CA2136087C - Sequence of human dopamine transporter cdna - Google Patents

Abstract

The cloning and characterization of a human dopamine transporter (HUDAT) cDNA
is described. RFLP analysis is used to determine the distribution of HUDAT alleles in two ethnic backgrounds. The means by which the association between HUDAT alleles and behavioral disorders which have altered HUDAT expression as a basis for their etiology is discussed. Methods for evaluating the expression of HUDAT are described.

Description

SEQUENCE OF HUMAN DOPAMINE TRANSPORTER cDNA
BACKGROUND OF THE, INVENTION
Field of the Invention The invention relates to a cloned cDNA which encodes the human dopamine transporter protein. The cloned cDNA provides a means of expressing human dopamine transorter protein in a variety of contexts and also provides a means of diagnosing and treating diseases presenting abnormal expression of dopamine transporter protein.
The dopamine transporter that acts to take released dopamine back up into presynaptic terminals WO 93/24628 PCT/US93/05179 ~~:..,:

has been implicated in several human disorders. Cocaine binds to the dopamine transporter and blocks dopamine reuptake in a fashion that correlates well with cocaine reward and reinforcement (M. C. Ritz et al., Science 237, 1219 (1987)). Neurotoxins that cause Parkinsonian syndromes are concentrated in dopaminergic neurons by this transporter (S.H. Snyder and R.J. D~Amato, Neurology 36, 250 (1986): G. Uhl, Eur. J. Neurol. 30, 21 (1990)). Binding to the dopamine transporter is l0 altered in brains of patients with Tourette's syndrome (H. S: Singer et al., Ann. Neurol. 30, 558 '(1991)).
These clinical links enhance interest in the structure and function of the human dopamine transporter (HUDAT).
Vulnerability to .these disorders may have genetic 1S components (E. J. Devor and C.R. Cloninger, Annu. Rev.
Genet. 23,.19 (i989): D: Pawls and J. Leckman, New Eng.
J. Med: 315, 993 (1986); R. Dickens at al., Arch. Gen.
Psychiatry ~8, 19 (1991)): thus identification of linkage markers for the human DAT is also of interest.
2'o Dopamine transporters act to terminate dopaminergic neurotransmission by sodium- and chloride-dependent reaccumulation of dopamine into pre-synaptic neurons (L.L. Iversen, in Handbook of Psychopharmacology, L.L. Iversen, S.J. Iversen, & S.H.
2-5 Snyder, Eds. (Plenum, N~w York, 1976), pp. 381-442:
M.J. Kuhar 'and M.A. Zarbin, J. Neurochem. 31, 251 (1978): A.S. Horn, Prog. Neurobiol. 34, 387 (1990)).
Cocaine and related drugs bind to these transpartera ~iD ~ fash~icn that , correlates well with 30 their behavioral reinforcing and psychomotor stimulant properties; these transporters are thus the principal brain "cocaine receptor8" related to drug abuse (M. C.
Ritz, R.J. Lamb, S.R. Goldberg, M.J. Kuhar, Science 23,1219 ('1987); J. 8ergman, B. K. Madras, S. E.

WO 93/24628 ,~ PCT/US93/05179 > ~3~~a l .

Johnson, R. O. Spealman, J. Pharmacol. Exp. Ther. 251, 150 (1989)). The transporters accumulate neurotoxins with structural features resembling dopamine their ability to concentrate the parkinsonism-inducing toxin MPP~ (1-methyl-4-phenylpyridinium) is key to this agent's selective dopaminergic neurotoxicity (S. H.
Snyder, and R. J. D'Amato, Neurology 36(2), 250 (1986):
S. B. Ross, Trend. Pharmacol. Sci. 8, 227 (1987))~
Studies of the dopamine transporter protein suggest that it is an 80 kDa glycoprotein, but have not yet yielded protein sequence data (D. E. Grigoriadis, A.A.
Wilson, R. Lew, J.S. Sharkey & M.J. Kuhar, J. Neurosci.
9, 2664 (1989)). Binding of cocaine analogs such as [3H]CFT to membranes prepared from dopamine-rich brain regions reveals two sites with differing affinities (F.
Javory-Agid, and S:Z. Langer, Naunyn-Schmiedeberg's Arch: Pharmacol. 329, 227 (1985)? J.W. Boja, and M.J.
Kuhar, Eur. J. Pharmacol. 173, 215 (1989): B.K. Madras et al., Mol. Pharmacol. 36, 518 (1989); M.J. Kuhar et al. , Eur. J. Neurol. 30 (1) , 15 (1990) : M.C. Ritz, E.J.
Cone, M.J. Kuhar, Life Sci. 46, 635 (1990).; D.O.
Calligaro, and M.E. Eldefrawi, J. Pharmacol. Exp. Ther.
243, 61 (1987): B:K. Madras et al., J. Pharmacol. Exp.
Ther: 251(1), 131 (1989) M.C. Ritz et al., J.
Neurochem: 55, 1556. (1990)).
Recent elucidation of cDNAs encoding dopamine transporters from experimental animals (B. taros et al. , FEBS Lett. 295, 149 (1992): J.E. Kilty et al., Science 254:, 578 (19~~,) : ; S.. ;~Sk~;imada et ,al. , Science 254, 576 (1991)': T.B. Usdin et al., Proc. Natl. Acad. Sci. USA
88, 11168 (1991) provides hybridization pxobes useful for isolation of their human cognate.

WO 93/2462$ PCT/US93/05179 ~-,.-:
~. ~ b ~ '~.''.i _Summarv of the Invention Described herein is a cDNA (pcfiUDAT), which encodes the human dopamine transporter protein (HUDAT).
Also described are unique features of the nucleotide sequence of the pcHtIDAT predicted for its encoded mRNA
and protein, restriction fragment length polymorphisms (,RFLPs) and Variable Number Tandem Repeats (VNTRs) identified by this cDNA and estimates of race-specific population frequencies of these RFLPs and VNTRs.
By virtue of its representation of the human dopamine transporter sequence, the pcFiUDAT is advantageous over those clones isolated from other species in that better results in applications having a human context would be expected.
The cDNA encoding the human dopamine transporter protein (FiUDAT) provides a means for diagnosing and treating disorders that arise by expression, of abnormal amounts . of or dysfunctional dopamine transporter molecules in a human being.
It is one object of the invention to produce a cDNA that encodes the human dopamine transporter protein, a product of dopaminergic neurons that binds dopamine, cocaine and cocaine analogs and will transport dopamine and MPP+ into mammalian cells expressing it on their surface. It is a further object of the invention to utilize the cDNA to produce cell lines that express human DAT on their surface and to provide a method for the screening of compounds that influence the,,binding,az~d/or transport of dopamine or, cocaine or functional analogs thereof to (into) the cells. Such cell lines may also find therapeutic application for treatment of diseases caused by depletion of cell populations which normally provide for uptake of dopamine.

VI~O 93/24628 . PCT/US93lOS179 %.~ r~~l~~'l A third object of the invention is to provide diagnostic means for assessing HUDAT expression in patients by DNA- or antibody- based tests anc'~ for assessing the onset or progression of disease by assay 5 of HUDAT degradation.
These and other objects are accomplished by providing a cDNA encoding the dopamine transporter protein and a gurified polypeptide conferring upon cells the phenotype of dopamine uptake from the l0 surrounding extracellular medium. Further, the invention is embodied in cell lines, created by stable transformation of cells by a vector encoding the dopamine transporter protein, expressing the dopamine transporter protein on their surface. Another aspect of the invention relates to a method of using such lines to screen pharmaceutical compositions for their ability to inhibit the binding of dopamine, cocaine or analogs of these compounds to the transporter protein.
Such a screening can also be accomplished by use of cells transiently expressing dopamine transporter cDNA.
The invention also relates to diagnostic applications of the dopamine transporter cDNA and anti-human DAT
antibodies and to therapeutic applications of the HUDAT
cDNA.
BRIEF DESCRIPTION OF THE DRAWINGS
Figure 1 (SEQ. ID. NO. 1) shows the nucleotide sequence of the pcHUDAT cDNA encoding the human dopamine tra,nsporter~ , ,pxotein: the sequence is a composite derived from the sequence of clones pHCDAT2, pHCDAT3 and pHCDAT7.
Figure 2 shows the sequences of the repeat elements in the 3~ untranslated portion of the pcHUDAT
cDNA. Also shown is the consensus sequence of the WO 93/24628 PCT/US93/45179 ;~., ...

repeats.
Figure 3 shows a comparison of the amino acid sequence of the human dopamine transporter (Hdat) protein with the amino acid sequence of the rat DAT
(Datl) and also with the sequences of the human norepinephrine transporter (Hnat) and of the human gamma-amino-butyric acid transporter (Hgabat).
Figure 4A shows a representative RFLP analysis of human genomic DNA from nine unrelated individuals digested with TaqI and hybridized with the insert portion of the pHCDAT7 plasmid. Figure 48 shows the same DNA, but hybridized with the Taq492 probe, which corresponds to nucleotides 301-793 of the pcHUDAT
sequence.
~15 Due' AILSD. INSCRIPTION OF THE INVENTION_ For many of the applications described in the examples below subfragments or variants of the HUDAT
protein disclosed in the present application wherein ~e .original amino acid sequence is modified or changed v 20 by~ insertion, addition, substitution, inversion or deletion of one or more amino acids are useful so far as they retain the essential binding or transport specificity for dopamine, cocaine, or functional analogs thereof. Thus, such variants of the HLTDAT are 25 considered to fall within the scope of the present invention: Such variants are easily produced by mutagenic techniques well developed in the art of genetic engin~egri~g.., ,; , , : ; , , .
Expression of heterologous proteins in ~. coli is 30 often utilized as a means of obtaining large quantities of a polypeptide: The product is an unglycosylated protein, which may be made as insoluble "inclusion bodies" in the bacterial cells. Alternatively, some WO 93/24b28 PCT/US93/OSi79 proteins can be secreted into the perplasmic space by fusion to a leader sequence that directs the secretion of the translation product. Other useful fusion sequences are those which allow affinity purification of the product, such as the pGE~ system (Pharmacia), which allows purification by use of a glutathione-Sepharose column.
The promoter to be employed is dependent upon the particular protein to be expressed. Some proteins are not detrimental to the physiology of the bacteria and may be expressed using a high-level constitutive promoter. Others are somewhat toxic and so are best expressed from an ~inducible promoter which keeps synthesis of the heterologous protein repressed until growth of the culture is complete. The promoter is then switched on and the heterologous protein is produced at a high level.
Other considerations in bacterial expression include the use of terminator seqences in the transcription unit and,the use of sequences in the 5' untranslated portion of the mRNA to abolish secondary structure which might impede translation. Also the choice of bacterial strain can be important. Some heterologous proteins 'are susceptible to proteolytic degradation and so are best expressed in strains of bacteria which lack proteolytic functions. Also, strains of bacteria other than E. coli are often useful as hosts for expression systems. The best-developed alternative currently being Ba_c~,illus strains.
Expression of proteins in bacteria is well-reviewed in "Current Protocols in Molecular Biology", which is published with quarterly updates by Wiley Interscience.

