPOLAR DISORDER CANDIDATE REGION ON CHROMOSOME 18Q22
McMahon FJ, Chen Y‐S, Schulze TG, Badner JA, Potlouri S, Akula N, and Singh G
University of Chicago, 924 E 57th Street Chicago, IL 60637, Phone: 773‐834‐2973; Fax: 773‐834‐2970; E‐mail: [email protected]
Linkage of bipolar disorder to chromosome 18 has been suggested by several studies, but the results have been inconsistent and poorly‐localized. Our analysis of the relationship between clinical features and allele‐sharing on chromosome 18q22 has identified a region linked to bipolar disorder with a lod of 4.85 and a 1‐lod confidence interval of ∼9 cM. We have anchored this interval to the human draft sequence, and have identified a ∼5 MB candidate region. Our strategy for identifying the gene that accounts for this linkage finding uses SNP‐based association analysis to sample essentially all of the common genetic variation occurring in and near each gene in the region. Our preliminary computational annotation of the human draft sequence has identified approximately 50 known or predicted genes. Our analysis of 32 SNPs mapped to a finished BAC in the region indicates that background linkage disequilibrium is detectable in our study populations at ∼60 kb. Based on this, we have developed an initial set of 50 common SNPs that are being used to screen all genes in the region for trait‐marker association in 3 samples of case‐parent triads. Our data indicate that SNP‐based association analysis is feasible, and illustrate one strategy for systematically evaluating marker‐trait association within a candidate region.
O96 REPLICATION OF SUGGESTIVE LINKAGE ON CHROMOSOMES 5 AND 16 IN THE NIMH GENETICS INITIATIVE BIPOLAR PEDIGREES
Dick DM, Nurnberger JI Jr, Edenberg H, McInnis MG, Reich T, Gershon ES, and Foroud T
Indiana University, Department of Medical & Molecular Genetics, Department of Psychology, Indiana University School of Medicine, 975 West Walnut Street, IB‐130 Indianapolis, IN 46202‐5251, Phone: (812) 855‐4101; Fax: (812) 855‐4691; E‐mail: [email protected]
Families who had a bipolar I (BP1) proband and at least one BP1, or schizoaffective‐bipolar type (SA/BP) first‐degree relative were ascertained through the NIMH Genetics Initiative. A series of hierarchical models of affection were utilized in linkage analyses. Model I considered as affected only individuals with BP1 or SA/BP; model II included all individuals in Model I as well as bipolar II individuals; and Model III included individuals diagnosed under Model II, and those with unipolar recurrent depression. An initial genome screen was completed in 540 subjects from 97 families. Genotyping at Indiana University was subsequently performed on chromosomes 3, 5, 15, 16, 17 and 22 in a replication sample of 353 individuals from 56 families. Nonparametric linkage analyses were performed using both affected sibling and relative pair methods. Analyses in the new sample on chromosome 16, with the broadest definition of affection, replicated previously reported suggestive linkage to the marker D16S2619 (lod ∼2.0). In addition, evidence of linkage was also found on chromosome 5q for models II and III (lod ∼2.5) in the same chromosomal region reported in the initial sample. Additional marker genotyping is currently underway to further delineate these linked regions.
XII. Autism
O98 GENDER‐SPECIFIC GENETIC INFLUENCES ON AUTISTIC TRAITS: EVIDENCE FROM A TWIN STUDY
Constantino JN and Todd RD
Washington University School of Medicine, 660 South Euclid Ave., Box 8134, Saint Louis, MO 63110, Phone: (314) 747‐6772; Fax: (314) 747‐6777; E‐mail: [email protected]
In this study we examined the genetic structure of reciprocal social behavior (RSB)—a core component of the autistic phenotype—as a function of gender.
Methods: The sample consisted of 232 pairs of male twins, 324 pairs of female twins, and 126 opposite sex twin pairs, all subjects age 7–15 years. One parent of each pair of twins completed the Social Reciprocity Scale (SRS) on their children. The data were subjected to structural equation modeling using the statistical software, Mx.
Results: Scale scores for RSB in males were stongly influenced by additive genetic factors (accounting for approximately 76 percent of the total trait variance), exhibited minimal measurement error, and were not significantly influenced by age, rater bias or rater contrast effects. For females, the magnitude of additive genetic influences on RSB was 0.33. For opposite sex pairs, the best fitting model was one which incorporated gender‐specific genetic influences, the magnitude of which was 0.26.
Conclusion: Given these findings, and given the fact that autism spectrum disorders are more commonly observed in boys than in girls, it appears possible that susceptibility loci accounting for a substantial share of the variance in subthreshold autistic traits may reside on the X chromosome.
