Arpana Agrawal1,3, Michael C. Neale1,2,3, Carol A. Prescott2,3, and Kenneth S. Kendler1,2,3.  The influence of early onset cannabis use on use and abuse/dependence of other illicit drugs: discordant twin design versus a model-fitting method.

1Department of Human Genetics, Virginia Commonwealth University, Richmond, VA, 2Department of Psychiatry, Medical College of Virginia, Virginia Commonwealth University, Richmond, VA, 3Virginia Institute of Psychiatric and Behavior Genetics, Virginia Commonwealth University, Richmond, VA

Address:  Virginia Institute of Psychiatric and Behavior Genetics, Virginia Commonwealth University, Box# 980003, 800 E. Leigh Street, Richmond, VA 23298 Phone:  804-828-8157 Fax:  804-828-1471 Email:  aagrawa@mail2.vcu.edu

 

Following-up a recent report (M. T. Lynskey, A.C. Heath, K.K. Bucholz, W.S. Slutske, P.A. Madden, E.C. Nelson, D.J. Statham, and N.G. Martin, 2003, JAMA, 289(4), 427-33), we used the discordant twin design to analyze the influence of early onset of cannabis use on use and abuse/dependence of other illicit drugs. Twins who used cannabis before the age of 18 years versus non-user co-twins had odds ratios of 2.0-7.5 for using other illicit drugs.  However, the discordant twin design does not utilize all the available data and has limited ability to examine the genetic etiology of the relationship between cannabis and other illicit drug use. Using 1191 male and 934 female same sex twin pairs from the Mid-Atlantic Twin Registry (K.S. Kendler and, C.A. Prescott, 1999, Arch. Gen. Psychiatry, 56, 39-44), we fit three models to our data (M.C. Neale and, Kenneth S. Kendler, 1995, Am. J. Hum. Genet., 57, 935-953). The correlated liabilities model most suitably explained the data with a very high correlation of 0.91 between the genetic factors influencing early use of cannabis and other illicit drug use. A model that only allowed for specific environmental influences to be correlated (re=0.87) fit poorly. Also, a genetic model of the gateway hypothesis was not supported. The relationship between early cannabis and other illicit drug use does not appear to be due to causation at the phenotypic level (A. Golub and, B.D. Johnson, 1994, J. Stud. Alcohol, 55(5), 607-14) but due to correlations between factors that influence the individual liabilities. Thus, prevention of cannabis use may not be successful in curbing use of other illicit drugs.


Maricela Alarcón1, Rita M. Cantor2, Jussi Niemi3, Mari Qvarnstrom4, Markku Ryynanen5, Taisto Leppasaari4, Aarno Palotie2, and Daniel H. Geschwind1.  Genomewide scan in Finnish families with dyslexia/dysphasia.

1Department of Neurology, University of California, Los Angeles CA, 2Department of Human Genetics, University of California, Los Angeles CA, 3Department of Linguistics, University of Joensuu, Joensuu, Finland, 4Department of Otolaryngology, Kuopio University Hospital, Kuopio, Finland, 5Department of Obstetrics and Gynecology, University of Oulu, Oulu, Finland

Address:  UCLA Department of Neurology, 710 Westwood Plaza 1-145, Los Angeles, CA 90095 Phone:  310 794 4090  Fax:  310 267 2401  Email:  malarcon@ucla.edu

 

Dyslexia affects approximately 5 to 10% of the population in the United States, making this disorder an important public health problem (G. R. Lyon, D. Alexander, and S. Yaffe, 1997, Learning disabilities:  a multidisciplinary perspective, 8, 1-6).  Although the genetic etiology of the disorder is well established, phenotypic and genetic heterogeneity have impeded the identification of susceptibility genes for dyslexia.  To circumvent these problems, our project aims to identify loci responsible for dyslexia in families from a genetic isolate in Central Finland.  The genetic homogeneity and phonetic language of the Finnish families makes them ideally suited for the identification of genes involved in dyslexia.  Linguists and neuropsychologists easily recognize the phonological deficit in Finnish children with dyslexia or dysphasia.  Thus, disagreements regarding diagnostic criteria are uncommon and clinical diagnosis of the disorder is highly consistent.  Three large and 7 nuclear families have been ascertained from phoniatric clinics in Finland.  The families were required to have at least two children and one parent with reading performance two or more levels below grade, and difficulty with phonological processing tasks.  Results of preliminary analyses suggest that candidate regions in chromosomes 6, 15 and 18 may not contain major dyslexia genes in these families.  Thus, a whole genome scan analysis is currently underway to identify novel regions of interest in the affected families.  The single- and multi-point analyses are based on the dyslexia and dysphasia diagnosis provided by the expert phoniatrists. Since the mode of inheritance for most of the families appears to be autosomal dominant with reduced penetrance and variable expressivity, we are using parametric as well as nonparametric approaches for the analysis.  We will also attempt to construct haplotypes linked to the disorder and, subsequently, we will conduct nonparametric linkage analyses.  Complete results of the genome scan will be presented at the Behavior Genetics Association meeting.


Juko Ando1 and Yutaka Ono2.  Genetic structure of Cloninger's seven factor model revisited.3

1Faculty of Letters, Keio University, Mita Tokyo, 2Health Center, Keio University, Yokohama Kanagawa, 3Supported by a Grant-in Aid for Scientific Research(A)from the Ministry of Education, Science, Sports and Culture

Address:  Faculty of Letters, Keio University, 2-15-45, Mita, Minatoku, Tokyo, 108-8345, Japan Phone: 81 3 3453 4511 Fax: 81 3 5427 1578 Email: juko@msa.biglobe.ne.jp

 

Ando et al. (2001, Journal of Personality, 70,584-609) investigated the genetic structure of R. Cloninger's seven factor model of personality with 296 pairs of Japanese twins with TCI (Cloninger et al. 1993) and revealed that three temperamental dimensions (novelty seeking: NS, harm avoidance: HA, reward dependence; RD) are genetiucally independent.  The current study is its replication with larger sample (513 pairs : 330 pairs of MZs , 116 pairs of same sex DZ, 67 pairs of opposite sex twins).  Univariate analysis shows that AE model is the best fir for all the dimensions but self-directedness(SD) and additive genetic contributions are 21% for NS, 45% for HA, 39% for RD, 34% for PS (persistence) 46% for CO (cooperativeness) and 48% for ST (self-transcendence ). SD can be explained either AE or CE model. Multivariate analyis by Cholesky decomposition indicates again that genetic latent factors for three temperamental dimentions are independent.


Andrey P. Anokhin1, Angela Ralano1, and Erin Myers1.  Heritability of prepulse inhibition of startle reflex in humans.2

1Department of Psychiatry , Washington University School of Medicine, St.Louis, MO, 2Supported by grants DA00421-02 from the NIH and CRTG-00-300-01-PBP from the American Cancer society

Address:  Washington University School of Medicine, 18 South Kingshighway, Suite 2T, St. Louis, MO 63108

 

Prepulse inhibition (PPI) is a suppression of the startle reflex that occurs when intense startle startling stimulus is preceded by a weaker “prepulse” stimulus. PPI is viewed as an index of sensorimotor gating, a fundamental process of sensory input regulation. PPI deficits have been implicated in the biological bases of schizophrenia and some other neuropsychiatric conditions including substance use disorders and proposed as a possible biological marker ("endophenotype") for genetic studies. However, little is known about the genetic determination of PPI in humans. Accordingly, the purpose of this study was to estimate heritability of PPI in normal human subjects. PPI of acoustic eyeblink startle reflex was assessed in 170 young adult female twins (49 MZ and 36 DZ pairs). Response magnitude was measured as square root of the area under the curve of a rectified, averaged, and smoothed orbicularis oculi EMG signal in the time window of 20-120 ms after stimulus onset. Twin intrapair correlations were significant in MZ (r=0.52, P<0.01), but not in DZ twins (r=0.14, n.s.). Genetic analysis using structural equation modeling showed significant heritability of PPI, suggesting that about 50% of the PPI variance is determined by genetic factors.


Andrey Anokhin1 and Angela Ralano1.  Wisconsin Card Sorting Test performance in female twins.2

1Department of Psychiatry, Washington University School of Medicine, St.Louis, MO, 2Supported by grants DA00421-02 and 5P50 AA11998-03 from the NIH

Address:  Washington University School of Medicine, 18 South Kingshighway, Suite 2T,

St.Louis, MO 63108

 

Wisconsin Card Sorting Test (WCST) is one of the most widely used assessments of executive functioning related to prefrontal cortex.  Performance on the WCST has been suggested to indicate familial vulnerability to psychiatric disorders characterized by executive deficits, such as schizophrenia. However, little is known about genetic and environmental determinants of individual differences in WCST performance in the general population. A previous study based on a very small twin sample did not find significant genetic influences (A. Campana, F. Macciardi, O. Gambini, S. Scarone S., 1996, Neuropsychobiology 34, 14-17).  The present study assessed heritability of standard WCST scores in a sample of 166 young female twins including 58 MZ and 25 DZ pairs. A computer-administered version of WCST was used. Several WSCT indices including total number and percent of errors, and the number of perseverative responses showed modest, but significant heritability ranging from 37 to 46 %, whereas other indices such as the number of categories completed and failure to maintain set did not show evidence for genetic influences. The results suggest that selected aspects of executive functioning measured by the WCST are moderately influenced by genetic factors.
Jessica H. Baker1, Kyle L. Gobrogge2, and Kelly L. Klump.  A Twin Study of Similarities in Self and Co-twin Reports of Personality.

1Department of Psychology, Michigan State University, East Lansing, MI, 2Supported by NIH Grant MH 65447

Address:  129 Psychology Research Building East Lansing, MI 48824 Phone:517 432 9861  Email: klump@msu.edu

 

Objective:  Several studies have examined the agreement between self and informant ratings of personality (Ready 2002; Ready & Clark 2002; Hill et al., 1995; Harkness et al., 1995).  However, few studies have examined the relative influence of genetic factors on similarities of self and informant reports of personality.  Thus, the purpose of the current study was to examine genetic and environmental influences on these reports in male and female twins.  Method:  Subjects included 228 MZ and 58 DZ male and female twins participating in the Michigan State Twin Study.  Personality was measured with the Neuroticism, Extraversion, and Openness to Experience Personality Inventory Revised (NEO-PI-R) (Costa & McCrae, 1985).  Twins completed two versions of this instrument, the first asking them to rate themselves, and the second asking them to rate their co-twin. Biometric model fitting analyses were used to examine genetic and environmental influences on similarities between self and informant reports.  Results:  Model-fitting analyses indicated significant additive genetic and non-shared environmental influences on most of the NEO-PI-R subscales.  Discussion:  Findings suggest greater similarity between self and co-twin reports of personality in MZ relative to DZ twin pairs that can be accounted for by genetic and nonshared environmental factors.


Bojan Basrak.  Copulas in QTL mapping.

Eurandom, Eindhoven, The Netherlands

Address:  Eurandom (TUE), P.O.Box 513, 5600 MB Eindhoven, The Netherlands Phone: +31 40 2478123 Fax: +31 40 2478190 Email: basrak@eurandom.tue.nl

 

Detecting linkage of a marker to a quantitative trait locus (QTL) in human genetics essentially means detecting a connection between genetic similarity and similarity of phenotypes. The first similarity is typically measured using the identity by descent (IBD) status and the second one by the notion of linear correlation. This also holds for the two most popular methods--the Haseman-Elston regression method and the variance components likelihood-ratio test. It is well known that the correlation coefficient is the most natural measure of stochastic dependence (similarity) in the world of multivariate normal distributions, but it can be less suitable when the normality assumption is not met. In practical linkage analysis this means that the methods that are derived on the basis of normality have to be used with a great care when this assumption is violated. Not surprisingly, this problem has attracted a lot of attention recently, see for instance J. Blangero, J.T. Williams, and L. Almasy, 2000, Genet. Epidemiol. 19, S8-S14 or P.C. Sham, J.H. Zhao, S.S. Cherny, and J.K. Hewitt, 2000, Genet. Epidemiol. 19, S22-S28.  The most general way of expressing statistical dependence between variables is via copulas. We show how this well established statistical tool can be applied in QTL linkage analysis with a little extra effort and potentially many benefits. One particular copula, namely the bivariate normal copula is discussed in more detail. In particular, we demonstrate how a statistical analysis based on the normal copula model deals with problems of non-normality that appear in many practical studies. Finally we demonstrate applicability and usefulness of this approach by simulation studies.


Daniel M. Blonigen1, Brian M. Hicks1, Robert F. Krueger1, Christopher J. Patrick1, William G. Iacono1, and Matt McGue1.  Psychopathic Personality Traits & Externalizing Psychopathology: Etiologic Connections.

1Department of Psychology, University of Minnesota, Minneapolis, MN, 2The Minnesota Twin Family Study is supported in part by U.S. Public Health Service Grants AA00175, & AA09367, & DA05147, & MH65137

Address:  Department of Psychology 75 East River Rd. University of Minnesota, Minneapolis, MN 55455 Phone: 612-624-0499   Fax: 612-626-2079  Email: bloni001@umn.edu

 

Psychopathic Personality (psychopathy) is defined as a constellation of interpersonal, affective, and behavioral features. Our recent factor analytic work with the Psychopathic Personality Inventory (PPI), a self-report psychopathy measure for non-incarcerated populations, has shown that it is represented by two orthogonal factors. The first factor (PPI-I) is related to social dominance and stress immunity, core features of the affective/interpersonal dimension of psychopathy. The second factor (PPI-II) is marked by negative emotionality and low constraint, and is strongly related to a dimension of externalizing psychopathology, defined as the covariance among child and adult antisocial behavior symptoms, alcohol and drug dependence, and disinhibitory personality traits (R. F. Krueger, B. M. Hicks, C. J. Patrick, S. R. Carlson, W.G. Iacono, and M. McGue, 2002, Journal of Abnormal Psychology 111, 411 - 424).  Although the two factors exhibit distinct relations with personality and behavioral measures, it is not clear if they are etiologically distinct as well. In this study we conducted a biometric analysis of the relationship between psychopathy and externalizing in a sample of male and female twins from the Minnesota Twin Family Study. The Multidimensional Personality Questionnaire was used to index the two PPI factors, and an externalizing factor was derived from a principal components analysis of symptom counts of child and adult antisocial behavior, and alcohol, drug, and nicotine dependence. We performed a Cholesky decomposition of the PPI factor scores and externalizing to index the amount of shared variance among the observed phenotypes that is determined by genetic or environmental contributions.  Ultimately, the results will address the question of whether or not the two phenotypically distinct dimensions of psychopathy are preferentially linked to the externalizing dimension at an etiologic level.


Stefan M. Brudzynski.  Quantitative parameters of the ultrasonic communication in rats.1

Department of Psychology and Centre for Neuroscience, Brock University, St. Catharines, ON, Canada, 1Supported by the Natural Sciences and Engineering Research Council of Canada

Address:  Department of Psychology, Brock University, 500 Glenridge Avenue, St. Catharines, Ontario, L2S 3A1 Canada, Email: sbrudzyn@spartan.ac.brocku.ca

 

There is ample evidence that ultrasonic calls emitted by rats are used for intraspecies communication, which plays an adaptive role in improving  chances of individual and group survival.    Ultrasonic calls are emitted in biologically significant situations and they have acoustic features, which are not only specifically recognized by the recipients but also contain a quantitative dimension reflecting the magnitude of the emitter's response. The goal of the presentation is to review the features of the ultrasonic calls in search of the acoustic parameters, which could reflect the quantitative aspect of the rat vocal communication in addition to the commonly recorded rate of calling. Isolation calls of rat pups have a variable sound frequency and call duration, and have changing sonographic structure over time. It has been concluded that the "message" is encoded in one or more of the alternative sweeps of frequency, which facilitate attending and retrieval of the pup by the dam. Isolation calls are replaced with age by two other ultrasonic call patterns. The, so called 22-kHz calls, or alarm calls, are emitted predominantly in agonistic and other aversive situations. They are characterized by a relatively constant sound frequency, and remarkable variability in the duration of single calls. This variability and the results of pharmacological studies suggest that the quantitative aspect in the vocal expression of these calls seems to be encoded in the length of individual calls. Finally, the 50-kHz calls have

been reported in positive, non-agonistic behavioural situations. They are characterized by a very short and relatively constant single call duration, and by a variable sound frequency. Results of the pharmacological studies and the variability of sound frequency suggest that the peak sound frequency is likely to reflect the quantitative aspect in the vocal expression in this type of calls.


Susan A. Brunelli.  Selective breeding for an infantile phenotype (rat pup ultrasonic vocalizations): A window on developmental processes.1,2

Department of Developmental Psychobiology, New York State Psychiatric Institute & Columbia University, New York NY, 1Supported by NIMH Grant MH40430, 2Supported by NIMH Grant MH54027

Address:  New York State Psychiatric Institute Box 40 Room 4917, 1051 Riverside Drive, New York NY 10032 Phone: 212 543 5711 Fax: 212 543 5467 Email: sab9@columbia.edu

 

A series of studies posed two questions: (1) How generations of selective breeding of rats for high (High USV line) or low (Low USV line) levels of USV responses at 10 days of age may affect the time course of USV response development over the postnatal period; and (2) How the development of other behavioral and psychological traits may have been altered over the same period in association with changes in the selected trait.  Longitudinal and cross-sectional studies of rat pups from litters of High, Low and Random lines were conducted in which two cohorts of litters were tested:  one at 3, 10 and 18 days postnatal age, and the second at 7, 14 and 21 days.  In addition to the USV response to two minute isolation tests, other isolation-induced behaviors, physical markers of maturation and measures of thermoregulation were studied. The results support the idea that the developmental trajectories of High and Low lines have been altered differently, i.e., the two lines are not mirror images, and developmental paths are independent of each other.


Tanya Button1, Anita Thapar2, and Peter McGuffin1.  Maternal smoking during pregnancy and antisocial outcomes in young people.3

1Social, Genetic and Developmental Psychiatry Research Centre, Institute of Psychiatry, King's College London, 2Department of Psychological Medicine, University of Wales College of Medicine, 3Supported by MRC PhD Scholarship

Address:  PO 80, Social, Genetic and Developmental Psychiatry Research Centre, Institute of Psychiatry, King's College London, De Crespigny Park, London, SE5 8AF Phone: +44 (0) 20 7848 0629 Fax: +44 (0) 20 7848 0575 Email:  t.button@iop.kcl.ac.uk

 

Maternal smoking during pregnancy is associated with conduct problems in young people. One possible explanation for this association is the effects of the cigarettes on the brain of the developing fetus, and there is evidence that nicotine exposure affects brain development. An alternative explanation is that smoking during pregnancy is an index of maternal antisocial behavior, and offspring conduct problems may result from genetic transmission for the propensity to antisocial behavior.  This study examines the relationship between maternal prenatal smoking and antisocial behavior using data from 1896 twins pairs from the Greater Manchester and Lancashire Twin Registry. Twins behavior and mothers prenatal smoking habits were analysed using structural equation modelling.  Two models were fitted. In the first it was assumed that maternal smoking had a direct environmental effect (the environmental mediation model), whilst in the second it was assumed that the correlation between conduct disorder and maternal smoking was mediated via the mothers' genotype (the genetic mediation model).  The results confirmed that adolescents whose mother smoked during pregnancy scored significantly higher for antisocial behavior than those whose mothers were non-smokers. This increased with the number of cigarettes smoked daily. The genetic mediation model was a slightly better fit than the environmental model calling into question the frequent assumption that the effects of maternal smoking on later antisocial behavior in offspring result from direct toxic effects on the developing fetus.


Michèle Carlier1,2, Charles Cohen-Salmon1,3, Chabane Chérif1, Fatima Marouf Verray1, Paricia Arechi1, Fabienne Godin1, Franz Sluyter1,4, Brice Marcet5, Bernard Verrier5, and Pierre L. Roubertoux1,6.  Development of congenic strains for mitochondrial DNA in mice.  Implication of mtDNA in pre-weaning development and motor behavior.

1CNRS Génétique Neurogénétique Comportement, France, 2Present address: PsyCLÉ (EA3273), Université de Provence, France, 3Present address: UMR 7593 CNRS and Université Paris VI, Paris, France, 4Present address: SGDP Centre, Institute of Psychiatry, London, UK, 5INPC CNRS 31, Marseille, France, 6Present address: INPC CNRS 31, Marseille, France.  Supported by the CNRS to Génétique Neurogénétique Comportement and to Institut des Neurosciences Physiologiques et Cognitives. F. Sluyter received a fellowship from the Fondation Fysen and Cohen Salmon a financial support from Laboratoire Ipsen.

Address:  Centre de recherche en Psychologie de la Cognition, du Langage et de l’Émotion (PsyCLÉ EA 3273), Université de Provence, 29 avenue Robert Schuman, 13624 Aix en Provence cedex 1, France Phone: 33 (0) 442933999 Fax: 33 (0) 442389170 Email:  mcarlier@up.univ-aix.fr

 

Mammalian mitochondrial (mt) genome that encodes 13 subunits and specifies 22 tRNAs and 2 rRNAs is exclusively maternally transmitted. The implication of the 13 proteins in respiratory chain and ATP synthase has been demonstrated. Their possible impact on cognitive or other behavioral processes has been previously suspected. As the number of mtDNA molecules is large per one somatic cell (more than 1,000) the gene targeting strategy cannot be used to address the implication of mtDNA genes in cognition. For this reason, we developed congenic strains for mtDNA. We selected NZB/BlN and CBA/H mice that carry mtDNA from different origins. We present here the development of this quartet of congenic strains, controls for mtDNA cross-transfer and for the isogenicity of the nuclear background.  mtDNA modulates motor development as soon as the pre-weaning period in interaction with nuclear DNA. This effect persist in adults and an interaction between mtDNA, nuclear DNA and age is observed for a wide set of motor measurements.