8 . PCT/US93/05179 s,,:~.:
s Expression of "foreign" proteins in mammalian cells can be accomplished in two general fashions.
Transient expression refers to the creation of a.~ool of transfected cells which harbor plasmids that are not stably maintained in the cell and so are gradually diluted out of the population. Transient expression is by nature a short term method. For repraducible expression of a heterologous protein, stable expression systems are preferable.
The current state of this art includes a variety of vector systems: both integrative and autonomous vectors are available. Inducible expression of heterologous proteins in mammalian cells is difficult to achieve at the current time. Some systems have been described, but they are not yet in general use. More commonly used are vectors bearing moderate to high ~.level constitutive gromoters. Plasmid vectors are relatively easy to use. Retroviral vectors, which rely upon packaging into infective viral particles and integration into the host cell chromosome are more difficult to use, due to the extra steps involved in creating the recombinant viruses and cell lines which secrete them, but have the advantage that they Pffectively introduce exogenous DNA into human cell lines. Vaccinia virus vector systems are also in widespread use. Other viral vectors are under development for gene therapy systems, including adenovirus-derived vectors.
Tk~e pre~~~red :embodiments ,of the ~,nvention are described by means of the following examples. These examples are intended to be illustrative, rather than limiting in scope. It is understood that variations in the materials and techniques described below will be apparent to those skilled in art and such are to be considered to fall within the scope and spirit of the instant application.
Example 1 Isolation and sgguencincr of cDNA encoding human dopymine trar~orter To isolate human cDNAs for the dopamine transporter, cDNA libraries prepared from "substantia nigra" and "brainstem" dissections containing cells known to express the transporter were screened with l0 hybridization probes prepared from the rat cDNA, pDATl (S. Shimada et al., Science 254, 576 (1991)).
Sequences from the 3' untranslated region of the rat cDNA were not used because of the presence of CA
dinucleotide repeats. Human brain stem and substantia nigra cDNA libraries (Stratagene, Ia Jolla, CA) were plated and blotted onto duplicate replica nitrocellulose (Schleicher and Schuell, Keene, NH) filters, which were incubated for 1 hour at 37°C with proteinase K (50 pg/~1 in 2 x SSPE/0.1%SDS) to reduce filter background, washed in 5 x SSC/0.5% SDS/imM EDTA, prehybridized and hybridized at 42°C, and washed at 54°C
in 0.4 x SSC/0.5% SDS. The hybridization probe was a 2300 by Eco RI fragment of the rat dopamine transporter cDNA6 (S. Shimada et al. , Science 254, 576 (1991) ) [~P]
labeled by random priming (Prime It*kit, Boehringer Mannheim)., and hybridized at approximately 106 cpm/ml.
Positively-hybridizing cDNA clones were purified from the brainstem library, autoexcized according to protocols provided by the manufacturer (Stratagene), and termed pHCDAT2, pHCDAT3, and pHCDAT7. Sequencing was performed on an Applied Biosystems automated sequences as described (S. Shimada et al. , Science 254, 576 (1991) ) . Sequence analysis was performed using the *Trade-mark GCG software package (J. Devereaux, et al.., Nucleic Acids Res. i2, 387 (1984)).
Screening of more than 2 x 106 plaques from the substantia nigra library produced no positives.
5 Screening 1 x 106 plagues from the brainstem library yielded 11 positively-hybridizing plaques, three of which were identified as human DAT clones by sequence analysis. These clones were identified as representing the 5'-half (pHCDAT2, bases 1-1733), the 3'-half 10 (pHCDAT3, bases 1679-3919), and an internal portion (pHCDAT7, bases 653-1434) of the human DAT cDNA whose reconstructed full-length sequence is shown in Figure 1. The structure of this cDNA resembles the structure of the rat cDNA DAT1, with a modest 5' untranslated region and a long 3' untranslated region. Both 5' and 3' untranslated regions are longer than those of the rat cDNA pDATl, however, making the length of the predicted human mRNA greater than the 3.7 kb observed for the rat mRNA (S. Shimada et al., Science 254, 576 (1991)). A striking difference between rat and human cDNAs is found in the 3' untranslated region where the human cDNA displays 10 copies of a 40 by repetitive element that are arrayed in head-to-tail fashion and are absent from the rat cDNA (Figs 1,2). These elements are highly stereotyped. The sequence of each element is more than 90% identical to the consensus sequence listed at the bottom of Figure 2, although the seventh repeat displays a 5 base pair insertion from its 24th to 28th nucleotides. The consensus element found here is 68% G+C. No exact match is found in searches of the EMBL/genbank* data base, release 70.
However, sequences conferring up to 70% nucleic acid identity over up to 37 of these bases are found in * Trade-mark ~~ r~~~~~~~

viral sequences, especially with herpesvirus sequences (e. g. locus HS1US).
The open reading frame predicted by the HUDAT cDNA
encodes 620 amino acids, identical in size to the rat DAT1 cDNA except for an additional amino acid (199) not found in the rat sequence (Fig. 3). This open reading frame predicts amino acid sequences that are 94%
identical to those encoded by the rat dopamine transporter cDNA (S. Shimada et al., Science 254, 576 (1991)). This high degree of conservation, and the weaker identities with the human norepinephrine and GABA transporter cDNA (H. Nelson et al., FEBS Lett.
269, 181-184 (1990): T. Pacholczyk et al., Nature 350, 350-354 (1991)) (Fig. 3), identifies this as the human homolog of the rat DAT1.
The amino acid sequence predicted by the HUDAT
cDNA reveals interesting differences from the rat cDNA.
It lacks one of the 4 consensus sites for N-linked glycosylation noted in the rat DAT1 cDNA (Fig 3, +
symbol): Three adjacent amino acids distinguish the human from the rat proteins at this locus: no other portion of the molecule differs by this extent.
Human DAT amino acids predicted to lie within hydrophobic, putative transmembrane domains show 97%
amino acid identity between the rat and human transporter cDNAs. This conservation is higher than the 87% conservation in regions thought not to span the membrane, and is consistent with the high conservation in these xegiQns; among different sodium dependent, transporter family members. The most striking difference between the rat and human transporters occurs in the putative second extracellular domain, at which each of the transporters cloned to date displays consensus sites for N-linked glycosylation. The V!,'O 93/24628 PCT/U593/05179 ..

glycosylation of the rat dopamine transporter has been defined in biochemical studies that suggest 20 to 30 kd of the molecular weight of the mature protein "may consist of sugar (R. Lew et al . , Brain Research 539, 239 (1991). Four potential N-linked glycosylation sites indicated in the rat transporter contain classic asparagine - X - serine/threonine sequences. Three of these sites are conserved among the rat and human sequences, but a middle glycosylation site, potentially the most distant from the embedding membrane, is absent in the human transporter. The amino acids surrounding this site provide the largest area of amino acid sequence divergence between the rat and human transporters. If glycosylation is evenly°distributed among the different potential sites far N-linked glycosylation, these observations would predict that the human dopamine transporter might display less glycosylation than the rat, and that its molecular weight might be correspondingly smaller. The function of the glycosylation has not been identified to date;
changes in ligand recognition, membrane targeting of the molecule, or even in cell/ cell recognition might conceivably result from these differences in glycosylatibn.
The repeated motifs in the 3' untranslated regions of these cDNAs are another interesting difference from the rat sequences Smaller polymorphic repeated elements have gained recent attention due to their 'implication in>;the fragile X syndrome and myotonic dystrophy (J. D. Brook, et al., Cell 68, 799-808 (1992) Y. Fu et al., Cell 67, 1047-1058 (1991) ; V.A.
McKusick, MendeTian Inheritance is Man, 9th edn. , Johns Hopkins University Press, Baltimore, 1990, 2028 p.).
The rat sequence does demonstrate 25 copies of a small ,, ;.

dinucleotide CA repeat from bases 2476 to 2525 of the 3' untranslated region of its mRNA: CA repeats are absent from the human cDNA (S. Shimada et al., Science 254, 576-578 (1991)). The sequence of the longer hDAT
repeated element is not found the rat cDNA, nor in searches of other sequences found in databanks. The significance of the partial matches in viral genomes is unclear. These repeated elements might alter mRNA
properties, perhaps including secondary structure and/or half-life, in ways that could contribute to the regulation of this genes expression. Search of this sequence using the stemloop program yields more than 150 possible loops with as many as 18 stabilizing hydrogen bonds. Conceivably, population variants in the number of these repeats could also contribute to heterogeneity in DAT function.
~xam~le 2 :Restriction Fragment Length Polvmorohism IRFLPI
analysis DNA was obtained from leukocytes, digested with TaQI,; and analyzed by Southern blotting using pHCDAT7 as the initial hybridization probe: Simpler patterns were also obtained using two other hybridization probes. Taq 120 corresponds to bases 668 to 787 of the HDAT (see below), and was generated by hybridizing 65 and 72 base oligonucleotides of opposite sense and extending the product using large fragment of DNA
polymerase~ I .~~nd ; [32P~]~dCTP or by random priming o~
these two hybridized oligonucleotides, as described (A.
Feinberg and B. Vogelstein, Analyt. Biochem. 132, 6-9 (1982): 5. Shimada et al., Science 254, 576-578 (191)). Identical results were also obtained using a random-priming labeled 492 base pair cDNA fragment (Taq ,m.v.:..; _ . . ' ; .. . <, , ~.;:. , :; , ..
WO 93124628 PCl'/US93/05179 ; .: , 492) corresponding to bases 301 to 793 of this sequence. Probes were hybridized to filters containing DNA from unrelated individuals at 42°C in hybridization solution containing 50% formamide as described.
Identical results were obtained with final washes at 68°C in 0.2 x SSC/0.2% SDS or at 54°C in 0.4 x SSC/0.5%
SDS. Patterns from these Southern blots were analyzed by two independent observers.
Digestion of DNA from 20 unrelated individuals with nine different restriction endonucleases revealed Southern blot patterns in each case that were consistent with the presence of a single gene. There were no clear interindividual differences in Southern blot restriction patterns using radiolabeZed pHCDAT 7 after digestion with Alu I, Bam HI, Eco RI, Hae III, Hind III, I~sp I, and Rsa I. Three other enzymes, Pst z, Hinf I and Tag I revealed polymorphisms. We focused on the polymorphisms identified by Taa I. When probed with radiolabeled pHCDAT 7, more than six bands were obtained from Taa I restricted DNA, many of which showed polymorphic patterns (Fig 4A). Hybridization survived washes of up to 68°C, consistent with specificity. A simpler pattern was rEVealed when hybridization was performed . with the Taq 120 hybridization probe or with the cDNA hybridization probe Taq 492 (Fig. 4B). Two hybridizing bands of 7 and 5.6 kilobases were observed and termed A1 and A2. Tag I A1 and A2 RFLP frequencies are presented the Table.
Of 272:"chromosomes fxom: 136 individuals,examined 36%
showed the A1 form, 64% showed the A2 form. There was a significant racial dimorphism in these distributions such that 26% of Caucasians, but 42% of blacks displayed the A1 RFLP (xz=7.45, p< 0.01).

