Association of polymorphisms of dopamine D 2 receptor (DRD 2 ), and dopamine transporter (DAT 1 ) genes with schizoid/avoidant behaviors (SAB)

Transcription

1 Molecular Psychiatry (1997) 2, Stockton Press All rights reserved /97 $12.00 ORIGINAL RESEARCH ARTICLE Association of polymorphisms of dopamine D 2 receptor (DRD 2 ), and dopamine transporter (DAT 1 ) genes with schizoid/avoidant behaviors (SAB) K Blum 1,4, ER Braverman 2,SWu 3, JG Cull 4, TJH Chen 5, J Gill 4, R Wood 7, A Eisenberg 8, M Sherman 8, KR Davis 8, D Matthews 9, L Fischer 9, N Schnautz 10, W Walsh 11, AA Pontius 12, M Zedar 4, G Kaats 13 and DE Comings 3 1 UTHSC School of Public Health at Houston, Department of Behavioral Sciences, 7703 Floyd Curl Drive, San Antonio, Texas ; 2 Department of Psychiatry, New York University School of Medicine, 550 First Avenue, New York, NY 10010; 3 City of Hope National Medical Center, Department of Medical Genetics, Duarte, California 91010; 4 Kantroll, Inc, 2402 Toftrees Drive, San Antonio, Texas 78209; 5 Chang Jung University, 100 Chang Jung Road, Tatan Kway Jen, Tainan, Taiwan, Republic of China; 7 Department of Computing Resources, UTHSC at SA, 7703 Floyd Curl Drive, San Antonio, Texas; 8 Department of Pathology, DNA Identity Laboratory, University of North Texas Health Science Center, 3500 Camp Bowie Blvd, Ft Worth, Texas 76107; 9 Comprehensive Neurobehavioral Systems, 515 S Capital of Texas Highway, Suite 240, Austin, Texas 78246; 10 San Marcos Treatment Center, Bert Brown Road, San Marcos, Texas 78667; 11 Health Research Institute, Pfeiffer Treatment Center, 1804 Centre Point Drive, Suite 102, Naperville, Illinois 60563; 12 Harvard Medical School, Massachusetts General Hospital, Department of Psychiatry, Fruit Street, Boston, Massachusetts 02114; 13 Health and Medical Research Foundation, Broadway Street, Suite, San Antonio, Texas 78209, USA The dopaminergic system, and in particular the dopamine D 2 receptor, has been implicated in reward mechanisms in the brain. Dysfunction of the D 2 dopamine receptors leads to aberrant substance-seeking behaviors (ethanol, drugs, tobacco, and food) and other related behaviors (pathological gambling, Tourette s disorder, attention-deficit/hyperactivity disorder). This is the first study supporting a strong association between the dopamine D 2 receptor Taq A 1 allele with schizoid/avoidant behavior (SAB). Additionally, an albeit weaker association between the 480-bp VNTR 10/10 allele of the dopamine transporter (DAT 1 ) gene with SAB was similarly found. Keywords: schizoid/avoidant behavior; polymorphisms, dopamine D 2 receptor gene (DRD 2 ); dopamine transporter gene (DAT 1 ) Introduction Correspondence: Dr K Blum, UTHSC School of Public Health at Houston, Department of Behavioral Sciences, 7703 Floyd Curl Drive, San Antonio, Texas , USA Received 13 August 1996; revised and accepted 22 January 1977 Among the eleven personality disorders in the DSM- IV, only the schizotypal, borderline, and antisocial have been examined to any considerable extent for gene effects. 1 Most recently, dopamine D 4 Exon bp repeat region (2 8 repeats) was found in two inde- pendent studies to associate with novelty-seeking behavior. 2,3 As previously suggested, human person- ality traits which can be measured adequately by standardized rating scales show a considerable heritable component. 4 Moreover Bouchard, 5 following an exhaustive review of the current twin and adoptive studies, suggested that 30 60% of the variance in a number of personality traits is due to inherited factors. Expression of complex human behavior such as personality involves both genetic and environmental effects. Where these complex traits do not show simple Mendelian patterns of inheritance, one would not expect simple genetic answers caused by a single gene. In this regard as suggested by Plomin et al, 6 the one- gene, one-disorder hypothesis (OGOD) does not look for a single gene for complex traits, but rather assumes that complex traits comprise several subtraits each influenced by a single gene. The OGOD strategy has already been successful for some complex behavioral disorders: mental retardation, 7 Alzheimer s, 8,9 impul- sive violence, 10 and severe alcoholism and drug abuse. 11,12 While controversy exists with regard to the appropriate use of linkage vs allelic association studies, in terms of complex diseases having polygenic inherit- able components for net expression, there are advantages for association studies, which can be complemen- tary to linkage. Allelic association can provide the statistical power needed to detect small effect size. Evidence for the involvement of dopaminergic dysfunction in personality traits including antisocial behaviors comes from a number of studies showing improvement with methylphenidate 13 and monamine