Penny L. Biersach Clark, Carol A. Van Hulle, Erri Hewitt, and H. Hill

Goldsmith1.  Exploring the genetic architecture of social and non-social fear in infancy.2

1Department of Psychology, University of Wisconsin, Madison, WI, 2Supported by NIH Grant MH-50560

Address:  Department of Psychology, 1202 W. Johnson St., Madison, WI 53706 Phone: 608-263-4735 Fax: 608-265-3649 Email: plclark@wisc.edu

 

Many researchers have examined the development of fear in infancy. The majority of this research has focused on social fear, which has been shown to be moderately heritable. Additionally, genetic influences account for the largest portion of the covariation in parent-report and observed measures of social fear (Goldsmith et al., 1999, Developmental Psychology 35(4), 972-985). Fewer studies have examined the development of non-social fear in infancy. Recent research suggests social and non-social fear are differentially related to vulnerabilities for disruption in emotion regulation (Locke& Goldsmith, 2002, poster presented at ICIS, Toronto, Canada) highlighting the importance of considering different aspects of fear in infancy.  The current study examines social and non-social fear in 150 6 and 12 month-old twin pairs participating in an ongoing longitudinal study of emotional development. Non-social fear was assessed via observation of infants' reactions to four novel masks. Social fear was assessed via maternal report as well as observation of infants' reactions to a stranger approach. At 6 months, the three measures were not significantly correlated. At 12 months, however, observed social fear correlated with maternal report of social fear (r=.261, p<.05) and observed non-social fear (r=.231, p<.05). All measures exhibited moderate stability over time (r=.248 to .341, p<.05). A bivariate Cholesky model was fit to the data on maternal report of social fear and observed non-social fear for each age.  At 6 months, there was complete overlap in additive genetic and shared environmental influence on social and non-social fear, but non-shared environmental effects were independent. At 12 months, there was no overlap in additive genetic influence, a slight overlap in non-shared environment (re=.15), and a single shared environment factor influenced both measures.  Future analyses will include a larger sample size and multivariate analyses with multi-method assessments of social and non-social fear.


R. P. Corley1, S. E. Young1, M. C. Stallings1, J. K. Hewitt1, A. Smolen1, and D. Huizinga2.  Sibling resemblance for adolescent problem behavior: National Youth Survey and CADD.3

1Institute for Behavioral Genetics, University of Colorado, Boulder, CO, 2Institute of Behavioral Science, University of Colorado, Boulder, CO, 3Supported by grants DA-05131 and DA-11015, HD-10333, MH-43899, and AA-11949

Address:  Institute for Behavioral Genetics, Campus Box 447, University of Colorado,

Boulder, CO 80309-0447 Phone: 303 492 5189   Fax: 303 492 8063  Email: Robin.Corley@colorado.EDU

 

The National Youth Survey (D. S. Elliott, D. Huizinga, and S. Menard, 1989, Multiple Problem Youth: Delinquency, Drugs and Mental Health Problems. New York, NY: Springer, 1989) is a longitudinal study of adolescent problem behavior and its sequelae.  The foundation sample of the National Youth Survey (NYS) consisted of 1725 individuals drawn from a national probability sample of 1044 households, with 574 households represented by only a single adolescent between the ages of 11 and 17, and the remaining 470 households represented by at least two such individuals, when initially recruited in 1976.  Ten waves of assessment on the original respondents have been completed through 2002. The sibling relationships in the NYS sample have been used previously by David Rowe and colleagues (e.g., Rowe & Britt, 1991, Journal of Quantitative Criminology, 7:315-331), but are central to the extension of the NYS into a three-generation family study of problem behavior, a collaborative research effort between the Institutes of Behavioral Science and for Behavioral Genetics at the University of Colorado.  We review the pedigree structures of NYS households and estimate sibling resemblance for indicators of problem use of substances and other problem behaviors, and compare them with estimates from data collected beginning in 1993 from genetically informative adolescent and young adult siblings (twins, adoptive siblings, and full siblings) assessed as part of the Colorado Center for the Genetics and Treatment of Antisocial Drug Dependence (CADD).


Danielle M. Dick1, Richard J. Viken1, Jaakko Kaprio2, Lea Pulkkinen3, & Richard J. Rose1.  Parents: The Anti-Drug?  Clarifying the role of parental influence on adolescent smoking and drinking using data from FinnTwin12.4

1Department of Psychology, Indiana University, Bloomington IN, 2Department of Public Health, University of Helsinki, FINLAND, 3Department of Psychology, University of Jyvaskyla, FINLAND, 4FinnTwin12 is supported by NIH (AA 12502), and by the Academy of Finland.

Address:  Department of Medical and Molecular Genetics, Indiana University School of Medicine, 975 West Walnut St., IB-130, Indianapolis, IN  46202-5251 Phone: 317-274-2218 Fax: 317-274-2387 E-mail: ddick@indiana.edu

 

In 1998, the White House Office of National Drug Control Policy launched the National Youth Anti-Drug Media Campaign, an initiative that has emphasized the role parents can play in preventing their adolescents from using drugs.  We sought to better clarify the influence of parents on their adolescents' substance use using data from a longitudinal twin-family study on the development of smoking and drinking patterns across adolescence.  The sample for this study, FinnTwin12, consists of five consecutive birth cohorts of Finnish twins, born 1983-1987.  Twins were assessed with questionnaires at ages 11-12 and reported on several aspects of their relationship with their parents, such as the amount of time spent engaged in activities with their parents, and the degree to which their parents monitor their behavior.  The twins were reassessed at age 14, at which time they reported on their use of cigarettes and alcohol.  Nearly all measures of parental relations at age 12 predicted smoking and drinking behavior at age 14:  children who spent more time with their parents, reported more parental monitoring and a better home atmosphere were less likely to be smoking or drinking at age 14.  However, twins' reports of parenting were under a modest degree of genetic influence, as were their substance use patterns at age 14.  Therefore, to clarify the role of genetic and environmental influences on the relationship between parenting and substance use, we fit bivariate models to reported parental supervision and smoking/alcohol use.  For all measures of parenting and substance use, shared genetic influence could be dropped from the model.  This suggests that the association between perceived parenting and substance use is not confounded by a shared genetic liability.  Genetically informative twin designs provide a useful method for differentiating these developmental influences.


Brian D_Onofrio1, Eric Turkheimer1, Robert Emery1, Jane Mendle1, Stacy Lynch1,

Wendy Slutske2, Andrew Heath3, Nick Martin4.  Association Between Parental Divorce and Offspring Life Transitions:  Selection or Causation?

1Department of Psychology, University of Virginia, Charlottesville, VA, 2Department of Psychological Science, University of Missouri, Columbia, MO, 3Department of Psychiatry, Washington University, St. Louis, MO, 4Genetic Epidemiology Section, Queensland Institute of

Medical Research, Brisbane, QLD, Australia

Address:  Psychology Department, University of Virginia, 102 Gilmer Hall, PO Box 400400, Charlottesville, VA 22904-4400, Phone: 804-982-4750, Email:  briand@virginia.edu

 

Previous research has shown that children of divorced parents are more likely to initiate harmful behaviors at younger ages and engage in untraditional family transitions than children from intact households.  However, the causes underlying these associations are unclear.  Genetically informed designs, including the Children of Twins design, help delineate the possible genetic and environmental processes that account for the association between parental characteristics and child outcomes. In order to disentangle the relation between parental divorce and life transitions in the offspring, a longitudinal study of twins and their adult children was utilized. The sample included 2,554 offspring from 889 twin pairs from the Australian Twin Registry. Results indicated that children from divorced families started to use alcohol, drink heavily, smoke cigarettes, try marijuana, and engage in sexual intercourse at earlier ages than children from intact families. They were also more likely to have a child before they were 18, live with a cohabiting partner, and experience marital separation themselves.  Genetically informed analyses revealed that environmental processes within nuclear families that are associated with parental divorce accounted for the relations. Shared genetic factors and environmental factors which influence twin families could did not mediate the intergenerational associations.


Cathy Fernandes1, Jose Paya-Cano1, Frans Sluyter1, Ursula D'Souza1, Robert Plomin1, and Leonard C. Schalkwyk1.  The power of gene expression profiling and mouse models to unravel  behaviour.

1Social, Genetic and Developmental Psychiatry Research Centre,Institute of Psychiatry, King's College London

Address:  Social, Genetic and Developmental Psychiatry Research Centre, PO 82, Institute of Psychiatry, King's College London, De Crespigny Park, London SE5 8AF, U.K. Tel: +44(0)207 848 0279 Fax: +44(0)207 848 0801 Email spjgcaf@iop.kcl.ac.uk

 

Understanding the genetic component of complex traits such as behaviour has long been a formidable task as it has been difficult to study multiple genes and their products in a single system.  Gene expression profiling using microarrays is an exciting new technology that can take a snapshot of the simultaneous gene expression across thousands of genes.  The mouse is an excellent model for such functional genomics research as there is considerable genetic overlap with humans, behavioural phenotypes have been well characterized and a great deal of genomic information is available. Moreover, both genes and environment can be manipulated or controlled in mice, which makes it possible to study the relationship between genes and environment at the level of gene expression.  I will present data on a study comparing hippocampal gene expression profiles across eight different inbred mouse strains (A/J, C3H/HeJ, FVB/NJ, DBA/2J, C57BL/6J, Balb/cByJ, 129S1/SvImJ and SJL/J) using the Affymetrix GeneChip system.  I will focus on the ways in which the microarray data can be analysed to identify novel candidate genes and potential gene interactions, using examples of gene expression profiles that replicate previous findings and those that nominate novel candidate genes in this study.  The potential of using a range of inbred strains of mice for quantitative  analysis of genetic and environmental influences on gene expression profiles will be considered.


Tom A Fowler1, Jane Scourfield1, Anita Thapar1, and Anne Farmer2.  Does Disabling Fatigue in Children & Adolescents Share Common Genes with Depression?

1Department of Psychological Medicine, University of Wales College of Medicine, Cardiff, Wales, 2Social, Genetic & Developmental Research Centre, Institute of Psychiatry, London, England, 3Supported by PPP Healthcare Medical Trust Grant 1206/192

Address:  Department of Psychological Medicine, University of Wales College of

Medicine, Cardiff, Wales, CF14 4XN Phone: 44 (0)29 20742201 Fax :44 (0)29 20747839 Email:

fowlerta@cf.ac.uk

 

INTRODUCTION:  Due to the overlap of symptomology between chronic disabling fatigue and depression it has been suggested that Chronic Fatigue Syndrome (K. Fukuda, S.E. Straus, I. Hickie, et al. 1994, Annals of Internal Medicine, 121, 953-9) is simply a form of depression. A twin method has been employed to examine whether chronic fatigue and depression in adolescence overlap in terms of genetic aetiology.  METHOD:  A screening questionnaire for lifetime ever disabling fatigue was completed by the parents of 1457 twins, aged 8-17 years, identified from the UK population based twin register CASTANET (Cardiff Study of All Wales & Northwest England Twins). Parents and children over 11 were also asked to complete the Mood and Feelings Questionnaire (E.J. Costello and A. Angold, 1988, Journal of the American Academy of Child and Adolescent Psychiatry, 27, 726-37). To examine whether the co-morbidity between disabling fatigue and  depression is due to shared aetiology, a multivariate twin analysis, including a Cholesky decomposition, was undertaken.  RESULTS:  The study showed that fatigue lasting at least one month was familial and had a substantial genetic aetiology. While there were some risk factors that were shared with depression, nonetheless, most of the genetic and environmental risk factors for chronic fatigue were specific and different to those for depression.  CONCLUSION:  Chronic disabling fatigue in adolescents is partly genetically influenced and this genetic aetiological component is largely independent of the genetic risk factors for depression.


Amber L. Gahagan1 and Irwin D. Waldman1.  Pubertal Maturation differences in Genetic and Environmental Influences on Conduct Disorder.

1Department of Psychology, 532 N. Kilgo Circle, Emory University, Atlanta, GA  30322.

Address:  Department of Psychology, 532 N. Kilgo Circle, Emory University, Atlanta, GA  30322 phone:  (770) 986-7715 fax:  (404) 727-0372 e-mail:  agahaga@emory.edu

 

Theories of the development of antisocial behavior have posited greater genetic influences and lesser environmental influences on early- as compared with late-onset antisocial behavior.  In this study, we examine the effects of pubertal maturation (PM) on the genetic and environmental influences on Conduct Disorder (CD) in children and adolescents.  The participants were twin pairs from the Georgia Twin Registry, a registry of all multiple births recorded in the state birth records of Georgia from 1973 to 1991.  In the present study, questionnaire data were available for ~500 twin pairs.  Parent-report questionnaire measures assessed symptoms of DSM-IV childhood disorders, including the symptoms of CD.  For each CD symptom, the parent indicated on a 0 to 4 scale how frequently the child exhibited the symptom during the past year.  A CD symptom scale score was calculated for each child by summing their scores for each of the symptoms that constituted CD.  The questionnaire also contained a set of sex-appropriate items regarding the Tanner stages of PM.  We addressed a number of hypotheses regarding pubertal differences on genetic and environmental influences on CD.  First, univariate behavior genetic analyses were conducted in order to explore the genetic and environmental influences on PM and CD, respectively.  Second, we conducted univariate behavior genetic analyses of CD separately for pre- and post-pubertal children in order to test whether the genetic and environmental influences on CD differed as a function of PM.  Third, we repeated these analyses separately for boys and girls to see whether any PM differences in the genetic and environmental influences on CD differed by sex.  Finally, we conducted bivariate behavior genetic analyses to estimate common and unique genetic and environmental influences on PM and CD.  In this final set of analyses, we explored to what extent the genetic and environmental influences on PM also underlie the etiology of CD.


Jody M. Ganiban1.  Mothering  and Adolescent Externalizing Behavior: A Behavioral Genetic Exploration.2

1Department of Psychology, The George Washington University, Washington, DC, 2The Swedish Twin Mothers Project was supported by a grant from the NIMH (R01 MH54610, PI: Dr. David Reiss)

Address:  Department of Psychology, The George Washington University, Washington, DC

20052. Telephone: (202) 994-7571 Fax: (202) 994-1602 Email: ganiban@gwu.edu

 

Previous research indicates that distal family-level factors such as marital dissatisfaction and parents’ mental health are related to adolescents’ behavior problems, and that these associations are mediated by parenting style. Although these findings are well-established, the mechanisms that account for them are less clear. This paper will examine the degree to which associations between distal family factors and maternal negativity are explained by genetic and environmental factors. Additionally, exploratory analyses will examine whether associations between family factors and adolescents’ externalizing behaviors are best explained by passive genetic transmission or by evocative gene-environment processes.  Data from the Swedish Twin Mothers Project will be presented (Reiss, D., Cederblad, M., Pedersen, N., et al, 2001, Family Process, 8, 40, 261-272). This study included female twins, their partners, and their children. The sample consisted of 150 MZ and 176 DZ female twin pairs (mean age = 44 + 4.5 years), and 552 children (322 male cousins, and 330 female cousins; average age = 15 + 2.2 years). Self-report measures were used to assess the women’s depressive symptoms, marital dissatisfaction, and parenting style. Mothers also rated their adolescents’ externalizing behaviors. Factor analyses were used to create composite scores for marital dissatisfaction and maternal negativity towards her child.  Preliminary analyses indicate that maternal depressive symptoms and marital dissatisfaction contribute to maternal negativity through different channels:  while genetic factors primarily explained associations between depressive symptoms and maternal negativity, associations between marital dissatisfaction and maternal negativity were explained by nonshared environment factors.  Exploratory analyses will capitalize upon the inclusion of adolescent cousins who vary in genetic relatedness in this study, and estimate the degree to which associations between maternal negativity and adolescents’ externalizing behaviors are explained by the passive transmission of genes from mother to child, or by the adolescents’ unique genetic makeup.


Heather L. Gelhorn1, Michael C. Stallings1, Susan E. Young1, Robin P. Corley1, John K. Hewitt1, Thomas J. Crowley2.  A Detailed Investigation of DSM-IV Conduct Disorder Symptoms: Heritability of Individual Items and a Comparison of Selected and Unselected

Samples.3

1Institute for Behavioral Genetics, University of Colorado, Boulder, CO, 2Dept. of Psychiatry, U. of Colorado Health Sciences Center, Denver, CO, 3Supported by NIDA grants DA-05131, DA-11015, DA-12845

Address:  Institute for Behavioral Genetics, Campus Box 447 University of Colorado, Boulder, CO 80309-0447 Phone: (303) 735-2428 Fax: (303) 492-8063 Email: gelhorn@colorado.edu

 

Research on antisocial traits from twin and adoption studies has consistently implicated genetic factors in the etiology of conduct disorder (CD). However few studies have looked at individual symptoms to determine the extent to which these behavioral problems may be genetically or environmentally influenced.  Additionally, researchers frequently employ case-control designs where CD symptoms are compared between selected samples and unselected controls. Such strategies assume that the selected cases simply represent the extremes of an underlying distribution that is shared with the controls.  There has been very little research to directly support this assumption.  The current study uses both patient and community-based adolescents participating in the Center for Antisocial Drug Dependence at the University of Colorado.  The aim of the investigation is to characterize individual DSM-IV CD symptoms in terms of heritability, age trends, severity, and the underlying factor structure of the disorder.  Twin analyses were conducted to estimate heritability at the item level, yielding a wide range of estimates (.00 - .80). Guttman scaling was carried out to examine possible heterogeneity in the relative severity of symptoms between the selected and community samples.  Binary factor analyses were used to explore factor structures for the two groups.  Results indicate that selected and control samples likely share a single underlying distribution for CD symptoms.  In general the single distribution appears to be a robust assumption and there are few items that show important variation between clinical and control adolescents in our sample.  The results of this study will be useful for future studies aimed at finding genetic variants that contribute to risk for CD.


Kyle L. Gobrogge1 and Kelly L. Klump1.  Homosexual Mating Preferences from an Evolutionary Perspective.2

1Department of Psychology, Michigan State University, East Lansing, MI, 2Supported by the Michigan State University Dean’s Assistantship Undergraduate Grant

Address:  5400 Mall Drive West Apt. 3111, MI 48917 Phone:  517/881 3776 Email: gobrogge@msu.edu

 

Objective:  Several studies have examined mating behavior from an evolutionary perspective in both heterosexual and homosexual male populations (Silverthorne & Quinsey, 2000; Bailey et al., 1994; Jankowiak et al., 1992; Freund et al., 1973).  These investigations have shown that both heterosexual and homosexual men prefer younger partners compared to heterosexual women.  However, most studies relied on laboratory-based assessments that required subjects to hypothesize about their preferences and did not specify what types of relationships were desired for each preference.  The purpose of the present study was to examine these issues naturalistically by comparing partner preferences of homosexual and heterosexual men who placed internet personal advertisements.  Method:  Subjects included 338 homosexual and 265 heterosexual men placing internet personal ads on lavalife.com in each of the 50 states in the U.S.  Only subjects who mentioned a preferred age of partner were included.  Ads were coded for the subject’s age, the relationship type they were seeking (i.e., sexual encounter versus longer-term relationship), and their partner age preferences.  Chi-square analyses and ANOVAs with Tukey’s post hoc t-tests were used to analyze differences in general preferred age (i.e., younger, older, same, or open), mean preferred age of partner, and relative difference in age between ad placer and partner sought.  Results:  Heterosexual and homosexual men seeking sexual encounters only sought significantly older partners and a wider age range of partners compared to men seeking longer-term relationships.  Discussion:  These findings indicate that both heterosexual and homosexual men are less selective about their partner’s age when seeking a sexual encounter.  These results suggest that sex, age, and relationship type preferences may develop independently.  Additional research examining a range of mating preferences and relationship types sought in heterosexual and homosexual men and women is needed in order

to replicate the current findings.


Detre A. Godinez1, Michael C. Stallings1, Susan E. Young1, Robin P. Corley1, John K. Hewitt1.  Investigating Genetic and Environmental Influences on Age at Onset of Alcohol Use and the Latency from First Use to Regular Use in the Colorado Adolescent Twin Sample.2

1Institute for Behavioral Genetics, University of Colorado, Boulder, CO, 2Supported by NIDA grants DA-05131, DA-11015 and DA-12845

Address:  Institute for Behavioral Genetics, Campus Box 447 University of Colorado, Boulder, CO 80309-0447 Phone: (303) 735-2428 Fax: (303) 492-8063 Email:  godinez@Colorado.edu

 

Although there is a considerable body of literature investigating genetic and environmental influences on drinking patterns, much less is known about the determinants of age at initial alcohol use and how quickly one progresses from initial use to higher levels of use.  The current study investigates the genetic and environmental influences on age of alcohol initiation, age of regular (at least monthly) use, and the latency between initiation and regular use in an adolescent sample.  Participants were 567 monozygotic and 598 dizygotic twin pairs between the ages of 12 and 18, drawn from a community based twin sample (Colorado Twin Registry).  Two analyses were conducted.  First, we used complete data where both twins had:  1)initiated use; 2)progressed to regular use; or 3)reached both use milestones—for the transition latency analysis. Age of onset was treated as a continuous variable and was corrected for age trends in the sample. In a second analysis we used methods described by Pickles et al. (1994, Behav. Genet., 24, 457-468), allowing for censored cases (i.e., that some adolescents have not yet initiated use or progressed to regular use).  Results of analysis-1 indicated moderate heritability for age at first use, age at regular use, and the transition latency.  Among twins who had both initiated use (ignoring monthly use status), shared environmental effects were estimated at approximately 20%.  However, among twin pairs that had progressed to monthly use there was little evidence for shared environmental influences for age of onset or the transition latency.  Results of analysis-2, utilizing all of the twin data, confirmed moderate heritability for age at initiation and age at monthly use, but suggested the importance of shared environmental effects (estimated at approximately 30%).  Additional survival analyses will examine the transition latency and potential telescoping effects due to retrospective reporting.