.~ ar': ~ ~ ., . .,., ,.. , .~..~. ~...... ...... ,.... .. . . ..;. _.... . ' ~.: ~ . .",.,.. ~ ..,~. ....,; .. , ...,;.-:

t.. .
The rich patterns of Taq I RFLPs identified with this cDNA sequence could relate to the fact that the clone itself contains three sites for Taq I cleavage.
Further studies are thus likely to detect other 5 polymorghisms, because extreme variability of bands in the initial Tact I restriction digestions has already been documented.
The tandem repeat in the 3'region of this gene also provides a Variable Number Tandem Repeat (VNTR).
10 The means for examining the distribution of alleles of the VNTR is set forth at the end of Example 3 below.
The hybridization probes that we have described provide useful markers for linkage analysis that would help to exclude the regions around the dopamine 15 transporter gene from involvement in familial disorders. Human dopamine systems are involved in a number of human disorders, with specific implication of involvement of transporter mechanisms in psycl~ostimulant abuse, Parkinsonism, and Tourette°s syndrome (E. J. Devor and C.R. Cloninger, Annu. Rev.
Genet. 23 , 19-36 (1989): D. Pawls and J. Leckman, New Eng. J. Med. 315, 993-997 (1986) : R. Dickens et al. , Arch. Gen. Psychiatry ~8, 19-28 (1991): M.C. Ritz et al., Science 23?, 1219-122 3 (1987); S. Shimada et al., Science 254, 576-578 (1991): H.S. Singer et al., Ann.
Neurol. 30, 558-562 (1991): S.H. Snyder S.H. and R.J.
D'Amato, Neurology 36., 250-258 (1986): G. Uhl, Eur. J.
Neurol. 30, 21-30 (1990)). The human dopamine transporter cD~l.As;and.RFLP information described here should provide useful tools to study its possible role in these and other human disorders.

WO 93/2462f3 PC1'/US93/OS179 Example 3 Predictive) A genetic component of substance abuse behavior identified by RFLP analysis of tie human DAT ger~e~
Abuse of substances, including drugs and alcohol, is currently viewed as arising from a combination of biological, psychological, and social factors (J. S.
Searles, J Abnorm Psychol. 97,153-167 (1988); E.J.
Devor and C.R. Cloninger, Annu Rev Genet. 23, 19-36 (1989); K.R. Merikangas, Psychological Medicine 2a, 11-22 (1990)). Genetic contributions to susceptibility to alcoholism are supported by family, twin, and adoption studies. (D. S. Goodwin, Arch Gen Psychiatry 36, 57-61 (1979); C.R. Cloninger et al., Arch Gen Psychiatry 38, 861-868 (1981); C.R. Cloninger, Science 236, 410-416 (1987)). A genetic component of vulnerability to drug abuse has also been suggested in both twin and adoption studies (R. J. Cadoret et al., Arch Gen Psychiatry ~3, 1131-1136 (1987); R.W: Pickens et al:, Arch Gen Psychiatry 48, 19-28 (1991)).
2o A number of substances which share the potential for abuse by humans also share the ability to enhance dopamine activity in mesolimbic/mesocortical circuits thought to be important for behavioral reward and reinforcement (A. S. Lippa et al., Pharmacol Biochem Behav. i, 23-28 (1973); G. Di Chiara and A. Imperato, Proc Natl Acad Sci USA 85, 5274-5278 (1988); R.A. Wise and P.P. Rompre, Annu Rev Psychol. 40, 191-225 (1989)).
Cocaine's ability to inhibit re-uptake of dopamine, for example, points ,strqngly towafid a possible direct action for this highly-reinforcing drug in these dopaminergic circuits (M.C. Ritz et al., Science 23~, 1219-1223 (1987); D.E. Grigoriadis et al., J Neurosci.
9, 2664-2670 (1989)).

Blum, Noble and co-workers first reported that the "Ai" T~gI restriction fragment length polymorphism (RFLP) of the human dopamine D2 receptor gene (DRDZ, D. K. Grandy et al . , Am J Hum Genet. 45, 778-785 ( 1989 ) ) was associated with alcoholism (K. Blum et al., JAMA
263, 2055-2060 (1990)): 69% of alcoholics displayed this RFLP compared to 20% of non-alcoholics. 42% of 504 Caucasian alcoholic individuals reported in literature to date display this RFLP, while only 27% of l0 461 Caucasian "control" individuals are A1 positive (G.R. Uhl et al:, Arch Gen Psychiatry 49, 157-160 (1992): E. Turner et al., Biol Psychiatry 31, 285-290 (1991): These data come from eight previous studies, five of which find significant associat~.ons between RFLP and alcoholism, and provide evidence for a ignificant association between gene markers and behavior.
Examination of gene marker/behavior associations in drug abusers raises several methodological concerns.
Relatively few individuals who abuse drugs abstain from alcohol, and many individuals who use drugs often self--administer multiple substances (D.R. Wesson et al., eds. Polvdrug Abuse~ The Results of a~National~
Collaborative Study. New York, NY: Academic Press, 25'Tnc: 19?8): Drug-using populations may also differ from -one another and from the general population in racial, ethnic and' other features that might be associated with altered distributions of the alleles for different:;, genes ~ (~I,.;~t.: Gillmore et al . , Am J Drug Alcohol Abuse 16, 185-206 (1990)). Also, some clinical assessments may not focus on the heritable features of the disorder (R. W. Pickens et al., Arch Gen Psychiatry 48, 19-28 (199I)):

WO 93/24628 ~ ~ ' " r PCT/US93/05179 t~ ;' ~'.:' A study of D2 dopamine receptor gene markers in polysubstance users and control subjects provides a useful model for investigating the association between alleles of the DAT gene and substance abuse behaviors or other behavioral disorders such as Tourette's syndrome. We have investigated the 3' Tacrl A1 RFLP
examined in previous studies of alcoholics, and a more 5' TaaI RFLP ("B") located closer to regulatory and structural/coding regions of the gene (X.Y. Hauge et al., Genomics 10, 527-530 (1991)). Only Caucasian individuals were included in this study because of evidence for different distributions of Ta~I A and B
markers in white and black individuals (Dr Bruce O'Hara et al, wnpublished data). Substance. users were identified according to two approaches. One group of users met criteria for lifetime DSM-III-R jDiaq_nostic and Statistical Manual of Mental Disorders, Revised Third Edition. Washington, DC: American: Psychiatric Association: 1987) psychoactive substance use 20~ disorder(s)' . A second group of users was identified based on their peak lifetime use of psychoactive substances. This quantity-frequency approach was chosen because of evidence that heavy use of alcohol may disglay significant heritability in males and females (R. W. Pickens et al., Arch Gen Psychiatry ~t8, 19-28 (1991)). Control subjects were free of significant lifetime substance use.
i), Subiect Recruitment:; : X88 Caucasian substance-using and control subjects were recruited from three sources:
21% were female. 224 drug-using and control volunteers consenting to research protocols at the Addiction Research Center (ARC) in Baltimore, Maryland were studied. The ARC is the major federal drug abuse WO 93/24628 . PCT/US93/05179 . fl n research facility that recruits through advertisement and word of mouth for participation in treatment and non-treatment studies. 12 volunteers from a chronic hemodialysis unit on the same campus, both users and controls, augmented this sample. A third group of users consisted of 52 HIV seronegative participants in an ongoing east Baltimore study of HIV infections in intravenous drug users (D. Vlahov et al. , Am J Egid.
132, 847-856 (1990)).
Each subject was individually interviewed to elicit information characterizing substance use. 192 users and 56 controls were assessed according to a quantity-frequency approach. 137 users met criteria for.DSM-III-R psychoactive substance use disorders. 9.7 users received both assessments. Written informed consent was obtained from all subjects:
quantity-Fre~ency Approach: Trained interviewers assessed subjects with the Drug Use Survey (DUS) interview (see below) in a confidential setting. The amount, frequency, and/or dollar cost at the time of l~.fetime Beak use were recorded for each of 15 different psychoactive drugs or drug classes used more than five times. Blinded ratings of lifetime peak use ,of each individual substance were made on a four-point scale: 0=absent,, 1=minimal, 2=moderate, or 3=heavy use as indicated in Table I. A composite "Total Use" index was constructed from the pooled ratings of use of all individual substances as follows: "0" = up to minimal use : of, alGOho,l,, marijuana, or nicotine ~c no , use of other drugs: ''1" = moderate use of alcohol or nicotine and/or minimal use of other drugs: "2" = heavy use of alcohol or nicotine , moderate use of marijuana, andjor up to moderate use of other drugs; "3",_ heavy use of any illicit drug: Thus, neither heavy use of alcohol WO 93/24628 PCf/US93J05179 or nicotine was sufficient to confer a rating of heavy total drug use. Control subjects were identified as those individuals with Total Use scores of 0 or 1;
substance abusers were individuals with Total Use 5 scores of 2 or 3.
DSM-III-R Diactnoses: Trained interviewers administered the Diagnostic Interview Schedule Version III Revised (DIS-III-R, L.N. Robins et al.,NIMH
Diacxnostic Tnterview Schedule Version III Revised 10 (Version 11/7/89). Department of Psychiatry, Washington University School of Medicine, St. Louis, MO.) to provide lifetime DSM-ITT-R diagnoses of psychoactive substance use disorders including nicotine and alcohol.
15 Reliability. and Validity of Drug Use Information:
Drug Use Survey (DUS) ratings were evaluated in subjects who'were: (a) assessed with the DUS on two different occasions at 3 to 13 months apart (n=31) , (b) tested for lifetime DSM-ITI-R psychoactive substance 20 'use disorders by the DIS-III-R~ (n=18) and the Structured Clinical Interview for DSM-III-R (R. L.
Spitzer et al., Structured clinical interview for DSM-III-R - patient version (with nsychotic screen) SCIL~~P (W/Psychotic Sereenl - 5/1/89). Biometrics Research Department, New York State Psychiatric Institute, New York, New York)(SLID; n=17), and (c) checked for urinary excretion of psychoactive drugs and metabolites on the day of the DUS (n=56). For the 18 DIS-III-R-assg,ssed sub~j~cts and, the 17 SCID-assessed subjects, DUS ratings were completed without knowledge of psychiatric assessment information. Genotypes were not available for 17 sub j ects assessed with the SCID
and were not included in the genetic analyses.

Table I. Drug Use survey - Rating Criteria Substance Cigarettes 0 = never smoked cigarettes 1 = 1 to 15 cigarettes per day 2 = 16 to 25 cigarettes per day 3 = more than 25 cigarettes per day Alcohol 0 = newer used alcohol 1 up to 4 drinks per drinking occasion, fewer than 10 drinking occasions/month 2 - up to 4 drinks per drinking occasion, more than 10 drinking occasions/month, OR, more than 4 drinks per drinking occasion, but fewer than 1~0 drinking occasions/month 3 - 5 or more drinks per drinking occasion; more than 10 drinking occasions/month Heroin: 0 never used heroin/other opiates illicitly Other 1 used 1 time/week or less than $30/w~ek Opiates 2 _ 2 to 6 times/week , spending $30 to $100/day 3 - daily use, typically spending >