2 240 oxidase inhibitors, 14 suggesting a strong relationship orbid drug or alcohol abuse were selected for both genotyping between these behaviors and reward deficits. 15 Molecular and assessment with the Millon Clinical Multibetween genetic evidence on the role of dopamine in reward axial Inventory (MCMI-II). The composition of the behaviors such as severe alcoholism, polysubstance group consisted of the following: dysthymia (22.1%), abuse, carbohydrate bingeing, smoking behavior, generalized anxiety disorder (8.4%), unipolar (24.5%), pathological gambling, as well as other related bipolar (12.9%), schizophrenia (5.2%), attention-defi- behaviors (ie attention-deficit/hyperactivity disorder cit disorder (21.9%), and substance use disorder (ADHD) and Tourette s disorder) have been confirmed, (18.1%). The demographic breakdown of our sample 16,17 in spite of a number of negative base is illustrated in Table 1. reports Polymorphisms of the dopamine D 2 receptor The Millon Clinical Multiaxial Inventory was gene (DRD 2 ) have associated with impulsive- employed to assess the eleven known personality dis- addictive-compulsive behaviors, the 480-bp VNTR orders including the schizoid/avoidant cluster in each 10/10 allele of the dopamine transporter gene (DAT 1 ) subject. This test contains 175 items and most people associated with not only alcoholism, 42 but with ADHD can complete it in minutes. Each of its 22 clini- and Tourette s disorder, 43,44 and the B 1 allele of dopa- cal scales was constructed as an operational measure mine beta hydroxylase gene (D H) also associated with of a syndrome derived from a theory of personality and Tourette s disorder and a number of behavioral sub- psychopathology. The clinically oriented scales are traits. 44 coordinated directly with the official diagnostic system The DRD 2 gene is localized to chromosome 11 q22 and its syndrome categories (DSM-IV). Separate scales 23, the D H gene is localized to chromosome 9 q34 and have been constructed in line with the DSM-IV model the human dopamine transporter gene first cloned by to distinguish the more enduring personality characteristics Vandenberg et al 45 has been mapped to chromosome 5 of patients (Axis II) from the acute clinical dis- pl 5.3. For association with schizoid/avoidant person- orders (Axis I). Actuarial base rate data, rather than ality, we utilized the Taq A 1 allele of the DRD 2 gene analyzed standard score transformations, were and the Taq B 1 allele of the D H gene. The presence employed in calculating and quantifying scale measures. of a polymorphism 40-bp repeat in the 3 untranslated A base rate of 60 is suggestive of a trait. Base rates region (UTR) of the gene with repeat numbers ranging between 75 and 83 indicate a chronic or moderately from 3 to 11 and Taq I RFLPs, allowed researchers to severe disorder and greater than or equal to 84 signifies carry out a number of linkage and association studies a pathological disorder. In this study we chose to identify for a number of conditions. The 10 and 9 repeats were the experimental group by using Millon s Schizoid found to be the most common, accounting for over 90% Avoidant cluster of behaviors in subjects achieving a of the alleles. 46 Therefore for the above reason we percentile rank of 84 or higher on this particular decided to utilize the 10/10 genotype of the DAT 1 gene cluster. for potential association with schizoid/avoidant personality. We classified subjects according to their MCMI-II It is noted that the simultaneous approach of scores, grouping the probands into four distinct base all three dopaminergic genes is based on the concept rates according to the following divisions: (1) scores of polygenic inheritance and this method was success- below 60; (2) equal to 60 up to 73; (3) equal to 74 up fully utilized most recently by Comings et al 47 to assess to 83; and (4) equal to 84 up to 100. The demographics allelic prevalence in Tourette s disorder and a number of the subject population are described in Table 2. of behavioral subtraits. The MCMI-II test scores were validated on a different inpatient population by comparing 104 patients with Materials and methods Minnesota Multi-Phasic Personality Inventory (MMPI) personality disorder scales. Conservative significance Subject selection and test administration levels were used to ensure valid conclusions. Schizoid, Caucasian volunteers were recruited from the patient avoidant, dependent, histrionic, and narcissistic scales population attending the PATH Clinic, Princeton, NJ, were correlated significantly, passive-aggressive, schi- USA. Volunteers gave informed consent and the proto- zotypal, and borderline scales did not correlate with col was approved by PATH Research Foundation Institutional Review Board Committee and the City of Hope Table 1 Diagnostic characteristics of a psychiatrically-ill National Medical Center Review Board. There were 58 population utilized in this study males and 71 females and the average age was 40.9 ± 1.8 and 47.3 ± 1.5 (mean ± s.d.) years, respectively. Prior to taking the second edition of the Millon Clinical Category of diagnosis % of Subjects Multiaxial Inventory (MCMI-II) computerized test, the Dysthymia 27.1 volunteers donated 15 cc of blood by venipuncture. It Generalized anxiety disorder 8.4 is noteworthy, because of the concerns about racial and Unipolar 24.5 ethnic stratification, that we drew blood from non- Bipolar 12.9 Hispanic, Northern and Western European Caucasians Schizophrenia 5.2 with some exceptions since the most extensive data on Attention deficit disorder 21.9 controls come from this group. For this study a total of 129 psychiatrically ill patients with and without com- Substance use disorder 18.1

