Molecular Psychiatry (2000) 5, 548–551  2000 Macmillan Publishers Ltd All rights reserved 1359-4184/00 $15.00 www.nature.com/mp ORIGINAL RESEARCH ARTICLE Attention-deficit hyperactivity disorder and the gene for the dopamine D5 receptor CL Barr1,2, KG Wigg1, Y Feng1, G Zai1, M Malone3, W Roberts3, R Schachar2, R Tannock2 and JL Kennedy4

1Department of Psychiatry, The Toronto Western Hospital; 2Department of Psychiatry, The Hospital for Sick Children and the University of Toronto; 3Division of Neurology, The Hospital for Sick Children; 4Neurogenetics Section, The Centre for Addictions and Mental Health, Clarke Division, University of Toronto, Toronto, Canada

Keywords: dopamine receptor D5; genetics; transmission parent. There were 24 affected siblings genotyped in disequilibrium test; attention-deficit hyperactivity disorder the sample for a total of 116 ADHD children. A recent study has suggested a possible association of We calculated the allele frequencies using the par- a polymorphism near the dopamine D5 receptor gene ental chromosomes in our sample (Table 1). The allele (DRD5) and attention-deficit hyperactivity disorder.1 The frequencies were similar to published reports for Cau- polymorphism studied was a (CA)n repeat located in the casian populations.2,6 We used the transmission dis- cosmid containing the D5 receptor gene2 and the allele equilibrium test (TDT) to determine if biased trans- that was reported to be associated with attention-deficit mission occurred for the alleles at this polymorphism hyperactivity disorder (ADHD) was the 148-bp allele. In (Table 2). Using the allele-wise test (compares the this study we sought to replicate this finding by testing number of transmissions of a particular allele vs any for biased transmission of the alleles at this same poly- morphism in a sample of 92 families with an ADHD pro- other allele compared to the number of times that the band. We did not observe significant evidence for allele is not transmitted), we did not observe significant biased transmission of the 148-bp allele, however we evidence for the biased transmission of the 148-bp did observe biased transmission of two other alleles, allele as previously reported, however, there was a the 136-bp allele and the 146-bp allele. For these two trend for the transmission of this allele to the ADHD alleles the bias was for these two alleles not to be trans- children. The 148-bp allele was transmitted 48 times mitted to the ADHD children. The number of informative compared to 38 times that it was not transmitted (TDT transmissions for these two alleles was small, therefore ␹2 = 1.163, P = 0.281). it would be premature to make any conclusions from our We observed significant evidence for biased trans- study concerning the role of DRD5 in ADHD. Molecular mission of two alleles, the 136-bp allele and the 146- Psychiatry (2000) 5, 548–551. bp allele. For the 136-bp allele there were only 11 informative transmissions, of these the 136-bp allele The dopamine receptor D5, located on chromosome was transmitted twice compared to nine times that it 4p15.1–p15.3,3 is a member of the guanine nucleotide- was not transmitted (allele-wise TDT ␹2 = 3.568, binding (G-protein) receptors family and is close in P = 0.059). For the 146-bp allele there were only 29 structure to the D1 receptor.4,5 Both the D1 and D5 informative transmissions, eight times the 146-bp receptors stimulate adenyl cyclase activity via G-pro- allele was transmitted compared to 21 times that it was tein coupling however the D5 receptors have a 10-fold higher affinity for dopamine than do the D1 recep- tors.4,5 Table 1 Allele frequencies of the (CA)n repeat An association with an allele at the dopamine D5 receptor locus (DRD5) and ADHD has been reported Allele size in bp Allele frequency using a sample of 91 nuclear families (69 trios and 22 parent-proband pairs) using the haplotype relative risk 134 0.022 analysis and the transmission disequilibrium test.1 136 0.017 Preferential transmission was reported for the 148-bp 138 0.088 allele of the (CA) repeat polymorphism located in the 140 0.053 n 142 0.039 region of the gene and within the same cosmid.2 This 144 0.022 allele was reported to be transmitted 90 times com- 146 0.079 pared to 54 times that it was not transmitted (P value 148 0.446 Ͻ0.01). 150 0.119 In this study, we tested this same polymorphism. We 152 0.071 genotyped 92 families, 80 consisting of a proband and 154 0.019 both parental DNAs genotyped and 12 families with a 156 0.019 proband with DNA available and genotyped for a single ADHD and DRD5 CL Barr et al 549 Table 2 TDT results for DRD5