$loo/day Cocaine 0 _ never used cocaine 1 _ less than 2 grams per week (up to $150/we~k):

typical use - about 1 gram per month 2 - 2 to 4 grams per week (more than $150/week , butless than $300/week) ' 3 = more than 4 grams per week, usually up to 7 to 10 grams/week: (more than $300/week, usually much higher: daily use common) WO X3124628 PCTlUS93105179 ~~a6~~~r~
2z rlarijuana 0 = never used marijuana 1 = up to one joint/day 2 = 2 to 3 joints ger day 3 = 4 or more joints per day Minor Tranquilizers, 0 = never used substance Amphetamines, 1 = fewer than 1 use per week Barbiturates, 2 = 1 to b uses per week (4 to 24 uses/month) Hallucinogens, 3 = 7 or more uses/week (more than 24 Inhalants, PCP, uses/month) Antidepressants, Other Tobacco products, Other Substances ii) DNA Extraction and Anal~rsis: Blood was obtained in EDTA-containing evacuated sterile tubes from each subject and stored at 4°C and/or frozen at -70°C in polypropylene tubes. DNA was extracted from non-frozen samples after initial isolation of nuclei and from frozen blood by selective white blood cell sedimentation followed by standard extraction methods (J. Sambrook et al., eds. "Molecular cloning: a laboratory manual" (2nd edition). Cold Spring Harbor (NY) Laboratory Press; 1989). 5-10 ~g of this DNA was digested with ~agI as recommended by the manufacturer, or with 20-fold excess of this enzyme for several individuals displaying A3 alleles. DNA fragments were electrophoresed using 0.8% agarose gels containing ethidium bromide at 1-2 volts per centimeter for 16 hours, transferred to nylon membranes, and immobilized by UV crosslinking.
Hybridization was performed for 16-24 hours at 42°C
in 50% formamide, 5xSSC, 50 mM NaPO' (pH 6.8), 1% SDS, 1mM EDTA, 2.5 x Denhardt's solution, 200 ~g/ml herring sperm DNA, and 4X106 cpm/ml of radiolabelled DNA (see below). Washing for 20 minutes in 2XSSC at room temperature was followed by two 30 minute washes in 0.4 x SSC/0.5%SDS at 55°C. Washed blots were exposed to Kodak XAR*film 1-6 days with an intensifying screen at -70°C. Band sizes were compared to .1 DNA molecular weight standards, and with patterns previously defined (K. Blum et al., JAMA 263, 2055-2060 (1990); A.M. Bolos et al., JAMA 264, 3156-3160 (1990). After ~gI A RFLP
status was determined, 3zP decay allowed re-hybridization of the same blots with hybridization probe for ~gI B ascertainment. When background levels of radiation were not reached, filters were incubated at 65°C for 30 min in 2 mM TRIS (pH 8), 1 mM EDTA, and * Trade-mark WO 93/24628 ' ~ ~~ PCT/US93/OS179 i,:
z4 0.1~ SDS to remove residual hybridized probe. RFLP
status was assigned by two independent raters unaware of the clinical status of the subjects. M
iii) Hybridization probes: A 1.7 kb BamHI fragment of the human genomic clone encoding the dopamine DZ
receptor (.lhD2Gl) was subcloned into the ,~amHI site on bluescript SK+ to produce phD2-9, which was used to detect A1, A2, and A3 patterns in the Southern analyses, as described (K. Blum et al., JAMA 263, 2055-2060 (1990); A.M. Bolos et al., JAMA 264, 3156-3160 (1990)) (Dr Bruce O'Hara et al, unpublished data). JlhD2G2 was~used to detect the TaQI "B"
patterns. DNAs were radiolabelled using random priming and 32P-CTP to specific activities of approximately 109 cpm/~g (A. Feinberg and 8 Vogelstein, Anal Biochem.
137, 266-267 (I984)).
iv) 'nalyses:
a~ Association analyses: A two-tailed Pearson chi square test (with Yates' correction for continuity) was used to evaluate the association between A1 RFLP presence and substance use/abuse;
the 'same analysis were repeated for the B2 RFLP.
Association was first tested contrasting controls and substance users meeting criteria for any lifetime DSM-III-R substance dependence disorder.
Next, controls,,wer~ contrasted with substance users who had been assessed with the DUS. Data for both groups of substance-using subjects were pooled and compared to RFLP frequencies for controls. b) Comparisons with Qther data~ Pooled ~gI A1 RFLP data from ARC users was compared with WO 93/24628 ~ ~ ~ ~ ~ j r' PCT/US93/05179 values obtained for Caucasian controls in other studies.
_c) Subtractinct heavy alcohol users: DUS-assessed users free of heavy alcohol use were compared with 5 controls to test whether the associations observed might be attributed solely to alcohol.
TaQI A and B RFLPs were assigned with 100%
agreement between two independent raters.
Substance use assessment by means of the Drug Use 'l0 Survey showed several features suggesting validity and reliability. For 31 subjects whose DUS was elicited twice, interrater reliability correlations for severity ratings ranged from 0 . 83 to 1. 00 (median = U . 94 ) , while test-retest reliability correlations for individual 15 drugs ranged, from 0.53 to 0.94 (median = 0.78) . For 35 subjects with DIS-III-R or SLID assessments and independent DUS ratings, analysis of the correspondence between a positive lifetime DSM-III-R Substance Use diagnosis and moderate to heavy substance use on the 20 DUS yielded' a kappa value of 0.68 (91% agreement).
Finally, drugs .tested as positive in urine drug screening were reported used 84% of the time (n= 56) in the DUS assessment.
ac;I A and B RFLP frequencies for substance-using 25 and control subjects are presented in Table II. For the Tacrl Bl RFLP, a significant association was found comparing users with at least one lifetime DIS-III-R
Substance Use ., Di;sorder~ ;diagnos,is arid DUS-assessed , controls (XZ = 6.74, p < 0.01). For the ~gI A1 RFLP, analysis of the same groups revealed a significant association (x2 - 3.98, p < 0.05). Comparison of DUS-assessed users to DUS-assessed controls revealed a significant association for the TaQI B1 RFLP (x2 = 5.45, WO 93/24628 PC 1'/US93/05179 >1~~~J~~~

p < 0.02) and a trend towards significant association for the TaQI A1 RFLP (x2 = 3.14, p < 0.08). Table III
presents TaQI A and B genotypes (homoxygotes "'and heterozygotes) for DUS-assessed controls and users.

;: , - ;, .;,, .. ~ , ~ ....
S'VO 93/24628 PCT/US93/05179 ~~I~~~ ~ P
z~
r. ,~. _. ..

N iL1 ~ N N

N N t~ O'~M ~. .~..se ea ~-1 ltde-ii41 O 1D ip N I~ t'~O 1t1 \ \

~tt'M N ~ t' M Lt7 v v ar v vc ~r ..irao d' lP ch C'1!y! O 01 r9 . . a s a .
~ its d' 01 si~e-iN tc1 c'WP
~ M N C7 C~ t'~1 e-i e~-1rl M Pi 6r" ri U , CLi .-..-~ ..~e.
N !f7 ..rN ~1 N N !~ O~ M

rl ri t0 ri iV O ~D In r-r-\1tf1 O It b \ \ 'w y !n d' c'ht~.01 to 00 ri s Ws v a.iv W .r oI
.~

~ ~ tp O CO GO ~P O N O
~ . . . . a . .
~ ri tC d' O ri O N 41 d' N1 d~ ep ~P C'9 N

~ ~

~ .

A O ...
U' ~ c~ ~ Cl III O
Np A
~ ~ ~ ~ J N
V ~ ~ ~ ~ AI
J ~ O (~ dro-ii~C
~ t~ V .~~,yH . ~
.i~v UI ch m ~ N .N
o ~ ~ D ~ N ~ a v ~n ~ N b a~ o A ;~ . ,>, ~ o ~ o R~i'x ~ ~ v a H H

~ H a .7 c . n ' ~ a~, a o A ~ v A ca v ro . ci b H ~ st i H Ch 1-! i-a WO 93/24628 ~ ~ PCT/US93/05179 ,-:
i.a, 2s .. + . 1 ...
H fa D O
Hvrtt...~~
..
b0 O ~ .~ p r-~i_~ ~ N
!.: ~ U ~
N Q p ,N ,p cd 1A ,~' O ~.. ~r1 b ~v 3 O
ra ~ ~ V
... O ~ ,p 't3 .p 3 ~ c~ rtf o. N O ''"T N C"
~ H ~ ~
o.. r-! sis ~ d ~ p G3 r.
~ rw...~ \O N O CO ~ M O u1 tA
~' r-I O tn e!' W OO ~D N d'm-1 \ O -E ~ ~O" ' NMNN\\\\~\\ N
\\\\~ ~ ~ d' \1a O It1 O
~re~c°~cn~NC~Nrnr~ch .~9 ~
vvvvW rvvy....vv v l~ d, 0 0 10 Q1 01 M i!'f N h ~ 10 ~ ~ H
s yv a rs w a : s v aU ~ ~ V
V3 tf1 !n N N t'~ sHf tn h t~ t?1 !'~ O ~ H 1"~ p M
ri ~-i e-1 r-i M d"9 N M r-1 C1 N N N ri H N
N N r!
O ~ ~ %~ .L e~~1 ,.:1 b ~ !~f H N c~f ~ ,~,~ N ~ L'a ~d ~ ~ p H ~ p t0 a ~ N H p 'C3 ~ ~ of ...OV p O H W LT
H * * irl r-1 tj~ 1 .C'., ~ri ,~ ~d ... d1 ~-' N CJ~ .~
r~l rl ei 01 .-w-. c°1 H O tp ~ h U .' * * 01 01 ..e C1 ~-i M 01 N Ip N ~ .O
~ ~~ 01 f31 * O 01 ~ . CO Q1 ,I,f II r..l O w ~ ~ ~. _e-i _e~'! "°' ~ ~ ~ ~ H o ::'. Cl 'Cj V ~
E _H HM a ~ ~H''M~ ,..j y. w.. ~ p ~°G,J -~ ~ ~ YU b O ~, ~ ~-i r-i ''!N ~ ro a~ P
.- ~ f~ q~ t ~ td ~N ~ r-1 r-1 H .1.1 C!
~AI
r'I
ri r-1 .1~ ~ ri N ~ C. ~"'~ ~ ~ ~ N ~, 'O ~ U Cue.
of rtf a1 01 rtf p fa ~.~i .N W
a .~ ~ m s~ +~ ~ N .~ O p p GD A1al~-Nd1''~t~~u;~~;, oa ~, pepN O
d . .' S ,,~ ~ ~ TJ O ~r1 CD x >~ b V tf .~ ~~ O r-I
H ~ I~ r~~s~l ~~Iw bx ~ o'~ ~ ~ ~rn.aau ~~ao~doooarwf~' wa oacauwHGAUC9uL'10 p H A~ ~~~~ro-t.ro4 0~1~~ , W ..~...:..... ~.:.........-.....-. ~rl O II N p ~ ~ O ~ 'CS
b .p U ~Cf p W !T .G W ~n aC '~ N ~ y.~ ~1 "'~1~ p tv Q ~ ~ p ~ H ~.
~ V ANN OH ~~U
~ ~' co w t-~ o * f WO 93124628 ~ ~ ~ ~ ~'~ ~ PCT/US93/OSi79 O d' Q~ 1D N r-I N

II ~l1 t~ t~ N tL) I' O rl O e-1N O ri . . . a a .

~~ II a a a a to r-r cr o ~ ~~ ~ i ~ i ~

,. c, u , ~

N ' N ~

!d 1 00 C'1 ~
H D

N M
' ,~7 b ~

p . ~, O

v ~ e-1O d1 CO O r-i d' d O O O O O O ~
U

~ ~p N
~
"~

~

rl II II II II II II ~.
rl o\ O

O

H ~ ~ ~ ~ P

O

ri rl ri V . '~ ~ _ .

~J ~

N ~ ~

~ ~j M d! N eP 3 M N d' r7 N

: s . . . . . N

~ ~ .11 ~I allnII il lol ~ U

.

~ O \
W

O I N sf' ~
E d' , W

A1 r~
.: .

C1 D: G!