3 Table 2 Association of DRD 2 and DAT 1 with SAB Demographics of three dopaminergic genes and schizoid/avoidant behavior (SAB) cluster patients 241 SAB Avg. age %M %F DRD 2 (%) a DAT 1 (%) D H (%) ± S.E. A 1 /A 1 A 1 /A 2 A 2 /A 2 10/10 10/9 9/9 B 1 /B 1 B 1 /B 2 B 2 /B ± (n = 27) (0) (6) (19) (5) (1) (0) (1) (2) (1) ± (n = 37) (1) (8) (26) (6) (65) (2) (4) (4) (6) ± (n = 15) 7 8 (1) (6) (7) (3) (3) (2) (2) (1) (2) ± (n = 32) (8) (18) (17) (17) (1) (6) (10) (4) a Parentheses indicate percentages. Genotyping The total number of subjects genotyped for one or more dopaminergic genes in this study was 271. All subjects were genotyped based on a neutral identification number and read without knowledge of the individual being typed. 1) DRD 2 polymorphism The D 2 A 1 and D 2 A 2 genotyping was performed by hybridization of Southern blots as described previously. 11,50 A number of samples were also genotyped by a PCR technique. 51 corresponding MCMI-II scales, therefore validating the use of this test to reliably measure abnormal personality traits 48,49 such as schizoid and avoidant behaviors. For this study 30 super control subjects attending the PATH Clinic, screened to exclude a number of reward deficit behaviors including alcoholism, poly- substance dependence, smoking behavior, carbo- hydrate bingeing, a BMI 25, family history of sub- stance use disorder, ADHD, pathological gambling and an Axis II diagnosis (including SAB), were genotyped for DRD 2 A 1 and A 2 alleles. Additional controls came from three sources of volunteers attending the City of Hope Medical Center: A) adopting, foster or step- parents of Tourette s disorder patients; B) subjects from an endocrinology clinic with thyroid cancer or noninsulin dependent diabetes mellitus; and C) hospital personnel including professionals, technicians, and maintenance workers. All 142 controls were screened to exclude ADHD, alcohol drug and tobacco abuse. We genotyped 91 of these controls for the DAT 1 9 and 10 alleles and 51 controls for the D H B 1 and B 2 alleles. One important caveat to consider involves the subject selection to evaluate association of dopaminergic alleles with SAB. While we are cognizant of possible statistical confounds with subjects having comorbid RDS behaviors (ie, substance use disorder at 18.1% in this population) with which associations have already been proposed at dopaminergic loci, we are equally cognizant that it would be quite difficult to exclude all RDS behaviors in patients presenting with SAB. Thus, we must await additional studies in the future to address this issue specifically. 2) D H polymorphism d Amato and associates 52 reported the presence of two Taq D H polymorphisms entitled A and B. A D H cdna clone All 53,54 was used consisting of a 2.7-kb insert at the EcoRI site. To improve labeling the vector was digested with BamHI and SalI to produce five bands. A 3.5-kb fragment was labeled for testing the polymorphism. Digestion with Taq I restriction endonuclease, electrophoresis in aga- rose, Southern transfer to a nylon filter, hybridization with 32 P-labeled probe, and autoradiography, demonstrated fragments of 2.8 kb (B 1 ) and 1.4 kb (B 2 ). 3) DAT 1 repeat polymorphism The alleles at the 3 UTR were determined by PCR using the oligomers and PCR conditions reported by Vandenberg et al. 45 Fol- lowing PCR amplification the products were electroph- oresed in an 8% acrylamide gel with a set of size markers. Statistics Proportions were analyzed for 2 by 2 tables using Pear- son s chi-square and Fisher s exact tests. Larger tables such as 2 by 3 tables were tested using Pearson s chi- square and Mantel Haensel test for linear trend. Differences between means were tested using Student s t- test. Multiple Logistic regression was used to analyze the contribution of one or several variables to predict the probability of having a high schizoid/avoidance score. The logistic model was assessed using the Hosmer Lemeshow goodness-of-fit test, the c statistic which represents the area under the receiver operating characteristic (ROC) curve and the r 2 or presence of variance accounted for by the SAS computer program Proc Logistic. Odds ratios were also computed for the predictor variables in any logistic regression model. 55 Results For the schizoid/avoidant data set we decided to analyze the data utilizing the chi-square approach limiting

