WO1994004573A1 - cDNA CLONE ENCODING AN EXPRESSIBLE GABA TRANSPORTER - Google Patents

Abstract

DNA molecules encoding a GAT-B transporter activity are disclosed. The GAT-B activity is sensitive to β-alanine, is expressed predominantly in neurons, and exhibits other characteristics distinguishing it from other GABA transporter activities. Also disclosed are cloned cDNAs encoding GAT-B activity that can be incorporated into expression vectors and used to transfect various types of cells; the nucleotide sequence of GAT-B cDNA; an amino acid sequence of the GAT-B protein; nucleotide sequences complementary or substantially homologous to either strand of the GAT-B DNA; various primers useful for probes of GAT-B-specific RNA and DNA sequences; various host cells transfected by GAT-B-specific DNA and/or RNA; and methods for transforming non-neural cells to cells that express GAT-B.

Description

CDNA CLONE ENCODING AN EXPRESSIBLE GABA TRANSPORTER

FIELD OF THE INVENTION This invention pertains to the cloning of a DNA sequence. In particular, this invention pertains to the cloning of a complete complementary DNA (cDNA) sequence for a neurotransmitter transporter protein.

ACKNOWLEDGEMENT The inventors' laboratory was supported in part by grant no. ROl DA07595 from the National Institute of Drug Abuse. Therefore, the United States government may have rights in this invention.

BACKGROUND OF THE INVENTION Transmission of a nerve impulse across a synapse involves the secretion of neurotransmitter substances by the presynaptic neuron into the synaptic cleft. This facilitates the transmission of a chemical signal across the synaptic cleft to the postsynaptic neuron. Transmission of the chemical signal is normally transient. Otherwise, if the neurotransmitter substances persisted in the synaptic cleft, a new signal would not get through.

Nervous tissue normally disposes of soluble or unbound neurotransmitter in the synaptic cleft by various mechanisms, including diffusion and enzymatic degradation. In addition, at most synapses, chemical signaling is terminated by a rapid reaccu ulation of neurotransmitter into presynaptic terminals. This reaccumulation is the result of re-uptake of the neurotransmitter by the presynaptic neuron. Of the various known disposal mechanisms for neurotransmitters, re-uptake of the neurotransmitter from the synaptic cleft is probably the most common mechanism used for terminating the chemical signal. At presynaptic terminals, the various molecular apparatuses for re-uptake are highly specific for such neurotransmitters as choline and the biogenic amines (low molecular-weight neurotransmitter substances such as dopamine, norepinephrine, epinephrine, serotonin, and histamine) . These apparatuses are termed "transporters" because they transport the corresponding neurotransmitter from the synaptic cleft back across the cell membrane of the presynaptic neuron into the cytoplasm of the presynaptic terminus.

Certain psychotropic drugs are effective because they block these re-uptake processes by, for example, interfering with action of one or more transporters. The administration of such drugs to block re-uptake prolongs and enhances the action of neurotransmitters such as the biogenic amines. γ-Aminobutyric acid (GABA) is a major inhibitory neurotransmitter in the mammalian brain, with exceptionally high levels found in the substantia nigra, globus pallidus, and hypothalamus. Fahn et al., J. Neurochem. JL5:209-213 (1968); and Perry et al., J. Neurochem. 1}:513-519 (1971). Rapid and efficient termination of GABA neurotransmission is achieved, in part, by a high affinity, Na⁺-dependen , presynaptic GABA transport (re-uptake) process. Iversen et al., J. Neurochem. 15:1141-1149 (1968); Iversen, J. Pharmacol. 11:571-591 (1971); Iversen et al., J. Neurosci. 18:1939-1950 (1971); and Gottesfeld et al., J. Neurochem. 18_.:683-690 (1971). A similar GABA transport process has been identified in glia, and is thought to also function in signal termination as well as in control of neuronal excitability. Schrier et al., J. Biol. Che . 249:1769-1780 (1974); Schousboe et al., Neurochem. Res. 2.:2l7-229 (1977); and Wilkin et al., Dev. Brain Res. 10:265-277 (1983).

GABA transport systems have often been classified as being either glial or neuronal on the basis of their pharmacological sensitivities to certain GABA uptake inhibitors. Neuronal GABA transport is effectively inhibited by σis-3-aminocyclohexanecarboxylic acid (ACHC) , Bowery et al., Nature 264:281-284 (1976), and L-2,4-diaminobutyric acid (L-DABA) . Iversen et al., Biochem. Pharmacol. 2_.:933-938 (1975); and Larsson et al., Brain Res. 260:279-285 (1983). GABA transport processes in both central and peripheral glia, Schon et al., Brain Res. 66:289-300 (1974); Schon et al., Brain Res. 86:243-257 (1975); and Gavrilovic et al., Brain Res. 303:183-185 (1984) are generally characterized by their ability to transport 0-alanine, and their sensitivity to inhibition by this GABA analog. Hence, on the basis of pharmacological responses, there appear to be at least two significantly different types of GABA transport function operable in the central nervous system.

GABA transport can also exhibit a differential characteristic not explained by pharmacological differences. Studies of a variety of glial cell types in primary culture, including rat retinal Muller ^;ells (Iversen et al., Biochem. Pharmacol. 4_.:933-938 (1975)), cerebellar stellate astrocytes (Cummins et al.. Brain Res. 239:299-302 (1982); and Levi et al., Dev. Brain Res. 10:227-241 (1983)), and oligodendrocytes (Reynolds and Herschowitz, Brain Res. 371:253-266 (1986)), have shown that not all glial GABA transport activities are sensitive to ,9-alanine, nor is _/3-alanine a substrate for all glial GABA transport systems. In fact, GABA transport in rat retinal Muller cells and cerebellar stellate astrocytes is sensitive to the putative neuronal transport-selective agents ACHC and L-DABA. Iversen et al., Biochem. Pharmacol. 2.:933-938 (1975); and Levi et al., Dev. Brain Res. 1.0:227-241 (1983). The GABA transport studies discussed above showing multiple distinctive GABA transport activities indicate that more than one type of GABA transporter is operable in mammalian CNS tissue. Hence, we believed, it was probable that more than one GABA transporter structural gene was present in the genomes of, at least, mammalian cells. Finally, since previously cloned transporter cDNAs had nucleotide sequences that were substantially different in a number of regions (Pacholczyk et al., Nature 350:350-354 (1991); Kilty et al., Science 254:578-579 (1991); and Yamauchi et al., J. Biol. Chem. 267:649-652 (1992)), there was no reason to expect that knowledge of the gene sequence of one type of GABA transporter gene (see Guastella et al., Science 249:1303-1306 (1990), describing the cloning of the ACHC-sensitive GABA transporter, also termed the "GAT-A" transporter) would provide a reliable prediction of the structure and properties of any other GABA transporter gene.