!1 ~ O 0 ! ~:
c h c t c0 t ' 7 ~"' ~ ~ lt ' '~ !n er ~ 1 tt~

.
~N

II il II II II <
II

0 ~ ~ ~ ~
ro M ~

r r-1O~ !n ~ 11 N c/~O\, ~ ~ ~ ~ \
~

Gi ~ ~ ,, ~ , ; ~ ~
,~-h ~ 0 c,~ iM .
, ;

~ O

!~

M rl ~ S

O

r. 1 0 ~ ~ CO al ~
~' d) ~ ~ ~ ~ ~ ~
~

E U': G

~'O 93/24628 ~ . fCI"1US93/05179 ~~~3C~~r~'~
No significant differences in RFLP
frequencies were found between DUS-assessed substance users and DIS-III-R-assessed users ( for TaaI Al, X2 = 0.005, ns; for TaQI B1, x2 = 0.331, 5 ns). We thus reanalyzed the data by pooling substance users assessed in both fashions for comparison with controls. Analysis of TaaI B1 data (x2 = 6.31, p < 0.02) and T~crl A1 data (x2 =
4:46, p < 0.04) again revealed significant 10 associations.
Comparisons of TaQI A1 RFLP frequencies in ARC users and controls from all other published studies revealed significant associations when the controls were assessed (XZ -. 15.41, p <
i5 O.OOZ), unassessed (xz 9:'31, p < 0.003), or pooled (7C2 _ 14.80, g < 0.001) To examine whether the effects noted could be attributed chiefly to the extent of alcohol intake, heavy alcohol users achieving DUS alcohol 20 ratings of 3 were eliminated from the user group and reanalysis performed. Omitting heavy alcohol users did not significantly alter the elevated frequencies of the TaaI~ B1 RFLP found in the user gxoup . 3 2 . 3 % of al l polysubstance users and 3 2 . 9 %
25 of the 73 polysubstance users free of heavy alcohol use displayed the ~agI ~Bl marker.
The hypothesis that individual differences in substance abuse ;may: be due, in ;part, to different dopamine D2 receptor alleles marked by 30 fag L RFLPs at the gene s 3~ and 5~ ends arises from initial work in alcoholics (K. Blum et al., JAMA 263, 2055-2060 (1990)). The hypothesis is WO 93/2462$ ,..

strengthened by a compelling biological rationale for interactions between abused drugs and brain dopamine systems (G. Di Chiara and A. Imperato, Proc Natl Acad Sci USA 85, 5274-5278 (1988); R.A.
Wise and P.P. Rompre, Annu Rev Psychol. ~0, 191-225 (1989)). In the current example, significant associations With heavy substance use or abuse were found consistently for the TacsI B1 RFLP and less consistently for the ~aqI A1 RFLP.
These findings provide preliminary evidence that a more 5' Taal RFLP (B1) may represent a better marker for a DRD2 gene variant possibly predisposing carriers to heavy substance use or abuse.
Selection of drug using and control populations provides opportunities for different approaches that could influence the results obtained. Many substance abusers use multiple psychoactive substances ( D.R. Wesson et al., eds. ~olydrug Abuses The Results of a National Collaborative Studv. New York, NY: Academic Press, Inc.; 1978): 71% of the 192 DUS-assessed users in the current study reported moderate to heavy use of three or more different substances.
To reflect this fact., we first studied subjects who frequently use multiple drugs, attempted to characterize each drug used by each subject, and analyzed .data on the: basis of overall; lifetime peak use. This approach might provide a weaker test of linkage if only a single abused substance displayed such genetic association. For example, if only alcohol abuse contributed to the WO 93/24628 , - PCT/US93/05l'79 >~~~~~~'~-1 associations noted here, we might anticipate a weaker association between- Ba RFLPs and substance abuse if individuals with heavy alcohol consumption were eliminated from our sample. In fact, elimination of heavy alcohol users (DUS
rating=3) resulted in no decrease in the differences between the remaining DUS-assessed drug-abusing and control individuals for the TaaI
B RFLP.
The.characterization of these subjects also raises important issues of assessment type, validity and reliability. Errors in clinical assessment would weaken tests of the allelic association hypothesis. In addition, studying behaviors that could contribute to features of clinical diagnosis but might not reflect the behavioral impact of a DRD2 gene variant could yield false-negative results.
We originally began work with the DUS, an interview-based assessment of substance use that enabled approximate quantification of peak lifetime use for several types of substances and appeared to provide an assessment of a basic feature of substance abuse: level of substance consumption. Psychiatric genetic work using classical methods suggests that heavy substance use can show substantial genetic determinants (~R.W. Dickens ; et :al. , Arch Gen Psychiatry ~8, ;
19-28 (1991); C.R. Cloninger and T. Reich, In:
Kety SS, Rowland LP, Sidman RL, Matthysse SW, eds. Genet3.cs of neurolociaal and ps~rchiatric disorders. New York, NY: Raven Press: 1983; pp.

Wt~3 93/24628 ~ ~ ~ ~ d~ ~ ~~~ PCT/US93/05179 145-166). Reliability and validity of the quantity-frequency approach to subj acts' drug use were supgorted by the correlations between drug use assessments made on two occasions, assessments made with multiple instruments, and correlations with results of urine drug screens.
However, several individuals who reported heavy use of various drugs did not fulfill criteria for DSM-III-R diagnoses of dependence or abuse on SLID or DIS-III-R assessments of the same drugs (S. S. Smith et al., "Validation of an instrument for quantifying drug use self-report: The ARC
Drug Use Scale" . Presented at the 53rd Annual Scientific Meeting, The Committee on Problems of Drug Dependence, June 16-20, 1991, Palm Beach, FL) We also evaluated subjects by determining lifetime psychiatric diagnoses of psychoactive substance use disorders using a structured psychiatric interview, the DIS-III-R, which can also demonstrate reliability and validity (J. E.
Helzer et al., Arch Gen Psychiatry 42, 657-666 (1.985) E N. Oskooilar et al. , DIS Newsletter 8,' 9-10 (1991)).
Comparison of Tactl A and B RFLP frequencies in substance-using subjects failed to indicate significant differences between the quantity-frequency and psychiatric diagnosis approaches.
These results suggested that we could combine . subjects meeting criteria for DSM-III-R Substance Use diagnsoses with subjects reporting moderate to heavy drug use. It is still conceivable, WO 93/24628 ~ PCT/US93/05179 however, that behavioral effects of a gene might be differentially reflected in quantity/frequency or in disease/disorder approaches to def fining the affected group.
The RFLPs studied here are the result of polymorphic Taal restriction sites in which °'A"
RFLPs are located ca. 9 kb 3' to the final axon of the D2 receptor gene (O. Civelli, personal communication) and "B" RFLPs are located near the l0 first coding axon (X: Y. Hauge et al., Genomics 10, 527-530 (1991)). These polymorphisms could have functional relevance if base pair differences directly influenced the gene's regulation. Alternatively, they could provide markers for structural or regulatory changes in other regions of the gene if these other,changes and the TactI variations were maintained together by linkage disequilibrium resulting in specific haplotypes (G. R. Uhl et al., Arch Gen Psychiatry 20. 49, 157-160 (1992)): This linkage disequilibrium does exist (X. Y. Hauge et al., Genomics 10, 527-530 (1991): Dr. Bruce 0'Hara et a1, unpublished data) . In our data, ~ for example, the expected frequency of the A2/A2-B2/82 haplotype would be 43% based on the frequencies of the A2 and B2 allelic markers. However, the observed frequency of this haplotype was 61% (x2 - 16.33, p <
0.0;001) (76% fpr 'c,~ntrols and ~57% for users, X~ _ , 5.15, p < 0.03), indicating substantial linkage disequilibrium.
The lack of strong association between DZ
receptor gene RFLPs and substance use evident in W'O 93/24628 ~ ~ ~~ ~ ~ ~ ~'~ PGT/US93105179 this study is consistent with estimates of the heritable components of alcoholism and drug abuse (E. J. Devor arid C.R. Cloninger, Annu Rev Genet.
23, 19-36 (1989).; R.J. Cadoret et al., Arch Gen 5 Psychiatry 43, 1131-1136 (1987)). One recent study of concordance rates for alcohalism in twin populations suggests that between 20 and 30% of the vulnerability to abuse or dependence on this substance may be genetic in origin (R. W. Dickens 10 et al., Arch Gen Psychiatry 48, 19-28 (1991)).
Attempts to link familial alcohol susceptibility to specific chromosomal markers and patterns of inheritance in families have not been consistent with a single genetic locus (S.B Gilligan et al. , 15 Genet Epidemiol. 4, 395-414 (1987); C.E. Aston and S:Y. Hill, Am J Hum Genet. 46, 879-887 ( 19!90) ) . The strong association between a single gene RFLP .and alcoholism found by Blum et al. (K.
Blum et al., JAMA 263, 2055-2060 (1990)) would 20 thus fit poorly with this extent of heritability.
The large environmental influences on expression of alcoholism, and their study of unrelated individuals rather than defined pedigrees also make the strength of their findings surprising.
25 To investigate the association between the DAT gene and substance abuse behaviors, one can make use of the variable number tandem repeat (~Tg) at ~ the; 3' ~er~d ; o,f the., ,mRNA described in , example 1. Alternatively the Taql RFLP described 30 in example 2 could be utilized. In general, examination of VNTR markers is preferred, as such markers have a larger number of alleles and hence are "more informative", i.e. VNTR markers identify more subtypes than a regular "site-no site" RFLP marker. The same methodology described above for the study of the D2 dopamine receptor gene can be employed. As shown above, particular attention must be paid to the diagnostic criteria for identifying the abuse behavior if the results are to be meaningful.
To assess frequencies of the VNTR, DNA is obtained from leukocytes from research volunteers as described above. Genomic DNA (40 ng) is subjected to 35 cycles of amplification using AmpliTaq*~NA Polymerase (1.25 U) and polymerase chain reaction with denaturing for 1 min at 93°C, and annealing/extension for 1 min at 72°C in buffer supplied by the manufacturer (Perkin-Elmer). Oligonucleotides T3-SLONG (5'-TGTGGTGTAGGGAACGGCCTGAG-3', SEQ. ID. NO. 4) and T7-3aLONG (5'-CTTCCTGGAGGTCACGGCTCAAGG-3', SEQ.
ID. N0. 5) are used at 0.5 uM final concentration. Reaction products are separated by 5% polyacrylamide gel electrophoresis, and product sizes estimated by comparison to molecular weight standards (BRL).
242 of the 254 chromosomes examined displayed either 9 or l0 copies of the 40 basepair repeat. Two chromosomes showed three copies, two showed 5 copies, three showed 7, four showed 8 and one showed 11 copies of the VNTR.
Among individuals with 9 and/or 10 copies per chromosome there were racial differences in copy number frequencies. Whites displayed 30% and * Trade-mark 11-'O 93124628 PCT/US93/05779 Blacks displayed 20% of the 9-copy variant, The 3' VNTR marker defined by 9 versus 10 copies csf the 40 basepair repeat displayed no significant linkage diseguilibrium with the more 5' TaqI RFLP
(X2 values were 5.51 and 4.62 for White and Black subjects, respectively, with 8 degrees of freedom, p > 0.2.) Examble 4 (predictive) Expression of HUDAT protein in Escherichia coli and purification of the bacterially ext~ressed protein Any of several expression systems can be utilized to obtain HUDAT protein expression in ~.
co i: For example, the plasmid vector pFLAG
system (International Biotechnologies, Inc., New Haven, CT) produces the polypeptide of interest attached to a short protein sequence that allows purification of the fusion protein by use of a monoclonal antibody directed against a hydrophilic, and thus surface localized, octapeptide. The open reading frame midportion of the HUDAT cDNA is obtained by digestion of the pHCDAT7 plasmid with EcoRI and purification of the inse=t fragment encoding the HUDAT protein by electrophoresis and elution from an agarose gel by standard techniques. Oligonucleotides having the: v sequences ' ~ 5'' -GGGTCTAGACG-3' , and 5' - , AATTCGTCTAGACCC-3' are annealed to form an adaptor and the adaptor is ligated to the ends of the insert DNA. The ligation product is digested with Xbal and cloned into the XbaI restriction WO 93/24b28 PCT/US93/05179 :., ~v-~ ~~ ', L~.~~~

