Tandem Duplication Polymorphism Upstream Of The Dopamine .

American Journal of Medical Genetics (Neuropsychiatric Genetics) 88:705–709 (1999)Tandem Duplication Polymorphism Upstream ofthe Dopamine D4 Receptor Gene (DRD4)M.I. Seaman,1,2 J.B. Fisher,3 F.-M. Chang,2,4 and K.K. Kidd2,5*1Department of Anthropology, Yale University, New Haven, ConnecticutDepartment of Genetics, Yale University School of Medicine, New Haven, Connecticut3Jonathan Edwards College, Yale University, New Haven, Connecticut4Department of Obstetrics and Gynecology, National Cheng Kung University Medical College, Taiwan5Department of Psychiatry, Yale University School of Medicine, New Haven, Connecticut2The dopamine D4 receptor (DRD4) is a member of the D2-like dopamine receptor family.Polymorphisms at the DRD4 gene have beenexamined for association with a wide rangeof neuropsychiatric disorders and normalbehavioral variation. The DRD4 gene is unusual in its high amount of expressed polymorphism in humans. Here we study theidentification of a polymorphic tandem duplication of 120 bp located 1.2 kb upstreamof the initiation codon. The duplicated region contains consensus sequences of binding sites for several known transcriptionfactors, suggesting that different allelesmay differ in their transcriptional activity.Because chimpanzees, gorillas, and orangutans lack the duplication, the duplicated allele is inferred to be derived. The frequencyof this derived duplication allele rangesfrom 0.40–0.81 in the 11 populations fromaround the world typed for this polymorphism. Am. J. Med. Genet. (Neuropsychiatr.Genet.) 88:705–709, 1999. 1999 Wiley-Liss, Inc.KEY WORDS: schizophrenia; neuropsychiatric disorders; gene regulationINTRODUCTIONThe dopamine D4 receptor (DRD4) is a member ofthe D2-like dopamine receptor family. The D4 receptorContract grant sponsor: USPHS; Contract grant number:AA09379-05; Contract grant number: MH30929; Contract grantnumber: MH39239; Contract grant number: GM57672; Contractgrant sponsor: National Science Foundation; Contract grantnumber: SBR9632509.*Correspondence to: Kenneth K. Kidd, Department of Genetics,Yale University School of Medicine, 333 Cedar Street, New Haven, CT 06520-8005. E-mail: kidd@biomed.med.yale.eduReceived 26 October 1998; Accepted 13 April 1999 1999 Wiley-Liss, Inc.is particularly relevant to the study of psychosis because clozapine, an atypical neuroleptic used to treatschizophrenics, has a much higher affinity for D4 receptors than other dopamine receptors [Van Tol et al.,1991]. In addition, the DRD4 gene displays an unusualamount of expressed polymorphism with a VNTR inexon 3, a polymorphic duplication of 12 bp in exon 1, apolymorphic deletion of 13 bp in exon 1, and singlenucleotide polymorphisms resulting in amino acid substitutions in exon 1 and exon 3 [Van Tol et al., 1992;Lichter et al., 1993; Catalano et al., 1993; Nöthen et al.,1994; Cichon et al., 1995; Seeman et al., 1994; Changand Kidd, 1996].Polymorphisms at the dopamine D4 receptor (DRD4)gene have received considerable attention because oftheir possible role in mental illness [Catalano et al.,1993; Grice et al., 1996; LaHoste et al., 1996; Cruz etal., 1997] and substance abuse [Muramatsu et al.,1996; Kotler et al., 1997], as well as in normal variationin human behavior [Ebstein et al., 1996; Benjamin etal., 1996]. Identification of new, and possibly functional, variants in the 5 upstream region is particularly intriguing because the up-regulation of D4 receptors may play a role in schizophrenia rather than anychanges in the protein itself as the levels of dopamineD4 receptors in the brains of schizophrenics may beincreased compared with nonschizophrenics [Seemanet al., 1993, 1997].Two polymorphisms in the flanking DNA upstreamof DRD4 have been reported as restriction fragmentlength polymorphisms (RFLPs) with SmaI [Petronis etal., 1994] and PstI [Paterson et al., 1996]. But only theSmaI site has been identified in sequence at position 11 relative to the initiation codon, and it does not alterany known regulatory sequences. Here, we study apolymorphic tandem repeat that is likely responsiblefor the PstI RFLP reported by Paterson et al. [1996].The duplicated sequence contains the consensus binding sequences for several known transcription factors,and the two alleles might therefore differ in transcription of DRD4. Using a rapid polymerase chain reaction(PCR) typing assay, we typed 11 human populationsfrom different parts of the world. Three species of apeswere also typed and sequenced as the most reliable

706Seaman et al.method to determine which human allele is ancestraland which is derived [Iyengar et al., 1998].MATERIALS AND METHODSSamplesDNA was extracted from lymphoblastoid cell lines of623 individuals from 11 populations from Africa(Biaka, Mbuti), Europe (Druze, Danes, unrelated Europeans), Asia (Chinese, Japanese, Yakut), Melanesia(Nasioi), and the Americas (Mayans, Rondonian Surui). Details describing these populations and the collection of samples have been discussed elsewhere [Castiglione et al., 1995; Calafell et al., 1998; see s.html〉].DNA from a common chimpanzee (Pan troglodytes),bonobo (P. paniscus), gorilla (Gorilla gorilla), andorangutan (Pongo pygmaeus) was extracted from lymphoblastoid cell lines established from blood generously provided by the Yerkes Regional Primate Center(Atlanta, GA) and