Julia D. Grant1 and Kathleen K. Bucholz1.  Early alcohol use as a risk factor for later alcohol dependence:  Genetic and environmental influence and overlap.2

1Department of Psychiatry, Washington University School of Medicine, St. Louis, MO, 2Supported by NIH Grants DA14632, DA14363, AA07728, AA11998

Address:  Dept. of Psychiatry, Washington Univ. School of Medicine, 40 N. Kingshighway, Ste. 3, St. Louis, MO 63108 Phone: (314) 286-2255  FAX: (314) 286-2243  Email: grantj@msnotes.wustl.edu

 

Adolescent behaviors such as alcohol use have the potential to impact adult development. The present analyses use data from a sample of Vietnam-era veteran male twin pairs to examine genetic and environmental contributions to early alcohol use, and how early use relates to adult alcohol dependence.  Data from 3,452 twin pairs (MZ=1,923; DZ=1,529; mean age=41.9 years) who participated in a 1992 telephone diagnostic interview for the Vietnam Era Twin Study (VETS) were analyzed.  Eighty-nine percent (n=6,112) of the respondents had used alcohol regularly (i.e., at least once a month for 6+ months).  Age of onset for regular drinking (M = 18.4 years) was used to create a 6-level age of use variable (with never drinkers included as the "latest onset" category).  Additionally, lifetime alcohol dependence (AD) was assessed using DSM-III-R criteria; 40% of regular drinkers met AD criteria.  Structural equation modeling (using Mx) indicated that age of onset for regular alcohol use was heritable, with genetic influences explaining 34% of the variance (95% CI = 22-46).  Shared environmental influences were also significant, explaining 15% of the variance (95% CI = 5-25).  Nonshared environmental influences explained 51% of the variance (95% CI = 47-55).  AD was also heritable, with genetic influences accounting for 52% of the variance (95% CI = 34-58).  Shared environmental influences on AD were non-significant, explaining 1% of the variance (95% CI = 0-17); nonshared environmental influences accounted for 47% of the variance (95% CI = 41-53).  Interestingly, the overlap between the two measures was almost entirely attributable to genetic influences:  the genetic correlation of -0.71 (95% CI = -0.54 to -0.97) indicated that 50% of the genetic variance in AD was overlapping with genetic influences on age of onset for regular alcohol use.  Neither the shared environmental nor the nonshared environmental correlation was significant.


Guang Guo, Glen Elder, and Nathan Hamilton.  Role of Peers, Heritability, & Adolescent Drinking.

Department of Sociology, University of North Carolina, Chapel Hill

Address:  Department of Sociology, CB# 3210, UNC, Chapel Hill, NC 27599 Phone: 919/962 1246 email: guang_guo@unc.edu

 

Using same-sex twin and sibling data from the National longitudinal Adolescent Health Study based in the University of North Carolina at Chapel Hill, we investigate how heritability for drinking is influenced by friends’ drinking behavior. Heritability is estimated using traditional twin and sibling methods. Drinking scores were used to measure adolescents’ drinking behavior (0: never drinks; 1: once in the last yr; 2:  once a month; 3: 2-3 times last month; 4: 1-2 a week; 5: 3+ a week). Unlike most previous studies on friends, in which the information on friends are usually reported by the ego him or herself, the friends in our study are respondents themselves. We classify one’s drinking friends into three types:  concordant/high (when both siblings’ friends score two or greater), concordant/low (when one or both score one or zero), and discordant (when friends’ drinking score is two points away from each other 1 vs. 3 or 2 vs. 4. Our analysis shows consistent patterns of evidence for relatively high heritability for concordant/high, lower heritability for concordant/low, and the lowest heritability for discordant. Moreover, the gap between concordant/high and discordant is significantly larger than zero; but gap between concordant/high & concordant/low is not.  These results are robust for three measures of friends context: average friends (drinking context is measured by the average drinking scores of all nominated friends); best friend (drinking context is measured by the drinking score of ego’s best friend); and maximum exposure (drinking context is measured by the drinking score of highest score among the nominated friends). Larger samples with twins and full sibs yield more significant results.  Selection can’t explain away the results. A likely explanation for the pattern in the data:  Friends’ drinking behaviors moderate the expression of genetic propensities on drinking. For analysis focusing on gene-by-environment interactions using DNA data (when available), one candidate environmental effect to look at: friends’ drinking behaviors.


Haberstick, B.C., Schmitz, S., Young, S.E., and Hewitt, J.K.  Individual differences in behavioral problems as rated by different teachers across middle childhood and early adolescence.

Institute for Behavioral Genetics, University of Colorado, Boulder

Address:  Institute for Behavioral Genetics, Campus Box 447, University of Colorado,

Boulder CO 80309 Email:  Brett.Haberstick@Colorado.edu

 

In the current study, we report findings from the first large scale longitudinal twin study of siblings rated by different teachers.  We examined teacher ratings on the Teacher Report Form (TRF, Achenbach, 1991b) collected across six ages during middle childhood and early adolescence for a large sample of monozygotic (MZ) and dizygotic (DZ) twin siblings in different classrooms.   As an index of enduring characteristics of a child, we took an average of these observations for each of the 358 MZ (194 girls and 164 boys) and 310 DZ twin siblings (144 girls and 166 boys).  For these averaged scores, boys, on average, were rated as having more social, thought, and attention problems, as well as displaying more delinquent, aggressive, and overall externalizing behaviors than girls.  Furthermore, MZ twins were rated as more similar than DZ twins on each of the behavioral and broadband scales on the TRF.  Univariate model fitting to observed individual variances and pair covariances was conducted in order to estimate the genetic and environmental contributions to individual differences in problem behavior within gender.  Heritability estimates were found not to differ between girls and boys and ranged between 54% and 61% for overall externalizing behaviors and its behavioral scales while ranging between 27% and 73% for overall internalizing behaviors and its behavioral scales.  Individual differences in stable attention and social problems were also found to be

strongly heritable across six years of observations, with estimates of 67% and 61%, respectively.   Environmental influences shared by members of a twin pair did not show a significant influence on any scale.   From these findings we conclude that individual differences in persistent problem behaviors during middle childhood and early adolescence within the school setting are due to a combination of genetic contributions and non-shared environmental influences.


 Noa Heiman1, Mats Larsson2, Michael C. Stallings1, Susan E. Young1, Andrew Smolen1, and John K. Hewitt1.  Novelty seeking and dopamine receptor polymorphisms in adolescence.

1Institute for Behavioral Genetics, University of Colorado, Boulder, CO, 2Center for Developmental Research, Orebro University, Sweden

Address:  Institute for Behavioral Genetics, Campus Box 447, University of Colorado, Boulder, CO, 80309 Phone: 303 492 2817 Fax: 303 492 8063 E-mail: noa.heiman@colorado.edu

 

The importance of genetic contributions to personality traits is well established, as indexed by moderate heritability estimates found in twin studies.  In addition, several studies have reported associations for particular variants of the dopamine D4 receptor gene (DRD4) with the personality trait of novelty seeking.  The DRD4 gene has a variable number of tandem repeats (VNTR) in its third exon, which have been shown to affect the function of the D4 receptor efficiency in vitro.  However, replication studies have shown mixed results.  The aim of our study was to investigate whether such an association replicates in a large adolescent sample.  Our sample consisted of 1227 adolescents between the ages of 11 and 19. Adolescents were drawn from the Colorado Twin Registry, a community-based sample of twins residing in Colorado (N=1083). The sample also included community non-twin participants who were drawn from the Colorado Adolescent Substance Abuse (ASA) family study (N=144).  Novelty seeking was measured using the Tridimensional Personality Questionnaire (TPQ) in adolescents aged 16 and older.  For children under age 16 the Junior Temperament and Character Inventory (J-TCI) was used.  Scores were standardized within these samples and then combined.  We found no evidence for an association between novelty seeking and the DRD4 VNTR.  Analysis of variance yielded non-significant results both when defining the risk allele as having seven or more repeats, and when comparing the presence or absence of seven repeats only.  Including covariates such as age, gender, and ethnicity yielded non-significant results as well.  A sub-sample composed of one twin randomly selected from each family yielded non-significant results as well.  Allele frequencies for the risk allele using both definitions were 0.19 and 0.21, respectively.  Additional analyses investigate potential associations between novelty seeking and other polymorphisms in the dopamine system, as well as potential epistatic effects between these candidate loci.


Brian M. Hicks, Linde Althaus, Robert F. Krueger, Matt McGue, and William

G. Iacono.  Adolescent sexual behavior and externalizing psychopathology:

Longitudinal, cross-sectional, and biometric analyses.

Department of Psychology, University of Minnesota, Minneapolis, MN 55455-0344, The Minnesota Twin Family Study is supported in part by USPHS grants AA00175, AA09367, DA05147, and MH65137

Address:  Department of Psychology, University of Minnesota, Elliott Hall, 75 E. River

Rd., Minneapolis, MN 55455-0344 Telephone: 612 625 5624  Fax: 612 626 2079  Email: hick0131@tc.umn.edu

 

Sexual behavior, particularly sexual behavior of a risky or irresponsible nature, has been linked with substance abuse and tendencies toward behavioral disinhibition.  Previously, we have reported that disinhibitory syndromes, including adult and child antisocial behavior and alcohol and drug dependence, can conceptualized as a spectrum of disorders indicative of a broad and highly heritable psychopathological process called Externalizing (EXT).  In this investigation, we sought to delineate the relations between sexual behavior and various facets of the EXT spectrum using data from the 11 and 17 year-old cohorts of the Minnesota Twin Family Study.  Both sexual behavior and EXT disorders were assessed via in-person, structured interviews.  For our measure of sexual behavior, we employed a dichotomous variable for engaging in sexual intercourse as well as a dimensional variable of sexual behavior that included items covering dating behavior, frequency of sexual intercourse, fear of being pregnant (or fear of getting someone pregnant),  and actual pregnancy (or actually getting someone pregnant).  We conducted longitudinal, cross-sectional, and biometric analyses.  First, we investigated whether childhood EXT disorders at age 11 (e.g., conduct disorder, oppositional defiant disorder, attention deficit hyperactivity disorder) predicted the emergence of sexual behavior at age 14.  Second, we examined cross-sectional associations between sexual behavior and adult EXT phenotypes (adult antisocial behavior and substance disorders) at age 17.   Finally, we examined the etiological sources of covariation between sexual behavior and EXT for both the longitudinal and cross-sectional analyses.       


Robert Hitzemann1, Cheryl Reed1, Barbara Hitzemann1, Robert W. Williams1, James Sikela1, Kari Buck1, Jonathan Flint1, and Christopher Talbot1.  On the intergration of QTL gene expression and sequence analyses.2

1Department of Behavioral Neuroscience, Oregon Health and Science University, Portland, OR 97239, 2Supported in part by AA11034, MH 51372 and the Vetrans Affairs Medical Research Service.

Address:  Department of Behavioral Neuroscience Oregon Health Sciences University 3181 SW Sam Jackson Park Road Portland OR 97201-3098 Telephone: (503) 404-2858 FAX: (503) 494-6877 E-Mail: hitzeman@ohsu.edu

 

Although hundreds if not thousands of quantitative trait loci (QTL) have been described for a wide variety of complex traits, only a very small number of these QTLs have been reduced to quantitative trait genes (QTGs) and quantitative trait nucleotides (QTNs). A strategy will described for detecting QTGs and QTNs that is based on leveraging the information contained within the haplotype structure of the mouse genome. The strategy utilizes the 6 F2 intercrosses that can be formed from the C57BL/6J (B6), DBA/2J (D2), BALB/cJ (C) and LP/J (LP) inbred mouse strains. Focusing on the phenotype of basal locomotor activity, it was found that in all three B6 intercrosses, a QTL was detected on distal chromosome 1; no QTL was detected in the other three intercrosses and thus, it was assumed that at the QTL, the C, D2 and LP strains were identical by descent. These intercross data were used to form a simple algorithm for interrogating microsatellite, SNP, brain gene expression and sequence databases.  The results obtained point to Kcnj9 (which has a markedly lower expression in the B6 strain) as being the likely QTG.  Further, it is suggested that the lower expression in the B6 strain results from a  polymorphism that disrupts the binding of at least 3 transcription factors - Ikaros 1, MZF1 and C/EBPbeta.  For these factors, the homology of the binding site motif to that found in Kcnj9 was highest for Ikaros 1 and poorest for C/EBPbeta . Overall, the method described should be widely applicable to the analysis of QTLs.  


Yoon-Mi Hur.  Two decades of assortative mating for education in Korea.

Graduate School of Education, Hansung University, Seoul, Korea

Address:  Graduate of Education, Hansung University, Seoul, Korea Phone:  +82 2 760 4167 Fax:  +82 2 760 4167 Email:  maryhur@hananet.net

 

Assortative mating affects heritability estimates in twin and family studies because it increases genetic resemblance between first-degree relatives.  Recently increasing numbers of twin studies on the basis of Asian samples have been published.  Degrees of assortative mating in Asian samples, however, are little known.  The present study examined assortative mating for educational attainment in Korea using birth record data from 1981 to 2000.  Implications of assortative mating for education in heritability estimates of intelligence and related traits in Asian samples are discussed.


N Jacobs1, F Rijsdijk2, C Derom3,4, M Danckaerts5, E Thiery4,6, R  Derom4, R Vlietinck3,4, and J van Os1,7.  Child problem behaviour and lower cognitive ability: a twin study of the causes of association.

1Department of Pyschiatry and Neuropsychology, Maastricht University, European Graduate School of Neuroscience, P.O. Box 616, 6200 MD Maastricht, The Netherlands,  2Social, Genetic & Developmental Psychiatry Research Centre, Institute of Psychiatry, London SE5 8AF, United Kingdom, 3Center for Human Genetics, Catholic University Leuven, Herestraat 49, 3000 Leuven, Belgium, 4Association for Scientific Research in Multiple Births, Kwadenplasstraat 12, 9070 Destelbergen, Belgium, 5Department of Child Psychiatry, University Hospital Gasthuisberg, Herestraat 49, 3000 Leuven, Belgium, 6Department of Psychiatry and Neuropsychology, University Ghent, De Pintelaan 185, 9000 Ghent, Belgium, 7Division of Psychological Medicine, Institute of Psychiatry, De Crespigny Park, London SE5 8AF, United Kingdom

Address:  Dept. Psychiatry and Neuropsychology,Section Social Psychiatry and Psychiatric Epidemiology Maastricht University, P.O. Box 616, 6200 MD Maastricht, The Netherlands Phone: +31-43-3299785  Fax: +31-43-3299708   Email: nele.jacobs@sp.unimaas.nl

 

Previous work has demonstrated associations between lower cognitive ability and childhood and adult non-psychotic psychopathology. As both cognitive ability (CA) and child psychopathology (CP) are influenced by genetic factors, one explanation for the association is that they are the pleiotropic manifestations of the same underlying genetic factors. The present paper examines three possible causes of the association: additive genetic factors, common environmental factors and individual-specific environmental factors.  376 twin pairs from the East Flanders Prospective Twin Survey were examined with the Child Behaviour Checklist and the Wechsler Intelligence Scale for Children-Revised. The cross-twin within-variable, within-twin cross-variable and cross-twin cross-variable correlations were calculated. Using structural equation modelling, bivariate models were fitted. The best fitting model was chosen, based on likelihood and parsimony.  The observed phenotypic correlation between CP and CA was -0.19 (95% CI:-0.09,-0.27), with genetic factors accounting for about 84% of the observed correlation. Bivariate model fitting quantified the genetic correlation between CP and CA at -0.27 (95% CI:-0.12,-0.42) and the individual-specific environmental correlation at -0.17 (95% CI:  -0.03,-0.31).  In children, three different genetic factors may exist: one that solely affects the liability to CP, one that has only an effect on CA and one that influences both CP and CA. While individual-specific environmental factors can influence the liability to both traits, our results suggest that most of

the environmental factors that increase the risk of CP do not influence CA and vice versa.


N Jacobs1, I Myin-Germeys1, L Krabbendam1, C Derom2,3, and J van Os1,4.  Do genes make people feel blue after stress?  An experience sampling twinstudy.

1Department of Pyschiatry and Neuropsychology, Maastricht University, European Graduate School of Neuroscience, P.O. Box 616, 6200 MD Maastricht, The Netherlands, 2Center for Human Genetics, Catholic University Leuven, Herestraat 49, 3000 Leuven, Belgium, 3Association for Scientific Research in Multiple Births, Kwadenplasstraat 12, 9070 Destelbergen, Belgium, 4Division of Psychological Medicine, Institute of Psychiatry, De Crespigny Park, London SE5 8AF, United Kingdom

Address:  Dept. Psychiatry and Neuropsychology,Section Social Psychiatry and Psychiatric Epidemiology Maastricht University, P.O. Box 616, 6200 MD Maastricht, The Netherlands Phone: +31-43-3299785  Fax: +31-43-3299708   Email: nele.jacobs@sp.unimaas.nl

 

People show different emotional responses to stressors.This study examined to what degree individual differences  in emotional reactivity to small daily life stressors could be explained by genetic and/or environmental factors.  152 female twin pairs (99 MZ and 53 DZ) participated in an Experience Sampling study (ESM). ESM is a structured diary technique used to assess stressors and mood at 10 random times for 5 consecutive days.  An increase in self-reported subjective stress was associated with an increase in negative affect and a decrease in positive affect. For each subject, the stress-induced increase in negative affect and  decrease in positive affect was calculated.  Structural equation modelling was applied to these variables. The best fitting model was chosen, based on fit and parsimony.  For the stress-induced decrease in positive affect, the best fitting model was the model with a common environmental factor, explaining 19% of the variance (95%CI 0.03-0.34) and an individual-specific environmental factor, explaining 81% of the variance (95%CI 0.66-0.97).  For the stress-induced increase in negative affect, the best fitting model was the model with a dominant genetic factor, explaining 35% of the variance (95%CI 0.17-0.51) and an individual-specific environmental factor explaining 65% of the variance (95%CI 0.49-0.83).  This was the first study ever to examine the role of genes and environment in emotional reactivity to daily stressors. It was found that i)stress-induced increase in negative affect was influenced by genetic and environmental factors, ii)stress-induced decrease in positive affect was influenced only by environmental factors.


Kristen C. Jacobson1, Andrew Crider2, William S. Kremen3, Hong Xian4, Brian Waterman4, Seth A. Eisen4, Ming T. Tsuang5, and Michael J. Lyons6.  Genetic and environmental influences on electrodermal lability and its association with antisocial behavior.

1VIPBG, Department of Psychiatry, Virginia Commonwealth University, 2(emeritus) Department of Psychology, Williams College, 3Department of Psychiatry, UC Davis, 4School of Medicine, Washington University in St Louis, 5Harvard Institute of Psychiatric Epidemiology and Genetics, Harvard University, 6Department of Psychology, Boston University

Address:  VIPBG; Department of Psychiatry, Virginia Commonwealth University; 800 E Leigh Street, PO Box 980126, Richmond, VA 23298-0126 Phone: 804 828 8126 Fax: 804 828 1471 Email: kcjacobs@hsc.vcu.edu

 

The present study used data from 345 twin pairs from the Alcohol Vulnerability Study to examine genetic and environmental influences on three measures of skin conductance response (SCR): SCR lability during rest, number of trials to habituation, and SCR lability during a digit transformation task situation. Participants were male-male twins aged 41-58 from the Vietnam Era Twin Registry. Although the three variables were highly correlated ( r = .57-.70), and univariate analyses revealed heritabilities of ~ .40 for each of the three variables, multivariate analyses indicated that there were significant genetic influences on the measures that were unrelated to genetic influence on the underlying latent trait. Specifically, 30% of the total heritabilities for trails to habituation and SCR lability during task were due to genetic influences not shared with the other measures. Only a small proportion (8%) of the genetic influence on resting SCR lability was not shared with genetic influence on the latent trait. Variation in the latent trait was due approximately equally to genetic and nonshared environmental influences. Common environmental influences were not significant. These results indicate that genetic factors play an important role in accounting for variation in an underlying latent phenotype that is defined by related measures of SCR. However, there are additional genetic factors that influence variation in individual measures of SCR, especially SCR sensitivity and SCR during task situations. Further analyses will examine the extent to which these three measures of SCR are related to conduct-disordered and adult antisocial behaviors, and will additionally examine the extent to which these associations between SCR and antisocial behavior are influenced by common genetic factors.


Jang, K.L.1, Dick, D.M.2, Taylor, S.1, Livesley, W.J.1, Stein, M.B.3.  Psychosocial adversity and personality function:  Exploration of the mechanisms of gene-environment interplay.

1Department of Psychiatry, University of British Columbia, Canada, 2Department of Psychology, University of Indiana, USA, 3Department of Psychiatry, University of California—San Diego, USA

 

The study tested for evidence for gene-environment interplay on the development of personality dysfunction:  specifically environmental control of genetic expression and experiential moderation of preexisting environmental conditions.  A battery of continuously measured psychosocial stressors was assessed in three independent samples of twins (141, 194, and 86 MZ pairs and 129, 184, and 77 DZ pairs, respectively) who reported on the environment of the childhood home, parental bonding, traumatic events and traits delineating personality disorder.  Genetic moderator path models evaluated the magnitude of heritable and non-heritable effects on personality function across levels of each of the psychosocial variables.  Results showed evidence for gene-environment interaction but most of the psychosocial variables tested impacted on the environmental conditions influencing personality function.


Wendy Johnson, Matt McGue, and William G. Iacono1.  Disruptive behavior and school achievement: Genetic and environmental relationships in 11-year-olds.2

1Department of Psychology, University of Minnesota, Twin Cities, Minneapolis, MN, 2The Minnesota Twin Family Study is supported by National Institute on Drug Abuse Grant #DA05147 and National Institute on Alcohol Abuse and Alcoholism Grant #AA09367.