site of the pFLAG vector (International Biotechnologies, Inc.). The appropriate E. coli host is transformed and colonies containing the HLTDAT cDNA may be screened by colony hybridization using the pcHUDAT as probe.
Positive clones are grown as large-scale cultures and the fusion protein is obtained in pure form by use of the monoclonal antibody affinity column as described by the manufacturer of the system, except that the elution buffer is modified by the addition of 0.5% CHAPS (3-[(3-Cholamidopropyl.)-dimethylammonio] 1-propane-sulfonate). Authentic DAT protein lacking the FLAG octapeptide is obtained by enterokinase cleavage of the fusion protein as described by the supplier of the FLAG
system.
Example 5 (,predictive) Purification of DAT from tissues or from transformed mammalian cells.
As protein asolated from transformed bacterial cells lacks post-translational modifications, such as sugar additions, that occur in mammalian cells, the purification of the protein from tranformed COS cells is discussed.
COS cells transformed as described in (S.
Shimada et al., Science 254, 576 (1991)) are subjected to: , a puri:f ioation protocol as , described for the purification of the GABA transporter (Radian, et al., J. Biol. Chem. 261, 15437-15441 (1987) with the modification that binding of labelled CFT is used to assay for the presence of WO 93/24628 ~ ~ ~ r; P~CT/US93/U5179 ~13~~~~

DAT in the sample rather than labelled gamma-amino butyric acid. The protocol is modified a~
required to allow the isolation of DAT as a distinct protein by techniques known to a practitioner of the art.
Examx~le 6 (predictive) Diagnosis of deficiency, mutant or overexoression of dobamine transporter by PCR
mRNA obtained from tissue biopsy from a patient is converted subjected to quantitative reverse-transcript PCR (for example, see A. M.
Wang,~et al. PNAS USA 86:9717 (1989)) utilizing as primers oligonucleotides derived from the cDNA
sequence of pcI~JDAT. Use- of the 5' 19-mer, GCTCCGTGGACTCATGTCTTC, bases 118 through 139 of Fig: 1 (SEQ: I.D. NO. 1) as the upstream primer and CACCTTGAGCCAGTGGCGG, the reverse complement of bases 1942 to 1960 of Fig. 1 (SEQ. I.D. NO. 1) as the downstream primer allows examination of the character of the protein coding region of the HUDAT mRNA. Variance in the expression level can be ascertained by comparison of product yield with a normal control: Abnormal mRNA structures can be diagnosed by observation of a product band of a length different from the normal control.
Point 'mutant's can 'be''dbserved~ by use of primers , and conditions appropriate for detection of the mismatch between the mutant and normal alleles.
For example, the "reverse dot blot" procedure for screening. the expression of several mutant WO 93!24626 ,.. f PCT/US93/O5i 79 ~..
f..:.
alleles in a single experiment, which has been described for the CFTR gene, mutants of which cause cystic fibrosis (Erlich, H.A., et a1 Science 252:1643 (1991).
5 The FiUDAT mRNA also contains a variable number tandem repeat element in the 3' untranslated portion of the mRNA which can be amplified for examination of an association between specific VNTR alleles and substance abuse 10 behavior or diseases associated with expression of particular HUDAT alleles (See example 3).
Example.? (predictive Use of dopamine transporter expression to incorporate as Bart of overexpression of a panel 15 of dopaminerctic crepes to reconstruct a dooamineraic cell fine for therany in human diseases result~na from defective dopamine transuorter expression.
cDNAs for' the human dopamine transporter, 0 and for tyrosine hydroxylase and aromatic ammino acid decarboxylase (DOPA decarboxylase) are transfected,into cell types including COS cells as described above. Cells are cotransfected with the neomycin resistance marker, selected by 2S growth in G418,r and then ;tested for their ability to synthesize and accumulate dopamine.
Individual subclones may be able to take up dopamine, without the ability to synthesize it.
However, individual subclones are also likely to WO 93/2Q628 ~~ '~ ~~ ~ .~'~ J ~/d PC'~'/US931OS179 integrate several of the plasmids. If the plasmids cannot be introduced serially or together in this direction, serial edition of tyrosine hydroxylase and DOFA decarboxylase to stable cell lines already expressing the dopamine transporter stably should be employed (see above). The ability of cells to incorporate tritiated tyrosine into tritiated dopamine is tested via HFLC analysis and radiochemical l0 detection as described (Uhl et al., Molecular Brain Research, 1991), their ability to take up ~

tritiated dopamine is performed as described in the same reference.

These same procedures are used in transfecting cells obtained from an individual with a disease state caused by defects in dopamine transporter expression, either in the amount expressed or due to expression of a defective protein, so that stable immortalized cell lines expressing human dopamine transporter could be constructed with immunologic identity to the patient. Means of controlling the replication of these cells by encapsulating them in a matrix that is not porous to cell bodies, but able to be permeated by cell processes, or by use of inhibitory growth factors, can also be employed: A third strategy, temperature sensitive ~ cel'1 mutants that would not divide under physiologic temperatures (e. g. temperature sensitive COS cells variants) could be used to be able to express the dopaminergic cDNA stably, in a fashion that would produce dopaminergic cells.

.'~-'!'.: .'. '.: ..~-: .... : w ;..,..;.,'.., .."'.. ~.,,...,.;.'. ~. .....
.::'.~. ~. ....;.;,, i. ,.~~..: ~"'...~,'.. "....:.;. -;.;..., .,....".
,;...,.....,, ;..,.. .,..,."..,,..., ..~A,:",.,..,. , .. ,.,, _ ; ."... ".... ..,,~,: :.. ~..~. . .. .. ~. ",.,. .
...,._..~. ,: ~; .':~, :.:..~..."..._.,._. .' .,.,...,.; ,. , ; .1 .,.,:
WO 93/24628 ~ ~ PCT/iJS93/Oa179 ~136~~~~

Each of these cell types are potential candidates for use in transplantation into striatum in individuals with striatal dopamine depletion in Parkinson's disease. Alternatively, genes could be incorporated with retroviral vectors as well-known for practitioners of the art.
Example 8 (predictive) Production of variant secLuences in HUDAT urotein and testing of their biological function Site directed mutagenesis using olgonucleotides is used to introduce specific single- and multiple-base changes into the HUDAT

cDNA that change specifis amino acids in the HUDAT protein. The ability of mutant transporters to take up [3Fi] dopamine, [3H]

I~iPP+, and to bind' [3Fi] cocaine and cocaine analogues (especially [3H] CFTj is tested as described previously (S. Shimada et al., Science 254, X76 (1991)). The Amersham mutagenesis 2n system (version 2.1, technical bulletin code RPN1523j can be used. Initial studies of mutants of' the aspartic acid residue in txansmembrane domain 1, and the serine residues in transmembrane domain 7 of the rat DAT protein have revealed substantial effects on dopamine . transport, and more modest effects on cocaine b~ind~ing. ' 'vThess ~r r'esvlts document ; that the ;

residues key to dopamine transport are not identical to those crucial to cocaine binding:

thefirst transmembrane residue change of aspartic acid (residue 79j to glycine .reduces ~~1~~10.8~ ~ ~~
a , , ~~, ,~:

cocaine binding by 10%, but reduces dopamine transport by over 95%. Mutations in the secozid extracellular domain in glycosylation sites help elucidate the role of glycosylation in the functions of this molecule (See also Example 9).
Selective removal of the N and C terminal intracellular and second extracellular loop, and production of chimeric molecules with replacement of these regions with the corresponding regions of the GABA transporter further confirm the molecular features of DAT that are essential for dopamine transport and cocaine binding and allow development of agents dissociating° the two processes.
ExamoZe'9 lpredictivel alteration of carbohydrate structure in the extracellu_~domain of the HUDAT protein As noted in example 1, the largest difference in the structure of the proteins predicted by the human and the rat dopamine transporter dDNA sequences is the absence of one of the four consensus sites for N-linked glycosylation of the protein (See figure 3). By virtue of their location in the same domain of the protein expected to most influence substrate binding, that is in the extracellular portion of the~''protein~ it i's~'of' interest to inve$tigate the contribution of the sugars to substrate binding.
Site directed mutagaenesis can be performed as described for Example~8 introduce into the human DAT cDNA the asparagine residues to which N-1V0 93/24628 ~ ~ ~ ~ ~~~''1 ; . ~ . ....
U~~._ ~ ,, ~ ~~

linked sugars are attached and the remaining amino acids which constitute the glycosylation signal for that site that are found in the rat, but not the human cDNA. The result of expression of such mutant proteins can be evaluated by photo-affinity labelling of the protein and analysis by SDS-PAGE. Digestion of the protein with various glycosidases can be performed to assess the degree to which the pattern of glycosylation has been altered, as described by Lew et al. (R. Lew et al., Brain Research 539, 239 (1991)). For instance, compararison of the wild--type and mutant protein, both untY~eated and digested with N-glycanase, would should show similar sized proteins for the digested protein, but a larger protein for the untreated mutant, compared to the untreated wild-type protein if the introduction of the asparagine glycosylation signal resulted in successful incorporation of sugar into the protein at that site. More detailed information regarding the sugar structure can be obtained by exoglycosidase digestion experiments. For example, the presence of sialic acid residues in the polysaccharide can be-detected by digestion with neuraminidase. The influence of the polysaccarhide structure on function of the protein is then assessed by testing the propdrt'ies. of ' the the transporter:
using either stably transfected cells expressing the mutant protein, or by using cells transiently expressing the mutant transporter on their surface. The means for carrying out such WO 93/24b28 ~ ~''_ ~? PCT/US93/05~79 ~13~~'8~
functional studies are described by Shimada et al . (S . Shimada et al . , Science 254, 576 ( 1991) )~~.
Exam»le 10 (predictive) Gell lines expressing HtTDAT protein on the 5 cell surface can be used to screen candidate compounds for efficacy as dopamine (or cocaine or functional analogs thereof) agonists or antagonists by evaluating the influence of the candidate compound upan the binding of dopamine 10 (or cocaine or functional analogs thereof) to the surface of such cells. Another assay for dopamine agonist or antagonist activity is to measure the cytotoxicity to such cells of MPP+ to such cells in the presence and absence of the 15 candidate compound: Such assays are described using cells expressing the rat DAT cDNA in Shimada et al. (S. Sh~:mada et al., Science 254, 576 (1991) ) and can be applied as. well to cells expressing the human DAT cDNA.
2p Example 11 ~~redictive~
~roductior of antibodies to ~IUDAT and use of same os is test o do a 'ne 'c a 1 deat .
A. Production of polyclonal antibodies.
HLTDAT pro~ein,.obtained as.'. described above or 25 synthetic polypeptides of amino acid sequence derived from the HUDAT sequence are used as immunogens in an appropriate animal. The serum is obtained from the immunized animal and either WO 93/24628 . ~ PCT/US93/85179 ~..~
,;,: . :.
213~~'~~ ~~'~ ~ ~ : .

utilized directly or the antibody may be purified from the serum by any commonly utilizdd techniques. Polyclonal antibody directed only toward HUDAT can be isolated by use of an affinity column derivatized with the immunogen utilized to raise the antibody, again using techniques familiar to one knowledgable,in the art:
Production of monoclonal antibodies to HUDAT

Monoclonal antibodies to HUDAT or to particular epitopes of giUDAT may be produced by immunization of an'appropriate animal with HUDAT

protein obtained as above or with peptides of amino acid sequence derived from the HUDAT amino acid sequence. Hybridoma cultures are then established from spleen cells as described by Jaffe and McMahon-Pratt (Jaffe, C.L. and MacMahon-Pratt, D. J. Immuriol. 131, 1987-1993 ( 1:983 ) )' : Alternatively, peripheral blood l~aphocytes maybe isolated and immortalized by transformation with Epstein-Barr virus. These cells produce monoclonal antibodies, but if desired, hybridomas can then be made from the transformed lymphocytes (Yamaguchi, H. et al.