4 242 group. Whereas, the group having only DRD 2 A 2 representation showed the lowest percentage of SAB (linear trend analysis: P 0.005, A 1 /A 1 = 83%, A 1 /A 2 = 41%, and A 2 /A 2 = 23% [See Figure 2]). However, unlike the DRD 2 A 1 allele data, chi-square analysis failed to reveal association of D H B 1 allele and DAT 1 10/10 allele with schizoid/avoidant behaviors. Utilizing multiple variable associations with the dichotomized SAB scores using logistic regression testing for significant relationships with the DRD 2 alleles, age and sex, both DRD 2 A 1 allele and sex were significant predictors of SAB severity. With DRD 2 A 1 allele we found an odds ratio of 2.79 (P = 0.018) and with sex 3.6 (P = ), with a Hosmer Lemeshow goodness-of-fit at P = Table 1 and Figure 2 show the data relevant to 172 screened controls for either DRD 2 A 1,D HB 1 and Figure 1 The percent carriers of the DRD 2 A 1 allele, DAT 1 DAT 1 10 alleles. 10/10 allele, and D H B 1 allele in schizoid and avoidant In 30 screened super controls (exclusion of alcoholbehaviors (MCMI-II scores 84). The number of subjects are ism, polysubstance dependence, body mass index less in parenthesis. The asterisk denotes significance between the controls and allelic prevalence: * = 16.8, d.f. = 1, P = than 25, smoking behavior, family history, pathological ; ** = 6.3, d.f. = 1, P = gambling, ADHD, schizoid/avoidant behavior), the prevalence, as shown in Table 3, of the DRD 2 A 1 allele was 1/30 or 3.3%. In 91 screened controls the prevapotential association of the dopaminergic alleles to lence of the DAT 1 10/10 allele was 34/91 or 37.4% as MCMI-II schizoid and MCMI-II avoidant scores above well as in 51 screened controls where the prevalence 84 to ensure severity of the proposed phenotype. Fig- of the D H B 1 allele was 27/51 or 53%. With regard ure 1 shows that carriers of the DRD 2 A 1 allele were to the DRD 2 A 1 allele (18/37 or 48.6%), a significant found in 50% of schizoid (11/22) and 44% avoidant association was found when compared to both litera- (12/27) subjects; DAT bp (VNTR 10/10 allele) was ture controls 185/714 or 26% ( 2 = 9.2, d.f. = 1, P = found in 72% of schizoid (13/18) subjects, and 62% of ; OR = 2.71) and super controls ( 2 = 16.8, d.f. = avoidant (13/21) subjects; D H B 1 allele was found in 1, P = ; OR = 27.5) and SAB 84. Moreover, a 81.3% of schizoid (13/16) subjects and 82.4% of significant association was also found between the avoidant (14/17) subjects. DAT bp (VNTR 10/10 allele) in those individuals With chi-square, we found that the DRD 2 A 1 allele diagnosed with SAB (18/28 or 64.3%) when compared significantly associated with patients to have the to screened controls ( 2 = 6.3, d.f. = 1, P = 0.012; OR schizoid/avoidant cluster (MCMI-II score 84) com- 3.0). A similar trend was found with carriers of the pared to DRD 2 A 2 allele ( 2 = 7.6, d.f. = 1, P 0.006). D H B 1 allele assessed as having SAB (17/23 or 73.9%) Homozygotes of the DRD 2 A 1 allele showed the highest compared to screened controls ( 2 = 2.89, d.f. = 1, P = percentage of subjects having membership in the SAB 0.09; OR 2.52). cluster ( 84). Heterozygotes had approximately half of Comparing all cases below MCMI-II scores of 84 and the percentage of SAB subjects as the homozygote all cases of MCMI-II scores above 84 (severity score), utilizing a statistical technique called logistic regression analysis, DRD 2 A 1 allele accounts for 8.3% of the variance which is statistically significant ( 2 = 7.5, d.f. = 1, P = ). In contrast, DAT 1 (10/10 allele) accounts for 1.6% of the variance; and the D H B 1 allele accounts for % of the variance and are not significant contributors to the variance. When sex is examined as a univariate in a logistic regression model, we find a contribution to the overall variance of sex alone to be 9.9% ( 2 = 9.4, d.f. = 1, P = ). In a logistic regression model predicting high schizoid/avoidant MCMI-II scores above 84, utilizing the DRD 2 A 1 allele and gender as predictors, odds Figure 2 Linear trend analysis of the DRD 2 percent prevaa ratios for these predictors were 2.79 for DRD 2 gene and 3.55 for male gender. The Hosmer Lemeshow goodness-of-fit P value = and the C statistic (area under the ROC curve) equals and the combined contribution to the variance was 17.9%. Moreover, in lence as a function of genotype in schizoid/avoidant cluster pilot study of 67 subjects genotyped for the DRD 2 A 1 behaviors (MCMI-II scores 84, P 0.005). and A 2 alleles, using the procedure discussed above,

5 Table 3 The three dopaminergic genes genotyped in controls, schizoid and avoidant behaviors 243 Genotype Percent prevalence Controls Schizoid Avoidant P value (at SAB) a DRD 2 A 1 1/30 11/22 12/ (3.3) (50) (44) DAT 1 10/10 34/91 13/18 13/ (37.4) (72) (62) D H B 1 27/51 13/16 14/ (53) (81.3) (82.4) a Data set utilized a cluster score of schizoid/avoidant behaviors 84. we found no association with any of the other ten per- behaviors, it is intriguing that unlike the DRD 4 sonality traits assessed by the MCMI-II test. gene, the DRD 2 A 1 allele was found to also associate with individuals who showed lower TPQ novelty Discussion scores, lower extroversion, more withdrawal depression and more neuroticism. 