by the Milwaukee County Zoo (Milwaukee, WI).PCR-Based Typing100 ng of genomic DNA is amplified in a total volumeof 25 l with the following primers at 0.2 M:D4upstrFor2 (5 -GTTGTCTGTCTTTTCTCATTGTTTCCATTG-3 ) and D4upstrRev3 (5 -GAAGGAGCAGGCACCGTGAGC-3 ) with 0.2 mM dNTPs,5%DMSO, 0.5 U Taq polymerase in standard PCRbuffer containing 1.5 mM MgCl2. PCR cycling conditions are 5 min at 95 C followed by 30 cycles of 30 s at94 C, 30 s at 58 C, and 60 s at 72 C with a final extension for 10 min at 72 C. Ten microliters of the PCRproduct is electrophoresed through a 1.5%agarose geland stained with ethidium bromide. The PCR reactionyields distinct bands at 429 bp (“S”, short allele) and549 bp (“L”, long allele). A faint diffuse band migratingmore slowly than the 549 bp band is also seen in someheterozygotes and is presumed to be composed of heteroduplex molecules.DNA SequencingThe original human sequence was derived from anEMBL3 clone, generously provided by Dr. Van Tol, using a primer-walking strategy. Genomic DNA of humans determined to be homozygous for each allele using the above typing procedure, as well as DNA of achimpanzee, gorilla, and orangutan, was sequenced byPCR-amplifying an approximately 1-kb fragment surrounding the tandem duplication using additionalprimers and sequenced using the same D4upstrFor2and D4upstrRev3 primers used for the typing, as wellas additional primers. All sequencing was performedon an ABI 373 automated sequencer using the ABIPrism Cycle Sequencing kit (Perkin Elmer Applied Biosystems, Foster City, CA).Database SearchingThe 120 bp of sequence, which is tandemly duplicated in the long allele, was searched for known transcription factor binding sites using the TESS webbased search tool to search the TransFac database[Schug and Overton, 1997; Heinemeyer et al., ] aswell as potentially functional sites by comparing thesequence with the eukaryotic promoter database (epd)using BLAST [http://www.ncbi.nlm.nih.gov/].RESULTSWhile examining the initial phage sequence, we noticed what appeared to be a 120-bp perfect tandem duplication 1.24–1.48 kb upstream of the initiation codon(Fig. 1). PCR amplification of genomic DNA from several humans with primers flanking this region yieldedpolymorphic fragments differing by approximately 120bp. Direct sequencing of PCR products of each apparent homozygote confirmed that the different size PCRfragments were caused by the presence or absence ofthe 120-bp tandem duplication.Figure 2 clearly shows the two different size fragments obtained from the PCR. Chimpanzees, gorillas,and orangutans h alters the thirdintercytoplasmic loop, other polymorphisms deserveconsideration. As one study has shown the number ofD4 receptors to be dramatically increased in the brainsof schizophrenics [Seeman et al., 1993], the possibilitythat variation in the upstream region may be involvedin the etiology of this disease is intriguing becausevariation in the regulation of the gene may be moreimportant than any structural changes in the proteinitself. What role, if any, variation at DRD4 may play inbehavioral variation, normal and abnormal, remains atantalizing but controversial area of research.As shown in Figure 1, there are now three lengthpolymorphisms in coding regions of DRD4, and this120-bp polymorphic tandem duplication is a fourthlength polymorphism around the gene. The polymorphism reported here is approximately 1.2 kb relativeto the initiation codon. Interestingly, this duplicationdoes not appear to be present in the recently reportedsequence generated independently and from a differentclone [Kamakura et al., 1997; GenBank accessionU95122]. However, there are other single nucleotidedifferences between that sequence and our sequence ofVan Tol’s clone that may represent single nucleotidepolymorphisms. Kamakura et al. [1997] reported thattranscription begins between 501 and 436 with anegative modulator between 770 and 679 and thepromoter between 591 and 123, suggesting that theregion in which the duplication occurs may not be directly involved in the regulation of transcription. However, further studies would be necessary to confirmthis, including expression assays using constructs containing each allele because additional regulatory sequences could exist further upstream. Even if the region harboring the duplication is not involved in transcription regulation, this new polymorphism could be auseful marker for further association studies as it maybe in disequilibrium with functional variants in the 5 flanking region of the gene. This polymorphism is easily and rapidly typed with the PCR-based assay described above.Though less than at many other loci, the allele frequencies for this polymorphism do vary sufficientlyamong populations that any association studies usingthis marker (or any marker) need to be adequately controlled for ethnicity. Absence of the duplication in theother hominoids shows that the nonduplication state isancestral. The derived allele is more common in most ofthe populations studied. This polymorphism probablyarose prior to the spread of modern humans out of Africa, because all populations have both alleles. The absence of any evidence (based on FST) for natural selection affecting the frequency distributions is not astrong argument against functional relevance. Many ofthe differences in human behavior hypothesized to be