Allele 134 bp 136 bp 138 bp 140 bp 142 bp 144 bp 146 bp 148 bp 150 bp 152 bp 154 bp 156 bp

Transmitted 8 2 15 6 6 4 8 48 17 15 3 3 Not transmitted 2 9 12 9 8 4 21 38 16 11 2 3

Chi-squares 3.600 4.455 0.333 0.600 0.286 5.828 1.163 0.030 0.615 P values 0.058 0.035 0.564 0.439 0.593 0.016 0.281 0.862 0.433

not transmitted (allele-wise TDT ␹2 = 5.828, P = 0.016). missions suggest that we should examine this gene in For both of these alleles there were not many informa- further experiments as a potential susceptibility factor tive transmissions therefore the significance of the in ADHD. If there is indeed an association with ADHD, results should be interpreted with extreme caution as then the most likely explanation is that the alleles at this could easily be the result of chance. The other the (CA)n repeat are in linkage disequilibrium with the important observation to note is that for these two alleles of the functional DNA variant that is contribu- alleles, the bias was in the direction of not being trans- ting to the phenotype. Although there is the possibility mitted to the ADHD child therefore a protective allele that the (CA)n repeat is located in a transcriptional would need to be hypothesized. In the Daly et al1 regulation region, and the alleles directly affect the report the transmission of the other alleles were com- transcription level of the gene, we are currently work- bined as a group so we are not able to determine if ing with the assumption that the functional variant has there was a similar trend for these alleles. yet to be identified. We have recently begun to screen The extended TDT program (ETDT)7 can be used to the DRD5 gene for additional DNA variants located in examine the pattern of preferential transmission of cer- the coding region that may explain the reported associ- tain alleles across genotypes (allele-wise TDT) and also ation starting with previously identified variants. Thus can consider every heterozygous parental genotype far, we have tested for the DNA change A1051G, that separately and examines whether each allele of the results in a change of the highly conserved amino acid genotype is transmitted to affected offspring on 50% of asparagine to aspartic acid at amino acid position occasions (genotype-wise analysis). The ␹2 for the 351.8,9 The asparagine residue at 351 is conserved in allele-wise TDT was 16.659, 11 df, P = 0.119 and the ␹2 all dopamine receptors in all species sequenced thus for the genotype-wise TDT was 53.310, 33 df, P = 0.014. far and in other G-protein coupled receptors.8 Func- This study cannot be considered a clear replication tional studies have shown that the substitution of of the original finding because we did not observe aspartic acid for asparagine results in an approximately biased transmissions of the 148-bp allele. One factor 10-fold decrease in dopamine binding affinity resulting influencing our result may be the power of our sample. in a dopamine affinity equivalent to the D1 receptor.10 Even though there was a similar number of families This amino acid change is rare and is found in less in both studies, there were not as many informative than 1% of a Caucasian population.8 A sample of 28 transmission of the 148-bp allele in our sample. In our ADHD subjects has been previously screened for this sample of 92 families there were only 86 informative DNA variant and this variant was not found.9 We transmissions of the 148-bp allele whereas in the study screened 30 ADHD probands in our sample and did of 91 families by Daly et al1 there were 144 informative not observe this change in any of the probands there- transmissions of this allele. fore this variant is unlikely to be contributing to ADHD. The different ethnic composition of our sample com- We are now directly sequencing this gene in probands pared to the previous study with a predominantly Irish from our sample to identify other DNA variants that composition may also have been a factor in the differ- may be linked to ADHD. ent results of the two studies. The majority (96%) of the families in our sample described themselves as mixed Methods European Caucasian descent. The most common ethnic groups were English, Scottish, Irish, German, French, Diagnostic criteria Italian, Polish, and Dutch. Of the families describing The assessment and diagnostic criteria for the subjects themselves as having an Irish ancestor, only one family in this study have been previously described.11–14 described all four grandparents of the proband as Irish. Briefly, subjects in this study were between 7 and 16 Five families described themselves as being of mixed years of age and met DSM-IV criteria for one of the or non-Caucasian descent. The non-Caucasian ethnic three ADHD subtypes (inattentive, hyperactive-impul- groups consisted of African Canadians and Native Can- sive, combined). Subjects were excluded if they had adians. evidence of neurological or chronic medical illness, Although we were not able to replicate the orginal bipolar affective disorder, psychotic symptoms, Tour- report finding biased transmission of the 148-bp alelle ette syndrome, chronic multiple tics, or had a comor- in our sample, the findings of some biased trans- bid anxiety, depressive or developmental disorder that