Proc. Natl. Acad. Sci. 84, 2416-2420 (198?)).

Cell lines producing anti-HUDAT antibodies are ' ' ider~ti~ied'~ ~ by ;' oontimdnly employed screening ;

techniques. Monoclonal antibody is then purified by well known techniques from the supernatants of large-scale cultures of the antibody producing cells.

~136~8'~ ~ ~~ ~ ~~

C. Diagnosis of dopaminergic cell death. in yivo by immunoassay of cerebrospinal fluid of ~'a patient using anti-HUDAT antibodies.
The death of dopami.nergic neurons in the brain of a patient should result in the accumulation in the cerebrospinal fluid, which bathes these cells, of membrane debris as a product of lysis of the dead cells. Other pathologic conditions, short of cell death that result in the release of DAT protein, or degraded peptide fragments of HTJDAT protein into the surrounding medium can also be imagined. The cerebospinal fluid can be sampled by lumbar puncture of a patient. The presence of degradation products of HUDAT protein is detected by immunoassay, using as the primary antibody at least one of the products obtained as described above: Elwated bevels. of HUDAT protein detected in the cerebrospinal fluid., compared with the range seen in normal controls is indicative of Parkinsons's disease or drug-induced neurotoxicity. Alternatively, disease progression can be monitored by the assessment of HUDAT levels in serial samples from the same Patient.

WO 93/24628 PCT/US93/051.79 SEQUENCE LISTING
M
(1) GENERAL INFORMATION:
(i) APPLICANT: Uhl, George R.
Vandenbergh, David Persico, Antonio (ii) TITLE OF INVENTION: Sequence of Human Dopamine Transporter (iii) NUMBER OF SEQUENCES: 5 (iv) CORRESPONDENCE ADDRESS:
(A) ADDRESSEE: Birch, Stewart, Kolasch & Birch (B) STREET: 301 N. Washington St.
(C.) CITY: Falls Church (D) STATE: Virginia (E) COUNTRY: USA
(F) ZIP: 22046-3487 (v) COMPUTER READABLE FORM:
(A) MEDIUM TYPE: Floppy disk (B) COMPUTER: IBM PC compatible (C) OPERATING SYSTEMa PC-DOS/MS-DOS ' (D) SOFTWARE: Patentln Release #1.0, Version #1.25 (vi).CURRENT APPLICATION DATA:
(A) APPLICATION NUMEER:
($) FILING DATE:
(C) CLASSIFICATION:
(viii) ATTORNEY~AGENT II~tF'ORM~LTION' ~ - , ~ ' (A) NAME: Murphy Jr., Gerald M.
(B) REGISTRATION NUMBER: 28,977 ~13~~8~

(ix) TELECOMMUNICATION INFORMATION:
(A) TELEPHONE: 703-241-1300 .~
(B) TELEFAX: 703-241-2848 (C) TELEX: 248345 (2) INFORMATION FOR SEQ ID NO:1:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 3919 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: double (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA
(iii) IiYPOTHETICAL: NO
(v~) ORIGINAL SOURCE:
(A) ORGANISM: Homo sapiens (F) TISSUE TYPE: brainstem (ix) FEATURE:
(A) NAME/KEY: CDS
(B) LOCATION: 102..1961 (D) OTHER INFORMATTON: /function= "dopamine transport"
/Product= '~HUDAT polypeptide"
(ix) FEATURE,:
(A) NAME/KEY: misc RNA
(B).LOCATION: 2724..3117 (D) OTHER INFORMATION: /function= "unknown"
/label= VNTR region ;, j ~ t i :, , WO 93/2462$ PCT/US93/d5179 (xi) SEQUENCE DESCRIPTION: SEQ ID NO:1:

Met Ser Lys Ser Lys Cys Ser Val Gly Leu Met Ser Ser Val Va1 Ala Pro Ala Lys Glu Pro Asn Ala Val Gly Pro Lys Glu Val Glu Leu Ile Leu Va1 Lys Glu Gln Asn G1y Va1 Gln Leu Thr Ser Ser Thr Leu Thr Asn Pro Arg Gln Ser Pro ValGlu AlaGln AspArgGlu ThrTrp GlyLysLys IleAsp Phe Leu LeuSer ValIle GlyPheAla ValAsp LeuAlaAsn ValTrp Arg Phe ProTyr LeuCys TyrLysAsn GlyGly GlyAlaPhe LeuVal g5 90 95 100 Pro Tyr LeuLeu PheMet ValIleAla GlyMet ProLeuPhe TyrMet ,,. 105 , , ; 110, , 115 Glu Leu AlaLeu GlyGln PheAanArg G1uGly AlaAlaGly ValTrp PCI'/US93/~5179 WO 93/~4b28 Lys Ile Cys Pro I1e Leu Lys Gly Val Gly Phe Thr Val Ile Leu Ile 135 ~ 140 145 Ser Leu Tyr Val Gly Phe Phe Tyr Asn Val Ile Ile Ala Trp Ala Leu His Tyr Leu Phe Ser Ser Phe Thr Thr Glu Leu Pro Trp Ile His Cys Asn AsnSer TrpAsn SerProAsn CysSer AspAla HisProGly Asp Ser SerGly AspSer SerGlyLeu AsnAsp ThrPhe GlyThrThr Pro Ala AlaGlu TyrPhe GluArgGly ValLeu HisLeu HisGlnSer His GGC ATCGAC GACCTG GGGCCTCCG CGG,TGGCAGCTC ACAGCCTGC CTG 833 Gly IleAsp AspLeu GlyProPro ArgTrp GlnLeu.ThrAlaCys Leu Val LeuVa1 IleVal LeuLeuTyr PheSer LeuTrp LysGlyVal Lys 245 250 255 260.

Thr Ser Gly Ly's Val Val'Trp!Ile.~Thr Ala Thr Met Pro TyrVal Val WO 93/2.628 PCT/US93/05179 ,~ t, Leu Thr Ala Leu Leu Leu Arg Gly Val Thr Leu Pro Gly Ala Ile Asp Gly Tle Arg Ala Tyr Leu Ser Val Asp Phe Tyr Arg Leu Cys Glu Ala 295 . 300 305 Ser Val Trp Ile Asp Ala Ala Thr Gln Val Cys Phe Ser Leu Gly Val Gly Phe.Gly Val Leu Ile Ala Phe Ser Ser Tyr Asn Lys Phe Thr Asn Asn Cys Tyr Arg Asp Ala Ile Val Thr Thr Ser Ile.Asn Ser Leu Thr Ser Phe Ser Ser Gly Phe Va1 Val Phe Ser Phe Leu Gly Tyr Met Ala CAG A~.G CAC AGT GTG CCC ATC GGG GAC GTG GCC AAG GAC GGG CCA GGG 1265 Gln Lys His Ser Val Pro Ile Gly Asp Val Ala Lys Asp Gly Pro Gly 375 380 385 , Leu Zle Phe Ile Ile Tyr Pro Glu Ala Ile Ala Thr Leu Pro Leu Ser Ser Ala '~rp Ala Val Va1 ' Phe~ Phe 'Ile Nlet Leu Leu Thr Leu , Gly Ile v :A
. n ., ,.

GAC AGC GCC ATG GGT GGT,ATG GAG TCA GTG ATC ACC GGG CTC ATC GAT 1409 Asp Ser A1a Met Gly Gly Met Glu Ser Val Ile Thr Gly Leu Tle Asp Glu Phe Gln Leu Leu His Arg His Arg Glu Leu Phe Thr Leu Phe Ile Val Leu Ala Thr Pha Leu Leu Ser Leu Phe Cys Val Thr Asn Gly Gly Ile Tyr Val Phe Thr Leu Leu Asp His Phe Ala Ala.Gly Thr Ser.Ile Leu Phe GlyVal LeuIle GluAla IleGlyVal AlaTrp PheTyr Gly Val Gly GlnPhe SerAsp AspIle GlnGlnMet ThrGly GlnArg Pro 5~5: 510 515 AGC CTG TACTGG CGGCTG TGCTGG AAGCTGGTC AGCCCC TGCTTT CTC 169?

Ser Leu TyrTrp ArgLeu CysTrp LysLeuVal SerPro CysPhe Leu CTG: TTC GTGGTC GTGGTC AGCATT GTGACCTTC AGACCC CCCCAC TAC 1745 Leu Phe ValVa1 ValVal SerIle Va1ThrPhe ArgPro ProHis Tyr Gly Ala TyrI1'ePhePro 'Asp'Trp'Alal~snAla LeuGly TrpVal Ile . . -., n ~.. , t ~ I ~e ~~ .-Ala Thr Ser Ser Met Ala Met Val Pro Ile Tyr Ala Ala Tyr Lys Phe Cys.Ser Leu Pro Gly Ser Phe Arg G1u Lys Leu Ala Tyr Ala Ile Ala Pro G1u Lys Asp Arg Glu Leu Val Asp Arg Gly G1u Va1 Arg Gln Phe Thr Leu Arg His Trp Leu Lys Val ACCAAGGAAA

TGTAATARCG ACG.TAGATCTGTGCAGCGAGGTCCACCCCGTTGTTGTCCCTGCAGGGCAG 2231 ~ I ' j _ i'' . , 213608' =,q~;.