63 Furthermore, In conclusion, we have performed an association study dopamine transporter receptor sites significantly that employed the MCMI-II self-report computerized increased in alcoholics compared to non-violent test in order to reveal a contribution of polymorphisms alcoholics, 64 and others 65 have shown a correlation of three dopaminergic genes to an abnormal person- between low D H levels and sensation-seeking ality trait referred to as schizoid/avoidant cluster. The behaviors. observed association did not appear to be due to popu- While schizoid/avoidant individuals initially tend to lation stratification since it was independent of the ethnicity, be languid, remote, non-affective (passionless), deper- sex, or age of the subjects. Moreover, the fre- sonalized, conflicted, hypersensitive, phobic and self- quency for schizoid/avoidant behaviors based on the deserting, the literature indicates that, especially in general population is 1 4%. The lack of significant subjects scoring above 84 on the MCMI-II, a significant contribution of both the DAT 1 (10/10 repeat allele) and number of probands tend to alleviate these dysphoric the D H B 1 allele to the overall variance cannot be symptoms and seek out pleasure through outrageous ruled out as yet due to sample size. In fact with a sample acts of violence with a pervasive pattern of social dis- size of only 44 (due to missing values), these three comfort followed by extreme comorbid substance use gene polymorphisms when combined in multivariate disorder. 66 regression analysis account for roughly 7.6% of the These association studies, given the size of the contribution genetic variance, as might be anticipated if there are of genetic variance, support the notion that multiple genes and other factors like gender involved human personality traits are inheritable to some degree in this complex behavioral disorder. and dopaminergic gene variants are involved in both As we previously suggested, association studies of temperament and at least two closely linked personality this sort provide a powerful strategy for detecting the disorders, schizoid and avoidant behaviors. subtle effects of genes on multifactorial quantitative Additionally, Cloninger s proposal that personality traits, especially where known candidate loci are available traits may have distinct neurochemical and genetic and specific dopaminergic alleles have been substrates mediated by genetic variability in dopamine shown to associate with other behaviors tied to the limbic transmission as well as other neurotransmitters is sup- regions of the brain. 56,57 ported by the work of both Benjamin 2 and Epstein 3 as According to Benjamin et al, 3 a similar approach well as this study. may be useful for detecting genes that influence abnor- The observed association in our study raises the mal psychological processes and health risk-related question of the causal relationship between the struc- behavior such as tobacco smoking and excess alcohol ture of three dopaminergic genes (DRD 2, DAT 1 and consumption. Thus, the work from the laboratories of D H) and how their gene products interact with the Benjamin 3 and Epstein 2 provide support for the earlier developing brain and with environmental factors to work of others 58 who found a strong association generate the complex mix of actions that comprise between variants of the D 4 gene and alcoholism, and abnormal personality traits such as schizoid and avoidant nicotine dependence. behaviors. In this regard, little is known about the Although DRD 4 receptor mrna shows a distinct resultant expression of polymorphisms linked to either neuro-anatomical distribution in comparison to D 2 and the DAT 1 10/10 allele and the D H B 1 allele except D 3 mrnas with a concentration in limbic areas asso- for the work of Malison 67 showing increased dopamine ciated with cognitive (P300 wave) 59,60 and emotional transporter sites in Tourette s disorder patients by

6 244 SPECT scanning techniques, and the work of Tiihonen ancial support of Kantroll, Inc and the PATH Foundation et al. 64 However, more is known about structure and and the Wacker Foundation, Dallas, Texas. function related to the DRD 2 A 1 allele, where naive carriers of the A 1 allele had significantly lower DRD 2 References receptors compared to the more frequent A 2 allele Dahl AA. Heredity in personality disorders an over- Based on a plethora of genetic research, polymorview. Clin Genet 1994; 46: phisms in 3 UTR could affect protein expression via 2 Ebstein R, Novick O, Umansky R, Priel B, Osher Y, Blaine reduced translation due to RNA unstability. 69 The D, Bennett ER, Nemanov L, Katz M, Belmaker RH. Dopa- DRD 2 gene is a particularly attractive candidate as a mine D 4 receptor (D 4 DR) Exon III polymorphism associaquantitative trait locus for the schizoid/avoidant clus- ted with the human personality trait of novelty seeking. ter, because of several reasons: (i) DRD 2 A 1 allele has Nature Genet 1996; 12: been found to associate with a number of reward 3 Benjamin J, Li L, Patterson C, Greenberg BD, Murphy DL, behaviors including severe alcoholism, polysubstance Hamer DH. Population and familial association between dependence, crack/cocaine addiction, tobacco smoking the D 4 dopamine receptor gene and measures of novelty and carbohydrate bingeing or generalized to DSM-IV seeking. Nature Genet 1996; 12: Cloninger CR, Svrakic DM, Przybeck TR. A psychobiologisubstance use disorder; 16,70 (ii) the MCMI-II assessed cal model of temperament and character. Arch Gen Psych schizoid/avoidant cluster compared to other Axis II 1993; 50: diagnostic clusters (antisocial, narcissistic, paranoid) 5 Bouchard TJ. Genes, environment, and personality. significantly correlated with alcohol abuse scales; 71 Science 1994; 264: (iii) clusters of patients with MCMI-II