Molecular Psychiatry ADHD and DRD5 CL Barr et al 550 could better account for the behaviors (as specified by Statistical analysis DSM-IV). Subjects were also excluded if they scored Statistical analysis was performed using the ETDT pro- below 80 on both the Performance and Verbal Scales gram.7 of the Wechsler Intelligence Scale for Children, 3rd edition.15 Subjects with comorbid oppositional defiant Acknowledgements disorder, reading disabilities, or conduct disorder were not excluded from this study. This work was supported by grants from The Hospital Diagnosis was based on information obtained from a for Sick Children Psychiatric Endowment Fund, The semi-structured interview of parents (Parent Interview National Alliance for Research in Schizophrenia and for Child Symptoms, PICS-IV; Schachar and Ickowicz, Depression (CLB and JLK), National Health Research unpublished) and teacher (Teacher Telephone Inter- Development Program of Health Canada (6606-5612- view-IV, TTI; Tannock and Schachar, unpublished). 401, RS) and the Medical Research Council of Canada Both interviews have established reliability and (MT14336 and PG11121). validity.16 This information was supplemented by evidence about behavior, development and medical References history derived from standardized parent and teacher questionnaires; the Conners Parent and Teacher Rating 1 Daly G, Hawi Z, Fitzgerald M, Gill M. Mapping susceptibility loci Scales – Revised17 and the Ontario Child Health Survey in attention deficit hyperactivity disorder: preferential trans- 18 mission of parental alleles at DAT1, DBH and DRD5 to affected Scales – Revised. Academic achievement in arith- children. Mol Psychiatry 1999; 4: 192–196. metic, reading, and spelling were measured using the 2 Sherrington R, Mankoo B, Attwood J et al. Cloning of the human Wide Range Achievement Test – III.19 Oral and recep- dopamine D5 receptor gene and identification of a highly polymor- tive language skills were tested with the Clinical Evalu- phic microsatellite for the DRD5 locus that shows tight linkage to 20 the chromosome 4p reference marker RAF1P1. Genomics 1993; 18: ation of Language Fundamentals, 3rd edition. Chil- 423–425. dren were assessed using self-report for anxiety 3 Eubanks JH, Altherr M, Wagner-McPherson C, McPherson JD, Was- (Children’s Manifest Anxiety Scale)21 and depression muth JJ, Evans GA. Localization of the D5 dopamine receptor gene (Children’s Depression Inventory).22 Children were free to human chromosome 4p15.1–p15.3, centromeric to the Hunt- of medication 24 h before their assessment. Of the ington’s disease locus. Genomics 1992; 12: 510–516. 4 Grandy DK, Zhang YA, Bouvier C et al. Multiple human D5 dopam- cases that were used in this study, 57% of the children ine receptor genes: a functional receptor and two pseudogenes. were of the combined subtype, 19% were of the Proc Natl Acad Sci USA 1991; 88: 9175–9179. hyperactive/ impulsive subtype, and 24% were of the 5 Sunahara RK, Guan HC, O’Dowd BF et al. Cloning of the gene for primarily inattentive subtype. a human dopamine D5 receptor with higher affinity for dopamine than D1. Nature 1991; 350: 614–619. This protocol was approved by The Hospital for Sick 6 Barr CL, Wigg KG, Zovko E, Sandor P, Tsui LC. Linkage study of Children Research Ethics Board and informed written the dopamine D5 receptor gene and Gilles de la Tourette syndrome. consent was obtained for all participants. Am J Med Genet 1997; 74: 58–61. 7 Sham PC, Curtis D. An extended transmission/disequilibrium test Isolation of DNA and marker typing (TDT) for multi-allele marker loci. Ann Hum Genet 1995; 59: 323–336. DNA was extracted directly from blood lymphocytes 8 Sobell JL, Lind TJ, Sigurdson DC et al. The D5 dopamine receptor 23 using a high salt extraction method. The (CA)n poly- gene in schizophrenia: identification of a nonsense change and morphism located near the DRD5 gene was genotyped multiple missense changes but lack of association with disease. according to previous methods.2 For the detection of Hum Mol Genet 1995; 4: 507–514. 9 Feng J, Sobell JL, Heston LL, Cook EH, Jr., Goldman D, Sommer the A1051G DNA variant, a 651-bp PCR fragment of the SS. Scanning of the dopamine D1 and D5 receptor genes by REF DRD5 gene was amplified with the following primers: in neuropsychiatric patients reveals a novel missense change at a DRD5-G: GCC TTG GTC ATG GTC GGC CTG GCA; highly conserved amino acid. Am J Med Genet 1998; 81: 172–178. DRD5-T: CCA GCA GCT GGG CAA ACA CCT TCT GA 10 Cravchik A, Gejman PV. Functional analysis of the human D5 as previously described.8 Amplification proceeded dopamine receptor missense and nonsense variants: differences in ° dopamine binding affinities. Pharmacogenetics 1999; 9: 199–206. after an initial denaturation step of 4 min at 95 C fol- 11 Barr CL, Wigg K, Malone M et al. Linkage study of catechol-O- lowed by 35 cycles of 94°C for 30 s, 57°C for 30 s and methyltransferase and attention-deficit hyperactivity disorder. Am extension at 72°C for 30 s. A final extension step was J Med Genet 1999; 88: 710–713. added at 72°C for 10 min. Ten ␮l of the PCR product 12 Barr CL, Wigg KG, Wu J et al. Linkage study of two polymorphisms at the dopamine D3 receptor gene and attention-deficit hyperactiv- was digested with 8 units of the HinfI restriction ity disorder. Am J Med Genet 2000; 96: 114–117. enzyme (New England Biolabs, Mississauga, Ontario, 13 Barr CL, Feng Y, Bloom S et al. Identification of DNA variants in Canada) per reaction at 37°C for approximately 2 h. the SNAP-25 gene and linkage study of these polymorphisms and The fragments were separated on 1.3% agarose. Allele attention-deficit hyperactivity disorder. Molecular Psychiatry 2000; number 1 (A, asparagine) is identified as two bands of 5: 405–409. 14 Barr CL, Wigg KG, Bloom S et al. Further evidence from haplotype 480 bp and 171 bp. Although we did not observe the analysis for linkage of the dopamine D4 receptor gene and atten- rare allele, allele number 2 (G, aspartic acid) could be tion-deficit hyperactivity disorder. Am J Med Genet 2000; 96: detected as three bands of 387 bp, 93 bp, and 171 bp. 262–267. The non-polymorphic restriction site in the PCR frag- 15 Wechsler DI. Examiner’s Manual: Wechsler Intelligence Scale for Children, 3rd edn. Psychological Corporation: New York, NY, 1991. ment was used as an internal control for each sample 16 Schachar R, Tannock R, Marriott M, Logan G. Deficient inhibitory as a guarantee that the restriction enzyme did indeed control in attention deficit hyperactivity disorder. J Abnorm Child work and therefore we did not miss the rare allele. Psychol 1995; 23: 411–437.