Address:  Department of Psychology, University of Minnesota, 75 East River Road, Minneapolis, MN 55455 Phone: 952-473-1673 Fax: 952-473-1998 Email: john4350@tc.umn.edu

 

There is a well-established association between conduct problems and attentional problems. In addition, there is a well-established association between conduct problems and poor academic achievement including school drop out. Several studies (e.g. D. M. Fergusson and L. J. Horwood, 1995, Journal of Abnormal Child Psychology, 23, 183-199) have concluded, however, that, in the absence of attentional difficulties or below average IQ, conduct difficulties are not associated with later difficulties with academic achievement. This has largely been based on observations of clinic-referred males of high school age. Using the population-based Minnesota Twin Family Study, we investigated the relationships among conduct and attentional problems, IQ, and academic achievement in 11-year-old males and females. Following development of appropriate sex-specific phenotypic models, we investigated the genetic and shared and nonshared environmental influences on the relationships observed.


Wendy Johnson and Robert F. Krueger1.  Genetic and environmental structure of the big 5 personality traits in a national twin sample.2

1Department of Psychology, University of Minnesota, Twin Cities, Minneapolis, MN, 2The National Survey of Midlife Development in the United States is supported by the John D. and Catherine T. MacArthur Foundation Research Network on Successful Midlife Development and by National Institute on Aging Grant #AG20166.

Address:  Department of Psychology, University of Minnesota, 75 East River Road, Minneapolis, MN 55455 Phone: 952-473-1673 Fax: 952-473-1998 Email: john4350@tc.umn.edu

 

Using a nationwide sample of 275 same-sex dizygotic and 315 monozygotic twin pairs from the National Survey of Midlife Development in the United States (MIDUS), we investigated the etiologic basis for the phenotypic coherence of the five domains (Extraversion, Neuroticism, Agreeableness, Conscientiousness, and Openness) of the Big Five Model of personality.  We compared multivariate models specifying varying degrees of latent phenotypic structure and estimating genetic and environmental structural influences among adjectives describing each of the Big Five personality domains. Results showed both common and specific genetic and environmental influences for each domain, suggesting that all of the domains are etiologically complex. In addition, the models specifying the domains as latent phenotypic constructs fit more poorly than models specifying less coherent structure, particularly for Openness and Agreeableness. These results raise questions about the BFM as an explanatory model of personality or, alternatively, about the etiological unity of latent phenotypic personality trait constructs in general.


Kenji Kato1,2 and Nancy L. Pedersen1,3.  Personality and coping: A study of twins reared apart and twins reared together.4

1Department of Medical Epidemiology, Karolinska Institutet, Stockholm, Sweden, 2School of Health Science, Faculty of Medicine, Osaka University, Osaka, Japan, 3Department of Psychology, University of Southern California, Los Angeles, CA 90089, 4Supported by National Institute on Aging Grants AG-04563 and AG-10175

Address:  Kenji Kato, Dept. of Medical Epidemiology, Karolinska Institutet, Box 281, SE-171 77 Stockholm, Sweden Email: Kenji.Kato@mep.ki.se

 

This study aimed to explore the genetic and environmental factors influencing stress coping styles, age and gender differences, and the relationship between coping and personality in middle-aged and older adults, as part of the ongoing Swedish Adoption/Twin Study of Aging (SATSA). SATSA participants in the third questionnaire wave (58 MZ and 147 DZ twin  pairs reared apart, and 101 MZ and 140 DZ twin pairs reared together; the mean age was 61) were given the Billings and Moos Coping Measure. Three meaningful factors were derived from factor analysis and termed ‘Problem Solving,’ ‘Turning to Others,’ and ‘Avoidance.’ For the purposes of these analyses, Neuroticism and Extraversion from the EPI and Openness to  Experience from the NEO-PI were included. Univariate model fitting showed additive genetic influences for the coping scales (variance components ranged 16-56%). Turning to Others and Avoidance indicated gender differences in the relative importance of genetic and environmental influences. The best-fit models contained only additive genetic and nonshared environmental factors except Turning to Others for women, which also had shared rearing environmental influences. No age differences were found in this sample. Multivariate model fitting showed substantial genetic contributions to the phenotypic correlations in the associations between Problem Solving and Openness to Experience, between Turning to Others and Openness to Experience for women, and among Avoidance, Extraversion, and Neuroticism for women. The findings suggest that genetic influences on adults' coping are differentially mediated by genetic factors in common with personality traits.


Kelly L. Klump1, Matt McGue2, and William G. Iacono3.  Puberty and eating pathology: Shared genetic effects.

1Department of Psychology, Michigan State University, East Lansing MI, 2Department of Psychology, University of Minnesota, Minneapolis MN, 3Supported by NIH Grants MH 65447, DA 05147, AA 09367

Address:  Michigan State University, 129 Psychology Research Building, East Lansing, MI 48824 Phone: 517 432 9861 Fax: 517 432 2476 E-mail: klump@msu.edu

 

Previous research has found significant increases in rates of clinical eating disorders as well as disordered eating characteristics (K.L. Klump, M. McGue, and W.G. Iacono, in press, Int J Eating Disorders) in girls following puberty. Recent findings suggest this increase may be due to genetic factors, as genetic influences on weight preoccupation and overall disordered eating appear to be greater in pubertal than pre-pubertal girls (K. L. Klump, M. McGue, and W.G. Iacono, in press, Int J Eating Disorders).  The purpose of the present study was to directly examine this possibility by investigating shared genetic effects between puberty and these disordered eating characteristics in a population-based sample of female twins.  Participants included 276 fourteen year-old female twins from the Minnesota Twin Family Study.  Disordered eating was measured with the Minnesota Eating Disorders Inventory (M-EDI) (K. L. Klump, M. McGue, and W. G. Iacono, 2000, J of Abn Psychology, 109, 231-251), while pubertal status was measured with the Pubertal Development Scale (PDS) (A.C. Petersen, L. Crockett, M. Richards, and A. Boxer, 1988, J of Youth and Adolescence, 17, 117-133).  Model-fitting analyses indicated significant genetic and nonshared environmental influences on PDS and M-EDI scales, with significant overlap in genetic risk factors for the two phenotypes.  These findings corroborate previous research highlighting the importance of puberty for the development of disordered eating. Moreover, results extend previous work by suggesting significant genetic relationships between disordered eating and puberty that may account for observed phenotypic associations.


Laura B. Koenig1, Matt McGue1, Robert F. Krueger1, and Thomas J. Bouchard, Jr.1.  Examining the relationship between religiousness, prosocial and antisocial behavior: Genetic versus environmental mediation.

1Department of Psychology, University of Minnesota, Minneapolis, MN

Address:  Department of Psychology, University of Minnesota, Elliott Hall, 75 East

River Road, Minneapolis, MN 55455-0344  email: koen0099@umn.edu

 

Although religiousness is considered a major protective factor against the expression of substance abuse and other antisocial behavior, it remains unclear whether these associations are primarily genetically or environmentally mediated.  In order to investigate this and related issues, religiousness was assessed by self-report in a sample of adult male twins (169 MZ pairs and 104 DZ pairs, mean age of 33 years).  Twin correlations for retrospective reports of religiousness in adolescence showed little difference between MZ (r=.69) and DZ (r=.59) twins.  Twin correlations for reports of current level of religiousness, however, did show stronger MZ (r=.62) than DZ (r=.42) similarity.  Biometric analysis of the two religiousness ratings revealed that genetic factors were significantly weaker (12% vs. 44%) and shared environmental factors were significantly stronger (56% vs. 18%) in adolescence compared to adulthood.  Nonetheless, both current and adolescent religiousness independently contributed to the prediction of adult antisocial and prosocial behavior.  Results from a joint biometric analysis of religiousness and antisocial and prosocial behavior, undertaken to determine whether the relationships among these

variables were genetically or environmentally mediated, will be presented.


Sulev Kõks1, Kati Koido1, Eduard Maron2, Tiit Nikopensius3, Ants Kurg3, Jakov Šlik2, Eero Vasar1, and Andres Metspalu3,4.  Analysis of Single Nucleotide Polymorphisms in patients with mood disorders.5

1Department of Physiology, University of Tartu, Estonia, 2Department of Psychiatry, University of Tartu, Estonia, 3Institute of Molecular and Cell Biology, Tartu, Estonia, 4Asper Biotech Ltd., Tartu, Estonia, 5 Supported by Estonian Science Foundation Grants 4562, 4614, 5528

Address:  Department of Physiology, University of Tartu, 19 Ravila Street, 50411 Tartu, Estonia

Phone: +372 7 374 335 Fax: +372 7 374 332 Email: Sulev.Koks@ut.ee

 

Mood disorders are among the most prevalent forms of mental illnesses. Depression and bipolar disorder are both highly heritable disorders, with genetic factors comprising roughly 50% of the risk for depression and as much as 80% of the risk for bipolar disorder. Linkage studies have focused on genomic regions containing putative candidate genes involved in various neurotransmitter systems. Weak associations have been reported between depression and several single nucleotide polymorphisms, but results are inconsistent. The aim of the present study was to find possible associations between the most extensively studied polymorphisms and unipolar or bipolar affective disorder. Patients with panic disorder were also involved in this study. We genotyped 128 SNPs from altogether 22 genes with Arrayed Primer Extension (APEX) technology. The SNPs included in our study cover several neurotransmitter systems (serotonin, dopamine, opioid and cholecystokinin) and additionally one plausible candidate gene with unknown function – wolframin. We recruited 150 healthy controls, 105 patients with unipolar disorder, 108 patients with panic disorder and 38 patients with bipolar disorder. All subjects were interviewed with M.I.N.I. and SSP tests. We found that several putative SNPs were not polymorphic at all. Evidence for significant association was established for some SNPs. In addition, some SNPs were positive with both unipolar and bipolar disorder group, other SNPs were disorder-specific – positive association was established only with a respective group of patients. We found positive association in patients with panic disorder for SNPs in the CCK, CCK2R, POMC, PENK, 5-HT1A and 5-HT1B genes. Positive associations were established in patients with unipolar depressive disorder for SNPs in the CCK1R, TPH, DRD1, DRD2, PENK, mu-opioid receptor genes and wolframin gene. Patients with bipolar depressive disorder had positive associations with the SNPs in the wolframin, DRD4, delta-opioid receptor and mu-opioid receptor genes.


Robert F. Krueger and Wendy Johnson1.  Sense of Control, sociometric status, and health:  genetic and environmental connections.2

1Department of Psychology, University of Minnesota, Twin Cities, Minneapolis, MN, 2The National Survey of Midlife Development in the United States is supported by the John D. and Catherine T. MacArthur Foundation Research Network on Successful Midlife Development and by National Institute on Aging Grant #AG20166

Address:  Department of Psychology, University of Minnesota, 75 East River Road, Minneapolis, MN 55455 Phone: 612-624-8204 Fax: 612-626-2079 Email: krueg038@tc.umn.edu

 

Given the robust finding that people in higher SES groups tend to experience better physical health, there has been interest in identifying psychosocial variables that are related to health and may help to explain the social class health gradient. Sense of control of one_s life is one variable that has been identified as filling a possible moderating role. Lachman and Weaver (M. E. Lachman and S. L. Weaver, 1998, Journal of Personality and Social Psychology, 74, 763-773), for example, found that, though control was related to health across all income levels, those low SES individuals who expressed a high sense of control experienced levels of health comparable to those of individuals of higher SES. Using a nationwide sample of twin pairs from the National Survey of Midlife Development in the United States (MIDUS), we fit a series of biometric models including multivariate and environmental moderation models in an attempt to unravel the genetic and environmental bases of these relations.


Henrik Larsson1, Henrik Andershed2, and Paul Lichtenstein1.  Genetic and environmental influences on the psychopathic personality constellation.

1Department of Medical Epidemiology, Karolinska Institutet, Stockholm, Sweden, 2Department of Behavioural, Social and Legal Sciences, Örebro University, Örebro Sweden

Address:  Box 281, SE-171 77 Stockholm. Phone: +46 8 728 74 16, Fax: +46 8 31 49 75, Email: Henrik.Larsson@mep.ki.se

 

Individuals with a psychopathic personality constellation differ from other antisocial youth in terms of age of onset, the number of violent acts committed, the seriousness of their offenses, as well as violence while institutionalized. Little is known about the etiology of this personality constellation, even though a strong genetic effect often is assumed.  The aim of this study was to investigate the importance of genetic and environmental influences on the relationship among the three dimensions (grandiose/manipulative, callous/unemotional, impulsive/irresponsible) of the psychopathic personality constellation in adolescence.  Data from the Swedish Young Twins study, comprised of 1090 twin pairs (aged 16-17 year), was used. The psychopathic personality constellation was measured with a self-report questionnaire (The Youth Psychopathic traits

Inventory; YPI).  We used a common pathway model hypothesizing that the covariation between the YPI- dimensions is determined by a latent phenotype (i.e., psychopathic personality constellation). The results suggested that the psychopathic personality constellation is highly heritable (a2 = .63). The shared environmental factor on the latent phenotype was estimated to zero.  Significant unique genetic influences were found in the callous/unemotional and in the impulsive/ irresponsible dimension, but not in the grandiose/manipulative dimension.  These results suggest that the three dimensions are subsumed under a higher order factor (i.e., psychopathic personality constellation), with a strong genetic influence.


Maria J. Lavooy1 and Martin E. Hahn2.  A review of the methods of studies on maternal retrieval and infant ultrasonic vocalization in small rodents.

1Department of Psychology, University of Central Florida, Cocoa, FL, 2Department of Biology, William Paterson University, Wayne, NJ

Address:  Department of Psychology, University of Central Florida, 1519 Clearlake Rd, Cocoa FL 32922-6598 Phone:  321 632 1111 x65598 E-mail:  mlavooy@pegasus.cc.ucf.edu

 

Infant rodents produce at least three sounds that serve as signals regulating aspects of infant-adult interactions. Of those three, pure ultrasonic vocalizations have been the most studied since their discovery in the late 1950's. Most investigations conclude that ultrasonic vocalizations and olfactory cues are the basis of maternal retrieval of pups that are found outside the nest and thus are critical to the survival of pups who are poor thermoregulators. While ultrasonic vocalizations and their developmental time course are currently the focus of several types of investigation, questions remain about the role of ultrasonic vocalizations in mother-infant interactions. The purpose of this paper is to review the literature on small rodent ultrasonic vocalizations and on maternal retrieval, with a focus on methods employed. In the ultrasound section, we will catalog the stimulus conditions used to elicit ultrasonic vocalizations and the results achieved.  In the maternal retrieval section, we will focus on the settings in which pup retrieval has been studied and the role of genotype and other maternal variables as the sources of differences in retrieval performance. We hope that the results of this review will assist in two areas. First, we hope to clarify the relationship between ultrasonic vocalizations and maternal retrieval. Second, we hope to recommend some sets of standardized conditions for the elicitation of ultrasonic vocalizations so that studies on the effects of psychoactive drugs will have a common basis.


Lisa N. Legrand, Matt McGue, Margaret Keyes, Steve Malone, and William G. Iacono.  Do Rural Environments Constrain the Genetic Expression of Externalizing Psychopathology?

Department of Psychology, University of Minnesota, Minneapolis, MN

Address:  Dept. of Psych., U of MN, Elliott Hall, 75 East River Rd., Minneapolis, MN 55455, Email: llegrand@tfs.psych.umn.edu

 

The relative contribution of genetic and shared-environmental influences to alcohol use has been shown to vary by environmental circumstances, with higher heritability reported in urban settings (D.M. Dick, R.J. Rose, et al., 2001,<I> J. of Abnormal Psych.</I>, 110, 625-632).  We sought to replicate this gene-environment interaction with a Minnesota sample, expanding the phenotype to include additional indicators of externalizing psychopathology.  Our participants are 1216 male and female twins (397 MZ pair; 211 DZ pair) who visited the Minnesota Twin Family Study when they were approximately 17 years old.  Twin pairs were classified as urban or rural using zip codes and Census 2000 data.  Symptomatology for Oppositional Defiant Disorder, Conduct Disorder, and Alcohol Abuse/Dependence was assessed both through self-report and mother¡¯s report.  Symptomatology for Adult Antisocial Behavior as well as the extensiveness of substance experimentation was assessed through self-report.  These five behaviors were used as indices of a latent Externalizing factor.  Biometric factor modeling was then employed to determine whether the relative contribution of genetic and shared-environmental factors varied from urban to rural settings.  Preliminary analyses suggest that they do for the male sample but not for the female sample.  For males in urban environments, externalizing psychopathology is influenced relatively equally by genetic and shared-environmental factors.  However, for males in rural environments, the genetic component falls to near zero.  Final analyses will be presented at the conference along with a discussion of their implications for intervention.


Jeffrey M. Lessem1, Christian Hopfer2, Andrew Smolen1, John K. Hewitt1.  Genetic Analysis of Antisocial Behavior and Drug Abuse in the Add Health Study.

1Institute for Behavioral Genetics, University of Colorado, Boulder, 2Div. of Subst. Dependence, Univ. of Co. Health Sciences Ctr., Denver

Address:  Institute for Behavioral Genetics, 447 UCB, Boulder, CO  80309-0447 Phone: (303) 492-2843 Email: Jeff.Lessem@Colorado.EDU

 

Add Health uses a school based design to assess health related information on adolescents in grades 7-12.  Subjects were interviewed twice, about one year apart, between 1994 and 1996, creating Waves I and II.  Wave III of data collection was undertaken six years later, and will be ready in 2003.  Of the large amount of information collected as part of the Add Health study, this investigation is examining antisocial behavior and drug related behavior.  Genotyping of several candidate loci, including dopamine transporter, serotonin transporter, the dopamine D4 and D2 receptors, and monoamine oxidase-A is currently underway.  As a preliminary step in performing association analyses, the antisocial behavior and drug phenotypes are being screened for genetic and environmental influences.  This study reports the heritability of the relevant phenotypes on approximately 2100 families, with 200 MZ twin pairs, 400 DZ twin pairs, 1000 full-sibling pairs, and 350 half-sibling pairs.  Supported by P01-HD31921 (PI: JR Udry).


Paul Lichtenstein1, Jan-Olov Larsson2, and Henrik Larsson1.  The development of Attention Deficit Hyperactivity symptoms from childhood to early adolescence.

1Department of Medical Epidemiology, Karolinska Institutet, Stockholm, Sweden, 2Department of Child & Adolescent Psychiatry, The Karolinska Hospital, Stockholm, Sweden

Address:  Box 281, SE-171 77 Stockholm. Email: paul.lichtenstein@mep.ki.se, Phone:  +46 8 728 7424, Fax:  + 46 8 314 975

 

Attention Deficit Hyperactivity Disorder (ADHD) has shown considerable genetic effects in both childhood and adolescence. The core symptoms of ADHD may persist into adolescence and early adulthood, but could also diminish in severity over time. It has been reported that the rate of ADHD decline by 50% approximately every 5 years. There are no reports on how genetic and environmental effects change over puberty.  We have studied continuity and change of ADHD symptoms in a nation wide sample of 1,500 Swedish twins first contacted when they were 8-9 years old and followed up when they were 13-14 years old.  Sex-limitation models indicated that different sets of genes were important for the sexes. Genetic effects accounted for about 65% of the variation in ADHD symptoms in both boys and girls at both time points.  There was little evidence for shared environmental effects. In both genders, correlations for ADHD symptoms between age 8-9 and age 13-14 were around 0.50. About 80% of the stability in symptoms was due to genetic effects. Genetic and nonshared environmental effects contributed about equally to change.  The strong genetic influences on the stability of ADHD symptoms suggest that persistence of ADHD might mediate genetic effects to adult psychopathology. The new genetic effects in adolescence could be due to hormonal changes during puberty. The relatively strong new nonshared environmental influences in adolescence suggest that changes in the environment could be beneficial in the prevention and treatment of ADHD.


John C. Loehlin1.  Parents and their children: How alike are they in personality and attitudes, and why?

1Department of Psychology, University of Texas, Austin, TX

Address:  Department of Psychology A8000, University of Texas, Austin, TX 78712 Phone: 512 475 7008  Fax: 512 471 6175  Email: loehlin@psy.utexas.edu

 

A search of the literature revealed 1279 published parent-child correlations from 59 studies since the 1930s.  The tabulation was limited to correlations based on at least 50 pairs, and to Western societies.  Mean correlations were compared for three types of parent-child pairs: (1) pairs related both genetically and socially, i.e., parents and children from ordinary families; (2) pairs related only socially, i.e., parents and children in adoptive families; and (3) pairs related only genetically, i.e., parent and adopted-away child or MZ and twin's child.  Comparisons were made for personality traits versus attitudes and interests; for older and younger children; across the "Big Five" personality domains; and for different pairings by sex--fathers with sons, mothers with daughters, etc.  The results suggested that most of the fairly modest parent-child resemblance in personality, attitudes, and interests reflects shared genes, although there is a small social contribution for attitudes, and young children's personalities appear to have an appreciable social component.  Resemblances were similar across the Big Five, and across different sex combinations in the case of attitudes; in the case of personality, resemblance appeared to be somewhat more social for fathers and children, somewhat more genetic for

mothers and children.