O99 SYMPTOM DOMAINS IN AUTISM AND RELATED CONDITIONS: EVIDENCE FOR FAMILIALITY
Silverman JM, Smith CJ, Schmeidler JM, Buxbaum JD, Lawlor BA, and Fitzgerald M
Mount Sinai School of Medicine, Department of Psychiatry, Box 1230, Mt. Sinai, One Gustave L. Levy Pl. New York, NY 10029, Phone: 212 659‐8822; Fax: (212) 849‐2505; E‐mail: [email protected]
Heterogeneity in autism impairs efforts to localize genes underlying this disorder. As autism comprises severe but variable deficits and traits in 3 symptom domains—social, communication, and repetitive behaviors—and shows variability in useful phrase speech, different genetic factors may be associated with each. Sibling‐pairs (n=212), including a autistic proband and 1+siblings with autism or marked deficits in autism symptom domains, were assessed using the ADI‐R. Symptom domain scores were examined to determine within sibling pair similarity. Results showed reduced variance within sibling pairs for repetitive behavior and for delays in and the presence of useful phrase speech. These features and nonverbal communication provided evidence of familiality when only autism was used to define sibling pairs (pairs=136). These same features appeared familial for those with autism‐related conditions and their severity varied within sibling pairs independently. The features identified as familial replicate the combined set suggested in earlier, smaller studies. Furthermore, the familiality of these features extend to related but milder conditions and appear independent. Making symptom severity classification distinctions may be useful for molecular genetic studies of autism.
O100 CLINICAL FREQUENCY OF CHROMOSOMAL ABNORMALITIES OBSERVED IN A CONSECUTIVE SERIES OF PATIENTS WITH AUTISTIC DISORDER (AUTD)
Wolpert CM, Wright HH, Cuccaro ML, DeLong GR, and Pericak‐Vance MA
Duke University Medical Center, Center for Human Genetics, 3445 Carl Building, DUMC, Durham, NC 27710, Phone: 919‐684‐4446; Fax: 919‐684‐2275; E‐mail: [email protected]
Numerous reports cite the co‐occurrence of AutD with various chromosome anomalies suggesting the potential of a causal relationship in a subset of cases. We examined the frequency of chromosome anomalies in 333 consecutively ascertained AutD patients from 99 multiplex and 127 singleton families. All patients had their diagnosis confirmed using the Autism Diagnostic Interview‐ Revised (ADI‐R). 32% (N=106) of the AutD individuals had chromosome analysis done as part of their medical evaluation.
Seven different chromosomal anomalies were observed in 7 independent families. The anomalies included: 18q‐ (3 unrelated patients); __de novo,__partial duplication of 7p, familial paracentric inversion (7) (q22.1–q31.2), XO, 2q‐, familial 13;14 Robertsonian translocation, and isodicentric chromosome 15 anomalies (5 unrelated patients). All chromosome anomalies with the exception of the Robertsonian translocation and the inv 7q cases were de novo occurrences in the AutD individuals and observed in singleton families. Here we report the clinical, developmental, and cytogenetic results for these individuals.
O101 A GENOME‐WIDE SEARCH FOR AUTISM PREDISPOSING GENES IN AUTISTIC FAMILIES
Liu J, Nyholt D, Geschwind D, Lord C, Iversen P, Ott J, and Gilliam C
Columbia Genome Center, 1150 St. Nicholas Avenue, New York, NY 10032, Phone: 212‐304‐7998; Fax: 212‐304‐5515; E‐mail: [email protected]
Autism is a severe neurodevelopmental disorder with significant genetic etiology. Results from five genome‐wide mapping studies have provided moderate statistical support for several potential autism loci. Utilizing 110 pedigrees from the AGRE program in which at least 2 siblings were classified as affected under a broad diagnostic scheme (autism, Asperger's syndrome or other PDD), we conducted a genome‐wide search for predisposing loci using 335 microsatellite markers. Affected sibpair analysis yielded multipoint maximum lod scores (MLS) reaching the accepted threshold for suggestive linkage on chromosomes 5, X and 19. Nominal evidence for linkage (point‐wise P<0.05) was obtained on chromosomes 2, 3, 4, 8, 10, 11, 12, 15, 16, 18, 20 and secondary loci on 5 and 19. Analysis of families sharing alleles at the putative X chromosomal linked locus and one or more other putative linked loci produced a MLS of 3.56 for the DXS470‐D19S174 marker combination. In an effort to increase power to detect linkage, scan statistics were used to evaluate the significance of peak lod scores based upon statistical evidence at adjacent marker loci. This analysis yielded impressive evidence for linkage to autism and autism spectrum disorders with significant genome‐wide P‐values below 0.05.