Elucidating the cellular and molecular bases for different substrate and inhibitor specificities of various GABA transport systems has heretofore been hindered in part by the inability of researchers to isolate cells having only a single species of GABA transporter. For example, neural cells typically have multiple species of transporters and/or produce interfering enzymes. Studies with such cells require complicated kinetic studies and/or blocking protocols in an attempt to isolate the behavior of the transporter of interest. Thus, it would be advantageous to have cloned cDNAs encoding such different transport systems. Thus, the foregoing hindrances could be overcome because such cDNAs can be transfected into other cells, including non-neural cells, in which studies of GAT-B expression and function can be more carefully controlled.

The predominant mechanism by which GABAergic neurotransmission in the central nervous system (CNS) is terminated appears to be the Na⁺-dependent reuptake of GABA into presynaptic terminals and adjacent glia.

Iverson et al., Biochem. Pharmacol. 4:933-938 (1975); and Balcar et al., Neurochem. Res. 4.:339-354 (1979). Glial transport activities may have the additional important role of limiting the diffusion of GABA into adjacent synapses present on GABA-responsive neurons widely distributed throughout the CNS. Because disrupted GABAergic neurotransmission has been implicated in a number of neurological and psychiatric disorders including epilepsy (Meldrum, Int. Rev. Neurobiol. .17:1-36 (1975); Ribak et al. , Science 205:211-214 (1979); and Loscher et al., J. Neurochem. 1*5:1322-1325 (1986)) and schizophrenia (Reynolds et al., Biol. Psych. 22:1038-1044 (1990); and Benes et al.. Arch. Gen. Psvch. 48:996-1001 (1991)), it would be advantageous to acquire a detailed understanding of GABA transporters both as candidate etiologic sites of these disorders and as potential sites for therapeutic intervention.

In view of the foregoing, an object of the present invention is to provide an isolated DNA encoding the GAT-B transporter.

Another object is to provide a cloned DNA encoding the GAT-B transporter.

Another object is to provide a complementary DNA encoding the GAT-B transporter.

Another object is to provide nucleotide sequences that hybridize to GAT-B DNA, thereby enabling the construction of probes specific for polynucleotides containing GAT-B-specific sequences.

Yet another object is to provide a way to render cells, particularly non-neural cells, capable of transporting GABA, thereby allowing the structure and function of GAT-B to be unambiguously studied.

SUMMARY OF THE INVENTION The foregoing objects are achieved by various aspects of the present invention. According to one aspect, an isolated DNA molecule is provided encoding a GAT-B transporter. Said isolated DNA molecule encodes a GABA transport activity that is sensitive to /3-alanine; that is, the GABA transport activity is lower in the presence of /3-alanine than in the absence of β-alanine. In addition, said DNA molecule encodes a GABA transport activity that is expressed predominantly in neurons and to a lesser extent, if at all, in glia. This GABA transport activity also exhibits other differential sensitivity behavior that distinguishes the activity from other types of GABA transport activities.

According to another aspect of the present invention, a cloned DNA is provided which encodes a GAT- B transporter. Such a cloned DNA is amenable to incorporation in a vector, such as an expression vector, thereby allowing transfection of GAT-B-encoding DNA into other cells, including non-neural eucaryotic cells and even procaryotic cells. According to another aspect of the present invention, a cDNA encoding GAT-B is provided. The present invention also comprehends the nucleotide sequence of said GAT-B cDNA which comprises an open- reading frame encoding a protein herein termed the GAT-B protein. Hence, the present invention comprehends the GAT-B protein that is produced via translation of the GAT-B cDNA and the a ino-acid sequence of said GAT-B protein.

Since the present invention comprehends the nucleotide sequence of GAT-B cDNA, nucleotide sequences complementary to either strand of GAT-B cDNA are also comprehended, including nucleotide sequences that are substantially homologous to either strand of GAT-B cDNA and nucleotide sequences that hybridize to GAT-B cDNA under what are known in the art as "stringent conditions." Thus, said sequences that hybridize to GAT-B cDNA include DNA and RNA molecules that deviate from absolute complementarily to GAT-B cDNA by minor sequence changes, including nucleotide substitutions, deletions, and additions. Hence, such sequences would include, but not limited to, "degenerate primers" and the like. Such sequences can also be labeled to permit their use as probes of GAT-B-specific DNA sequences, and GAT-B-specific RNA sequences. The present invention also comprehends host cells, either procaryotic or eucaryotic, transfected by GAT-B-specific DNA or RNA according to the present invention. The present invention also comprehends methods for rendering non-GAT-B-producing eucaryotic cells capable of producing GAT-B transporter, involving coupling a DNA encoding the GAT-B transporter to a functional promoter in an expression vector which can be transfected into a susceptible eucaryotic host cell; transfecting the host cell with the GAT-B DNA-containing vector; then culturing the cell under conditions conducive for transcription and translation of the GAT-B DNA. The present invention also encompasses the rendering of such transfected cells capable of transporting GABA.

The foregoing and other features and advantages of the present invention will become more apparent from the following detailed description and accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 provides the nucleotide sequence of the GAT-B cDNA according to the present invention as well as an amino-acid sequence for the corresponding GAT-B protein as deduced from the nucleotide sequence.

FIG. 2 shows the amino-acid sequence of GAT-B aligned with the amino-acid sequences of GAT-A, DAT (dopamine transporter) , 5HT-T (serotonin transporter) , canine BGT-1 (betaine transporter) , and human NET (norepinephrine transporter) .

FIG. 3A shows time-course and Na⁺- and Cl^~- dependence of radiolabeled GABA transport by Int 407 cells transfected with GAT-B cDNA. FIG. 3B shows the results of saturation analyses of transport of radiolabeled GABA by Int 407 cells transfected with GAT-B cDNA, wherein the inset is an Eadie-Hofstee plot of initial transport velocity data. FIG. 4A is a plot illustrating the sensitivity of uptake of GABA by Int 407 cells expressing GAT-B cDNA in the presence of various transport inhibitors.

FIG. 4B is a plot illustrating, for comparison with FIG. 4A, the sensitivity of uptake of GABA by Int 407 cells expressing the prior-art transporter GAT-A in the presence of various transport inhibitors.