Hendrik Luuk1, Sulev Kõks2, Aleksei Nelovkov2, and Eero Vasar2.  A Screen for Genes Induced in the Amygdaloid Area During Cat Odor Exposure.3

1 Department of Zoology and Hydrobiology, University of Tartu, Tartu, Estonia, 2 Department of Physiology, University of Tartu, Tartu, Estonia, 3 Supported by Estonian Science Foundation Grant 4562

Address:  Institute of Physiology, Biomeedikum, Tartu 50411, Estonia Phone: +372 7 374 335 Fax: +372 7 374332 Email: hluuk@psych.ut.ee

 

The aim of the present study was to identify differentially expressed genes in the rat amygdala in response to exposure to cat odor. Cat odor was used to induce ethologically relevant anxiety reaction in male rats (J. Panksepp, 1998, Affective Neuroscience: the Foundations of Human and Animal Emotions, Oxford University Press, New York). The differential expression of genes was analyzed using cDNA Representational Difference Analysis (cDNA RDA) (M. Hubank and D.G. Schatz, 1999, Methods. Enzymol. 303, 325-349). Products of substractive hybridization (differentially expressed transcripts) were cloned and identified by sequencing and database search. From 576 sequenced clones 67 unique differentially expressed transcripts were identified. The results were subsequently confirmed by dot-blot analysis. According to their function the transcripts can be classified as neurotransmission related, metabolic enzymes, cell cycle regulating proteins, and transcription factors. This report provides preliminary clues for understanding the molecular basis of anxiety. Additional quantitative analysis of expression levels of the identified transcripts is needed to verify the present results.


Stacy Lynch1, Jane Mendle1, Brian D’Onofrio1, Robert Emery1, Eric Turkheimer1, Wendy Slutske2, Andrew Heath3, Nick Martin4.  Marital Status and Depression.

1Department of Psychology, University of Virginia, Charlottesville, VA Supported by William T. Grant Foundation, Grant #2054, 2Department of Psychological Science, University of Missouri, Columbia, MO, 3Department of Psychiatry, Washington University, St. Louis, MO, 4Genetic Epidemiology Section, Queensland Institute of Medical Research, Brisbane, QLD, Australia

Address:  Dept. of Psychology, Box 400400, University of Virginia, Charlottesville, VA 22903

Phone: (434) 982-5573 E-mail: akl8t@virginia.edu

 

Several studies have considered the relationship between depression and marital status.  Findings in this area have illustrated that interpersonal discord influences depressive symptoms.  What remains unclear is whether depression is a precursor or a byproduct of relational discord.  One strategy to address such questions involves the utilization of twin data.  Using longitudinal data from the Australia Twin Registry, a number of bivariate models will be fit in an effort to understand the direction of the association between depression and martial status and the gender differences in that relationship.  Examining 277 MZ and DZ twin pairs discordant for divorce on the outcome of depression will also help to unravel the genetic and environmental processes which account for this relationship.


Michael T Lynskey.  Early onset risky behaviors and risk for substance dependence:  A latent class approach.

Missouri Alcoholism Research Center, Department of Psychiatry, Washington University School of Medicine, St Louis, MO 63108

Address:  Department of Psychiatry, Washington University School of Medicine, 40 N. Kingshighway, Suite One, St Louis, MO 63108 Phone: 314-286-2228 Fax: 314-286-2213 Email: mlynskey@matlock.wustl.edu

 

Early life transitions, including early substance use and early school leaving often co-occur and are also associated with increased risks for the development of substance dependence and mental health problems.  To examine the extent to which associations between these transitions could be modeled as a function of a smaller number of latent classes a series of latent class models were fitted to data on these transitions using the program LCAP.  Data was collected on 6265 young

adult Australian twins (median age = 30 years) including retrospective reports of early transitions (early onset substance use behaviors, precocious sexual activity and role transitions, including early school leaving and leaving the parental home) and lifetime rates of substance abuse and dependence.  There was a moderate degree of association between these measures of early transitions and latent class analyses identified a four class model for both males and females. These classes corresponded to: a) a group with low probabilities of all transitions (60.2% males, 58.8% females); a group with elevated probabilities of tobacco use and early school leaving but not of the other transitions (18.2% males 11.4% females); c) those with elevated probabilities of early substance use and sexual activity only (18.5% males; 28.6% females); those with elevated

probabilities of all transitions (3.1% males, 1.1% females).  After control for retrospectively assessed measures of CSA, conduct disorder and other childhood disadvantages, those most prone to early transitions had odds of substance dependence, major depression and suicidal behaviors that were 1.8 to 11.4 times higher than individuals with low probabilities of these transitions.  It is proposed that the processes linking these early transitions to later risks are likely to be social in origin.


Hermine H. Maes, Patrick F. Sullivan, Cynthia Bulik, Michael C. Neale, Carol A. Prescott, Lindon J. Eaves, and Kenneth S. Kendler1.  Genetic analyses of smoking initiation, regular smoking and nicotine dependence.2

1Virginia Institute for Psychiatric &amp; Behavioral Genetics, Department of Human Genetics and Psychiatry, Virginia Commonwealth University, Richmond VA 23298, 2Supported by NIH grants HL60688, MH45268, AA06781, AA07728, AA07535, MH40828, RR08123 and MH01458

Address:  VIPBG, MCV/VCU, P.O. Box 980003, Richmond VA 23298-0003 Tel 804 828 8145 Fax 804 828 8801 Email: hmaes@vipbg.vcu.edu URL: http://www.vipbg.vcu.edu

 

Understanding the contribution of genetic and environmental influences to the various stages of smoking initiation to more severe forms of smoking behavior (regular smoking, smoking persistence, nicotine dependence) is an important first step in finding specific etiological factors for smoking.  We examined the relationship between genetic and environmental risk factors for smoking initiation, regular smoking, smoking persistence and nicotine dependence in twins from the population-based Virginia Twin Registry.  The smoking variables were assessed in female, male and opposite sex twin pairs by personal interview.  Model fitting results indicated that the liabilities for smoking initiation, regular smoking and smoking persistence/nicotine dependence were highly related, but did not lie on a single distribution of liability.  Although variance components could be equated across zygosity or gender without significant loss of fit; thresholds could not.   The contribution of shared environmental factors was not significant.  Heritability for smoking initiation was estimated at 75%.  About a quarter of the heritability for liability to regular smoking was explained by genetic factors independent of those influencing initiation.  The liability to smoking persistence was primarily accounted for by genetic factors (h<sup>2</sup>=.66).  As much as 26% of the variance in liability to persistence was due to genetic factors specific to smoking persistence.  These results were similar to those for liability to nicotine dependence (24%) and the Fagerstrom Test for Nicotine Dependence (FTND, 26%).  These results suggest that although there appears to be significant overlap between the genetic factors involved in starting to smoke and continuing to smoke, they may be genes specific to smoking addiction.


Matt McGue1, Margaret Keyes1, S. Brent Walden1, and William G. Iacono1.  Shared environmental influences on adolescent functioning: Parent and sibling effects.2

1Department of Psychology, University of Minnesota, Minneapolis, MN, 2The Sibling Interaction and Behavior Study is supported in part by U.S. Public Health Service Grants AA11886

Address:  Department of Psychology 75 East River Rd. University of Minnesota, Minneapolis, MN 55455 Phone: 612-625-8305   Fax: 612-626-2079  Email: mcgue001@umn.edu

 

Research has consistently shown that two domains of adolescent adjustment, problem behavior and intellectual ability, run counter to the general finding of minimal shared environmental influence on psychological outcomes.  Unclear is whether these shared environmental influences reflect parent effects or common exposure to other environmental factors.  The Sibling Interaction and Behavior Study (SIBS) was undertaken to identify and characterize shared environmental influences on adolescent adjustment.  We will report findings from the first 325 adopted and 120 biological sibling pairs to participate in SIBS.  We find significant sibling similarity, both adopted and biological, on a latent factor of adolescent problem behavior that loads on delinquency, negative peer models, substance use, and substance abuse/dependence.  Further, sibling similarity on this phenotype is moderated by sibling similarity in age and gender, such that like-sex siblings who are near in age are much more similar than unlike-sex siblings who are distant in age.  We contrast these findings with results for intellectual ability, where there is significant adopted and biological sibling similarity that is not moderated by sibling gender or age difference.  We conclude that the nature of the shared environmental influences differ in the two domains, with sibling effects being more salient for adolescent problem behavior and parent effects being more salient for intellectual achievement.


Jane Mendle1, Brian M. DOnofrio1, Stacy K. Lynch1, Eric N. Turkheimer1, Robert E. Emery1, Wendy Slutske2, Andrew C. Heath3, and Nicholas G. Martin4.  Association of paternal absence with age of menarche.5

1Dept. of Psychology, University of Virginia, Charlottesville VA, 2Department of Psychological Science, University of Missouri, Columbia, MO, 3Department of Psychiatry, Washington University, St. Louis, MO, 4Genetic Epidemiology Section, Queensland Institute of Medical Research, Brisbane, QLD, Australia, 5Supported by William T. Grant Foundation, Grant # 2054

Address:  Dept. of Psychology, Box 400400, University Of Virginia, Charlottesville, VA 22903

Phone:  (434) 982-5573 E-mail: jemendle@virginia.edu

 

Early onset of menarche is associated with increased risk for intercourse.  Since cognitive and physical development do not necessarily occur at the same rate, precocious sexual maturation increases the likelihood that girls will be forced to confront new environments, stressors, and social expectations before they are psychologically prepared to do so.  In recent years, theorists have offered an evolutionary, environmental explanation for early menarche, linking it with paternal absence in the home.  Girls raised either by single mothers or by mothers and stepfathers are primed to undergo physical maturation at an earlier age.  Although posited as an environmental consequence of family stress, this association could be mediated by genetic pathways.  Girls begin menstruating at roughly the same age as their mothers began menstruating.  Rowe (2002) has estimated the heritability of menarcheal age to be .50.  Therefore, the relationship between early menarche and paternal absence might be an artifact of genetic associations.  Since early menarche is associated with early sexual intercourse and consequent single motherhood, it seems possible that mothers predisposed to raise children without a biological father present bequeath an early menarcheal age to their daughters.  One way of clarifying the genetic composition of menarche is through the children-of-twins design.  If the association between menarche and paternal absence is actually a by-product of genetic influences, the age of menarche in children of discordant twins will be comparable -- despite the differing environmental circumstances in which these cousins were raised.  Using data from the Australian Twin Registry, this analysis will be accomplished through a series of hierarchical linear regressions modeling age of menarche at an individual, nuclear family, and twin family level.  Rowe, D.C. (2002):  On genetic variation in menarche and age at first sexual intercourse.  Evolution and Human Behavior, 23, 365-372.


Elizabeth A. Molloy1, Jonathan Blumenthal1, Liv S. Clasen1, Alex Zijdenbos2, Judith L. Rapoport1, and Jay N. Giedd1.  Brain Morphometry and IQ in Pediatric Monozygotic Twins.

1Section on Brain Imaging, Child Psychiatry Branch, NIMH, Bethesda, MD, 2Montreal Neurological Institute, McGill University, Montreal, Quebec, Canada

Address:  922 24th Street, NW, #519 Washington, DC 20037 Phone: 202-471-4286 Email: molloy@gwu.edu

 

Magnetic Resonance Imaging (MRI) studies have shown a modest correlation between brain size and general intelligence.  As part of an ongoing study of the effects of genes and environment on normal pediatric brain development we sought to explore the relationship between brain morphometry and IQ within monozygotic (MZ) twins.  Thirty-five sets of MZ twins ages 6.03 to 18.45 (mean age = 12.9, s.d. = 3.27) underwent MRI on the same GE 1.5 Tesla scanner.  Subjects were administered the Wechsler Abridged Scale of Intelligence (WASI).  Quantitative analysis of cortical gray matter volume was performed using the Montreal Neurological Institute automated method.  Spearman rank correlations showed that within-pair differences in IQ were positively correlated with differences in Total Cerebral Volume (TCV) (r = 0.35, p < 0.05).  Differences in IQ were positively correlated with differences in Total White Matter (r = 0.42, p < 0.01).  After controlling for TCV, differences in IQ were negatively correlated with differences in Total Gray Matter (r = -0.37, p < 0.05).  Within-pair differences in TCV were related to differences in birth weight (r = .33, p < 0.01); however, differences in IQ were not related to differences in birth weight (r = .16, n.s.). Birth order was not associated with TCV or IQ.  Understanding the relationship between pediatric brain changes and behavior changes is a fundamental objective of neuroscience and may shed light on the nature, pathophysiology, and treatment of neurodevelopmental disorders.  Our results suggest that non-genetic influences play a role in brain/IQ relationships, and may indicate that general intelligence is associated with developmental changes (increased myelination and gray matter pruning) known to occur in late childhood and adolescence. The relationship between brain size and IQ during childhood and adolescence appears to be influenced by environmental factors not accounted for by birth weight or birth order.


Brian S. Mustanski1, Richard J. Viken1, Jaakko Kaprio2, Torsten Winter2, and Richard J. Rose1.  Multivariate Behavior Genetic Analyses of Sexual Health.

1Department of Psychology, Indiana University, 2Department of Public Health, University of Helsinki, Finland Data collection was supported by NIH (AA 12502) and the Academy of Finland

Address:  Indiana University Department of Psychology, Bloomington, IN 47404  E-mail:

bmustans@indiana.edu

 

The field of human sexuality has received little attention from behavior geneticists relative to such areas as personality, psychopathology, and substance abuse.  That is unfortunate, because behavior genetic research can enhance understanding of the relative contribution of specific social, psychological, and biological influences that have been identified as important predictors of sexual health. For example, adolescents represent one of the fastest-growing risk groups for HIV in the U.S. and approximately ¼ of all new STD infection occur in teenagers. Age at first intercourse is a critical indicator of pregnancy and STD risk, and yet few genetically informative studies have been conducted on this variable. Beyond partitioning variance into genetic and environmental components, multivariate behavior genetic analyses can help elucidate the association between individual temperamental, developmental (e.g. pubertal timing), and environmental (e.g. inner city) characteristics that increase the likelihood that certain individuals will engage in risky sexual behavior.  With longitudinal, population-based data from Finnish twins studied since adolescence, we analyze two important sexual health variables measured at ages 23-25: number of sexual partners and age at first sexual intercourse.  Multivariate behavior genetic models are fit to the data to characterize the genetic and environmental sources of variance and covariance with longitudinal predictor variables (personality and pubertal timing). Sex limitation models suggested similar influences in males and females, but the magnitude of genetic effects was greater in males. Gene X environment interaction models suggested significant differences in heritability between urban and rural environments.


Michael C. Neale1.  A finite mixture distribution approach to the analysis of twin data.2

1Virginia Institute for Psychiatric &amp; Behavioral Genetics, Department of Psychiatry, Virginia Commonwealth University, Richmond VA 23298, 2Supported by NIH grants RR-08123, MH-01458 and MH-65322

Address:  VIPBG, MCV/VCU, P.O. Box 980126, Richmond VA 23298-0126. Phone:  804/828-3369 Fax:  804/828-1471 Email:  neale@hsc.vcu.edu Web:  http://www.vipbg.vcu.edu

 

The analysis of data collected from a classical twin study of monozygotic (MZ) and dizygotic (DZ) twins depends upon accurate diagnosis of zygosity.  However, large scale surveys frequently resort to questionnaire-based methods of diagnosis which classify twins as MZ or DZ with less than perfect accuracy, and in some cases no information on zygosity is available at all.  This paper describes a mixture distribution approach to the analysis of twin data when zygosity is not perfectly diagnosed.  Estimates of the probability that each pair is MZ (p(MZ) or DZ (p(DZ)) are used to obtain a weighted conditional likelihoods of the form p(MZ) L(data|MZ) + p(DZ) L(data|DZ). The performance of this method is compared to fully accurate diagnosis, and to the analysis of samples that include some misclassified pairs. Even for rates of misclassification as high as 15\%, the method shows relatively little loss of precision of estimates of proportions of variance due to additive genetic, common environment and specific environment sources.  However, considerable bias in parameter estimates is observed when misclassification is ignored, such that additive genetic variance is underestimated while common environment and to a lesser extent specific environment components are overestimated.  These biases are eliminated with the mixture distribution approach.


Jenae M. Neiderhiser1.  Genetic and Environmental Influences on Change and Continuity in Family Relations from Adolescence to Young Adulthood.2

1Center for Family Research, Department of Psychiatry and Behavioral Sciences, George Washington University, Washington, DC, 2Supported by NIH grant MH59014

Address:  2300 K St., N.W., 3rd Floor Warwick Building, Washington, DC 20037 Phone: (202) 994-2212 Fax: (202) 994-4812 Email: cfrjmn@gwumc.edu

 

Few studies have examined relationships among family members from adolescence to adulthood, and fewer still have employed genetically sensitive designs.  Current research indicates that genetic and shared environmental factors influence parent-child relationships during childhood and adolescence, and that genetic and nonshared environmental factors influence marital relationships during middle adulthood. The current study will examine associations among middle and later adolescent parent-child relationships and young adult relationships with their parents and with their romantic partners.  Subjects from the Nonshared Environment in Adolescent Development (NEAD) project will be used for the current report. NEAD included 720 families consisting of two parents married at least 5 years in nondivorced and step families with two siblings (< 4 year age difference). Average child age at Time 1 was 13.6 (+3) years and 15.0 (+2) years at Time 2. There were five types of sibling pairs: MZ and DZ twins, full, half and step siblings.  Measures included in these analyses index parent-child relationships as rated by mothers, fathers, adolescent self-reports and observer ratings for each parent’s behavior towards each adolescent. A follow-up of the NEAD sample is just being completed. The now young adult participants range in age from 23 to 31 years and over 90% are married, cohabitating or involved in an exclusive romantic relationship. Measures for this study include young adult and parent reports of current adult child-parent relationship and young adult reports of current romantic relationship quality.  We expect parent-child relationships from adolescence to young adulthood to show high levels of stability although the pattern of genetic and environmental influences should change over time. Specifically, shared environmental influences on parent-child relationships should decrease from adolescence to young adulthood. The expectations for the correlation between parenting and young adult relationships are less clear both phenotypically and in terms of genetic and environmental influences.


Jenae M. Neiderhiser1.  The Impact of Antisocial Behavior on Young Adult Parent-Child Relationships: Genetic and Environmental Influences.2

1Center for Family Research, Department of Psychiatry and Behavioral Sciences, George Washington University, Washington, DC, 2Supported by NIH grant MH59014

Address:  2300 K St., N.W., 3rd Floor Warwick Building, Washington, DC 20037 Phone: (202) 994-2212 Fax: (202) 994-4812 Email: cfrjmn@gwumc.edu

 

In the early 1980's David Rowe published some of first work in "environmental genetics" by looking at genetic and environmental influences on parenting in a sample of adolescent twins. This work inspired a generation of researchers to consider this sort of question and to attempt to understand its relevance for child and adolescent adjustment. In the current paper this question has been taken to a different level in that genetic and environmental influences on longitudinal associations between adolescent antisocial behavior and young adult parent-child relationships are considered.  Subjects from the Nonshared Environment in Adolescent Development (NEAD) project will be used for the current report. NEAD included 720 families consisting of two parents married at least 5 years in nondivorced and step families with two siblings (< 4 year age difference). Average child age at Time 1 was 13.6 (+3) years and 15.0 (+2) years at Time 2. There were five types of sibling pairs: MZ and DZ twins, full, half and step siblings. Measures included in these analyses index parent-child relationships as rated by mothers, fathers, adolescent self-reports and observer ratings for each parent's behavior towards each adolescent. A follow-up of the NEAD sample is just being completed. The now young adult participants range in age from 23 to 31 years and over 90% are frequent (approximately once/week) contact with their parents. Measures for this study include young adult and parent reports of current adult child-parent relationship.  Preliminary analyses have confirmed a phenotypic correlation between adolescent antisocial behavior and young adult-parent relationships. We expect that this association will be due primarily to genetic influences common to both adolescent antisocial behavior and young adult-parent relationships.


Michele L. Pergadia1, Andrew C. Heath1, Nicholas G. Martin2, and Pamela A.F. Madden1.  Genetic Analyses of DSM-IV Nicotine Withdrawal.

1Washington University School of Medicine, in St. Louis, MO, 2Queensland Institute of Medical Research, Australia.  Supported by NIH Grants: AA07535,AA07728, and AA11998 (A.C.H.), DA12854 (P.A.F.M.), AA07580 (M.L.P.), and the Australian National Health and

Medical Research Council.

Address:  Department of Psychiatry, Washington University School of Medicine, 40 N. Kingshighway, Suite One, St. Louis, MO 63108 Phone: 314 286 2270 Fax: 314 286 2213  Email michelep@matlock.wustl.edu

 

Less is known about the genetic influences on nicotine withdrawal relative to other smoking related behaviors. This study examined whether there are genetic influences on nicotine withdrawal, and whether there are genetic factors specific to nicotine withdrawal, after controlling for experimentation with cigarettes and lifetime regular smoking (at least weekly smoking). Data were obtained by telephone diagnostic interview in 1989 with 6257 individual Australian twins (3454 women, 2803 men; mean age=30). Analyses of relative genetic and environmental influence on nicotine withdrawal were surprising. Genetic modeling of nicotine withdrawal using the entire sample, including never smokers, fit the data well [Chi sq: 4.6, p=.80; A: 46% (23-55), C: 0% (0-17), E: 54% (45-64)].Restricting the sample to only pairs where both twins at least experimented with cigarettes, resulted in a good fitting model and no substantial reduction in genetic influence on nicotine withdrawal [(Chi sq: 4.8, p=.85; A: 42% (18-52) C: 0% (0-17), E: 58% (48-69)]. However, when we used only pairs where both twins have a lifetime history of regular smoking, genetic influences on nicotine withdrawal were substantially reduced [(Chi sq: 5.1, p=.82; A: 30% (0-43) C: 0% (0-27), E: 70% (57-85)],although familial influences remained significant. Preliminary results from fitting a two stage genetic model (first dimension of liability: regular smoking-second dimension of liability: DSM-IV nicotine withdrawal) suggested that genetic influences on nicotine withdrawal partially overlap with genetic influences on lifetime regular smoking, providing little evidence for genetic influences specific to withdrawal.


Carol A. Prescott, Jonathan W. Kuhn, and Kenneth S. Kendler.  Adolescent problem behaviors and risk for alcoholism: A causal association?