Molecular Psychiatry ADHD and DRD5 CL Barr et al 551 17 Conners CK. Conners’ Rating Scales – Revised. Multi-Health Sys- 22 Kovacs M. Manual: The Children’s Depression Inventory. Multi- tems: Toronto, Canada, 1997. Health Systems: Toronto, Canada, 1995. 18 Boyle MH, Offord DR, Racine Y, Fleming JE, Szatmari P, Sanford 23 Miller SA, Dykes DD, Polesky HF. A simple salting out procedure M. Evaluation of the revised Ontario Child Health Study scales. for extracting DNA from human nucleated cells. Nucleic Acids Res J Child Psychol Psychiatry 1993; 34: 189–213. 1988; 16: 1215. 19 Wilkinson GS. Wide Range Achievement Test 3 – Revision 3. Jastak Associates: Wilmington, DE, 1993. Correspondence: Dr CL Barr, The Department of Psychiatry, The 20 Semel E, Wing E, Secord W. Clinical Evaluation of Language Fun- Toronto Western Hospital, 399 Bathurst St, MP14-302 Toronto, Onta- damentals – Third Edition (CELF-3). The Psychological Corpor- rio, Canada M5T 1S8. E-mail: CBarrȰuhnres.utoronto.ca ation: San Antonio, TX, 1995. Received 3 May 2000; revised and accepted 29 June 2000 21 Reynolds CR, Richmond BO. What I Think and Feel (RCMAS). Western Psychological Services: Los Angeles, CA, 1985.

Molecular Psychiatry