O102 PHENOTYPIC HOMOGENEITY PROVIDES INCREASED SUPPORT FOR LINKAGE ON CHROMOSOME 2 IN AUTISTIC DISORDER
Shao Y, Raiford K, Wolpert CM, Ashley‐Koch A, Cuccaro ML, Gilbert JR, and Pericak‐Vance MA
Duke University Medical Center, Box 3445 Durham, NC 27710
A two‐stage genomic screen analysis of 99 autistic disorder (AutD) families revealed suggestive evidence for linkage to chromosome 2q (D2S116 MLS (nonparametric sibpair LOD score)=1.12 at 198 cM) (Shao 2001). In addition analysis of linkage disequilibrium (LD) for D2S116 showed an allele‐specific P‐value <0.01 (Bass 2000). Recently, Buxbaum et al. (2001) also reported linkage to the same 2q region (Heterogeneity LOD score (HLOD)=1.96). Their evidence for linkage increased (HLOD=2.99) when they restricted their analysis to the subset of AutD patients with delayed onset (>36 months) of phrase speech (PSD). We similarly classified our data set of 99 AutD patients identifying 37 AutD families with PSD. Analysis of this PSD subset increased our support for 2q linkage (MLS of 2.82 and HLOD of 2.05 for D2S116). These data support evidence for a gene on chromosome 2 contributing to AD risk and suggest that phenotypic homogeneity increases the power to find susceptibility genes for AutD.
O103 EVIDENCE FOR A SUSCEPTIBILITY GENE FOR AUTISM ON CHROMOSOME 2
Buxbaum JD, Davis KL, Greenberg DA, Kilifarski M, Reichert J, Silverman JM, and Smith CJ
Mount Sinai School of Medicine, One Gustave L. Levy Place, Box 1230 New York, NY 10025, Phone: (212) 659‐8862; Fax: (212) 828‐4221; E‐mail: [email protected]
Although there is considerable evidence for a strong genetic component to idiopathic autism, several genome‐wide screens for susceptibility genes have been carried out with limited concordance of linked loci, reflecting either numerous genes of weak effect and/or sample heterogeneity. We studied the effect of restricting a sample of autism affected relative pairs to those with delayed (>36 months) onset of phrase speech on evidence for linkage. In the second stage of a two‐stage genome screen for susceptibility loci involving 95 families with two or more individuals with autism or related disorders, we observed a maximal multipoint heterogeneity LOD score (HLOD) of 1.96 and a maximal multipoint NPL score of 2.39 on chromosome 2q. Restricting the analysis to the subset of families (n=49) with two or more individuals with a narrow diagnosis of autism and delayed onset of phrase speech generated a maximal multipoint heterogeneity HLOD score of 2.99 and an NPL score of 3.32. The increased scores in the restricted sample indicates that this sample is more genetically homogeneous, which could therefore increase the likelihood of positional cloning of susceptibility loci. We are repeating this study and carrying out association studies with candidate genes in the linked region.
O104 EXAMINATION OF CANDIDATE GENES FOR AUTISM ON CHROMOSOME 7
Hutcheson HB and Haines JL
Vanderbilt University Medical Center 1211, 22nd Avenue, South Nashville, TN 37232, 519 Light Hall, Nashville, TN 37232‐0700, Phone: 615‐936‐1671; E‐mail: [email protected]
Evidence from previous genetic and cytogenetic studies performed in autism research points to the existence of one or more autism genes residing on chromosome 7q. However, further localization using linkage analysis has proven difficult. To overcome this problem, we examined our CLSA dataset to identify only the families potentially linked to chromosome 7. 47 from a total of 86 families were identified and 17 markers were used to generate chromosomal haplotypes. We performed recombination breakpoint analysis to determine if any portion of the chromosome was predominately shared. The preponderance of this data identified a 6 cM region between D7S501 and D7S2847 as being most commonly shared. Additional markers at 1 cM intervals within this region were genotyped and association and recombination breakpoint analysis was performed. Although no significant association was found, the preponderance of the recombination breakpoint data points to a 3cM shared region between D7S496‐D7S2418 encompassing about 4.5 Mb of genomic DNA. This region contains more than fifty genes that can now be prioritized based on proposed function. Multiple SNPs within KIAA0716 have been examined and no association has been found. SNPs in other candidate genes in this region are being tested.