Virginia Institute for Psychiatric and Behavioral Genetics, Department of Psychiatry, Virginia Commonwealth University, Richmond VA.  Supported by NIH Grants  R01-MH/AA-49492, R01-AA/DA-09095 and K01-AA-00236. Access to subjects was provided by the Mid-Atlantic Twin Registry, currently directed by Linda Corey and Lindon Eaves. Sarah Woltz, Frank Butera, Patsy Waring, Lisa Halberstadt, and Barbara Brooke supervised data collection.

Address:  VIPBG, VCU, P.O. Box 980126, Richmond VA 23298-0126,  telephone: (804) 828-5968, fax: (804) 828-1471,  email: cprescot@hsc.vcu.edu

 

Risk for alcoholism is increased among individuals who begin drinking at an early age. In a prior paper (C. A. Prescott and K. S. Kendler, 1999, Alc: Clin Exp Res, 23, 101-107, we reported that this association was familial in origin and suggested that early drinking is a vulnerability indicator rather than a direct cause of problematic alcohol involvement. In the present study we examined the association between alcoholism and other indicators of adolescent problem behavior, including conduct disorder, school dropout, and early use of tobacco and illicit drugs.  METHODS -  Lifetime DSM-IV alcohol abuse/dependence (AAD) and five indicators of problem behavior (early use of alcohol, tobacco and illicit drugs, conduct disorder symptoms, and school dropout) were assessed in 1,958 women and 3,523 men aged 18 to 64 from same-sex twin pairs in the Virginia Twin Registry.   RESULTS - AAD was significantly associated with each of the problem behavior indicators in males and all but school dropout in females. Alcoholism in both males and females was predicted about equally well by early smoking and early drug use as by early drinking.  Results from twin pair analyses indicated that the majority of this association arose because these problem behaviors and AAD share familial (probably genetic) variation.  CONCLUSIONS - Alcoholism is preceded by multiple forms of behavioral deviance during early adolescence. Our finding that early use of tobacco and drugs predicts the development of alcoholism even in the absence of early drinking, combined with the evidence from twin analyses that the covariation between alcoholism and problem behaviors is due primarily to genetic and familial environmental factors, suggest that early drinking is a vulnerability indicator rather than a direct cause of alcoholism.


David C. Rettew1, William Copeland1, Catherine Stanger1, and James J. Hudziak1.  Associations between Temperament and DSM-IV Disorders in Children and Adolescents2.

1Department of Psychiatry, University of Vermont College of Medicine, Burlington, Vermont, 2Supported by a Physician Scientist Award from the University of Vermont College of Medicine and NIMH grant K08 MH01265.

Address:  University of Vermont College of Medicine, Given Building B227, Burlington, VT 05405 Phone: 802 656 8125  Fax: 802 656-0987 Email: David.Rettew@uvm.edu

 

Objective:  To investigate associations between child temperament and DSM-IV disorders in children.  Method:  A total of 156 probands (97 boys, 59 girls; mean age=10.78) and 154 randomly selected siblings (90 boys and 64 girls; mean age=10.80) were assessed using the Junior Temperament and Character Inventory (JTCI) and a structured DSM-IV interview as rated by mothers.  Subjects were placed into non-overlapping diagnostic groups of 1) attention-deficit/hyperactivity disorder (ADHD) only, 2) disruptive behavior disorders (DBD) only, 3) DBD plus an affective and/or anxiety disorder (DBD+Int), and 4) controls with no diagnosis.  Statistical analyses conducted separately with probands and siblings included analyses of variance and hierarchical logistic regressions.  Results:  Many JTCI scales were found to differ between diagnostic groups and controls.   Regression analyses showed independent associations between low persistence and ADHD-only group membership, high novelty seeking and the DBD-only group, and between high harm avoidance, high novelty seeking and low self-directedness, and DBD+Int group membership. The interaction of novelty seeking X harm avoidance was related to the ADHD-only group.  Conclusion:  Psychopathology in children is associated with temperamental differences, including possible interactions between different temperamental dimensions.


Sally-Ann Rhea1 and Brett C. Haberstick1.  Parenting and Peer Relationships in Early Adolescence: A twin study replication of adoption study data.2

1Institute for Behavioral Genetics, University of Colorado, Boulder, CO, 2Supported by grants HD10333, MH43899 and DA11015.

Address:  Institute for Behavioral Genetics, Campus Box 447, University of Colorado,

Boulder, CO 80309-0447 Phone: 303 492 2822 Fax: 303 492 8063 E-mail: rhea@colorado.edu

 

Family influences on peer relationships were explored in the University of Colorado's Longitudinal Twin Study (LTS; n=742). The procedures are a replication of an analysis from the Colorado Adoption Project (CAP) evaluating links among the quality of self-rated parenting style, child-rated family environment and teacher-rated popularity at ages 10 and 12 (T.G. O'Connor, J.M. Jenkins, J.K. Hewitt, J.C. DeFries. and R. Plomin, 2001, Marriage and Family Review, 33, 251-271). The adoption study found both parental influence on longitudinal change in teacher-rated popularity as well as evidence for genetic influences. As in the previous study, we find that parental ratings are stable (.65 to .74, p < .01). Child ratings are also stable but somewhat less so (.25 to .62, p < .01). We find evidence for shared environment and a greater tendency for girls to concur in their ratings. Parent-rated positive parenting style and child-rated family cohesion predict teacher-rated popularity. Additionally, although ratings were completed by different teachers, popularity at age 10 predicts popularity at age 12.


Soo Hyun Rhee1,2, John K. Hewitt2, Susan E. Young2, Robin P. Corley2, Thomas J. Crowley3, Michael C. Neale4, and Michael C. Stallings2.  The etiology of comorbidity in substance dependence in adolescents.5

1Department of Psychology, University of Colorado, Boulder, CO 80309, 2Institute for Behavioral Genetics, University of Colorado, Boulder, CO  80309, 3Department of Psychiatry, University of Colorado Health Sciences Center, Denver, CO  80262, 4Virginia Institute for Psychiatric and Behavioral Genetics, Richmond, VA  23298-0126, 5Supported by NIDA grants DA-05131, DA-11015, and DA-13956.

Address:  Institute for Behavioral Genetics, Campus Box 447, University of Colorado, Boulder, CO  80309 Telephone: (303) 492-4631 Fax: (303) 492-8063 E-mail: Soo.Rhee@colorado.edu

 

Knowledge regarding the causes of comorbidity among substance use disorders can have significant impact on future research examining the etiology of these disorders.  Unfortunately, the conclusions of past studies examining the comorbidity among substance use disorders are conflicting, with some studies emphasizing the importance of common familial influences and others emphasizing the importance of substance-specific familial influences.  Discrepancies in results may reflect different analytical approaches or differences in the samples examined.  Here, we address the particular question of the causes of comorbidity in substance dependence in adolescents.  We will examine the causes of comorbidity in substance dependence in a clinical sample of adolescents treated for antisocial substance problems and a control sample.  The Neale and Kendler model fitting approach will be used to test 13 alternative hypotheses for the causes of comorbidity.


Frances Rice1, Tom A Fowler1, Jane Scourfield1, and Anita Thapar1.  Anxiety and depression in childhood and adolescence: developmental pathways, genes and environment.

1Department of Psychological Medicine, University of Wales College of Medicine, Cardiff, Wales

Address:  Department of Psychological Medicine, University of Wales College of Medicine, Cardiff, Wales, CF14 4XN Phone: 44 (0)29 20742201 Fax : 44 (0)29 20747839 Email: ricef@cf.ac.uk

 

Depression and anxiety co-occur more commonly than would be expected by chance in children and adolescents.  This comorbidity has been reported in clinical samples and with sub-clinical symptoms in the general population (Brady & Kendall, 1992, Psychol Bull., 111, 244-55; Kovacs & Devlin, 1998, J Child Psychol Psychiatry.,39, 47-63). The association between anxiety and depression is of interest as results from family studies have suggested that anxiety may be a genetic precursor of depression (Rende, Warner, Wickramaratne & Weissman, 1999, Psychol Med., 29, 1291-298). This is thought to be the case particularly in children at high risk of depression, however family studies are unable to disentangle genetic and shared environmental effects.  The genetic and environmental architecture underlying the association between early anxiety and later depression can be assessed using a longitudinal twin design. All families of twin pairs born between 1980 and 1991 were identified from a sub-sample of CASTANET (Cardiff Study of All Wales and Northwest England Twins), a population-based U.K. twin register, and asked to participate. In the first wave a parent rated measure of anxiety was collected on 670 families. Of the families eligible at the time of the second wave, parent and twin rated measures of depressive symptoms were collected on 338 families.  Complete pairwise responses at time 1 and time 2 were available for 69% of time one responders. The underlying aetiological architecture was examined using bivariate and causal analysis. There was significant genetic covariation between anxiety and depression, however there were also significant genetic effects specific to depression. A causal model provided a significantly worse fit than the general bivariate model.  These results suggest that the association between early anxiety and later depression is due to a partly shared genetic aetiology and that the association is not mediated by the phenotype of anxiety.


Frances J. Rice1,2, Gordon T. Harold2, and Anita Thapar1.  Negative life events as an account of age-related differences in the genetic aetiology of depression in childhood and adolescence.

1Department of Psychological Medicine, University of Wales College of Medicine, 2School of Psychology, Cardiff University

Address:  Department of Psychological Medicine, University of Wales College of Medicine, Heath Park, Cardiff, CF14 4XN Phone: 044 (0)29 20743241 Fax: 044 (0)29 20747839 Email:

RiceF@cardiff.ac.uk

 

Many twin studies have reported that the genetic aetiology of depression differs according to age, with genetic influences being more important for adolescents than younger children.  We sought to examine whether this age related increase in the relative importance of genetic factors is due to an increase in gene-environment correlation specifically involving negative life events.

Questionnaires were sent to the families of CASTANET (Cardiff Study of All Wales and Northwest England Twins), a population-based U.K. sample of twins aged between 8 and 17 years.  Parents of all the twins and adolescents aged 11 and over were asked to complete a measure of depressive symptoms and a stressful recent life events checklist.  Responses were obtained from 1468 families and data were analysed using genetic model fitting.  Bivariate analysis of 1) negative life events and 2) behaviour-dependent life events and depression symptoms was undertaken separately for children (aged 8 to 10 years) and adolescents (aged 11 to 17 years).  Differences in genetic and environmental parameters across age groups were tested. Adolescence was associated with a greater number of behaviour-dependent life events.  Genetic covariation of negative life events and depression was greater for adolescents than for children, in particular for life events that were judged to be behaviour-dependent.  Parameter estimates were also significantly heterogeneous across age groups for behaviour-dependent negative life events.  Bivariate model fitting was consistent with the greater heritability of depression seen in adolescence being due to an increase in gene-environment correlation involving negative life events.  However, the effects of genes associated specifically with maturation in adolescence, the possibility of 'person' effects and the role of other environmental factors also need to be considered.


Frances J. Rice1,2, Gordon T. Harold2, and Anita Thapar1.  The transmission of depressive symptoms from mother to child: genetic and environmental pathways.

1Department of Psychological Medicine, University of Wales College of Medicine, 2School of Psychology, Cardiff University

Address:  Department of Psychological Medicine, University of Wales College of Medicine, Heath Park, Cardiff, CF14 4XN Phone: 044 (0)29 20743241 Fax: 044 (0)29 20747839 Email:

RiceF@cardiff.ac.uk

 

Although family studies have shown that maternal depression is associated with depression in children, it is uncertain to what extent this resemblance is due to genetic factors and how much is

explained by environmental adversity. The relative importance of genetic and environmental factors in the transmission of depression from mother to child was evaluated in the CASTANET (Cardiff Study of All Wales and Northwest England Twins) U.K. community sample of twins aged 8 to 17 years and their mothers (an extended twin-study design).  Mothers reported on both their own internalizing symptoms and their children's current depressive symptoms. Self reports of depressive symptoms were also obtained from children aged 11 and over.  Responses were obtained from 1468 families.  A modification of a familial transmission model (D.I. Boomsma, M.B.M. Vandenbree, J.F. Orlebeke, and P.C.M. Molenaar, 1989, Behav. Genet.,19, 123-141) was tested using LISREL (K. Joreskog, and D. Sorbom, 2001 Scientific Software Int. Inc.).  This model included genetic and environmental transmission paths between mother and offspring (twin). When controlling for genetic similarity between mothers and children, the transmission of maternal internalizing symptoms to child depression included significant environmental paths when maternal ratings were used. When cross informant data were used (maternal internalizing and child self-reports of depression), the environmental path was no longer significant. However, previous DF analyses of this cohort had shown that the aetiology of high self-rated depression scores was more environmentally influenced than self-rated depression scores within the normal range (F. Rice, G.T. Harold and A. Thapar, 2002, J Child Psychol. Psychiat., 43, 1039-1051). Therefore, genetic and environmental transmission paths were tested across rater for high scorers as it was possible that environmental transmission effects may have differed according to severity of symptoms.  Significant environmental transmission effects were found in cross-informant models for children with high depression scores.


Nathan K. Risk1, Ashley H. Hayden1, Krista L. Russel1, and Rumi K. Price1.  The interaction of ALDH@ and CYP2A6 with acculturation on alcohol and nicotine use and problem use among Vietnamese and Japanese:  Preliminary results from a pilot study.2

1Department of Psychiatry, Washington University, Saint Louis, MO, 2Supported by the Missouri Alcoholism Research Center (P50AA11998, Director—Andrew C. Heath)

Address:  Department of Psychiatry, Suite 2, 40 N. Kingshighway, St. Louis, MO 63108 Phone:   314-286-2294, E-mail:  nathan@rkp.wustl.edu

 

In the genetic literature, mutations of two metabolic genes, aldehyde dehydrogenase class 2 (ALDH@) and cytochrome P450 2A6 (CYP2A6), have been found to be protective against alcoholism and regular cigarette smoking uniquely among subgroups of Asian population.  However, population epidemiologic data often show findings inconsistent with this notion of uniform genetic protection among Asians.  The widely different rates among Asian subgroups are partially due to the size of mixed-race subgroups, among whom a fourfold increase in alcohol and nicotine use is observed.  Using pilot data for a larger genetic epidemiological project, we examine the relative strengths of the two candidate genes and sociocutural and psychiatric factors pertaining to alcohol and nicotine use and problem use in the Vietnamese and Japanese populations.  Genotype data on ALDH2 and CYP2A6 and self-report measures of ethnicity, substance use and problem use, acculturation, socioeconomic achievement, mental health, and deviance will be obtained from approximately 160 Vietnamese and 80 Japanese participants recruited through ethnic-specific community-based organizations and service providers in St. Louis City and County, Missouri.  We report on preliminary results  gathered to date:  1)genotypic distributions of mixed-heritage vs. unmixed Japanese and Vietnamese; 2)correlations of genotypes of the two candidate genes with self-reported measures of alcohol use and cigarette smoking; and 3)differences in genotypes and acculturation measures between the Japanese and Vietnamese in St. Louis.  We hope to show that genotype data, when combined with detailed ethnicity and acculturation data, can inform the debate about why Asians in the United States appear not to be at increased risk for substance abuse.


Joseph Lee Rodgers.  Reformulating and Simplifying the DF Analysis Model.

Department of Psychology, University of Oklahoma, Norman, OK 73019

 

DeFries-Fulker (DF) Analysis for unselected populations is reformulated as a no-intercept model with centered variables and only two independent variables.  The reformulation serves three purposes.  First, the original formulation implicitly estimated two different values for c2 and two values for h2.  The new formulation resolves this ambiguity.  Second, because the original formulation estimated h2 with the coefficient from a regression-interaction term, whether to center the interaction variables was unclear.  The new formulation explicitly resolves this issue.  Finally, the new formulation estimates fewer parameters, and therefore improves estimation efficiency and statistical power.


Pierre L. Roubertoux1,2, Brice Marcet3, Franz Sluyter1,4, and Bernard Verrier3.  Mitochondrial DNA (mtDNA) and behavior : interaction between mitochondrial and nuclear genes, preliminary results from micro arrays.

1CNRS Génétique Neurogénétique Comportement, France, 2Present address: INPC CNRS 31, Marseille, France, 3INPC CNRS 31, Marseille, France, 4Present address: SGDP Centre, Institute of Psychiatry, London, UK.  Supported by the CNRS to Génétique Neurogénétique Comportement and to Institut des Neurosciences Physiologiques et Cognitives. F. Sluyter received a fellowship from the Fondation Fysen.

Address:  INPC CNRS 31 rue Chemin Joseph Aiguier, 13402 Marseille France Phone: 33 (0) 491164565 E-mail: rouber@lnf.cnrs-mrs.fr

 

Nuclear and mitochondrial (mt) genomes interact. The expression of nuclear genes is modified by mt genes and mt gene expression may be modified by nuclear genes as recently reported. New or aggravated phenotypes can result from these interactions. We investigated the modifications of expression of nuclear genes, in the brain, after cross transfer of mtDNA in mice. For this purpose, we used congenic strains for mtDNA (see Carlier et al. and Sluyter et al. abstracts, same issue). The congenics were derived from NZB/BlN and CBA/H mice that carry mtDNA from different origins. We show that mitochondrial genes of complexes I and IV were polymorphic in NZB/BLNJ and CBA/H. The NZB/BlN mice were compared to NZB/BlN with mtDNA from CBA/H origin and CBA/H with CBA/H with mtDNA from NZB/BlN origin. We used Pan mouse DNA chips covering 30,000 genes, corresponding to the whole mouse genome (MWG SA, Biotech, Paris, France). The expression of no more than 200 nuclear genes was modified by mtDNA crossed transfer. Several of these nuclear genes had been previously reported for their implications in learning or other cognitive processes.


Angelica Ronald1, Francesca G. E. Happé1, and Robert Plomin1.  Genetic heterogeneity for social and nonsocial symptoms of the autism spectrum: A twin study of 3000 pairs of 7-year-old twins.2

1Social Genetic and Developmental Psychiatry Research Centre, Institute of Psychiatry, London, UK, 2The Twins' Early Developmental Study (TEDS) is supported by a programme grant from the UK Medical Research Council.  The authors are indebted to the parents of the twins in TEDS for making the study possible.

Address:  SGDP Research Centre Box PO83 Institute of Psychiatry Denmark Hill London

SE5 8AF UK Telephone: +44 (0) 20 7848 0039 Fax: +44 (0) 20 7848 0866 Email:

a.ronald@iop.kcl.ac.uk

 

We investigated the genetic relationship between the two major components of symptoms in the population that at the extreme characterise autism spectrum disorders: social features (social and conversational abilities, theory of mind skills) and nonsocial features (rigidity and stereotyped behaviours, central coherence).  Parents and teachers of over 3,000 7-year old twin pairs completed a booklet that included three questionnaires: a questionnaire based on DSM IV social and non-social criteria for Autistic disorder and Aspergers, the Strengths and Difficulties Questionnaire (R. Goodman, 1997, J Child Psychol Psychiatry, 38, 581-586), and a questionnaire on theory of mind in everyday life (U. Frith, F. Happé, and F. Siddons, 1994, Soc Dev, 3, 108-124).  Social and non-social symptom scales showed substantial heritability (54%-71%) and nonsignificant shared environmental influence, for both teacher and parent data and for both males and females.  One scale, parent reported social features, showed modest shared environmental influence (23%).  Similar results were found for the entire sample and for high-scoring extreme groups.  For both the entire sample and the high-scoring extreme groups, the social and non-social components were only modestly correlated phenotypically.  Bivariate genetic analysis indicated that the social and non-social components of autistic spectrum symptoms are largely influenced by independent genetic factors.  These findings suggest that there is genetic heterogeneity for social and non-social symptoms of the autism spectrum, which has implications for diagnosis and treatment of autism spectrum disorders.  For molecular genetic research, these findings suggest the social and non-social components of the autism spectrum should be studied separately rather than combined in a single syndrome.


Alan R. Sanders1 and Pablo V. Gejman1.  Human G protein-coupled receptors coding variation database.2

1Department of Psychiatry, University of Chicago, Chicago, IL 60637, 2Supported by a NARSAD Young Investigator Grant to ARS

Address:  924 East 57th Street, Room R-004 Chicago IL 60637 Phone: 773 834 3502 Fax: 773 834 2970 Email: asanders@yoda.bsd.uchicago.edu URL: http://mpbsl.uchicago.edu/

 

G protein-coupled receptors (GPCRs) play fundamental roles in regulating the activity of almost every cell, typically functioning to relay a signal from outside to within the cell, and are especially important for all fields of medicine because they are targets for most medications.  I propose to create a comprehensive and multipurpose database for human GPCR genes including information about naturally occurring genetic variations, and use this database to test a set of hypotheses of fundamental biological significance regarding the general characteristics of these genetic variations.  We will test hypotheses about the position at which naturally occurring mutations occur in the messenger RNA (mRNA) and the protein, the type of amino acid affected, induced alterations in the codon usage pattern, and their effect on computer modeled mRNA

structure.  The proposed database will be integrated into the existing GPCR Database and will provide information that will facilitate the testing of pathophysiological hypotheses with positional or etiological candidate genes (that are GPCRs) for major psychiatric disorders by the scientific community.  This avenue of research could have far-reaching implications for human health because of the insight gained by exploiting this knowledge (e.g., rational drug design) in pursuit of better understanding variation among genes relevant to most major mental disorders, such as schizophrenia and mood disorders, or their treatments.