FIG. 5 is a photograph of the results of a Northern blot analysis of the expression of GAT-B mRNA in various tissues of the rat central nervous system, wherein radiolabeled KW13 DNA (a 681-bp fragment amplified from rat midbrain cDNA by polymerase chain reaction (PCR)) was employed as a probe.

FIG. 6 is a series of direct-reverse prints showing the localization of GAT-B mRNA in various tissues of the rat central nervous system by in situ hybridization histochemistry.

FIGS. 7A-7E are photographs obtained from in situ hybridization experiments showing selective localization of GAT-B to particular cells and regions of the rat central nervous system.

DETAILED DESCRIPTION Nucleotide Sequence and Amino Acid Sequence of GAT-B A GABA transporter cDNA (complementary DNA) , herein termed GAT-B cDNA, was isolated from a rat midbrain genomic library using a sequence similarity- based PCR (polymerase chain reaction) strategy known by persons skilled in the art. Then, GAT-B cDNA was cloned using standard molecular biology techniques. Sambrook et al. , Molecular Cloning: A Laboratory Manual.

Vol. 1-3, Cold Spring Harbor Laboratory, N.Y. (1989).

It will be understood that the "GAT-B" GABA transporter is distinct from the neuronal GABA transporter known in the prior art as "GAT-A," as described in detail hereinbelow. GAT-B is a novel high- affinity Na⁺- and Cl^'-dependent GABA transport protein. The "B" suffix signifies the J-alanine sensitivity of this transporter.

As used herein, to be "isolated," a DNA must be removed from its original environment. For example, a naturally occurring DNA molecule present in a living organism is not isolated, but the same DNA molecule, separated from some or all of the coexisting materials in the natural system, is regarded as "isolated."

As used herein, "cloned" DNA is a population of substantially identical DNA molecules. According to conventional cloning practice (Sambrook et al., id.), a desired segment of DNA is cloned by introducing a length of DNA including the desired segment and a replication origin into a susceptible host organism such as a type of susceptible bacterium in which the sequence of DNA can undergo replication. The transfected bacterium is then cultured under conditions in which large numbers of genetically identical bacterial cells containing the sequence of DNA are obtained. Cloning of DNA is usually followed by isolating the DNA from the host organism. Generally, a "cDNA" is a piece of DNA lacking internal, non-coding segments and regulatory sequences which govern transcription. cDNA can be synthesized in the laboratory by reverse transcription from messenger RNA (mRNA) extracted from cells using techniques well- understood by skilled artisans.

DNA sequences as provided herein utilize base nomenclature as set forth in 37 C.F.R. § 1.822.

"PCR," as understood by skilled artisans, is a technique for selectively increasing the amount of DNA from a particular gene sequence, relative to other gene sequences, in a sample of DNA. PCR employs cycles of denaturation, annealing with primers, followed by extension of the primers as catalyzed by a DNA polymerase. See generally. Innis et al. (eds.), PCR Protocols, Academic Press (1990) .

For PCR, degenerate oligonucleotide primers were synthesized corresponding to transmembrane domains II and VI of the norepinephrine transporter (NET) , Pacholczyk et al., Nature 350:350-354 (1991), and the ACHC-sensitive GABA transporter (GAT-A) found in the central nervous system, Guastella et al., Science 249:1303-1306 (1990). (See also. FIG. 2 which shows various transmembrane domains of NET and GAT-A and other transporters labeled with roman numerals.) The primers had the following sequences:

Sense: CCGCTCGAGAAGAACGG(C/T)GG(C/T)GG(C/T)GC (C/T)TTC(C/T)T(G/A)AT(C/T)CC(A/G)TA

Antisense: GCTCTAGAAA(G/A)AAGATCTG(G/A)GT(G/T)G C (G/A)GC(G/A)TC(G/A/C)A(G/T)CCA

The "degenerate" character of these primers is reflected by the fact that more than one base can exist in locations indicated by parentheses in the sequences above. For example, "(C/T)" denotes that either of the bases C (cytosine) or T (thy ine) can exist at the particular location in the sequence occupied by the parentheses. Similarly, "(A/G)" denotes that either of the bases A (adenine) or G (guanine) can exist at the particular location in the sequence occupied by the parentheses.

The oligonucleotide primers and rat midbrain cDNA were used in PCR reactions performed using the thermostable Taq polymerase (Promega, Madison, Wise). Twenty-five cycles were performed at 94°C for 1 min, 45°C for 2 min, and 72°C for 3 min, with the final extension period lengthened to 12 min.

KW13, a 681-bp fragment of DNA that was produced by the PCR, was isolated from other PCR products using conventional DNA isolation and purification methods. KW13 was labeled with [³²P]dCTP (Amersham, Arlington Heights, 111.) using a random primer kit (Boehringer Mannheim, Indianapolis, Ind.), and used to screen a random-primed rat midbrain cDNA library constructed in the vector Bluescript SKII(-) (herein designated pBSSKII(-)) (Stratagene, La Jolla, Calif.), as previously described. Pacholczyk et al., Nature 150:350-354 (1991). The rat midbrain cDNA library was derived from a single fraction of size-fractionated poly(A)⁺ RNA which had been previously shown to have both GABA and L-glutamate transport activity when injected into Xenopuε oocytes. Blakely et al. , J. Neurochem. 56:860-871 (1991). Nitrocellulose filter lifts were made from ten plates of about 1,000 to 5,000 individual clones each, resulting in a total library size of about 30,000 clones. Filter lifts were processed according to standard protocols. Sambrook et al. , Molecular Cloning: A Laboratory Manual. Vol. 1-3, Cold Spring Harbor Laboratory, N.Y. (1989).

Screening was performed in hybridization buffer (50% formamide, 6X SSPE, 0.05X BLOTTO) at 42°C. (One liter of 6X SSPE is an aqueous solution of 52.59 g NaCl, 8.28 g NaH₂P0₄-H₂0, and 2.22 g EDTA per liter; pH 7.4. BLOTTO is Bovine Lacto Transfer Technique Optimizer, lx BLOTTO is an aqueous solution of 5% nonfat dried milk with 0.02% sodium azide.) Between 5 x 10⁵ and 1 X 10⁶ cpm of [³²P]-labeled KW13 were added per mL of hybridization buffer, and filters were incubated until 1 x C₀t_1/2 was achieved. Filters were washed 4 times for 5 min each in 2X SSPE/0.1% SDS (sodium dodecyl sulfate) at 25°C, air dried, and exposed to XAR-5 film (Kodak, Rochester, N.Y.) for 16 hr.