O105 LINKAGE AND ASSOCIATION OF THE GLUTAMATE RECEPTOR 6 GENE WITH AUTISM
Bourgeron T, Jamain S, Betancur C, Quach H, Philippe A, Gillberg C, and Leboyer M
Pasteur Institute, 25 rue du Docteur Roux 75015, Paris, France, Phone: 33 1 40 61 32 16; Fax: 33 1 40 61 31 53; E‐mail: [email protected]
A genome scan was previously performed and pointed chromosome 6q21 as a candidate region for autism. This region contains the glutamate receptor 6 (GluR6) gene, a functional candidate for the syndrome. We used two different approaches, the affected sib‐pair (ASP) method and the transmission disequilibrium test (TDT), to investigate the linkage and association between GluR6 and autism. The ASP method, conducted on 59 families, showed a significant excess of allele sharing, generating an elevated multipoint maximum LOD score (NPL=3.28; P=0.0005). Using 107 additional families with a single affected child, a significant maternal transmission disequilibrium was observed (TDT linkage P=0.0004). Furthermore, TDT and Haplotype Relative Risk (HRR) analyses showed significant association between GluR6 and autism (TDT association P=0.008; HRR P=0.01). Mutation screening was performed in 33 affected individuals, revealing several SNPs, including one amino acid change (M867I) found in 8% of the autistic subjects, in a highly conserved domain of the protein and seems to be more maternally transmitted than expected to autistic males (P=0.007). Taken together, these data suggest that GluR6 is in linkage disequilibrium with autism.
O106 CPG ISLAND IDENTIFICATION AND MAPPING IN THE AUTISTIC DISORDER REGION ON CHROMOSOME 15q11–q13
Kim S‐J, Menold M, Stajich J, Pericak‐Vance MA, and Gilbert JR
Duke University Medical Center, Department of Medicine and Center for Human Genetics, Duke University Medical Center, Durham, NC 27710, Genomic Research Laboratories, Center for Human Genetics, Department of Medicine, Duke University Medical Center, Box 2903, Research Park Building II, Room 102 Durham, NC 27710, Phone: 919‐681‐5546 or 919‐684‐6433; Fax: 919‐681‐7894; E‐mail: [email protected]
Chromosome 15q11–q13 has been implicated in the genetic etiology of autistic disorder (AutD). To identify candidate AutD genes, a physical map was generated from the GABRB3 receptor to the OCA2 gene. To identify AutD candidate genes within the genomic contig, 28 BAC, PAC and P1 clones containing numerous rare restriction sites were analyzed using Island Rescue PCR (IR‐PCR). 150 EagI, BssHII and SacII related CpG island sites were cloned, sequenced and analyzed. BAC/PAC sequence comparison analysis identified 45 unique CpG islands that met full CpG island criteria. 31 CpG island clones were mapped onto the human genomic draft contigs spanning the region from GABRB3 gene to the APBA2 gene. 14 CpG clones including GABRG3 and APBA2 showed expression in human fetal brain tissue. 38 IR‐PCR clones did not meet CpG island criteria. 13 clones showed expression in human fetal brain tissue. Five known genes including GABAA receptor subunits, APBA2, and numerous ESTs colocalized with CpG islands in this region and are candidates with AutD gene(s). Currently, we are investigating CpG island SNPs in this region for association with AutD. This island rescue system will allow us to investigate the methylation status and alterations of genes within the AutD region in tissues.
XIII. Special Session: Ethics
O107 ATTITUDES OF GERMAN PSYCHIATRISTS, PSYCHOLOGISTS, GYNAECOLOGISTS, HUMAN GENETICISTS AND PATIENTS TOWARDS PSYCHIATRIC GENETIC RESEARCH AND TESTING
Illes F,^1^ Rietz C,^2^ Fangerau H,^1^ v Widdern O,^1^ Schulze TG,^1^ Mueller DJ,^1^ Gross M,^1^ Angermeyer MC,^3^ Maier W,^1^ Rudinger G,^2^ and Rietschel M^1^
^1^Department of Psychiatry, University of Bonn, Sigmund‐Freud‐Str. 25, 53105 Bonn, Germany
^2^Department of Psychology, University of Bonn, Römerstr. 164, 53117 Bonn, Germany
^3^Department of Psychiatry, University of Leipzig, Liebigstr. 22, 04103 Leipzig, Germany
The soon‐coming availability of genetic testing in psychiatric disorders raises new and complex ethical issues. Psychiatrists, psychologists, gynaecologists, and human geneticists will be in the crucial position to transfer knowledge about psychiatric genetics to their mentally ill patients and their families. Significant differences in attitudes between counselors and their patients may cause misunderstandings and problems. Here we assessed similarities and differences of attitudes between future counselors and patients. In a study, which is conducted in the framework of the “German Human Genome Project,” 76 psychologists, 116 psychiatrists, 50 gynaecologists, 56 human geneticists and 120 patients were asked about their knowledge, attitudes and fears towards psychiatric genetics. Intra‐ and intergroup differences were analyzed. Whereas the consultants showed a relative conformity in their attitudes towards psychiatric genetics, they differed significantly from the attitudes of their patients. To avoid problems in future consultations, experts have to learn about the specific hopes, fears and objections of their patients.