Kimberly J. Saudino1,2.  Sex and the Single Teacher.  Sex Differences in Parent- and Teacher-rated Behavior Problems.3

1Boston University, Department of Psychology, Boston, MA 02215, 2Supported by NIH Grant R01 MH062375-01, 3Supported by a program grant from the UK Medical Research Council

Address:  Boston University, Department of Psychology, 64 Cummington Street, Boston,

MA 02215 Telephone: 617 353 3679  Fax: 617 353 6933  E-mail: ksaudino@bu.edu

 

Although sex differences in mean levels of behavior problems in childhood have been well-studied, the question of sex differences in the etiology of behavior problems has been relatively unexamined.  The present study explores whether genetic and environmental influences on parent- and teacher-rated problem behaviors differ for males and females in a sample of

1857 twin pairs (621 MZM, 725 MZF, 568 DZM, 641 DZF, and 1159 DZO; mean age 7.04 years) participating in the Twins Early Development Study (TEDS).  Parents and teachers rated the behavior problems of twins on the Strengths and Difficulty Questionnaire (SDQ).  Preliminary analyses indicated that there were significant differences in parameter estimates

for twins in which both members of a pair were rated by the same teacher versus twins who were both rated by different teachers.  Therefore, separate genetic sex-limitation models were fit to data for parents, same teacher, and different teachers.  There was no evidence of genetic sex-differences when twins' behavior problems were evaluated by parents or different teachers.  However, when both members of a twin pair were rated by the same teacher, qualitative sex-specific genetic effects were indicated for hyperactivity.  Similarly, quantitative genetic effects were indicated for prosocial behavior, peer problems and conduct problems.  For these three SDQ scales males displayed significantly higher heritabilities than females (prosocial behavior: Males h2 = .76, Females h2 =.51; peer problems: Males h2 = .83, Females h2 =.60; conduct problems:  Males h2 = .81, Females h2 =.57).  These findings suggest ratings of twins' behavior problems by the same teacher reveals subtle behavioral differences that can go undetected with parent or different teacher ratings.


Stephanie Schmitz1.  Attention and internalizing problems.2

1Institute for Bahavioral Genetics, University of Colorado, CO 80309, 2Supported by NIH grants MH-43099, HD-10333, and MH-62116.

Address:  Institute for Bahavioral Genetics, Campus Box 447, University of Colorado, CO 80309; Telephone: (303) 492-0835; Fax: (303) 492-8063; email: schmitzs@colorado.edu

 

Most of the literature on problem behaviors co-occurring with attention problems has been on those from the externalizing domain.  In recent years, however, research has also examined the joint observation of attention and internalizing behaviors, such as mania (Biederman et al., 1996) and depression (Biederman, Newcorn, \& Spric, 1991).  The current study combines two genetically informative data sets, namely a twin and an adoption study, both longitudinal and ongoing, to explore the genetic and environmental associations between attention and internalizing problem behaviors. Ratings from both the children's mothers and teachers are utilized at ages 7 and 12, spanning the grade school years.  The etiologies of mother- and teacher-rated behaviors, and their associations in particular, are different in that mother-rated behaviors point towards common genetic and shared environmental influences while teacher ratings reveal common non-shared environmental factors.


Jane E. Schreiber1, Carol A. Van Hulle1, Penny L. Biersach Clark1, Kathryn Lemery2, and H. Hill Goldsmith1.  Genetic and Environmental Influence on Afternoon and Evening Cortisol Levels.3

1Department of Psychology, University of Wisconsin, Madison, WI, 2Department of Psychology, Arizona State University, Tempe, AZ, 3Supported by NIMH Grant R01-MH59785

Address:  Wisconsin Twin Project Waisman Center 1500 Highland Ave Madison, WI 53705

Phone: 608 265-2674 Email: jeschreiber@students.wisc.edu

 

Basal cortisol activity typically follows a circadian rhythm, peaking in the morning and dropping throughout the day. At any given time, cortisol level is potentially affected by three processes (a waking effect, a circadian effect, and a stress-reactive effect), which vary in their influences. Genetic and environmental influences on basal cortisol activity apparently vary throughout the day with morning cortisol showing the highest heritability (Wust et. al., 2000, Psychoneuroendocrinology, 25, 707-720). Unpublished data from another of our studies also support a higher heritability for morning values in children. Research is needed to address influences on basal activity later in the day. This study estimates genetic and environmental influences on afternoon and evening cortisol.  Participants were 54 families (20 MZ, 34 DZ, age 6-9 years) participating in the Wisconsin Twin Project, an ongoing longitudinal study of children at risk for behavior problems. By June, 2003, the sample size should increase to approximately 200 families. Salivary cortisol samples were collected at home in the late afternoon and before bedtime over three days, from parents and twins. Cortisol values were averaged across days.  Using a scalar model that assumed the same factors influenced parents and twins, we found that shared environment accounted for 40% of the variance in twin evening levels; 49% of the variance was to due to non-shared environmental effects, and 10% was due to genetic factors. In general, variance of parental values was not different from the twin values. Similar results were found for the afternoon sample. With the enlarged sample, we shall test for cultural and genetic transmission and analyze the change between time points. This report should help disentangle the factors of time of day, source of cortisol (blood, saliva, urine), age of participants, and basal versus reactive cortisol measures that complicate the current literature.


Jane Scourfield, Frances Rice, Anita Thapar, Gordon T. Harold, Neilson Martin, Peter McGuffin.  Depressive symptoms in children and adolescents: changing aetiological influences with development.

University of Wales College of Medicine, Cardiff University & Institute of Psychiatry, London.

Address:  Department of Psychological Medicine (4th floor), University of Wales College of Medicine, Heath Park, Cardiff, CF14 4XN, UK. Tel (+44) 029 20 743241 Fax: (+44) 029 20 747839 email: scourfieldj@cardiff.ac.uk

 

Parent and self-report questionnaire data was used to examine the genetic and environmental influences on depressive symptoms in a UK sample of 670 twin pairs aged 5-17. Age effects were examined cross-sectionally and longitudinally using data collected over a 3 year period. Cross-sectional analyses showed significant shared environmental influences in younger children only, with genetic effects having more influence on depression scores for adolescents. Analyses of two waves of data showed that shared environmental effects had significant influence in younger children and acted across time to influence depressive symptoms measured 3 years later. New genetic influences emerged in adolescence but no new shared environmental influences. These findings support and extend earlier work which has shown increasing genetic influence on

depressive symptoms as children grow into adolescence. 


Nancy L. Segal1, William D. Marelich2, Rick Castillo2, Jack Mearns2, and Shirley A. McGuire2.  Problem Behaviors in Virtual Twins: Analysis of Shared Family Influences.3

1Psychology Department, California State University, Fullerton, CA 92834, 2Psychology Department, University of San Francisco, San Francisco, CA 94117, 3Supported by  NIMH Grant 1 R01 MH63351-01A1 (McGuire and Segal), NSF Grant SBR-9712875 (Segal) and CSUF faculty research grants (Segal).

Address:  Psychology Department, California State University, Fullerton, California 92834.

Phone: 714-278-2142, Fax: 714-278-4843, Email: nsegal@fullerton.edu

 

Behavioral-genetic studies of children’s problem behaviors have compared behavioral problem scores for monozygotic (MZ) and dizygotic (DZ) twins and adoptive siblings. Both genetic and environmental factors have been shown to influence individual differences across scales. The present study uses a unique kinship called virtual twins (VTs); VTs are same-age, unrelated children raised together, composed of two adoptees or one biological child and one adoptee. VTs, therefore, replicate twins’ rearing situation, but without the genetic link.They circumvent problematic features of ordinary adoption designs, such as children’s different age at family entry and different age at testing. The present analysis was conducted as part of the ongoing Virtual Twins Study at California State University, Fullerton. As part of the standard test battery, parents completed the Child Behavior Checklist (CBCL), developed by Achenbach (1991), for each child separately.  The sample included 100 VT pairs (76 adopted-adopted and 24 biological-adopted). Same-sex  (n = 53) and opposite-sex pairs (n = 47) were included. Mean

age was 5.51 years (sd = 2.23) and ranged between 4 - 16 years. Pairs included 106 males and 94 females. An earlier study of 40 pairs (1997) found that, with a few exceptions, VT means were generally comparable to those from a non-referred sample of children 4 - 11 years of age, as reported in the CBCL manual. In that analysis, VT intraclass correlations  (-.11 to .58 for problem behaviors and .28, .40 and .51, respectively, for internalizing, externalizing and total problem scores)  were significantly lower than those of MZ twins, and generally lower than those of DZ twins, suggesting both genetic and environmental influences. In the present study, with only a few exceptions, VTs’ mean scale scores were generally comparable to those of the non-referred sample. VT intraclass correlations ranged between -.02 and .50 for problem behaviors and were .35, .54 and .74, respectively, for internalizing, externalizing and total problem scores. Correlations for all scales (except thought problems) were significantly below those of MZ twins, suggesting genetic effects. In contrast, VT-DZ twin correlations did not differ. The VTs younger age may be partly responsible, given that within-family influences are more potent among younger children. Resemblance between biological-adoptive and adoptive-adoptive pairs will also be presented.


Wendy S. Slutske1, Andrew C. Heath2, Nikole J. Cronk1, Kathleen K. Bucholz2, Pamela A.F. Madden2, and Nicholas G. Martin3.  Familial transmission of alcoholism and antisociality: A study of twins and their adult offspring.4

1Department of Psychology and Missouri Alcoholism Research Center, University of Missouri, Columbia, MO 65211, 2Department of Psychiatry and Missouri Alcoholism Research Center, Washington University School of Medicine, St. Louis, MO, 63110, 3Epidemiology Unit, Queensland Institute of Medical Research, Brisbane, Queensland, Australia, 4Supported by NIH grants AA00264, AA07535, AA07728, and AA11998.

Address:  Department of Psychology, University of Missouri, 210 McAlester Hall, Columbia, MO 65211 phone: (573) 882-4043 fax: (573) 882-7710 email: wendy@martha.psyc.missouri.edu

 

Although it has been widely embraced by the treatment community, and certainly has a great deal of intuitive appeal, it has been difficult to demonstrate empirically a (non-genetic) consequence of being reared by an alcoholic parent. In particular, twin studies have generally led to the conclusion that family environmental influences do not play a major role in the familial transmission of alcoholism risk.  However, in the twin design the estimate of family environmental effects only includes those that are independent of genetic effects. One critical test for demonstrating an important environmental effect of being reared by an alcoholic parent is to compare the rates of adverse outcomes among the biological offspring of alcoholic parents to the rates of adverse outcomes among the biological offspring of the unaffected cotwins of alcoholic parents. In this paper, we examine the history of alcohol use disorders and conduct disorder among 774 offspring of 408 parents from same-sex twin pairs who were characterized by the history of alcohol use disorder and conduct disorder in themselves and their cotwin.  For example, of the 408 parents, 127 had a personal history of alcohol dependence, 28 were unaffected but had a monozygotic cotwin with a history of alcohol dependence, 33  were unaffected but had a dizygotic cotwin with a history of alcohol dependence, and 220 were unaffected with an unaffected cotwin.  Offspring of parents from these four groups were at varying degrees of genetic (high, high, intermediate, and low, respectively) and environmental (high, low, low, and low, respectively) risk for alcohol dependence.  Preliminary analyses of these data suggest that the only group of offspring who had significantly elevated rates of alcohol dependence and conduct disorder (compared to the low risk group) were those with both high genetic and high environmental risk.


Frans Sluyter1,2, Charles Cohen-Salmon1,3, Michèle Carlier1,4, Chabane Chérif1, Fatima Maarouf Verray1, and Pierre L. Roubertoux1,5.  Mitochondrial DNA and behavior: Implication of mitochondrial DNA in learning and memory.

1CNRS Génétique Neurogénétique Comportement, France, 2Present address: SGDP Centre, Institute of Psychiatry, London, UK, 3UMR 7593 CNRS and Université Paris VI, Paris, France, 4Present address: PsyCLÉ (EA3273), Université de Provence, France, 5Present address: INPC CNRS 31, Marseille, France.  Supported by the CNRS to Génétique Neurogénétique Comportement and to Institut des Neurosciences Physiologiques et Cognitives. F. Sluyter

received a fellowship from the Fondation Fysen and Cohen Salmon a financial support from Laboratoire Ipsen.

Address:  SGDP Centre, Institute of Psychiatry De Crespigny Park P083 London SE5 8AF UK

Phone: Tel: 44-(0)207-848-0028 Fax: 44-(0)207-848-0866 Email: F.Sluyter@iop.kcl.ac.uk

 

Although the role of mitochondrial DNA (mtDNA) in normal and abnormal development of the nervous sytem has been well established, only one study so far has hinted at an association between mtDNA and cognition. Using a congenic quartet of well-selected inbred strains (NZB/BlN and CBA/H male mice, see (Carlier et al. abstract), we here provide direct evidence for the  implication of mtDNA in learning and memory. We measured the congenic  quartet's performance in three distinct behavioral paradigms (Morris water  maze, 8-arm radial maze & Krushinsky task), all of which have been  validated to reflect 'mouse cognition' and all of which tap into different  combinations of reward mechanisms and locomotor demands.  mtDNA effects  mostly occurred in interaction with nuclear DNA and persisted with age, increasing in magnitude as the mice got older. In addition to other mtDNA effects on exploration, sensorial development and brain anatomy, the  present findings clearly show that mitochondrial polymorphisms are not as neutral as was previously believed and might, at least partly, explain the sometimes enigmatic observations in mammals generated by cloning and assisted reproduction.


Frans Sluyter1 and Leonard C Schalkwyk1.  Symposium:  Using  microarrays in the analysis of  behavior.

1Social, Genetic and Developmental Psychiatry Research Centre, Institute of Psychiatry

Address:  Social, Genetic and Developmental Psychiatry Research Centre, PO 82, Institute of Psychiatry, King's College London, De Crespigny Park, London SE5 8AF, U.K. tel: +44(0)207 848 0028 fax: +44(0)207 848 0801 email spjgfns@iop.kcl.ac.uk

 

Microarray methods are becoming increasingly widely used in many kinds of genetics and are even beginning to penetrate into the study of behavior. One application is high throughput SNP genotyping.  Other applications allow the function of genes to be explored.  As technological advances allow ever more ingenious applications to be dreamed up, the prospects are good for a chip-dominated decade.  This symposium combines three highly novel uses of microarrays in mouse behavioral genomics and one leading application of microarrays to human behavioral genetics.


Erica L. Spotts1 and Paul Lichtenstein1.  Early adolescent behavior problems and middle adolescent relationships:  Genetic and environmental influences.2

1Karolinska Institutet, Department of Medical Epidemiology, Stockholm, Sweden, 2The Young Twins project is supported by the Swedish Council for Working Life and Social Research (project 2001-2368) and the Swedish Research Council (2001-4231).  The first author is supported by NRSA grant MH65008.

Address:  Karolinska Institutet, Department of Medical Epidemiology, Box 281, Stockholm,

Sweden, SE 114-21.  Telephone:  +468 728 7412 Fax: +468 31 49 75  E-mail: Erica.Spotts@mep.ki.se

 

Research on adult relationships has found that mental health and adjustment are important associates of relationship quality (ie. Beach, S.R., Fincham, F.D. & Katz, J., 1998, Clinical Psychology Review, 18, 635-661.).   Relationships become increasingly important in adolescence so we wanted to see if earlier adjustment was a predictor of later relationship quality.  The few studies that have examined this association in adolescence have found associations between internalizing and externalizing behaviors and friend relationships (Klein, D.N., Lewinsohn, P.M, & Seeley, J.R., 1997, Journal of Affective Disorders, 42, 2-3), sibling relationships (Deater-Deckard, K., Dunn, J. & Lussier, G., 2002, Social Development, 11, 571-590; Dunn, J., Slomkowski, C., Beardsall, L & Rende, R., 1994, Journal of Child Psychology & Psychiatry & Allied Disciplines, 35, 491-504), and romantic relationships (Grello, C.M, Dickson, J.W., Welsh, D.P., et al, 2001, paper presented at Society for Research on Child Development conference,  Minneapolis, MN).  The current study will examine the genetic and environmental influences on associations between early adolescent internalizing and externalizing behaviors and mid-adolescent sibling, friend, and romantic relationships.  Data from three waves of the Swedish Young Twins study will be used.  Twins were 8-9 years at time 2, 13-14 years at time 2, and 16-17 years at time 3.  Behavior problems were assessed at Times 1 (parent reports) and 2 (parent, self, teacher reports), and the quality of sibling, friend and romantic relationships (self reports) was assessed at Time 3.  Analyses show that externalizing, but not internalizing, at Time 2 is associated with sibling relationships and some aspects of friend and romantic relationships at Time 3.  Genetic influences seem to account for most of the associations between externalizing and relationships, though these influences may differ by gender.  Findings suggest that the quality of adolescent relationships may originate in genetically influenced aspects of earlier adjustment.  


Michael C. Stallings1, Robin P. Corley1, Brianna Dennehey2, John K. Hewitt1, Kenneth S.

Krauter2, Jeff M. Lessem1, Susan K. Mikulich3, Soo Hyun Rhee1, Andrew Smolen1, Susan E. Young1, and Thomas J. Crowley3.  A Genome Scan for Quantitative Trait Loci Influencing Antisocial Drug Dependence in Adolosence.4

1Institute for Behavioral Genetics, University of Colorado, Boulder, CO 80309, 2Department of Molecular, Cellular and Developmental Biology, University of Colorado, Boulder, CO 80309, 3Division of Substance Dependence, University of Colorado School of Medicine, Denver, CO 80262, 4Supported in part by NIH Grants DA-05131, DA-11015, and DA-12845

Address:  Institute for Behavioral Genetics, Campus Box 447, University of Colorado Boulder, CO 80309-0447 Phone: 303 492 2826 Fax: 303 492 8063 Email: Michael.Stallings@Colorado.Edu

 

There is substantial comorbidity among substance use disorders and externalizing problem behavior, particularly in adolescence. Increasing evidence suggests that familial influences, including shared genetic risk factors, may account for part of this comorbidity. In this study we describe results from a genome-wide search for quantitative trait loci (QTL) influencing substance dependence vulnerability and externalizing problem behavior in adolescence. The gene mapping goals of the Center on Antisocial Drug Dependence at the University of Colorado are to: 1) identify QTL influencing substance dependence vulnerability in adolescence; 2) identify QTL influencing externalizing problem behavior; and 3) determine if there are overlapping chromosomal regions likely to contain genes influencing comorbid risk. Regression-based QTL mapping procedures designed for selected sibling pair samples were utilized. The selected sibling pairs included 250 proband-sibling pairs from 192 families. Patient-probands (13 to 19 years of age) were drawn from consecutive admissions to treatment facilities for substance abuse and delinquency in the Denver metropolitan area. Siblings of the selected probands ranged in age from 12 to 25 years. A community-based sample of over 3500 adolescents and young adults (12 to 25 years of age) were utilized to define clinically-significant, heritable, age- and sex-normed, indices of substance dependence vulnerability and conduct disorder—independent of our linkage results. Siblings and their parents were genotyped for 374 micro-satellite markers distributed across the 22 autosomes (average inter-marker distance = 9.2 cM). Non-parametric single-point and multipoint mapping provided preliminary evidence for linkage to regions on chromosome 3q24-25 (near markers D3S1279 and D3S1614) and chromosome 9q34 (near markers D9S1826 and D9S1838) for our quantitative index of substance dependence vulnerability. Gene mapping analyses utilizing a quantitative index of conduct disorder also indicated suggestive linkage to the same region on chromosome 9q34 (but not to 3q24-25), and preliminary evidence for a novel region on chromosome 17.


Anita Thapar1, Tom Fowler1, Frances Rice1, Jane Scourfield1, Marianne van den Bree1, Hollie Thomas1, Gordon Harold2, and Dale Hay2.  Maternal smoking in pregnancy and Attention Deficit Hyperactivity Disorder symptoms in offspring.

1Department of Psychological Medicine, University of Wales College of Medicine, Cardiff, UK, 2School of Psychology, Cardiff University, Cardiff, UK

Address:  Professor Anita Thapar, Professor of Child and Adolescent Psychiatry, Department of Psychological Medicine, University of Wales College of Medicine, Heath Park, Cardiff CF14 4XN, UK Phone: 044 (0) 29 2074 3241, Fax: 044 (0) 29 2074 7839, Email: thapar@cardiff.ac.uk

 

Maternal smoking in pregnancy has been found to be associated with Attention Deficit Hyperactivity Disorder (ADHD) in several clinical studies. Confounding factors do not appear to account for the association. However, with the exception of one study, this work has been based on clinical samples. Population-based studies have focused on the association with Conduct Disorder symptoms. The aim of the study was examine whether smoking in pregnancy is associated with ADHD symptoms in offspring and that these effects are additional to genetic influences, using a population-based twin sample. Children's ADHD symptoms (parent and teacher rated), maternal smoking in pregnancy, conduct disorder symptoms and family adversity were assessed using questionnaires in a population-based sample of 1,452 twin pairs aged 5 to 16 years from the CAST-A-NET sample (Cardiff Study of All Wales and North West England Twins). Although genetic influences accounted for most of the variants in offspring ADHD, smoking in pregnancy was still found to show a significant environmentally mediated association. Maternal smoking remained a significant influence when other potential confounders were taken into account. In conclusion, maternal smoking in pregnancy appears to show an association with ADHD symptoms in offspring that is additional to the effects of genes and not attributable to shared rater effects, clinical referral biases or co-variation with antisocial behaviour.