The screening yielded eight positive clones, five of which proved to be unique by restriction-enzyme mapping and sequencing of both the 5' and 3' ends of each clone. Positive clones were confirmed by Southern blot analysis using the KW13 fragment labeled with

[³²P]dCTP as described above. A single clone, designated 4C4a-l, was found to encode a potential start site and had no stop codon downstream. 4C4a-l also had a stretch of bases homologous to the Kozak consensus sequence (Kozak, Nucl. Acids. Res. 12:857-872 (1984)) preceding an in-frame methionine. Restriction digest analyses of 4C4a-l, and of a second clone designated 2C4-lc-2, revealed that both shared a unique Pst I site located within an 889-base pair overlap in the KW13 region of these two clones. After digesting with Pst I, the 3' end of 4C4a-l was ligated to the 5' end of 2C4-lc-2, resulting in the construction of a plasmid (designated pGAT-B) with a complete open reading frame as deduced by functional expression studies.

As used herein, an "open reading frame" is a region of DNA that contains a series of nucleotide triplets ("codons") encoding amino acids without any termination codons. Open reading frames are usually translatable into protein. DNA Sequencing and Analysis

Sequencing was performed using the dideoxy-chain termination technique known in the art, employing the "Sequenase 2.0" kit (United States Biochemical Corp., Cleveland, Ohio) on a set of overlapping exonuclease Ill-digested EcoR I, Sal I unidirectional deletions ("Erase-a-Base"; Promega, Madison, Wise). Double-stranded templates were sequenced using the vector primers T7 and T3. The sequence of the antisense strand was compiled from the sequences of the partial clones 4C4a-l and 2C4-lc-2. To confirm this sequence, pGAT-B was excised from the Xho I site of pBSSKII(-), relegated in the opposite orientation, and sequenced as described above. A region of poor resolution on the 5' end of the sense strand was unambiguously resolved on the other strand. "MacVector" DNA analysis software (IBI, New Haven, Conn.) was used for sequence assembly and analysis.

The GAT-B cDNA clone, as shown in FIG. 1, has a total of 2,063 bases. GAT-B includes a 1,897-base open reading frame defined by an initial ATG codon within a region that agrees well with the consensus sequence for translational start sites. Kozak, Nucl. Acids. Res.

±2.'857-872 (1984) . From the cDNA sequence, we deduced a protein having 627 amino acids and a relative molecular mass of 70,000 (M_r = 70K) .

Significant features in the deduced GAT-B polypeptide sequence include twelve putative transmembrane regions, shown within brackets in FIG. 1. A large extracellular loop is apparent between the third and fourth transmembrane domains wherein three potential sites for N-linked glycosylation are designated by asterisks. The lack of a strong candidate for an amino- terminal signal sequence (von Heijne, Eur. J. Biochem. 133:17-21 (1983)) suggests that both the amino-terminus and the carboxy-terminus of the protein are located in the cytoplasm. As shown in FIG. 2, the twelve transmembrane domains appear to be shared by other Na⁺- dependent transporters, including GAT-A (Guastella et al., Science 249:1303-1306 (1990)), norepinephrine transporter (NET) (Pacholczyk et al. , Nature 350:350-354 (1991)), dopamine transporter (DAT) (Kilty et al., Science 254:578-579 (1991)), and serotonin transporter (5HT-T) (Blakely et al., Nature 354:66-70 (1991)). FIG. 1 also shows, within the predicted intracellular portions of the carboxy terminus of GAT-B, potential sites for phosphorylation by Ca²⁺-calmodulin- dependent protein kinase II, each denoted by a triangle (Δ) . FIG. 1 also shows residues potentially phosphorylated by protein kinase C (PKC) , as denoted by double underline. Kemp et al., Tr. Biochem. Sci. 15:342-346 (1990). Studies using cultured primary astrocytes have suggested that activation of various second messenger systems is involved in the regulation of GABA transport. Rhoads et al., Biochem. Biophvs. Res. Comm. 119:1198-1204 (1984); Bouhaddi et al.,

Neurochem. Res. 13_.:1119-1124 (1989); Hansson et al., Life Sci. 11:27-34 (1989); and Gomeza et al., Biochem. J. 275:435-439 (1991). Such modulatory effects on astrocytic GABA transport, as well as the presence of potential phosphorylation sites in GAT-B (and GAT-A) , imply that intracellular signalling mechanisms and protein kinases may regulate GABA transport processes in neurons.

Turning to FIG. 2, the deduced amino-acid sequence of Rat GAT-B is aligned with deduced amino-acid sequences of GAT-A, DAT, 5HT-T, BGT-1 (canine betaine transporter) , and human NET. (The boxes enclose amino acids that are identical in all the transporter sequences shown or that are unique to the three GABA transporters shown; i.e., to GAT-B, GAT-A, and BGT-1.) The alignment of transmembrane domains (brackets numbered with Roman numerals) indicates a significant similarity in hydropathy profiles (Kyte et al., J. Mol. Biol. 157:105-132 (1982)) of these various transporters.

Another feature that GAT-B shares with certain other neurotransmitter transporters is a cluster of charged residues in the intracellular loop immediately preceding transmembrane domain IX. Clusters of charged residues are also found in a similar location in members of another symporter family which includes the Na⁺/glucose and Na⁺/proline transporters (Hediger et al., Proc. Natl. Acad. Sci. USA 86:5748-5752 (1989)), raising the possibility that these charged residues are involved in a Na⁺-cotransport process common to all Na⁺-dependent transporters.

Comparison of the amino acid sequence for GAT-B with that of GAT-A shows that these two transporters are about 50 percent identical with respect to amino acid residues, with an overall 61 percent similarity when conservative amino acid substitutions are considered (FIG. 2) .

The GAT-B protein is most similar to another member of the same transporter family, i.e., the Na⁺- dependent betaine carrier (BGT-1) from canine kidney (Yamauchi et al., J. Biol. Chem. 267:649-652 (1992)) which transports GABA as well as betaine. However, BGT-1 transports GABA and betaine with relatively lower affinities (K^ = 93 μM and 398 μM, respectively) than are generally observed for amino-acid neurotransmitter carriers in brain. Specifically, GAT-B appears to share about 63 percent identity with BGT-1, increasing to 72 percent with conservative amino acid substitutions are considered.