ACCCCAGGAC GCATGCAGGG CCCCCACAGG AGCGTGTACT ACCCCAGGAT GCATGCAGGG 3077.
CCCCCACAGG AGCGTGTACTACCCCAGGACGCATGCAGGGCCCCCFsTGCAGGCAGCCTGC3131 CGTGCAGGGC CAGTCATGGC.TGTCCCCTGCAAGTGGACGTGGGCTCCAGGGACTGGAGTG3551 ,. ; ~z ' ~ , ~ , WO 93/24628 P'C'1'/US93/OS179 .r "~ , ~ y .a TATTCAGCAT CGTGTGGGTC CCTAAGCACA ATAAAAGACA TCCACAATGG AAA.AAAAAA.A 3911 (2) INFORMATION FOR SEQ ID N0:2:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 620 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: protein (xi) SEQUENCE DESCRIPTION: SEQ ID N0:2:
Met Ser Lys Ser Lys Cys Ser Val.Gly Leu Met Ser Ser Val Val Ala 1~ 15 Pro Ala Lys Glu Pro Asn Ala Va1 Gly Pro Lys Glu Val Glu Leu Ile Leu Val Lys Glu Gln Asn Gly Va1 Gln Leu Thr Ser Ser Thr Leu Thr Asn Pro Arg Gln Ser Pro Val Glu Ala Gln Asp Arg Glu Thr Trp Gly 50 55 60, Lys Lys Ile Asp Phe Leu Leu Ser Val Ile Gly Phe Ala Val Asp Leu Ala Asn Val Trp Arg Ph'e'Pro'Tyr~'Leu Cy$ Tyr'Lys Asn Gly,Gly Gly Ala Phe Leu Val Pro Tyr Leu Leu Phe Met Val Ile Ala G1y Met Pro WO 93/24628 ... PCI'/US93/OS179 '~ 'v s7 Leu Phe Tyr Met Glu Leu Ala Leu Gly Gln Phe Asn Arg G1u Gly Ala 115 120 125 ", Ala Gly Val Trp Lys Ile Cys Pro Ile Leu Lys Gly Val Gly Phe Thr Val Ile Leu Ile Ser Leu Tyr Val Gly Phe Phe Tyr Asn Val Ile Ile Ala Trp Ala Leu His Tyr Leu Phe Ser Ser Phe Thr Thr Glu Leu Pro Trp Ile His Cys Asn Asn Ser Trp Asn Ser Pro Asn Cys Ser Asp Ala His Pro Gly Asp Ser Ser Gly Asp Ser Ser Gly Leu Asn Asp Thr Phe Gly Thr Thr Pro Ala Ala Glu Tyr Phe Glu Arg G1y Va1 Leu His Leu His Gln Ser His Gly Ile Asp Asp Leu Gly Pro Pro Arg Trp G1n Leu Thr Ala Cys Leu Val Leu Val Ile Val Leu Leu Tyr Phe Ser Leu Trp Lys Gly Val Lys Thr Ser Gly Lys Val Val Trp Ile Thr Ala Thr Met Pro Tyr Val Val Leu Thr Ala Leu Leu Leu Arg Gly Val Thr Leu Pro Gly Ala Ile Asp~Gly Ile Arg'Ala ~'yr Leu Ser Val Asp Phe Tyr Arg Leu Cys Glu Ala Ser Val.Trp Ile Asp Ala Ala Thr Gln Val Cys Phe ,; , . ; , ... ,> .'.:; _.:: . . ;. ;'' WO 93/24628 . PC1"/US93/0~1'79 213 6 ~ 8'~ ~ P ~~:

Ser Leu Gly Val Gly Phe Gly Val Leu Ile Ala Phe Ser Ser Tyr Asn 325 330 335 ..
Lys Phe Thr Asn Asn Cys Tyr Arg Asp Ala Ile Va1 Thr Thr Ser Ile Asn Ser Leu Thr Ser Phe Ser Ser Gly Phe,-Val Val Phe Ser Phe Leu Gly Tyr Met Ala Gln Lys His Ser Val Pro Ile Gly Asp Val Ala Lys Asp Gly Pro Gly Leu Ile Phe Ile Ile Tyr Pro Glu Ala Ile Ala Thr Leu Pro Leu Ser Ser Ala Trp Ala Val Val Phe Phe Ile Met ~Leu Leu Thr Leu Gly Ile Asp Ser Ala filet Gly Gly Met G1u Ser Val Ile Thr 420 . 425 430 Gly Leu Ile Asp GIu Phe Gln Leu Leu His Arg His Arg Glu Leu Phe Thr LeuPhe Ile Leu Ala Phe LeuSer LeuPhe Val Val Thr Leu Cys Thr AsnGly Gly Tyr Val Thr LeuAsp HisPhe Ala Ile Phe Leu Ala GIy ThrSer Ile Phe Gly Leu GluAla IleGly Ala Leu Val Ile Val Trp Phe Tyr Gly Val Gly'~Glri Phe''Ser Asp Asp ~~le Gln Gln,Met Thr Gly Gln Arg Pro Ser Leu Tyr Trp Arg Leu Cys Trp Lys Leu Val Ser WO 93/24628 PC.T/US93/U5179 ~I~v': v: i: .. .:.

Pro Cys Phe Leu Leu Fhe Val Val Val Val Ser Ile Val Thr Phe Arg 530 535 540 ", Pro Pro His Tyr G1y AIa Tyr Ile Phe Pro Asp Trp Ala Asn Ala Leu Gly Trp Val Ile Ala Thr Ser Ser Met Ala Met Val Pro Tle Tyr Ala Ala Tyr Lys Phe Cys Ser Leu Pro Gly Ser Phe Arg Glu Lys Leu Ala Tyr Ala Ile Ala Pro Glu Lys Asp Arg Glu Leu Val Asp Arg Gly Glu Val Arg Gln Phe Thr Leu Arg His Trp Leu Lys Val 610. 6~.5 620 (2) INFORMATION FOR SEQ ID NO:3:
(i) SEQUENCE CHARACTERISTICS:
(A)' LENGTH: 40 base pairs (B)- TYPE: nucleic acid (C) STRAN~RDNESS: double (D) TOPOLOGY; linear :(ii) MOLECULE TYPE: DNA
(iii) HYPOTHETICAL: XES
(iv) ANTI-SENSE: NO
(ix) FEATURE:
(~A) 'NAMEIKEY: ''- ' '1 (B) LOCATION: 1..40 (D) OTHER INFORMATION: /label= consensus /note= "consensus sequence of VNTR element in 3' untranslated region of HUDAT cDNA"

,. , .. .. .. : .. ' ~ ,.;; ;;
WO 93/24628 ' ' PCT/US93/U5179 (xi) SEQUENCE DESCRIPTION: SEQ ID N0:3: M

(2) INFORMATION FOR SEQ ID N0:4:
(i) SEQTJENCE CHARACTERISTICS:
(A) LENGTH: 23 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii).MOLECULE TYPE: DNA (synthetic) (iii) HYPOTHETICAL: NO
(iV) ANTI-SENSE: NO
(ix) FEATURE:
(A) NAME/KEY: --{B) LOCATION: x...23 (D) OTHER, INFORMATION: /label= oligonucleotide /note= "synthetic oligonucleotide T3-5LONG, upstream primer for PCR analysis of VNTR region of in 3~ untranslated region of HUDAT gene "
(xi) SEQUENCE DESCRIPTION: SEQ ID N0:4:
TGTGGTGTAG GGAACGGCCT GAG

( 2 ) INFORMATION FOR SEA' ID NO : ~ : ' ~ , w ( i ) SEQUENCE CF~AR.ACTERISTICS
(A) LENGTH: 24 base pairs (g) TYFE: nucleic acid sz (C) STRANDEDNESS: single ...
(D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (synthetic) (iii) HYPOTHETICAL: NO
(iv) ANTI-SENSE: YES
(ix) FEATURE:
(A) NAMELY: -(B) LOCATION: 1..24 (D) OTHER INFORMATION: /label= oligonucleotide /note= "synthetic oligonucleotide, T7-3aLONG;
downstream primer for PCR analysis of~ VNTR region of 3' untranslated region of HUDAT gene"
(xi) SEQUENCE DESCRIPTION: SEQ ID NO: S:

Claims (19)

The embodiments of the invention in which an exclusive property or privilege is claimed are defined as follows:

1. An isolated, purified protein comprising the amino acid sequence as set forth in SEQ ID NO:2.

2. An isolated cDNA comprising a nucleotide sequence which encodes the protein of claim 1.

3. An isolated cDNA comprising a nucleotide sequence as set forth in SEQ ID NO: 1.

4. An isolated cDNA comprising a DNA fragment having a nucleotide sequence which encodes the amino acid sequence of SEQ ID NO:2 and further comprising a repetitive element in a 3' untranslated portion of said cDNA, wherein said repetitive element comprises head-to-tail repeats of a nucleotide sequence that is at least 90%
identical to the sequence shown in SEQ ID NO:3.

5. An isolated cDNA of claim 4 having the nucleotide sequence of SEQ ID NO:1.

6. A DNA plasmid comprising the cDNA of claim 2, 3, 4 or 5, and a DNA sequence that provides for replication in a prokaryotic host cell and DNA sequences that provide for transcription of the cDNA in vitro, such that the resulting mRNA can be isolated and translated upon introduction into a eukaryotic cell type.

7. A eukaryotic cell line derived from a cell type that does not normally express dopamine transport activity at its surface that has been made to transport dopamine or to bind CFT by the introduction of a DNA into its genome, said DNA encoding a protein as defined in claim 1.

8. A eukaryotic cell line derived from a cell type that does not normally express dopamine transport activity at its surface that has been made to transport dopamine or to bind CFT by the introduction of the cDNA of claim 2, 3, 4 or 5 into its genome.

9. A eukaryotic cell line derived from a cell type that does not normally express dopamine transport activity at its surface that has been made to transport dopamine or to bind CFT by the introduction of the plasmid of claim 6 into said eukaryotic cell.

10. The cell line of claim 8 or 9, wherein the cell type is COS cells.

11. A method for screening a compound for cocaine antagonist activity, which comprises measuring the inhibition by said compound of binding of cocaine, or of functionally equivalent cocaine analogs, to the surface of cells or to membrane preparations obtained from said cells, wherein said cells express the cDNA of claim 2, 3, 4 or 5.

12. A method for screening a compound for cocaine antagonist activity, which comprises measuring the inhibition by said compound of cell toxicity due to MPP+
import into cells expressing the protein of claim 1 or transport across reconstituted membrane preparations obtained from such cells.

13. A method for screening a therapeutic agent effective in the prevention or treatment of Parkinson's disease which comprises measuring the inhibition of cell toxicity due to MPP+ import into cells expressing the protein of claim 1, or measuring MPP+ transport across reconstituted membrane preparations obtained from such cells.

14. A method for screening a compound for effectiveness in the prevention or treatment of Parkinson's disease which comprises measuring the inhibition by said compound of binding of cocaine, or of functionally equivalent cocaine analogs, to the surface of cells or to membrane preparations obtained from said cells, wherein said cells express the cDNA of claim 2, 3, 4 or 5.

15. A method for the screening a compound for effectiveness in the prevention or treatment of diseases caused by abnormal dopamine transport which comprises measuring either the inhibition or facilitation by said compound of binding of cocaine, or of functionally equivalent cocaine analogs, to the cell surface of cells of the cell line of claim 7, 8, 9 or 10, or to membrane preparations obtained from said cell line.

16. A method for the screening a compound for effectiveness in the prevention or treatment of diseases caused by abnormal dopamine transport which comprises measuring inhibition by said compound of the cell toxicity due to MPP+ import into cells of the cell line of claim 7, 8, 9 or 10, or transport across reconstituted membrane preparations obtained from said cell line.

17. Use of a recombinant DNA construction expressing the cDNA of claim 2, 3, 4 or 5, for the treatment of Parkinson's disease, Tourette's syndrome or other disease caused by abnormal human dopamine transporter (HUDAT) expression.

18. Use of a recombinant DNA construction expressing the cDNA of claim 2, 3, 4 or 5 for the manufacture of a medicament for the treatment of Parkinson's disease, Tourette's syndrome or other disease caused by abnormal human dopamine transporter (HUDAT) expression.

19. Use of the cDNA of claim 2, 3, 4 or 5 as a probe for diagnosing genetic variants in dopamine transporter protein.