Anita Thapar, Kate Langley, Lucy Marshall, Marianne van den Bree, Hollie Thomas, Michael Owen, and Michael O'Donovan.  ADHD children with and without the dopamine D4 receptor 7-repeat allele: evidence of differences in performance on neuropsychological tests.1

Department of Psychological Medicine, University of Wales College of Medicine, Cardiff, UK, 1Supported by Wellcome Trust Grant 059870/Z/99/Z

Address:  Professor Anita Thapar, Professor of Child and Adolescent Psychiatry, Department of Psychological Medicine, University of Wales College of Medicine, Heath Park, Cardiff CF14 4XN, UK Phone: 044 (0) 29 2074 3241, Fax: 044 (0) 29 2074 7839, Email: thapar@cardiff.ac.uk

 

Association between the 7-repeat allele of a variant in the DRD4 gene (a 48bp VNTR) and ADHD has been widely documented with evidence of significant association from a meta-analysis of 21 studies (Faraone, SV, Doyle AE, Mick E and Biederman J, American Journal of Psychiatry, 2001, 158:7,1052-1057).(1) We examine whether the DRD4 7-repeat allele is associated with performance on a variety of neuropsychological tasks. 133 drug-naive children aged 6 to 13 years who fulfilled diagnostic criteria for ADHD were assessed on several neuropsychological tests (continuous performance test, matching familiar figures test (MFFT), go-no-go task and stop task).  Activity level was assessed with an Actigraph. Those with at least one 7-repeat allele (7-present) and those without (7-absent) were compared. We also compared results from the clinical sample with existing comparison data. Compared to the 7-absent group, the 7-present group showed a significantly increased number of incorrect responses on the MFFT (16.1 vs. 14.3, p=0.001), decreased mean reaction times for incorrect responses on the MFFT (846.1ms vs. 1103.7ms, p=0.02) and on the Stop task (116.6 vs. 134.1ms p=0.029).  The 7-present group also displayed increased activity levels assessed using an actigraph (t=-2.115, p=0. 037). No significant differences were found on measures of sustained attention or percentage of inhibitions. Both ADHD groups were more impaired than comparisons. In conclusion children with ADHD, possession of the DRD4 -7-repeat allele appears to be associated with an inaccurate, impulsive response style on neuropsychological tasks that is not explained by ADHD symptom severity.


Anita Thapar1, Jacky Boivin2, Gordon Harold2, Dale Hay2, Frances Rice1, Adam Goody2.  A novel method for disentangling the effects of genes, intrauterine and environmental risk factors.3

1Department of Psychological Medicine, University of Wales College of Medicine, Cardiff, UK, 2School of Psychology, Cardiff University, Cardiff, UK, 3Supported by Wellcome Trust Showcase Grant  067527/Z/02/Z

Address:  Professor Anita Thapar, Professor of Child and Adolescent Psychiatry, Department of Psychological Medicine, University of Wales College of Medicine, Heath Park, Cardiff CF14 4XN, UK Phone: 044 (0) 29 2074 3241, Fax: 044 (0) 29 2074 7839, Email: thapar@cardiff.ac.uk

 

It is undisputed that genes and environmental factors affect health and behaviour. Even after specific genes are identified, methods for understanding how they work together with environmental factors will still be needed. Family, twin and adoption studies are widely used genetic epidemiological methods. We propose a new strategy based on families using in vitro fertilisation (IVF). Children conceived via these methods may be genetically related to both parents (homologous IVF), the mother only (sperm donation), the father only (egg donation) or to neither parent (surrogacy). In all cases the social mother experiences the pregnancy. With surrogacy both parents are genetically related to their child but the child is born to a genetically unrelated surrogate. By comparing the similarity of parent and offspring across each of these groups we can examine not only the contribution of genes but also disentangle the effects of genes, intrauterine and early environmental influences. It will also be possible to examine to what extent the association of specific environmental risk factors and behaviour is mediated by genes and environment. The main aims of the study will be to (1) test the separate and joint influences of genes, intrauterine and environmental effects on a variety of behaviours (2) explore the feasibility of this method to be used more widely.


Laura M. Thornton1, Martin E. Hahn2, and Norman Schanz2.  Genetic and Developmental Influences on Infant Mouse Ultrasonic Calling. III. Patterns of Inheritance of Call Characteristics.3

1Eating Disorders Research, University of Pittsburgh Medical College, Pittsburgh PA, 2Department of Biology, William Paterson University, Wayne NJ, 3Supported by the Center for Research of the School of Science and Health of William Paterson College Eating Disorders Research, Suite 600 Iroquois Building, 3600 Forbes Avenue, Pittsburgh PA 15260 Phone: 804 673 4949 Fax: 804 673 4949 Email: psalms329@hotmail.com

 

Infant mice produce ultrasonic calls that elicit retrieval by adult mice.  Age-related differences in some of the call characteristics have been noted (M. E. Hahn, L. Karkowski, L Weinrub, A. Henry, N. Schanz, E. M. Hahn, 1998, Behav. Genet., 28, 315-325), as have genetic influences.   Hahn et al. (M. E. Hahn, J. K. Hewitt, N., Schanz, L. Weinrub, A. Henry, 1997, Behav. Genet., 27, 133-143) found additivity and directional dominance for most call characteristics at 3 days of age.  Maternal effects were not significant for any call feature.  Roubertouz et al. (P. L. Roubertouz, B. Martin, I. Le Roy, J. Beau, C. Marchaland, F. Perez-Diaz, C. Cohen-Salmon, M. Carlier, 1996, Behav. Genet., 26</B>, 427-437) also found additivity and dominance for rate of calling but found a maternal effect.  However, little research has been conducted investigating how the influence of these genetic effects changes with age for the call characteristics.  This study explored developmental-genetic patterns of inheritance of 7 ultrasonic call characteristics from 16 genetic groups of mice, ages 2-12 days, derived from a complete diallel cross.  The results indicate that additive genetic variance contributes significantly to all characteristics for all ages.   Maternal effects appear to have the greatest influence at 2 days of age for ending, highest and lowest call frequency, but at 4-6 days of age for rate and range.  Perhaps the most interesting result is the role of dominance across age.  The dominance effects tend to decrease with age for rate of calling, range, and length of calls.  This may indicate less selection pressure toward higher rates of calling, greater range and longer calls as the animals age and increase in their ability to thermoregulate.


 

Alexandre A. Todorov1, Shimy Apoorva2, and Richard Todd1,3.  Sample size as a random variable and its effect on power computations.4

1Departments of Psychiatry and 3Genetics, Washington University School of Medicine, St. Louis MO, 2School of Public Health, Saint Louis University, St. Louis MO, 4Supported by NIH grants AA-12232

Address:  Department of Psychiatry, 40N Kingshighway Suite 1, St Louis MO 63108 Phone: 314 286 2301 Fax: 314 286 2213 Email: todorov@matlock.wustl.edu

 

Power computations are an integral part of study design and are generally done assuming a fixed or "expected" sample size (e.g., number of affected sib pairs). Oftentimes, however, the final sample size is not known in advance. We illustrate here power computations for a linkage study of ADHD (prevalence between 3 and 5%, sibling relative risk between 3 and 7) which uses a multistage sampling scheme, and faces three common constraints: (i) sampling from a finite population (here, N=11,502 large families in Missouri); (ii) budgetary limit on the total number of families that can be screened; and (iii) budgetary limit on the total number of individuals that can be genotyped. Thus, neither the exact size nor the final composition of the sample are known in advance. In our example, we determined that the study would end with 300-400 affected sib-pairs and that power would be acceptable (taking into account fluctuations in sample size and composition) to detect a gene accounting for about ~4% of the liability to ADHD.  The methods have been implemented in a program available from the authors. Six parameters (respondent cooperation rate, screening test sensitivity and specificity, individual-wise cooperation rate, diagnostic test specificity and sensitivity) are used to determine the probability distribution of the numbers of recruited affected and unaffected sibs, conditional upon the true number of affected and unaffected individuals in that family. For various genetic models, we then calculate the frequencies of families with a given number of affected and unaffected offspring. From these, we determine the probability distribution of family types types (defined by the number of  affected and  unaffected offspring that participate) under the multistage sampling scheme, taking into account practical constraints (population size, budgetary), and use this distribution to estimate power for the all-pairs test.


Marianne van den Bree1, Lucie Robinson1, Darko Turic1,3, Mary Duke1, Derek W. Morris1, Marian Hamshere1, Andrew Grierson1, Martha Easton1, Ruma Raha-Chowdhury2, Jeffery Gruen3, Jim Stevenson4, Michael Krawczak1,5, Michael J. Owen1, Michael C. O'Donovan1, Julie Williams1.  Linkage disequilibrium mapping provides further evidence of a gene for reading disability on chromosome 6p21.3-22.

1Department of Psychological Medicine, University of Wales College of Medicine, Heath Park, Cardiff, UK, 2Department of Neuroscience and Cognitive Development, Babraham Institute, Babraham, Cambridge, UK, 3Yale Child Health Research Center, OA464 Congress Avenue, New haven, USA, 4Centre for Research into Psychological Development, University of Southampton, Southampton, UK, 5Department of Medical Genetics, University of Wales College of Medicine, Heath Park, Cardiff, UK

Address:  Department of Psychological Medicine, University of Wales College of Medicine, 4th Floor, Heath Park, Cardiff CF5 3PU Phone: 029-20743242 Fax: 029-2074 7839 Email:

vandenbreemb@cardiff.ac.uk

 

Using a two-stage approach, we tested for association between reading disability (RD) and 22 microsatellite markers in two independent samples of 101 (Stage 1) and 77 (Stage 2) parent/proband trios in which RD was rigorously defined. The most significant replicated associations were observed between combinations of markers D6S109/422/1665 (Stage 1, P=0.002 (adjusted for multiple testing); Stage 2, P=0.0001) and D6S506/1029/1660 (Stage 1, P=0.02 (adjusted), Stage 2, P=0.0001). The only two-marker association observed in both samples was with D6S422/1665 (P=0.01, 0.04). No single marker showed replicated association but D6S506 produced values of P=0.01 and 0.08 which were significant when combined (P=0.02). We observed weaker and less consistent evidence of association in a region of confirmed linkage to RD in previous studies. The most consistently significant haplotypic association D6S109/422/1665, showed association with single-word reading, spelling, phonological awareness, phonological decoding, orthographic accuracy and random automised naming, but not with vocabulary or Attention Deficit Hyperactivity Disorder. Our findings strongly support the presence of a gene contributing to RD in a region of chromosome 6 between markers D6S109 and D6S1260, but do not rule out the presence of a gene between D6S1556 and MOG.


Edwin JCG van den Oord1.  A framework for controlling discovery rates and minimizing the amount of genotyping in LD studies.2

1Virginia Institute for Psychiatric and Behavioral Genetics, Medical College of Virginia of Virginia Commonwealth University, Richmond VA, 2Supported by grants from NIH (MH065320) and NARSAD.

Address:  Virginia Institute for Psychiatric and Behavioral Genetics, Department of Psychiatry, Medical College of Virginia of Virginia Commonwealth University, P.O. Box 980126, Richmond VA 23298-0126 Phone: 804 828 8127 Fax: 804 828 1471 Email: ejvandenoord@vcu.edu

 

I explore how linkage disequilibrium studies may be designed efficiently using a sequential design. The method allows researchers to set false and true discoveries at desired numbers, perform power calculations for LD studies involving a large number of SNPs, and compute required sample sizes while minimizing the genotyping burden. The approach is illustrated for a fine mapping study and a whole-genome scan.


Edwin JCG van den Oord1.  Ethnic differences in birth weight.2

1Virginia Institute for Psychiatric and Behavioral Genetics, Medical College of Virginia of Virginia Commonwealth University, Richmond VA, 2Supported by grants from NIH (MH065320) and NARSAD.

Address:  Virginia Institute for Psychiatric and Behavioral Genetics, Department of Psychiatry, Medical College of Virginia of Virginia Commonwealth University, P.O. Box 980126, Richmond VA 23298-0126 Phone: 804 828 8127 Fax: 804 828 1471 Email: ejvandenoord@vcu.edu

 

Ethnic differences in birth weight have remained largely unexplained. Birth weight is correlated with developmental outcomes and health. It is therefore important to identify the factors that might put specific ethnic groups at high risk. However, the large domain of possible influences, and the difficulty of measuring them in large surveys, makes a resolution of the causes underlying ethnic differences in birth weight difficult. We therefore apportioned the ethnic differences in general components by fitting a quantitative genetic model to data from the National Longitudinal Survey of Youth and studied biracial infants using the whole 1991 birth cohort as collected by the U.S. National Center of Health Statistics. Results were very consistent and provided guidelines how to advance research on ethnic differences in birth weight.


Richard J. Viken1, Danielle M. Dick1, Jaakko Kaprio2, and Richard J. Rose1.  Trajectories of Drinking from Adolescence to Young Adulthood: Genetic and Environmental Effects.3

1Department of Psychology, Indiana University, Bloomington IN, 2Department of Public Health, University of Helsinki, FINLAND, 3FinnTwin16 is supported by NIAAA (AA 08315), and by the Academy of Finland (44069).

Address:  Indiana University Department of Psychology, 1101 East 10th St., Bloomington, IN 47405 Telephone: 812 855 1697 Fax: 812 856 4544 E-mail: viken@indiana.edu

 

In most samples alcohol consumption levels change dramatically during the period from mid-adolescence to young adulthood.  Past research has found equally dramatic changes in the determinants of drinking, with shared environmental effects typically decreasing during this period and additive genetic effects increasing.  Although it is possible to investigate developmental changes in the magnitude of genetic and environmental effects by cross-sectional studies of twins varying in age, much more information can be obtained from longitudinal analyses of twins measured repeatedly across a crucial developmental period. Drinking in the 2500 twin pairs of the FinnTwin16 study has now been assessed on four occasions: age 16, age 17, age 18.5, and age 23.  These repeated assessments make it possible to evaluate changes in drinking patterns and their determinants in a longitudinal framework.  We used a variety of analytic techniques to evaluate genetic and environmental effects on mean levels of drinking as well as trajectories of developmental change in drinking from age 16 to age 23.  Because the twins were assessed on a variety of personality and other individual differences measures at age 16, we also evaluated genetic and environmental associations between individual characteristics at age 16 and both average drinking and patterns of change in drinking over the next seven years.


Holly C. Wilcox1,2, Jenae Neiderhiser1, and David Reiss1.  Genetic and Environmental Influence on Adolescent Cannabis Use: The Role of Family Relationships.3

1Center for Family Research, Department of Psychiatry and Behavioral Sciences, George Washington University, Washington, DC, 2Department of Mental Health, Bloomberg School of Public Health, Johns Hopkins University, Baltimore, MD, 3Supported by NIH grants MH43373, MH48825 and MH19833

Address:  2300 K St., N.W., 3rd Floor Warwick Building, Washington, DC 20037 Phone: (410) 563-9108 Fax: (202) 994-4812 Email: cfrhxw@gwumc.edu

 

Recent evidence, using genetically informed designs, indicates the importance of family environment on the use of cannabis, as compared with other drugs (K.S. Kendler and C.A. Prescott, 1998, Am. J. Psychiatry, 55, 1016-1022; M.T. Tsuang, M.J. Lyons, J.M. Meyer, et al., 1998, Archives of General Psychiatry, 55, 967-972; M.T. Lynskey, A.C. Heath, E.C. Nelson, et al., 2002, Psychological Medicine, 32, 195-207).  The current study will examine how aspects of parent-child relationships relate to adolescent cannabis use. Potential familial transmission and specific aspects of parent-child relationships, such as conflict/negativity and parental monitoring, will be studied as potential influences on the pathway to adolescent cannabis use. The relative contribution of genetic, shared and unique environmental influences on cannabis use and the role of family relationships will be assessed at two distinct points in adolescent development.  The data for this study is from the first two waves of the Nonshared Environment in Adolescent Development (NEAD) project, a longitudinal study designed to investigate the role of nonshared environment on adolescent development. The sample consists of a total of 720 two-parent families across the United States with at least two same-sex adolescent sibling pairs, approximately evenly divided by sex (51.6% boys). The families comprise a genetically varied sample of monozygotic and dizygotic twins in nondivorced families, full siblings from nondivorced and stepfamilies, half siblings, and genetically unrelated stepsiblings.  Prevalence proportions of affected siblings and parent-child pairs, concordances, tetrachoric correlations, as well as univariate and multivariate twin-family models will be presented to clarify the mechanism of intergenerational transmission of the vulnerability to cannabis use. Prevalence proportions of cannabis using sibling pairs are higher in divorced than intact families. Higher rates of concordance for monozygotic and dizygotic twins were observed (49.9% and 33.3%, respectively) than adolescents from divorced families with step and half siblings (28.2%, 23.9%, respectively). Associations between cannabis use and parent-child relationships will be reported, specifically conflict/negativity and monitoring, and genotype-environment interactions will be examined.


Robert. W. Williams1, Lu Lu1, Yanhua Qu1, Jintao Wang1, Elissa J. Chesler1, Hui-Chen Hsu1, John D. Mountz1, David W. Threadgill1, Kenneth F. Manly1.  Brain Transcriptional Networks.2

1Center of Genomics and Bioinformatics, University of Tennessee, Memphis, TN 38163, 2Supported by an NIH/NSF Human Brain Project and the Dunavant Chair of Pediatrics.

Address:  Department of Anatomy and Neurobiology, University of Tennessee Health Science Center, 855 Monroe Avenue, Memphis TN 38163 Tel: (901) 448-7018, or -7050, -7557 Fax: (901) 448-7193 or -1716 Email: rwilliam@nb.utmem.EDU

 

Variation in mRNA levels measured using microarrays is generated primarily by environmental differences and gene variants. This variation often has important functional and behavioral repercussion. We are exploiting recombinant inbred (RI) strains in combination with microarrays to map large sets of cis- and trans-acting modulators of transcriptional activity in mouse brain.

These same strains have been used extensively for behavioral analysis (see http://webqtl.roswellpark.org/search.html).  Arrays of most RI lines were hybridized in triplicate with pooled brain samples from females at three ages. Our current mapping panel consists of 30 BXD strains. We developed custom programs and procedures to map each of 12422 transcripts using a data set consisting of approximately 600 microsatellite markers. We have mapped ~650 cis- and transacting loci that modulate the expression of key CNS genes such a receptors, ion channels, and transmitter transporters. Given the large numbers of traits that we have tested, a sizable fraction of QTLs are false positives. Even by the most conservative Bonferroni correction, we have mapped 40 to 50 new transacting QTLs. The prior probability that expression of a transcript is modulated by a linked QTL (for example, a promoter variant) is far higher than that of a randomly interval. Transacting QTLs with LODs above 5 are likely to be genuine whereas a majority of cis-acting QTLs with LOD scores above 1.5 to 3.0 are likely to be correctly identified. The collection of QTLs have unexpected and intriguing patterns of distribution. For example, an interval on proximal Chr 9, associated with the RNA helicase gene Ddx25, harbors over ~8% of all significant loci. This massively parallel approach to mapping QTLs that control gene expression poses significant statistical challenges but offers an even more significant opportunities to dissect transcriptional networks.


Hong Xian1, Jeffrey Scherrer2, Seth A. Eisen1, and William R. True2.  The role of post-traumatic stress disorder in explaining the association between combat trauma and lifetime nicotine or alcohol dependence.

1VAMC Research Service and department of Internal Medicine, Washington University, 2VAMC Research and St. Louis University School of Public Health

Address:  Washington University School of Medicine, VAMC 151-JC, 915 N. Grand, St.

Louis, MO 63106-1621 Telephone: 314-289-6532 FAX: 314-289-7604 E-mail: hxian@im.wustl.edu

 

Trauma, including combat, has been shown to be associated with increased substance use. In particular, previous studies have found increased intake of nicotine and alcohol following severe trauma. Yet it is not known if these associations operate through the development of post-traumatic stress disorder (PTSD) or another mechanism. We investigated, after controlling for the genetic and environmental contributions to PTSD, whether and to what degree genetic and environmental contributions overlap between combat and nicotine dependence (ND), and between combat and alcohol dependence (AD). Subjects were 3,120 monozygotic and dizygotic male-male twin pair members of the Vietnam Era Twin (VET) Registry. Separate Cholesky trivariate models were fit to estimate the magnitude of genetic and environmental contributions to the co-occurrence of combat and ND and combat and AD after controlling for the genetic and environmental contributions to PTSD. The liability for combat was due to about 4.0% additive genetic and 10.2% unique environmental contributions common to PTSD and ND, and to PTSD and AD, respectively. After controlling for PTSD, we found no evidence of genetic and environmental overlap between combat and ND, or between combat and AD. Additive genetic influences specific to combat accounted for about 29.2% of the total variance in combat. The remaining variance for combat was due to unique environmental factors specific to combat,

which accounted for about 56.3% of the variance in combat. These results suggest that the association between combat and ND, and between combat and AD observed by previous studies, in fact, operate through the mediation of the development of PTSD.


Susan E. Young1, Andrew Smolen1, Michael C. Stallings1, Robin P. Corley1, John K.

Hewitt1.  Anxiety and depression from early to late childhood:  A sibling-based

association study of the serotonin transporter polymorphism.2

1Institute for Behavioral Genetics, University of Colorado, Boulder, CO, 2Supported by the following NIH Grants: MH-01865, MH-43899, DA-11015, DA-05131, HD-18426; and the John D. and Catherine T. MacArthur Foundation

Address:  Institute for Behavioral Genetics, UCB 447, University of Colorado, Boulder, CO  80309 Phone: 303 492 1235 Fax: 303 492 8063 Email: Susan.Young@Colorado.edu

 

Childhood internalizing problems are often precursors in the development of more serious psychiatric syndromes including anxiety and depressive disorders.  Twin studies of the etiology of these disorders suggest that the genetic risk factors underlying anxiety and depression are highly correlated.  We examined the association between childhood internalizing problems and a functional polymorphism in the serotonin transporter gene (5-HTTLPR) in 711 children participating in a longitudinal twin study of behavioral and emotional development.  Internalizing problems were measured at ages 4, 7, 9, 10, 11 and 12 years using the Child Behavior Checklist (CBCL) parent report form, and at age 12 using the Separation Anxiety,

Generalized Anxiety and Major Depressive Disorder modules of the Diagnostic Interview Schedule for Children.  We applied a sibling-based methodology for estimating allelic association with quantitative traits, while controlling for population stratification.