As indicated in FIG. 2, fifty residues unique to the three GABA transporter sequences (GAT-B, GAT-A, and BGT-l) are interspersed throughout the amino-acid sequences. These uniquely conserved residues include a number of uncharged polar residues which may be important for recognition by these transporters of common substrates and inhibitors, such as by hydrogen bonding.

In addition to residues uniquely conserved in the transporters that use GABA as a substrate, approximately one-fourth of all amino acid residues are identical in all the transporter proteins shown in FIG. 2. The greatest homology is found within the transmembrane domains, suggesting that these regions play an important role in functions common to all Na⁺- dependent neurotransmitter transporters. Less homology was seen in the intracellular and extracellular loops connecting the transmembrane domains. For example, a notable region of divergence was found in the large, glycosylated extracellular loop connecting transmembrane regions III and IV in each of the transporter sequences. Since this loop is unique for each transporter, it may contribute to substrate and/or inhibitor specificities. GAT-B contains proline residues in six of its twelve hydrophobic (i.e., transmembrane) domains. The majority of the proline residues appear to be conserved in all members of this neurotransmitter transporter family. It is noted that proline residues have been found in the transmembrane domains of many transport proteins and are presumed to facilitate the conformational changes necessary for transport of substrates across a lipid bilayer. Brandl et al., Proc. Natl. Acad. Sci. USA 83:917-921 (1986). Proline residues also introduce bends in membrane-spanning regions and generate domains that traverse the membrane multiple times before exiting the bilayer.

The cDNA of the present invention now permits researchers to formally test structural predictions for transporters and to examine the roles that various regions of the protein play in neurotransmitter transport processes. The GAT-B cDNA also permits the synthesis, using conventional techniques, of nucleotide probes (either DNA or RNA) exhibiting substantial homology with GAT-B DNA sequences. As will be understood by skilled artisans, such probes can be tagged (i.e., labeled) with radioactive labels, fluorescent labels, immunoreactive labels, electron-dense labels, and other diagnostic labels.

A search of the GenBank database revealed that GAT-B has no extended regions of sequence similarity to other proteins, including receptors, ion channels, other symporters or facilitated carriers. pGAT-B Encodes a g-Alanine-Sensitive GABA Transporter

Transport assays were performed using intestine 407 cells (Int 407; ATCC) , which are human embryonic intestine cells that possess the four HeLa markers as well as the A-subtype of glucose-6-phosphate dehydrogenase. The cells were cultured to a density of 1.5-2.0 x 10⁵ cells per well and infected with recombinant vaccinia virus strain VTF-7 encoding a T7 RNA polymerase. Fuerst et al., Proc. Natl. Acad. Sci. USA 83.:8122-8126 (1986). Infection was followed 30 minutes later by liposome-mediated transfection with pGAT-B or pGAT-A. (pGAT-A is also referred to in the art as pGAT-1. Guastella et al., Science 249:1303-1306 (1990)). pGAT-A was isolated as described in Blakely et al., Anal. Biochem. 194:302-308 (1991)). For transfection, pGAT-B and pGAT-A were separately diluted in pBSSKII(-), yielding individual solutions of 100 ng pGAT-A or pGAT-B + 900 ng pGSSKII(-) ("Lipofectin," BRL, Gaithersburg, Md.). Blakely et al., id. Control transfections were done with equivalent amounts of vector alone. Assays were performed 8 hours after transfection in a modified Krebs-Ringer-HEPES buffer (Blakely et al., id.). Cells were incubated with

15-30 nΛf [³H]GABA and the GABA transaminase inhibitor aminooxyacetic acid (100 μM) , with or without cold GABA transport inhibitors, for 20 minutes (pGAT-B) or 30 minutes (pGAT-A) at 37°C. Uptake of GABA by the cells was stopped by placing the cells on ice and washing them twice with 1 mL ice cold assay buffer. Cells were solubilized in 1% SDS, and the amount of radioactivity accumulated was determined by liquid scintillation counting. NA⁺- and Cl^'-dependence were determined by isotonic substitution of lithium chloride and sodium glucuronate, respectively, for sodium chloride in the assay buffer. Background levels of GABA transport were determined using control transfections of cells with pBSSKII(-) for each assay, and the values obtained were subtracted from the signals determined for pGAT-A and pGAT-B.

FIG. 3A shows time-course and Na⁺- and Cl^~- dependence of [³H]GABA transport by Int 407 cells transfected with pGAT-B. As can be seen, transient expression of pGAT-B in these cells resulted in a significant increase in cellular [³H]GABA accumulation above the background levels of cells transfected with vector alone. Isotonic substitution of LiCl for NaCl in the assay buffer resulted in a decrease in GABA uptake below that seen in cells infected with the vector alone, thereby revealing the presence of a low-level, endogenous, Na⁺-dependent GABA transport in these cells. Reduction of Cl^" concentration in the assay buffer by replacement of NaCl with sodium glucuronate, on the other hand, decreased GABA transport by these cells to background levels. Therefore, GABA transport resulting from pGAT-B expression in these cells is both Na⁺- and Cl^'-dependent.

FIG. 3B shows the results of saturation analyses of [³H]GABA transport by the Int 407 cells. The inset is an Eadie-Hofstee plot of initial velocity data. (Note: data shown in FIGS. 3A and 3B represent the results of at least three separate experiments, and each datum point in FIGS. 3A and 3B is the mean of said at least three determinations with a standard error of the mean of < 10 percent) . FIG. 3B shows that transport of GABA by GAT-B is saturable, with an apparent Michaelis constant (K of 2.3 μM) . No GABA transport above background was detected when [³H]L-glutamate (100 nM) , [³H]D-aspartate (100 nM) , [³H]L-glycine (100 nM) , or [³H]/3-alanine (up to 500 μM with cold 0-alanine) was substituted for GABA in the uptake assay (data not shown) .

The differential sensitivity of GAT-B or GAT-A transport to a variety of agents was examined in HeLa cells expressing these proteins. The results of these studies, presented in FIGS. 4A and 4B and in Table I, indicate that these two transport processes have very different pharmacological profiles.

TABLE I

Inhibitor Sensitivity of GABA uptake in HeLa Cells Transfected with pGAT-B or pGAT-A Transporter cDNA

K tf

In Table I, K_A values are presented as means ± one standard error of the corresponding mean. The mean values each represent an average of at least three separate experiments. Each K_ value was calculated from an IC₅₀ determination:

_c concentration of xadioligand ^{where S =} K_D of xadioligand ^{• Chen}^ ^{et al}- '

Biochem. Pharmacol. 22:3099-3108 (1973). FIG. 4A shows the sensitivity to various agents of [³H]GABA uptake by cells expressing GAT-B; FIG. 4B shows the sensitivity to the agents of [³H]GABA uptake by cells expressing GAT-A. Curves are representative of data obtained in three separate experiments, wherein each datum point is the mean of the three corresponding determinations.

Based on its ability to potently inhibit GABA transport, and serve as a substrate for transport in glia of rat sensory ganglia and cerebral cortical slices (Schon et al., Brain Res. j$ :243-257 (1975)), _/3-alanine has been assumed in the art to be a glial-selective agent. Although 3-alanine does not appear to be transported by GAT-B, it is a potent and selective inhibitor of GAT-B-mediated GABA transport. Referring to FIGS. 4A and 4B and Table I, 8-alanine was three orders of magnitude more potent at inhibiting GABA transport by GAT-B than by GAT-A (K_A = 6.7 μM and 2 mM, respectively) . Nipecotic acid, a nonselective competitor for GABA transport, was an order of magnitude more potent at inhibiting GAT-A transport than GAT-B transport. Guvacine and cis-4-hydroxynipecotic acid, compounds reported to be slightly more effective as competitors of GABA transport in astrocytes than in neurons (Krogsgaard-Larsen et al.. Epilepsy Res. 1:77-93 (1987)), showed similar differential potencies of inhibition of GAT-A transport relative to GAT-B transport.

In Table I, THPO showed similar K_A values for the inhibition of GAT-B and GAT-A (Ki = 233 μM and 378 μM, respectively) . Taurine has also been associated with GABA transport in both neurons and glia, although it is not clear whether taurine is transported by the same re- uptake system. Kaczmarek et al., J. Neurochem. 19:2355-2362 (1972); Schrier et al. , J. Biol. Chem. 249:1769-1780 (1974); Martin et al., J. Biol. Chem. 254:7076-7084 (1979); Borg et al., J. Neurochem.

11:1113-1122 (1980); and Holopainen et al., Neurochem. Res. 11:207-215 (1986). Taurine was a weak competitor for GABA transport by GAT-B, but did not inhibit GAT-A transport.

Table I and FIGS. 4A and 4B also include data pertaining to GAT-A and GAT-B inhibition by ACHC and L-DABA. Both these compounds are considered to be selective competitors of neuronal GABA transport. Bowery et al., Nature 264:281-284 (1976); Larsson et al., Brain Res. 260:279-285 (1983); Iversen et al., Biochem. Pharmacol. £1:933-938 (1975); and Krogsgaard- Larsen et al., Epilepsy Res. 1:77-93 (1987). ACHC and L-DABA were more potent at inhibiting GABA transport by GAT-A than by GAT-B (Ki = 15.4 μM and 28 μM versus 813 μM and 109 μM, respectively) . N0-711, a highly potent GABA uptake inhibitor with anticonvulsant properties (Suzdak et^'al., Eur. J. Pharmacol. In Press (1992)) showed the most marked selectivity between the two transporters: transport mediated by GAT-A was nearly five orders of magnitude more sensitive to inhibition by N0-711 than was transport mediated by GAT-B. Expression of GAT-B mRNA in Rat CNS

Studies of GAT-B mRNA expression in central nervous tissue of the rat were performed using the "Northern blot" methodology and in situ hybridization histochemistry methods as known in the art. To perform Northern blot analysis, the KW13 cDNA was labeled with [³²P]dCTP using random-prime synthesis and used as a probe to examine the tissue distribution of GAT-B mRNA. Poly(A)⁺ RNA was isolated from various regions of rat brain using the "Mini RiboSep" (Collaborative Research Inc., Bedford, Mass.) and "Fast Track" kits (Invitrogen, San Diego, Calif.), and from cell lines using standard protocols. Sambrook et al. , id. Brain dissections were performed as described in Blakely et al., J. Neurochem. E>6_:860-871 (1991), and Deutch et al., Brain Res. 333:143-146 (1985).

Approximately 1 μg of poly(A)⁺ RNA was separated on a 1% formaldehyde-agarose gel (Ogden, et al., Methods Enzvmol. 152:61-87 (1987)), and vacuum-transferred (LKB, Piscataway, N.J.) to a nylon membrane ("Zeta-probe," BioRad, Richmond, Calif.). Hybridization was carried out at 42°C in 50% formamide. Blots were washed twice at room temperature for 20 min each in 2X SSPE/0.1% SDS, followed by a one-hour wash at 65°C in 0.2X SSPE/0.1% SDS.

To perform in situ hybridizations, serial coronal sections through the rat neuraxis were cut at 16μm and thaw-mounted onto subbed slides. Sections were fixed by immersion in 4% paraformaldehyde/O.lM sodium phosphate buffer (pH 7.4) for 10 min, and then washed twice in phosphate buffer. Sections were then acetylated in 0.25% acetic anhydride/0.25% triethanolamine (pH 8.0) for 10 min, and subsequently dehydrated and dilapidated, then rehydrated and dried at room temperature.

Tissue sections were prehybridized for 2 hr at 37°C. The prehybridization buffer consisted of 50% formamide, 10 mM Tris (pH 8.0), 1 mM EDTA, 0.1 M DTT (dithiothreitol), 0.3 M NaCl, IX Denhardt's solution and

10% dextran sulfate. Sections were then spotted with pre-hybridization buffer containing either of the following 3'-end [³⁵S]-labeled "antisense" or "sense"

45-mer oligonucleotides: Antisense: CATTGAGATGCGCCTCGATCACCTGTGTTGCATAGC

CGATGCCTT

Sense: AAGGCATCGGCTATGCAACACAGGTGATCGAGGCGCATC TCAATG (500,000 cpm/50 μL) . The sections were then coverslipped with parafilm, and hybridized overnight at 37°C. The slides were washed 4 times at 15 minutes each in 2X SSC/50% formamide at 40°C, followed by 2 x 30 min in IX SSC at 42°C; sections were then dipped briefly in water followed by 70% ethanol and allowed to air dry before being apposed to Hyperfilm (Amersham, Arlington Heights, 111.) The films were exposed for 7 weeks. Separate sets of slides were dipped in 50% Kodak NTB-2 emulsion, placed in light-tight boxes, and exposed for 7 weeks prior to development.

In order to determine if GAT-B mRNA is expressed in neurons or glia, the midbrains of male Sprague-Dawley rats were lesioned with ibotenic acid (7.5 μg in 1.0 μL of 0.1 M PBS (phosphate-buffered saline), pH 7.4, delivered over 10 min) to deplete neurons and increase glial cells. Five days later, animals were perfused with 4% paraformaldehyde, and sections prepared for localization of GAT-B mRNA. Adjacent sections were stained with neutral red.

Northern blot results are shown in FIG. 5, showing the expression of GAT-B mRNA in rat central nervous system (CNS) tissue. In. FIG. 5, "C₆ glioma" is a rat glial tumor cell line; "PC12" is a rat adrenal pheochromocytoma cell line; "SKNSH" is a human neuroblastoma cell line; and "Y79" is a human retinoblastoma cell line.

FIG. 6 comprises direct-reverse prints showing localization of GAT-B mRNA in the rat CNS tissues by in situ hybridization histochemistry. The following abbreviations are used: AON, anterior olfactory nucleus; BST, bed nucleus of the stria terminalis; CER, cerebellum; CG, central gray; CIN, cingulate cortex; CMA, corticomedial amygdala; DCN, deep cerebellar nuclei; DBB, diagonal band of Broca; DHP, dentate gyrus of hippocampus; DMN, dorsomedial hypothalamic nucleus; ENT, entorhinal cortex; IC, inferior colliculus; 10, inferior olivary nucleus; IPN, interpeduncular nucleus; LHB, lateral habenula; LSN, lateral septal nucleus; MB, mam illary body; MSN, medial septal nucleus; NRP, nucleus reticularis paragigantocellularis; POA, preoptic area; PFC, prefrontal cortex; PHP, pyramidal cell layer of hippocampus; PN, pontine nuclei; PVN, thalamic paraventricular nucleus; PYR, pyriform cortex; SC, superior colliculus; SN, substantia nigra; VN, vestibular nuclei; VP, ventral pallidum; VPL, ventroposterolateral thalamus; ZI, zona incerta. Northern blot analyses performed using a PCR fragment corresponding to GAT-B as a specific probe demonstrated a 4.7 kb mRNA species that is expressed in tissues in the nervous system (FIG. 5) . Comparison of the regional distribution of expression of GAT-A and GAT-B mRNAs using Northern blot analysis suggested that there are substantial divergences between the patterns of expression of these two GABA transporters. For example, GAT-A is very abundant in striatum and cerebellular cortex, areas where GAT-B mRNA is essentially absent. Similarly, GAT- A mRNA levels are high in cortex and hippocampus where only low levels of GAT-B are evident. Some areas such as midbrain contain significant levels of both GAT-A and GAT-B.

The in situ hybridization results shown in^~ FIG. 6 demonstrate that GAT-B mRNA is heterogeneously distributed in brain tissue. In particular,_increased labeling was revealed in mediobasal structures in the forebrain, in contrast to the much weaker expression of GAT-B in the cortex and the virtual absence of GAT-B expression in the neostriatum. GAT-B mRNA was extensively distributed in the brainstem, but absent in the cerebellum. In the telencephalon, cortical GAT-B expression was relatively weak, and was most prominent in the temporoparietal regions (where a bilaminar appearance of GAT-B mRNA could be detected) and in the pyriform and entorhinal cortices. No neurons expressing GAT-B were seen in the neostriatum, and there was very weak expression in the ventral striatum (nucleus accumbens- olfactory tubercle) . GAT-B expression was very high along the entire rostrocaudal extent of the medial septum-diagonal band complex. In contrast, there was weaker labeling of neurons in the lateral septum.

Relatively high GAT-B expression was seen in the ventral pallidum, and extended into the medial and lateral preoptic regions. GAT-B expression was moderate in the bed nucleus of the stria terminalis. In the amygdala, GAT-B labeling was essentially restricted to the anterior cortical and medial amygdaloid nuclei.

GAT-B mRNA was expressed throughout the thalamus and epithalamus. Thus, a high density of GAT-B transcripts was seen in the lateral habenula and medially adjacent paraventricular thalamic nucleus. Moderate labeling was observed in the ventroposteromedial and ventroposterolateral nuclei of the thalamus. In the posterior thalamus, moderate labeling was present in the medial geniculate nucleus. The zona incerta was strongly labeled. More ventrally, GAT-B was expressed in a variety of hypothalamic nuclei. The anterior, dorsomedial, ventromedial, arcuate, and lateral hypothalamic nuclei all expressed GAT-B mRNA.

Hippocampal labeling was relatively low; but GAT-B mRNA was expressed in neurons of both the dentate gyrus and pyramidal cell layer.

In the brainstem, GAT-B labeling was seen in a large number of structures in moderate-to-high density. In the mesencephalon, labeling was present in neurons of the superior colliculus, central gray, and substantia nigra; in the latter structure, GAT-B mRNA was expressed both in the pars reticulata and pars compacta, with the dopamine-rich pars compacta being more densely labeled in the rostral substantia nigra. GAT-B expression was very high along the entire rostrocaudal extent of the interpeduncular nucleus. In the pons, GAT-B was strongly expressed in the pontine nuclei, with moderate- to-high degree of expression in the paralemniscal area, central gray, cuneiform, and dorsal nucleus of the lateral lemniscus. The dorsal pontine tegmentum (particularly the dorsal tegmental nucleus) was strongly labeled, as was the parabrachial region, the trapezoid area, and the superior and inferior olivary nuclei. The vestibular nuclei were prominently labeled.

The absence of GAT-B mRNA in neurons of the cerebellar cortex and vermis was in striking contrast to the more ventral brainstem sites. However, the deep cerebellar nuclei were strongly labeled.

FIGS. 7A-7E show the localization of GAT-B mRNA to neurons. Examination of emulsion-coated sections revealed the presence of silver grains over neurons (FIGS. 7A and 7B) ; no glial labeling was evident.

In order to confirm the neuronal expression of GAT-B mRNA, ibotenic acid lesions were made by infusing ibotenic acid into the mesencephalon. Such lesions result in severe neuronal loss with a concomitant reactive gliosis. Kδhler et al., Neuroscience 1:819-835 (1983). Thus, if the GAT-B transporter were expressed in glia, one would expect to observe intense labeling over the lesion, whereas neuronal localization would be reflected by the absence of specific hybridization.

Five days after infusion of ibotenic acid, animals were perfused and brains sectioned for in situ hybridization analyses.

The lesions resulted in a very dense gliosis (X) in the substantia nigra (SN) and overlying mesencephalic reticular formation. The more medial ventral tegmental area (VTA) and mam illary body (MB) were intact on both the lesioned and contralateral (FIG. 7E) sides, as was the myelinated medial lemniscus (ml) . The loss of specific hybridization in the gliotic zone (FIG. 7D) indicates that GAT-B is predominantly expressed in neurons.

Thus, the in situ hybridization studies demonstrated that GAT-B mRNA is distributed predominantly in neurons. While it is conceivable that GAT-B transcripts are present in very low abundance in certain types of glia, there is currently no indication of glial expression of this GABA transporter.

The distribution of GAT-B in CNS tissue is in many cases consistent with the localization of well- characterized populations of GABAergic cell bodies. For example, GABAergic neurons are present in high density in the medial septum-diagonal band complex and in the ventral pallidum, Onteniente et al., J. Comp. Neurol. 248:422-430 (1986), sites in which GAT-B mRNA production is very high. A more detailed comparison of the regional distributions of GAT-A and GAT-B is difficult because GAT-B positive cell bodies have been identified by in situ hybridization (FIGS. 7A-7E) while prior-art reports on the localization of GABA transport proteins has been limited to immunohistochemistry using antibodies generated against purified brain GABA transporter. Radian et al., J. Neurosci. 10:1319-1330

(1990) . However, our data (and data provided to us in a personal communication from Nicholas Breecha at the UCLA School of Medicine, Los Angeles, California) indicate that the patterns of expression of GAT-A and GAT-B are often complementary. For example, in the cerebellum, GAT-A is expressed in Purkinje cells while GAT-B is found in neurons of the deep nuclei.

Heretofore, the characterization of GABA transport processes as either glial or neuronal has been based on the ability of a few selective agents to inhibit transport. However, these pharmacologic distinctions were largely based on studies of GABA uptake in cortical tissue or in specialized glial populations, which would have excluded most GAT-B activity. The identification and cloning of the novel GABA transport activity encoded by pGAT-B provides evidence for a greater heterogeneity of GABA transport in the CNS than hitherto realized.

There are clear pharmacological distinctions between GAT-B transport and glial GABA transport.

Although 3-alanine is a potent inhibitor of GABA uptake by GAT-B, it does not appear to be transported by GAT-B. Thus, GAT-B is distinguished from glial GABA transporters because 3-alanine transport and inhibition by β-alanine of GABA transport have usually been associated with the same transport process in both brain slices and cultured astrocytes. Iverson et al., Biochem. Pharmacol. 21:933-938 (1975); and Schon et al.. Brain Res. 86:243-257 (1975).

Having illustrated and described the principles of the present invention, particularly with respect to isolating the GAT-B cDNA, preparing probes specific for GAT-B mRNA, and using said probes to ascertain the site- specificity in the CNS of GAT-B expression, it should be apparent to persons skilled in the art that the invention can be modified in arrangement and detail without departing from such principles. We claim all modifications that are within the spirit and scope of the appended claims.

Claims

1. An isolated DNA molecule encoding a GAT-B transporter.

2. An isolated DNA molecule encoding a GABA transporter having a GABA transport activity that is lower in the presence of 3-alanine than in the absence of S-alanine.

3. An isolated DNA molecule encoding a GABA transporter having a GABA transport activity that is expressed in neurons but is substantially not expressed in glia.

4. An isolated DNA molecule comprising a nucleotide sequence as disclosed in FIG. 1 and SEQ. ID N0:1.

5. A cloned DNA encoding a GAT-B transporter.

6. A cloned DNA as recited in claim 5 wherein said cloned DNA comprises at least a portion of a DNA vector.

7. A cloned DNA as recited in claim 6 wherein the vector is an expression vector.

8. A cloned DNA comprising a nucleotide sequence as disclosed in FIG. 1 and SEQ. ID N0:1.

9. A cloned DNA as recited in claim 8 wherein said cloned DNA comprises at least a portion of a DNA vector.

10. A cDNA encoding a GAT-B transporter.

11. A cDNA comprising a nucleotide sequence as disclosed in FIG. 1 and SEQ. ID NO:l.

12. A nucleotide sequence exhibiting substantial homology with either strand of a DNA sequence as disclosed in FIG. 1 and SEQ. ID N0:1.

13. A nucleotide sequence as recited in claim 12 comprising RNA.

14. A nucleotide sequence as recited in claim 12 comprising single-stranded DNA.

15. A nucleotide sequence as recited in claim 12 that includes a label attached to said nucleotide sequence.

16. A nucleotide sequence that hybridizes under stringent conditions to either strand of a DNA sequence as disclosed in FIG. 1 and SEQ. ID N0:1.

17. An isolated DNA molecule encoding a polypeptide as disclosed in FIG. 1 and SEQ. ID N0:1.

18. A DNA sequence that hybridizes to at least a portion of a GAT-B gene.

19. A DNA sequence having the identifying characteristics of pGAT-B.

20. A polypeptide encoded by the DNA of claim

1.

21. A polypeptide encoded by the DNA of claim 4.

22. A polypeptide encoded by the cloned DNA of claim 5.

23. A polypeptide encoded by the cloned DNA of claim 8.

24. A polypeptide encoded by the cDNA of claim 10.

25. A polypeptide encoded by the cDNA of claim 11.

26. A cell transfected with the DNA of claim 1.

27. A cell transfected with the DNA of claim 4.

28. A cell transfected with the cloned DNA of claim 6.

29. A cell transfected with the cloned DNA of claim 9.

30. A procaryotic cell comprising the DNA of claim 6.

31. A procaryotic cell comprising the DNA of claim 8.

32. A method for rendering a non-GAT-B- producing eucaryotic cell capable of producing GAT-B, comprising: coupling an isolated DNA encoding GAT-B to a functional promoter in an expression vector such that transcription of the GAT-B-containing vector can be initiated by the promoter in a susceptible non-GAT-B- producing eucaryotic cell; transfecting the GAT-B-containing vector into the susceptible non-GAT-B-producing eucaryotic cell; and culturing the transfected cell so as to facilitate transcription and translation of the GAT-B- containing vector in the cell.

33. A method for rendering a eucaryotic cell normally incapable of GABA uptake able to take up GABA, comprising: coupling an isolated DNA encoding GAT-B to a functional promoter in an expression vector such that transcription of the GAT-B-containing vector can be initiated by the promoter in a susceptible eucaryotic cell^' incapable of GABA uptake; transfecting the GAT-B-containing vector into the susceptible eucaryotic cell; and culturing the transfected cell so as to facilitate transcription and translation of the GAT-B- containing vector in the cell so as to cause the cell to synthesize GAT-B and take up GABA.