elevations that 6 Plomin R, Owen MJ, McGuffin R. The genetic basis of indicated schizoid and avoidant qualities tended to complex human behaviors. Science 1994; 264: stay in treatment fewer days and relapsed earlier; 72 7 Brunner HG, Nelen MR, van Zandvoort P, Abeling NGGM, (iv) high scores of schizoid/avoidant cluster correlated van Gennip AH, Wolters EC, Kutper MA, Ropers HH, van with inpatient male alcoholics 73 and cocaine-depenbehavioral disturbance: phenotype, genetic localization Oost BA. X-linked borderline mental retardation with dent patients; 74 (v) schizoid/avoidant behaviors includand evidence for disturbed monamine metabolism. Am J ing low levels of sensation were found to consume Hum Genet 1993; 52: more alcohol and to have higher MAST scores than 8 Corder EH, Saunders AM, Risch NJ, Strittmatter WJ, patients with high levels of sensation; 75 and Schmechel DE, Gaskell Jr PC, Rimmler JB, Locke PA, Con- (vi) avoidant personality is significantly correlated in neally PM, Schmader KE, Small GW, Roses AD, Haines subjects with severe binge eating disorder. 76 JL, Pericak-Vance MA. Protective effect of apolipoprotein It is noteworthy that Pontius et al 77 reported on a E type 2 allele for late onset Alzheimer s disease. Nature condition limbic psychotic trigger reaction supporting Genet 1994; 7: the link between limbic dysfunction, unexplained 9 Sandbrink R, Hartmann T, Masters CL, Beyreuther K. murders, and schizoid/avoidant personality traits. Genes contributing to Alzheimer s disease. Mol Psychiatry According to the work of Neiswanger et al, 78 research 1996; 1: Brunner HG, Nelsen M, Breakefield XO, Ropers HH, van results are more sharply differentiated if the researcher Oost BA. Abnormal behavior associated with a point uses super controls rather than a general population mutation in the structural gene for monoamine oxidase. of normals. In the current study it was not possible Science 1993; 262: to screen out those individuals in the control group 11 Blum K, Noble EP, Sheridan PJ, Montgomery A, Ritchie T, with some level of SAB using the DAT 1 and the D H Jadadeeswaran P, Nogami H, Briggs AH, Cahn JB. Allelic gene loci; consequently, the statistical significance of association of human dopamine D 2 receptor gene in the results in this study would vary inversely with the alcoholism. J Am Med Assn 1990; 263: number of SAB subjects in the control group. Due to 12 Uhl G, Blum K, Noble EP, Smith S. Substance abuse vul- the comments of Neiswanger et al, these findings nerability and D 2 receptor genes. Trends Neurosci 1993; should be viewed cautiously until controls can be 16: Hinshan SP, Heller T, McHale JP. Covert antisocial screened adequately to eliminate all associated behavior in boys with attention-deficit/hyperactivity disbehaviors as we have done for the DRD 2 gene controls. order: external validation and effects of methylphenidate. Given the significant heritability of human person- J Consulting Clin Psychol 1992; 60: ality disorders 79 and polygenic inheritance of complex 14 Fahlen T. Personality traits in social phobia, II: changes diseases, it is likely that additional genes that influence during drug treatment. J Clin Psychiatry 1995; 56: 569 these behaviors will be found in the future Johnson JG, Hyler SE, Skodol AE, Bornstein RF, Sherman Acknowledgements M. Personality disorder symptomatology associated with adolescent depression and substance abuse. J Personality We thank Sadie Phillips for typing the manuscript, Dis 1995; 9: Blum K, Cull JG, Braverman ER, Comings DE. Reward David Martin for his encouragement, dedicated to the deficiency syndrome. Am Scientist 1996; 84: work of EP Noble, R Cloninger, and Thomas Bouchard. 17 Blum K, Sheridan PJ, Wood RC, Braverman ER, Chen TJH, DEC is the recipient of the NIDA grant R01-DA08417 Cull JG, Comings DE. The D 2 dopamine receptor gene as a and Tobacco Related Research Disease Program grant predictor of reward deficiency syndrome (RDS) behavior: 4RT We are grateful to the National Computer Bayes Theorem. J Royal Soc Med 1996; 87: Systems for the Millon tests and acknowledge the fin- 18 Bolos AM, Dean M, Lucas-Derse S, Ramsburg M, Brown

7 GL, Goldman D. Population and pedigree studies reveal a 35 O Hara BF, Smith SS, Bird G, Persico A, Suarez B, Cutting lack of association between the dopamine D 2 receptor GR, Uhl GR. Dopamine D 2 receptor RFLPs, haplotypes and gene and alcoholism. JAMA 1990; 264: their association with substance use in black and Caucasian 19 Cook BL, Wang ZW, Crowe RR, Hauser R, Freimer M. research volunteers. Hum Hered 1993; 43: Alcoholism and the D 2 receptor gene. Alcohol Clin Exp 36 Parsian A, Todd RD, Devor EJ, O Malley KL, Suarez BK, Res 1992; 4: Reich T, Cloninger CR. Alcoholism and alleles of the 20 Gelernter J, O Malley SO, Risch N, Kranzler RR, Krystal J, human D 2 dopamine receptor locus. Arch Gen Psych 1991; Merikangas K, Kennedy JL, Kidd KK. No association 48: between an allele at the D 2 dopamine receptor gene (DRD 2 ) 37 Pato CN, Macciardi F, Pato MT, Verga M, Kennedy JL. and alcoholism. JAMA 1991; 47: Review of the putative association of dopamine D 2 recep- 21 Goldman D, Brown GL, Albaugh B, Robin R, Goodson S, tor and alcoholism: a meta analysis. Am J Med Gen Trunzo M, Akhtar L, Lucas-Derse S, Long J, Linnoila M, (Neuropsychiatric Genetics) 1993; 48: Dean M. DRD 2 dopamine receptor genotype, linkage dis- 38 Smith SS, O Hara BF, Persico AM, Gorelick DA, Newlin equilibrium, and alcoholism in American Indians and DB, Vlahov D, Solomon L, Pickins R, Uhl GR. Genetic vul- other populations. Alcohol Clin Exp Res 1993; 17: 199 nerability to drug abuse: the D 2 dopamine receptor Taq I 204. B 1 restriction of fragment length polymorphism appears 22 Goldman D, Dean M, Brow GL, Bolos AM, Trokola R, Virk- more frequently in polysubstance abusers. Arch Gen Psychiatry kunen M, Linnoila M. D 2 dopamine receptor genotype and 1992; 49: cerebrospinal fluid homovanillic acid, 5-hydroxyindole- 39 Smith SS, Gorelick DA, O Hara BF, Uhl GR. The dopamine acetic acid and 3-methoxy-4-hydroxy-phenylglycol in D 2 receptor gene and alcoholism. JAMA 1991; 256: alcoholics in Finland and the United States. Acta Psychi atr Scand 1992; 86: Uhl GR. Molecular and genetic studies of the targets of 23 Suarez BK, Parsian A, Hampe CL, Todd RD, Reich T, Clon- acute drug action, substrates for inter individual differences inger CR. Linkage disequilibria at the D 2 dopamine receptor in vulnerability to substance abuse, and candidate locus (DRD 2 ) in alcoholics and controls. Genomics mechanisms for addiction. In: Chiarello E (ed). NIDA 1994; 19: Monographs. Washington, DC, 1994, (in press). 24 Turner E, Ewing J, Shilling P, Smith TL, Irwin M, Schuckit 41 Comings DE, Gade R, MacMurray JP, Muhleman D, Johnson M, Kelsoe JR. Lack of association between an RFLP near P, Verde R, Peters WR. Genetic variants of the human the D 2 dopamine receptor gene and severe alcoholism. obesity (OB) gene: association with body mass index in Biol Psychiatry 1992; 31: young women, psychiatric symptoms, and interaction 25 Blum K, Wood RC, Sheridan PJ, Braverman ER, Chen TJH, with the dopamine D 2 receptor (DRD 2 ) gene. Mol Psychiatry Comings DE. Dopamine D 2 receptor gene variants: association 1996; 1: and linkage studies in impulsive-addictive-compul- 42 Muramatsu T, Higuchi S. Dopamine transporter gene sive disorders. Pharmacogenetics 1995; 5: polymorphism and alcoholism. Biochem Biophys Res 26 Amadeo S, Abbar M, Fourcade ML, Waksman G, Leroux Com 1995; 211: MG, Madec A, Selin M, Champiat J-C, Brethome A, Lucclaire 43 Cook EH Jr, Stein MA, Krasowski MD, Cox NJ, Olkon DM, Y, Castelnau D, Venisse J-L, Mallet J. D 2 dopamine Kieffer JE, Leventhal BL. Association of attention-deficit receptor gene and alcoholism. J Psychiat Res 1993; 27: disorder and the dopamine transporter gene. Am J Hum Genet 1995; 56: Arinami T, Itokawa M, Komiyama T, Mitsushio H, Mori 44 Comings DE, Wu H, Chiu C, Ring RH, Dietz G, Muhleman H, Mifune H, Hamaguchi H, Toru M. Association between D. Polygenic inheritance of Tourette syndrome, stuttering, severity of alcoholism and the A 1 allele of the dopamine attention-deficit/hyperactivity, conduct and oppositional D 2 receptor gene Taq I a RFLP in Japanese. Biol Psychiatry defiant disorder: the additive and subtractive effect of the 1993; 33: three dopaminergic genes DRD 2,D H and DAT 1. Am J 28 Cloninger CR. D 2 dopamine receptor gene is associated but Med Genet (Neuropsychiat Genet) 1996; 67: not linked with alcoholism. JAMA 1991; 266: Vandenbergh DJ, Persico AM, Hawkins AL, Griffin GA, Li 29 Conneally PM. Association between the D 2 domain recep- X, Jabs EW, Uhl GR. Human dopamine transporter gene tor gene and alcoholism: a continuing controversy. Arch (DAT 1 ) maps to chromosome 5 PI 5.3 and displays a Gen Psych 1991; 48: VNTR. Genomics 1992; 14: Cook CCH, Brett P, Curtis D, Holmes D, Gurling HMD. 46 Comings DE. The Discovery of the Tourette Syndrome and Linkage analysis confirms a genetic effect at the D 2 dopa- Human Behavior Genes. Hope Press, Duarte, CA, mine receptor locus in heavy drinking and alcoholism. 47 Comings DE, Rosenthal RJ, Lesieur MR, Rugle L, Muhle- Psych Genet 1993; 3: 130. man D, Chiu C, Dietz G, Gade R. A study of the dopamine 31 Devor EJ. The D 2 dopamine receptor and Tourette s syndrome. D2 receptor gene in pathological gambling. Pharmaco- JAMA 1992; 267: genetics 1996; 6: Flanagan SD, Noble EP, Blum K, MacMurray J, Comings 48 Schuler CE, Snibber JR, Buckwalter JL. Validity of the D, Ritchie T, Sheridan PJ, Lopatin GO, Gysin R. Evidence MMP1 personality disorder scales (MMPI-PF). J Clin Psychology for a third physiologically distinct allele at the dopamine 1994; 50: D 2 receptor locus (DRD 2 ). Annu Meet Am Psychopath 49 Zimmerman M. Diagnosing personality disorders. Arch Meet, New York, Gen Psychiat 1994; 51: Neiswanger K, Hill SY, Kaplan BB. Association between 50 Comings DE, Comings BG, Muhleman D, Dietz G, Shahbahrami alcoholism and the Taq I A RFLP of the dopamine D 2 B, Tast D, Knell E, Kocsis P, Baumgarten R, Kovalcoholism receptor gene in the absence of linkage. Psych Genet 1993; acs BW, Levy DL, Smith M, Kane JM, Lieberman JA, Klein 3: 130. DN, MacMurray J, Tosk J, Sverd J, Gysin R, Flanagan S. 34 Noble EP. The D 2 dopamine receptor gene: a review of The dopamine D 2 receptor locus as a modifying gene in association studies in alcoholism. Behav Genetics 1993; 23: neuropsychiatric disorders. J Am Med Assn 1991; 266:

8 Noble EP, Noble RE, Ritchie T, Grandy DK, Sparkes RS. D 2 thy adults. Res Commun Psychol Psychiat Behav 1981; 6: dopamine receptor gene and obesity. Int J Eating Disorder ; 15: O Boyle M. Personality disorder and multiple substance 52 d Amato T, Leboyer M, Malafosse A, Samolyk D, Lamouroux dependence. J Personality Disorders 1993; 7: A, Junien C, Mallet J. Two TaqI dimorphic sites at 67 Malison RT, McDougle CJ, van Dyck CH, Scahill L, Bald- the human P-lydroxylase locus. Nucleic Acad Res 1989; win RM, Seibyl JP, Price LH, Leckman JF, Innis RB. 17: [1231]beta-CIT SPECT imaging of striatal dopamine transporter 53 Lamouroux A, Vigny A, Faucon-Biguet N, Darmon MC, bindings in Tourette s disorder. Am J Psychiatry Frank R, Henry JP. The primary structure of dopamine- 1995; 152: beta-hydroxylase: insights into the relationship between 68 Noble EP, Blum K, Ritchie T, Montgomery A, Sheridan the soluble and the membrane-bound forms of the PJ. Allelic association of the D 2 dopamine receptor gene enzyme. AMBO J 1987; 6: with receptor-binding characteristics in alcoholism. Arch 54 Lamouroux A, Houhou L, Biguet NF, Serck-Hanssen G, Gen Psychiatry 1991; 48: Guibert B, Icard-Liepkalns C, Mallet J. Analysis of the 69 Boeller DM, Horowitz JA, Casey JL, Klansner RD, Itarford human dopamine -hydroxylase promoter: transcriptional JB. Translation and the stability of mrnas encoding the induction by cyclic AMP. J Neurochemistry 1993; 60: transferrin receptor and c-fos. Proc Natl Acad Sci USA ; 88: Dunn O, Clark V. Applied Statistics: Analysis of Variance 70 Blum K, Sheridan PJ, Wood RC, Braverman ER, Chen TJH, and Regression. 2nd edition, Wiley, New York, Comings DE. Dopamine D 2 receptor gene variants: association 56 Noble EP, Jeor ST, Ritchie T, Syndulko K, Jeor SC, Fitch and linkage studies in impulsive-addictive-compul- RJ, Brunner RL, Sparkes RS. D 2 dopamine receptor gene sive behavior. Pharmacogenetics 1995; 5: and cigarette smoking, a reward gene? Med Hypothesis 71 Corbisiero JR, Reznikoff M. The relationship between per- 1994; 42: sonality type and style of alcohol use. J Clin Psychology 57 Blum K, Braverman ER, Wood RC, Gill J, Li C, Chen TJH, 1991; 47: Taub M, Montgomery AR, Sheridan PJ, Cull JG. Increased 72 Fals-Stewart W. Personality characteristics of substance prevalence of the Taq I allele of the dopamine receptor abusers: an MCMI-II cluster typology of recreational drug gene (DRD 2 ) in obesity with comorbid substance use dis- users treated in a therapeutic community and its relationship order: preliminary findings. Pharmacogenetics 1996; 34: to length of stay of outcome. J Personality Assessment ; 59: George SR, Cheng R, Nguyen T, Israel Y, O Dowd BF. Polymorphisms 73 Matano RA, Locke KD, Schnartz K. MCMI-II personality of the D4 dopamine receptor alleles in chronic subtypes for male and female alcoholics. J Personality alcoholism. Biochem Biophys Res Commun 1993; 196: Assessment 1994; 63: Kranzler HR, Satel S. Apter A 1 personality disorders and 59 Blum K, Braverman ER, Wood RC, Sheridan PJ. Prolonged associated features in cocaine-dependent inpatients. Com- P300 latency in a neuropsychiatric population with the prehensive Psychiat 1994; 35: D 2 dopamine receptor A 1 allele. Pharmacogenetics 1994; 75 Ohannessian CM, Hesselbrock VM. Temperament and 4: personality typologies in adult offspring of alcoholics. J 60 Noble EP, Berman SM, Ozkaragoz TX. Prolonged P300 latency Studies Alcohol 1995; 56: in children with the D 2 dopamine receptor A 1 allele. 76 Yanovski SZ, Nelson JE, Dubbert BK, Spitzer RL. Associ- Am J Hum Gen 1994; 54: ation of binge eating disorder and psychiatric comorbidity 61 Comings DE, Muhleman D, Ahn G, Gysin R, Flanagan SD. in obese subjects. Am J Psychiat 1993; 150: The dopamine D 2 receptor gene. A genetic risk factor in 77 Pontius AA. Forensic significance of the limbic psychotic substance abuse. Drug Alc Dep 1994; 34: trigger reaction. Bull Am Acad Psychiatry Law 1996; 24: 62 Comings DE, Muhleman D, Gysin R. The dopamine D receptor (DRD 2 ) gene in posttraumatic stress disorder. A 78 Neiswanger K, Kaplan BB, Hill SY. What can the study and replication. Biol Psychiatry 1996; 40: DRD 2 /alcoholism story teach us about association studies 63 Compton PA, Anglin MD, Khalssa-Denison E, Paredes A. in psychiatric genetics? Am J Med Genet The D 2 dopamine receptor gene, addiction and personality: (Neuropsychiatric Genetics) 1995; 60: 272. clinical correlates in cocaine abusers. Biol Psy- 79 Cloninger CR. The genetic structure of personality and chiatry 1996; 39: learning: a phylogenetic model. Clin Genet 1994; 46: 64 Tiihonen J, Kuikka J, Bergstrm K, Hakola P, Karhu J, Ryn nanen O-P, Pohr J. Altered striatal dopamine re-uptake 80 Vanyukov MM, Moss HB, Yu CM, Tarter RE, Ranjan D. site densities in habitually violent and non-violent Preliminary evidence for an association of a dinucleotide alcoholics. Nature Med 1995; 1: repeat polymorphism at the MAOA gene with early onset 65 Umberkomen-Wita B, Vogel WH, Wita PJ. Some biochemi- alcoholism/substance abuse. Am J Med Genet cal and behavioral (sensation seeking) correlates in heal- (Neuropsychiatric Genet) 1995; 60: