DNA Encoding an Insect Gamma-Aminobutyric Acid (GABA) Receptor Subunit Cells Expressing It, and Pesticide Screening Methods Using Such Cells

The present invention relates generally to the ligand-gated chloride channel gene family. More particularly, the present invention relates to the characterization and isolation of an insect .gamma.-aminobutyric acid (GABA) receptor subunit gene from Drosophila melanogaster.

.gamma.-Aminobutyric acid (GABA) is the principal inhibitory neurotransmitter in the vertebrate brain, in the insect central nervous system and at insect neuromuscular junctions (Enna et al., 1986, In Benzodiazepine/GABA Receptors and Chloride Channels: Structural and Functional Properties, Alan R. Liss, Inc., New York, pp. 41-56; and Sattelle, D. B., 1990, Adv. Insect Physiol., vol. 22, pp. 1-113). Mammalian genes encoding at least 5 distinct classes of GABA receptor subunits, i.e., .alpha., .beta., .gamma., .delta., and .rho., have been cloned and characterized (Burr et al., GABA.sub.A receptor subtypes: from pharmacology to molecular biology, 1991, FASEB/J., vol. 5, 2916-2923). Genes encoding vertebrate GABA receptor subunits and the strychnine-binding subunit of the glycine-gated chloride channel comprise a family of homologous ligand-gated chloride channel genes, which is part of a larger superfamily of ligand-gated ion channel genes (Barnard et al., 1987, Trends Neurosci., vol. 10, pp. 502-509). Gene products of this superfamily have a conserved structural organization with four hydrophobic membrane-spanning domains. GABA stimulates chloride ion conductance through the associated chloride ion channel. The predominant effect of GABA is the interaction with a specific receptor protein which results in an increase of chloride ion conductance to produce an inhibition of neuronal firing. Heterologous expression in vitro of different combinations of GABA receptor subunit types (.alpha., .beta., .gamma., .delta., etc.) and subunit isoforms (.alpha.1, .alpha.2, etc.) results in heteromultimeric receptors with differing structure and pharmacology (Schofield, P. R., The GABA.sub.A receptor: molecular biology reveals a complex picture, 1989, Trends Pharmacol. Sci., vol. 10, pp. 476-478; and Burt et al., GABA.sub.A receptor subtypes; from pharmacology to molecular biology, 1991, FASEB/J., vol. 5, 2916-2923). It is thought that the differential expression of subunits in different cell types is the molecular basis of the known pharmacological diversity of GABA receptors in the mammalian central nervous system (Burt et al., 1991, cited elsewhere herein).

Electrophysiological studies of ligand-gated ion currents in insect nerve and muscle cells provide evidence for the existence of chloride channels gated by glutamate, histamine, and taurine, as well as those gated by GABA (Sattelle, D. B., 1990, cited elsewhere herein; and Lummis et al., 1990, Annu. Rev. Entomol., vol. 35, pp. 345-377). Although these findings imply the existence of a large and diverse gene family encoding ligand-gated chloride channels in insects, very little is known of about homologous channels of invertebrates. Only a single gene from insects having a significant degree of structural conservation and amino acid sequence identity with vertebrate ligand-gated chloride channel genes has been described (ffrench-Constant et al., 1991, Proc. Natl. Acad. Sci. USA, vol. 88, pp. 7209-7213). Here, a Drosophila melanogaster cDNA having significant predicted amino acid sequence identity to vertebrate ligand-gated chloride channel genes was isolated and mapped to a genetic locus (Rdl) that confers resistance to cyclodiene insecticides and other blockers of GABA-gated chloride channels. Rdl was shown to encode a GABA receptor subunit by the expression of functional homomultimeric GABA receptors in Xenopus oocytes following injection with Rdl mRNA (ffrench-Constant et al., 1993, Nature, vol. 363, pp. 449-451).

The only other example of a ligand-gated chloride channel gene from an invertebrate species is a GABA receptor .beta.-like subunit gene isolated from the pond snail, Lymnaea stagnalis (Harvey et al., 1991, EMBO/J., vol. 10, pp. 3239-3245). The functional relationship of the product encoded by this gene to vertebrate GABA receptor .beta. subunits was corroborated by the formation of a functional chimeric receptor with properties similar to vertebrate .alpha./.beta. heteromultimers when the gene was co-expressed with a vertebrate .alpha. subunit in Xenopus oocytes.

The .gamma.-aminobutyric acid (GABA) receptor-chloride channel complex mediates synaptic inhibition in both vertebrate and invertebrate nervous systems and is a target site for a variety of drugs, toxicants, and insecticides (Enna et al., 1986 and Sattelle, 1990, cited elsewhere herein). That is, GABA receptors of insects are known to be target sites for chemical agents having insecticidal or pesticidal activity. Furthermore, invertebrate GABA receptors have different pharmacological properties as compared to those of vertebrate GABA receptors. Therefore, the characterization and isolation of an invertebrate GABA receptor subunit gene(s) would be useful in developing screening techniques to identify insect-specific pesticides.

The primary object of the present invention is the isolation and characterization of an invertebrate (i.e., insect) .gamma.-aminobutyric acid (GABA) receptor subunit gene(s).

The present invention provides for the isolation of genomic DNA fragments from Drosophila melanogaster which encode for conserved amino acid sequence elements unique to the ligand-gated chloride channel gene family. A polymerase chain reaction (PCR)-based homology probing strategy was utilized to isolate the genomic DNA fragments of the invention. As described herein, these DNA fragments are designated "LCCHs", which refers to ligand-gated chloride channel gene homologues.

Using the PCR strategy, three discrete amplified sequence elements designated LCCH1, LCCH2 and LCCH3, were isolated. LCCH1, LCCH2 and LCCH3 contained open reading frames and >40% amino acid sequence identity to the corresponding regions of vertebrate ligand-gated chloride channel genes. Genomic DNA clones corresponding to each sequence element were isolated and sequenced, and predicted amino acid sequences corresponding to the second (M2) and third (M3) transmembrane domains of vertebrate genes were analyzed for identity or similarity to known sequences.

In another aspect of the present invention, there is provided the cloning and characterization of a cDNA encoding a novel member of the ligand-gated chloride channel gene family of insects isolated from Drosophila melanogaster, the LCCH3 genomic sequence of the invention. The 1488 nucleotide open reading frame of this cDNA encodes an amino acid sequence having structural features conserved among ligand-gated chloride channel subunit proteins, including four hydrophobic domains capable of forming transmembrane helices (M1-M4), an octapeptide "signature motif" occurring in M2, a large intracellular domain between M3 and M4, and cysteine residues postulated to form a disulfide-bridged loop structure in the extracellular domain. The LCCH3 gene of the present invention exhibited the greatest similarity to GABA receptor .beta. subunit genes. This high level of structural similarity suggests that its encoded protein may be a functional homologue of the .beta. subunit family in the Drosophila melanogaster nervous system.

The DNA sequences and corresponding amino acid sequences encoding the .gamma.-aminobutyric acid (GABA) receptor subunit gene(s) of the invention may be cloned into any suitable expression vector, such as, for example, plasmid DNA, viral DNA including human viruses, animal viruses and insect viruses and bacteriophages to form a recombinant expression system which directs the expression of the subunits of the invention.

The expression system, which directs the expression of the GABA receptor subunit(s) of the invention, in combination with at least one other vertebrate and/or invertebrate GABA receptor subunit gene, can be co-expressed in an appropriate host cell to form a functional recombinant GABA receptor.

A method of screening chemical agents for their effectiveness as pesticides using the .gamma.-aminobutyric acid (GABA) receptor subunit gene(s) of the invention is also disclosed.

FIG. 1 is a schematic diagram of a single ligand-gated chloride channel subunit depicting the inferred transmembrane (M1-M4) and hydrophobic domains, and the expanded M2-M3 region.

FIG. 2 is a composite photomicrographs of Drosophila melanogaster polytene chromosomes labelled with biotinylated DNA probes (Top: left arm of chromosome 3 showing labelling (arrow) of cytogenetic region 75A by a LCCH2 probe; Bottom: X chromosome showing labelling (arrow) of cytogenetic region 13F by a LCCH3 probe).

FIG. 3 is a schematic diagram illustrating positions of intron-exon splice junctions in Lymnaea and vertebrate GABA receptor subunit genes and in LCCH1, LCCH2 and LCCH3.

FIG. 4 illustrates the nucleotide and inferred amino acid sequences of LCCH3.

FIG. 5 is a hydropathy profile of the LCCH3 gene product calculated with a window size of 17 residues.

FIG. 6 illustrates the alignment of the predicted amino acid sequence of the present isolated gene fragment (LCCH3) with those of the Lymnaea stagnalis GABA receptor .beta.-like subunit ("Lymnaea .beta."), the rat GABA receptor .beta.1 subunit ("rat .beta.1"), and the Drosophila melanogaster Rdl locus ("Dm Rdl") using the GCG computer program.

The present invention provides for the isolation of genomic DNA fragments from Drosophila melanogaster. These fragments encode for conserved amino acid sequence elements unique to the ligand-gated chloride channel gene family. A polymerase chain reaction (PCR)-based homology probing strategy was utilized to isolate these genomic DNA fragments. Use of this PCR strategy yielded three DNA fragment sequence elements designated LCCH1, LCCH2 and LCCH3, from Drosophila melanogaster. These sequence elements appear to represent structurally divergent members of the ligand-gated chloride channel gene family. Two of these sequence elements, LCCH2 and LCCH3, are novel sequences.

It is understood that the sequence elements designated LCCH1, LCCH2 and LCCH3 are for purposes of illustration only, and the existence of a diverse family of genes in Drosophila melanogaster that are structurally related to the ligand-gated chloride channel gene family of vertebrates is supported by the present invention.

In another aspect of the present invention, there is provided the cloning and characterization of a cDNA encoding a novel member of the ligand-gated chloride channel gene family of insects isolated from Drosophila melanogaster. The cDNA corresponds to the LCCH3 genomic sequence of the invention. In this context, the novel sequence elements LCCH2 and LCCH3 of the invention exhibit all the conserved features commonly found in members of the family of ligand-gated chloride channel genes. More specifically, the sequence elements of the present invention exhibit: (1) a structural organization encompassing a large N-terminal extracellular domain, (2) four hydrophobic domains (M1, M2, M3, and M4) capable of forming transmembrane helices, as well as a large intracellular domain lying between transmembrane domains M3 and M4, (3) the conserved octapeptide -TTVLTMTT- (SEQ.ID.NO.1) "signature motif" for this gene family found in transmembrane domain M2, and (4) the conserved cysteine residues and associated sequence elements capable of forming a disulfide-bridged loop structure in the large extracellular domain.

In one embodiment of the present invention, there is provided an isolated gene and/or gene fragment or portion thereof comprising a DNA molecule encoding an .gamma.-aminobutyric acid (GABA) receptor subunit from Drosophila melanogaster. Preferably, the DNA molecule of the invention encodes for an amino acid sequence, or mutant thereof, corresponding to LCCH3, as shown in FIG. 4, SEQ. ID. NO. 2. The DNA molecule of the present invention preferably comprises a nucleotide sequence, or a mutant DNA sequence thereof, corresponding to LCCH3, as shown in FIG. 4, SEQ. ID. NO. 3.

In still another embodiment of the invention, the DNA molecule encodes for an amino acid sequence, or mutant thereof, corresponding to LCCH2, as shown in FIG. 1, SEQ. ID. NO. 16.

It is understood that any modifications i.e., insertions, deletions, mutations, recombinants, etc., of the DNA nucleotide and/or corresponding amino acid sequence(s) are within the scope of the present invention provided that the modified sequence(s) encode for a gene, its homologs or a fragment thereof producing an .gamma.-aminobutyric acid (GABA) receptor subunit from Drosophila melanogaster. Preferably, this receptor subunit exhibits pharmacological properties of .gamma.-aminobutyric acid receptors in insect tissue.

Recombinant DNA techniques are used to insert the DNA sequences of the invention (e.g. GABA receptor subunit from Drosophila melanogaster) into an appropriate expression vector, i.e., a vector which contains the necessary elements for the transcription and translation of the inserted coding sequences. A large number of vector systems known in the art can be used, such as, plasmids, bacteriophage virus or other modified viruses. Suitable vectors include, but are not limited to the following viral vectors such as lambda vector system gt11, gtWES.tB, Charon 4, and plasmid vectors such as pBR322, pBR325, pACYC177, pACYC184, pAR series, pKK223-3, pUC8, pUC9, pUC18, pUC19, pLG339, pR290, pKC37, pKC101 and other similar systems. The DNA sequences are cloned into the vector using standard cloning procedures in the art, as described by Maniatis et al., 1982, Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Springs Harbor, N.Y., which disclosure is hereby incorporated by reference.

The recombinant DNA molecule (e.g., vector+sequence of invention) can then be introduced into appropriate host cells, including but not limited to bacteria, virus, yeast, mammalian cells or the like. The vector system must be compatible with the host cell used. The recombinant vectors can be introduced into the host cells via transformation, transfection or infection using standard techniques in the art. A variety of host cell systems can be used to express the GABA receptor subunit gene of the invention. For example, host cell systems include, but are not limited to the following: bacteria transformed with bacteriophage DNA, plasmid DNA or cosmid DNA such as E. coli JM103, E. coli C600, E. coli CO4, E. coli DH20 and E. coli TB1; microorganisms such as yeast containing yeast vectors; mammalian cell systems infected with virus (vaccinia virus, adenovirus, etc.); insect cell systems infected with virus (baculovirus).

In order to obtain efficient expression of the GABA receptor subunit gene, a promotor must be present in the expression vector. RNA polymerase normally binds to the promotor and initiates transcription of a gene or a group of linked genes and regulatory elements (operon). Promoters vary in their strength, i.e., ability to promote transcription. For the purpose of expressing the gene of the invention, it is desirable to use strong promoters in order to obtain a high level of transcription and, hence, expression of the gene. Depending upon the host cell system utilized, any one of a number of suitable promoters can be used, such as, the lac promotor, trp promotor, recA promotor, ribosomal RNA promotor, the P.sub.R and P.sub.L promoters of coliphage lambda and others including but not limited to lacUV5, ompF, bla, lpp and the like, nos promoter, the small subunit ribulose bisphosphate carboxylase genes, the small subunit chlorophyll A/B binding polypeptide, the 35S promoter of cauliflower mosaic virus, and promoters isolated from plant genes, including the Pto promoter itself (Vallejos, et al., 1986, Genetics, vol. 112, pp. 93-105, which disclosure is hereby incorporated by reference) to direct high levels of transcription of adjacent DNA segments. Additionally, a hybrid trp-lacUV5 (tac) promotor or other E. coli promoters produced by recombinant DNA or other synthetic DNA techniques can be used to provide for transcription of the gene of the invention.

Bacterial host cell strains and expression vectors can be chosen which inhibit the action of the promotor unless specifically induced. In certain operons the addition of specific inducers is necessary for efficient transcription of the inserted DNA; for example, the lac operon is induced by the addition of lactose or IPTG (isopropylthio-beta-D-galactoside). A variety of other operons, such as trp, pro, etc., are under different controls. The trp operon is induced when tryptophan is absent in the growth media; and the P.sub.L, promotor of lambda can be induced by an increase in temperature in host cells containing a temperature sensitive lambda repressor, e.g., c1857. In this way, greater than 95% of the promotor-directed transcription may be inhibited in uninduced cells. Thus, expression of the gene of the invention can be controlled.

One such promotor/operator system is the so-called "tac" or trp-lac promotor/operator system (Russell and Bennett, 1982, Gene, vol. 20, pp.231-243, which disclosure is hereby incorporated by reference). This hybrid promotor is constructed by combining the -35 b.p. (-35 region) of the trp promotor and the -10 b.p. (-10 region or Pribnow box) of the lac promotor (the sequences of DNA which are the RNA polymerase binding site). In addition to maintaining the strong promotor characteristics of the tryptophan promotor, tac is also controlled by the lac repressor.

When cloning in a eucaryotic host cell, enhancer sequences (e.g., the 72 bp tandem repeat of SV40 DNA or the retroviral long terminal repeats of LTRs, etc.) may be inserted to increase transcriptional efficiency. Enhancer sequences are a set of eucaryotic DNA elements that appear to increase transcriptional efficiency in a manner relatively independent of their position and orientation with respect to a nearby gene. Unlike the classic promotor elements (e.g., the polymerase binding site and the Goldberg-Hogness "TATA" box) which must be located immediately 5' to the gene, enhancer sequences have the remarkable ability to function upstream from, within, or downstream from eucaryotic genes. Therefore, the position of the enhancer sequence with respect to the inserted gene is less critical.

Specific initiation signals are also required for efficient gene transcription and translation in procaryotic cells. These transcription and translation initiation signals may vary in "strength" as measured by the quantity of gene specific messenger RNA and protein synthesized, respectively. The DNA expression vector, which contains a promotor, may also contain any combination of various "strong" transcription and/or translation initiation signals. For instance, efficient translation in E. coli requires a Shine-Dalgarno (SD) sequence about 7-9 basis 5' to the initiation codon (ATG) to provide a ribosome binding site. Thus, any SD-ATG combination that can be utilized by host cell ribosomes can be employed. Such combinations include but are not limited to the SD-ATG combination from the CRO gene or the N gene of coliphage lambda, or from the E. coli tryptophan E, D, C, B or A genes. Additionally, any SD-ATG combination produced by recombinant DNA or other techniques involving incorporation of synthetic nucleotides can be used.

Any of the conventional cloning methods for insertion of DNA fragments into a vector can be used to ligate the promotor and other control elements into specific sites within the vector. Accordingly, gene sequences containing those regions coding for the GABA receptor subunit of the invention can be ligated into an expression vector at a specific site in relation to the vector promotor and control elements so that when the recombinant DNA molecule is introduced into a host cell the foreign genetic sequence can be expressed (i.e., transcribed and translated) by the host cell.

As previously mentioned, the recombinant DNA molecule can be introduced into appropriate host cells (including but not limited to bacteria, virus, yeast, mammalian cells or the like) by transformation, infection or transfection (depending upon the vector/host cell system). Transformants are selected based upon the expression of one or more appropriate gene markers normally present in the vector, such as ampicillin resistance or tetracycline resistance in pBR322, or thymidine kinase activity in eucaryotic host systems. Expression of such marker genes should indicate that the recombinant DNA molecule is intact and is replicating. Expression vectors may be derived from cloning vectors, which usually contain a marker function. Such cloning vectors may include, but are not limited to the following: SV40 and adenovirus, vaccinia virus vectors, insect viruses such as baculoviruses, yeast vectors, bacteriophage vectors such as lambda gt-WES-lambda BC, lambda gt-1-lambda B, M13mp7, M13mp8, M13mp9, or plasmid DNA vectors such as pBR322, pAC105, pVA51, pACYC177, pKH47, pACYC184, pUB110, pMB9, pBR325, Col E1, pSC101, pBR313, pML21, RSF2124, pCR1, RP4, pBR328 and the like.

The expression vectors containing the foreign gene inserts (e.g., DNA encoding the GABA receptor subunit of the invention) can be identified by three approaches: (1) DNA-DNA hybridization using probes comprising sequences that are homologous to the gene(s); (2) presence or absence of "marker" gene function and (3) expression of inserted sequences based on physical, immunological or functional properties. Once a recombinant which expresses the gene is identified, the gene product should be analyzed. Immunological analysis is especially important because the ultimate goal is to use the gene or recombinant expression systems that express the gene in assays for screening chemical agents. Once the GABA receptor subunit is identified, it is cultured under conditions which facilitate growth of the cells and expression of the gene as will be apparent to one skilled in the art, then isolated and purified by standard methods including chromatography (e.g., ion exchange, affinity, and sizing column chromatography), centrifugation, differential solubility, or by any other standard techniques. In addition, since the amino acid sequence is known from the DNA sequence of the invention, the GABA receptor subunit can be synthesized by chemical methods according to the procedure of Hunkapiller et al., 1984, Nature, vol. 310, pp. 105-111, which disclosure is hereby incorporated by reference.

In order to obtain efficient expression of a functional GABA receptor, the gene, or fragment thereof, of the present invention is co-expressed with at least one other vertebrate or invertebrate GABA receptor subunit gene. This other GABA receptor subunit gene preferably is any .alpha. (alpha) or .alpha. (alpha) subunit-like sequence(s), such as, for example, the rat GABA receptor .alpha.1 subunit, the Drosophila melanogaster Rdl locus or gamma subunit sequence(s).

The functional GABA receptor produced by co-expression of the GABA receptor subunit gene of the present invention with at least one other vertebrate or invertebrate GABA receptor subunit, can be used to screen for pesticides that are effective in the control of insects such as Drosophila melanogaster. It is known that the insect GABA receptor forms a particularly attractive site for insecticides due to pronounced differences in its pharmacology with that of vertebrates, as described in ffrench-Constant et al., 1991, cited elsewhere herein, which disclosure is hereby incorporated by reference. Due to those differences in insect and vertebrate GABA receptor pharmacology, cells transformed to include the insect GABA receptor formed in accordance with the present invention can be exposed to various potential insecticides and pesticides and evaluated for their susceptibility to the agents to develop and identify insect control agents that will not cause adverse effects to vertebrate species. Exemplary methods of screening are described in Eldefrawei et al., 1987, FASEB/J., vol. 1, pp. 262-271; and Rauh et al., 1990, Trends in Pharmacol. Sci., vol. 11, pp. 325-329, which disclosures are hereby incorporated by reference.

The present invention is further illustrated by the following examples.

1. DNA Amplification:

Two homology probing procedures based on the use of PCR with degenerate oligonucleotide primers were employed. The first, a "single site" procedure (Roux et al., 1990, Biotechniques, vol. 8, pp. 48-57; and Knipple et al., 1991, Mol. Gen. Genet., vol. 226, pp. 241-249, which disclosures are hereby incorporated by reference), employed three 512-fold degenerate target primer pools (TP1a, TP1b, and TP1c; Table Table 1).

These primer pools comprise all possible sequences (+ strand) encoding the first seven amino acids of a conserved octapeptide (TTVLTMTT (SEQ. ID. NO. 1) ) sequence element found in the second transmembrane domain (M2) of most known ligand-gated chloride channel genes. The second primer site was provided by ligation of an oligonucleotide adapter comprised of a tailed linker and an anchor template (Table 1) to Drosophila. melanogaster (Canton-S) genomic DNA digested to completion by BamH1 or BglII. Initial single strand synthesis was primed only by TP1a, TP1b, or TP1c, whereas subsequent cycles of amplification were primed by both the target primer and the anchor primer (AP). In this approach, selective geometric amplification occurs because the AP anneals only to DNA complimentary to the anchor template that is formed in the first strand synthesis. PCR reactions (50 .mu.l) included 50 ng template prepared as above, 200 .mu.M TP1a, TP1b, or TP1c, 2 mM AP, 250 .mu.M dNTP's, 50 mM KCl, 10 mMTris-HCl (pH 8.8), 1.5 mM MgCl.sub.2, 0.1% Triton X-100, 1 U Tag DNA polymerase (Perkin Elmer Cetus) and 0.5 U Perfect Match (Stratagene). Thirty-five cycles of amplifications were performed using the following cycle program: denaturation for 1 minute at 94.degree. C., annealing for 1 minute at 52.degree. C., and extension at 72.degree. C. for 0.5 minutes. DNA amplification products were detected by electrophoresis of 10 .mu.l of the reaction in 3% Nusieve GTG/1% Seakem ME (FMC) agarose gels, followed by ethidium bromide staining.

The second homology probing procedure employed two degenerate primers, TP2 (Table 1) and either TP1a, TP1b, or TP1c. The 256-fold degenerate TP2 comprises all possible sequences (- strand) that encode the CFVFVF (SEQ.ID.NO.28) motif, which is conserved in the third transmembrane domain (M3) of GABA receptor .beta. subunit sequences (Harvey et al., 1991, cited elsewhere herein, which disclosure is hereby incorporated by reference). Reactions were performed under similar conditions to those described above, except that TP2 (100 .mu.M) was used instead of AP, unmodified genomic DNA was used as the template, and the reactions were carried out with the following program: denaturation for 50 seconds at 92.degree. C., annealing for 30 seconds at 45.degree. C. (first 5 cycles) or 52.degree. C. (last 30 cycles), and extension for 30 seconds at 72.degree. C.

To facilitate subcloning of the products obtained from the single site PCR method an XbaI restriction site was incorporated at the 5' end of TP1a, TP1b or TP1c. Amplification products were digested with XbaI after precipitation from the reaction mix with ethanol. The XbaI-digested products were excised from 0.8% to 2.0% Seaplaque GTG (FMC) low melting temperature aqarose gels following electrophoresis in TAE buffer, and ligated with T4 DNA ligase (Promega) directly to XbaI- and SmaI-digested pBluescript KS+ plasmid DNA (Stratagene). The major product obtained by the standard PCR method was similarly isolated by cutting with XbaI and EcoRI at the restriction sites incorporated into the oligonucleotide primers and ligation into XbaI and EcoRI-digested pBluescript KS+ plasmid DNA. E. coli host strain XL1-Blue (Stratagene) was transformed with the in-gel ligation mixtures and individual transformant colonies were screened for the presence of insert-containing plasmids by PCR (Zon et al., 1989, Biotechniques, vol. 7, pp. 696-698, which disclosure is hereby incorporated by reference).

2. Recombinant Bacteriophage Library Screening:

.sup.32 P-Labelled probes were prepared using linearized plasmid DNAs containing the subcloned PCR products as templates in Taq Polymerase-mediated primer extension reactions. The reaction mixture (50 .mu.l) contained 50 mM KCl, 10 mMTris-HCl (pH 8.8), 1.5 mM MgCl.sub.2, 0.1% Triton X-100, 1 U Taq Polymerase (Promega), 20 .mu.M dATP, dGTP, and dTTP, 2 .mu.M dCTP, 200 .mu.Ci of 6000 Ci/mmol [.alpha.-.sup.32 P]dCTP (NEN), 1 .mu.g plasmid DNA, and 0.5 .mu.M primer. Reaction mixtures were incubated for 50-60 cycles according to the following program: denaturation for 50 seconds at 92.degree. C., annealing for 30 seconds at 52.degree.-60.degree. C. (depending on primer), and extension for 30 seconds at 72.degree. C. Unincorporated [.alpha.-.sup.32 P]dCTP was removed from the probe by chromatography using a G50 sephadex (Pharmacia) column. A .lambda. EMBL4 library containing Drosophila melanogaster (Canton-S) genomic DNA (Knipple et al., 1991, cited elsewhere herein, which disclosure is hereby incorporated by reference) was transferred to nitrocellulose (Millipore) using the plaque lift procedure (Benton et al., 1977, Science, vol. 196, pp. 180-192, which disclosure is hereby incorporated), and screened by hybridization overnight at 43.degree. C. in 4 ml hybridization buffer (50% formamide, [5X] Denhardt's solution, 0.1% SDS, [6X]SET, 100 .mu.g/ml calf thymus DNA (Sigma), and 20 .mu.Ci purified probe). The final high stringency wash was in [0.2X]SET and 0.1% SDS for 1 hour at 67.degree. C. Following plaque purification, phage DNA was extracted and purified by standard methods (Maniatis et al., 1982, Cold Spring Harbor Laboratories, Cold Spring Harbor, N.Y., which disclosure is hereby incorporated). Restriction fragments of the genomic DNA clones obtained in these procedures were identified by Southern analysis with the same probe and subcloned into pBluescript KS+ or KS- plasmid DNA.

3. DNA Sequence Analysis:

Double stranded plasmid templates were isolated by the alkaline lysis procedure (Birnboim, H. C., 1983, Meth. Enzymol., vol. 100, pp. 243-255, which disclosure is hereby incorporated), followed by PEG precipitation. Single stranded templates from XL1-Blue cells were isolated as recommended by Stratagene. All DNA sequencing was based on the dideoxynucleotide termination procedure (Sanger et al., 1977, Proc. Natl. Acad. Sci. USA, vol. 74, pp. 5463-5467, which disclosure is hereby incorporated), adapted for either .sup.35 S labelling and autoradiography or dye labelling for use with an Applied Biosystems Model 373 Automated DNA Sequencer. Open reading frames were searched for intron splice junctions by a combination of the following computer-aided sequence analysis approaches: 1) identification of consensus splice junction nucleotide sequences; 2) comparisons to available cDNA sequences; and 3) identification of divergence in deduced amino acid sequence from those in conserved regions of previously cloned gene family members.

4. Cytogenetic Localization of Cloned DNAs:

The cytogenetic location of each cloned DNA was determined by in situ hybridization to polytene chromosome preparations from third instar larval salivary glands using a standard procedure (Pardue et al., 1975, In Methods in Cell Biology (Edited by D. P. Prescott), vol. 10, pp. 1-16, which disclosure is hereby incorporated by reference) modified for use of biotinylated probes. Salivary glands were dissected from third instar larvae of the Canton S strain and polytene chromosome spreads prepared as described previously (Knipple et al., 1991, cited elsewhere herein, which disclosure is hereby incorporated by reference). Recombinant bacteriophage DNA was labeled by the mixed primer labeling procedure (Feinberg et al., 1983, Anal. Biochem., vol. 132, pp. 6-13, which disclosure is hereby incorporated by reference) using biotin-21-dUTP (Clontech) according to the protocol of the supplier. The biotinylated probe was ethanol precipitated in the presence of tRNA carrier and resuspended in 40 .mu.l of hybridization solution that was 0.6M NaCl, 50 mM sodium phosphate, pH 8.0, [1X] Denhardt's solution (0.02% BSA/0.02% Ficoll/0.02% polyvinylpyrrolidone), 4 mg/ml of carrier tRNA, and 5 mM MgCl.sub.2. Each slide containing chromosome spreads was heat-treated and denatured immediately prior to the addition of 10 .mu.l of denatured hybridization solution. Hybridization was performed at 58.degree. C. for 48 hours after which the following wash sequence was performed: three washes for 15 minutes at 53.degree. C. in [2X] SSC, two for 5 minutes in [1X] PBS at room temperature, and a brief rinse in [1X] PBS/0.1% Triton X-100. The hybridized and washed chromosome preparations were then incubated for 1 hour at 37.degree. C. with 100 .mu.l of 0.5% (v/v) Detek streptavidin/horseradish peroxidase conjugate (Enzo) in [1X] PBS/0.1% Triton X-100. Following three 5 minutes washes in [1X] PBS at room temperature, signal was developed by incubation for 1 hour at 37.degree. C. with 100 ml of 54 (w/v) diaminobenzidine (Sigma), 1% hydrogen peroxide in [1X] PBS/0.1% Triton X-100. Following a brief rinse in [1X] PBS and counter-staining with Giemsa for 30 seconds, orange-brown (lighter signal) to black stain deposition corresponding to sites of hybridization were detected visually as narrow bands against the background of light blue chromosome staining. Photomicrographs of labeled sites were made at 1000X amplification through differential interference contrast (Nomarski) optics using an Olympus BH-2 Microscope, Kodak Technical Pan 2415 film, and an Olympus PM-10ADS photomicrographic system.

5. Screening Of Recombinant Bacteriophage Libraries:

The Drosophila melanogaster cDNA libraries used were constructed in .lambda.gt10 and correspond to mRNAs isolated from discrete embryonic, larval, and pupal developmental stages as described by Poole et al., 1985, Cell, vol. 40, pp. 37-43, which disclosure is hereby incorporated by reference. A genomic library was made in .lambda.EMBL4 with DNA form a Canton S strain according to the procedures of Knipple et al., 1991, cited elsewhere herein, which disclosure is hereby incorporated by reference. To identify cDNA libraries containing .beta.-subunit-specific sequences, phage DNA was prepared from plate lysates of each library by a rapid small scale procedure as described by Benson et al., 1984, Biotechniques, vol. 2, pp. 126-27, which disclosure is hereby incorporated by reference and used as the template for PCR reactions (Saiki et al., 1988, Science, vol. 239, pp. 487-491 which disclosure is hereby incorporated by reference) employing oligonucleotide primers designed to amplify GABA receptor .beta.-subunit-specific sequences as previously described. Libraries were plated and transferred in duplicate to nitrocellulose (Millipore) using the plaque lift procedure as described by Benton et al., 1977, cited elsewhere herein, which disclosure is hereby incorporated by reference, and hybridized overnight with .sup.32 P-labelled proves at 43.degree. C. in 50% formamide, 5X Denhardt's solution, o.1% SDS, 6X SET, 100 .mu.g/ml calf thymus DNA, and 20 .mu.Ci purified probe.

To prepare high specific activity probes, 50 cycles of Taq polymerase-mediated primer extension were performed using a 266 kb gel-purified fragment of coding sequence (derived from a 2.26 kb LCCH3 genomic subclone) as template and specific primers directing the synthesis of either sense or antisense strands (5'-GCCGCAGTTCACCATCATTG-3'(SEQ. ID. NO. 11) and 5'-AGTGAGCTGCGAACACCCGTG-3'(SEQ. ID. NO. 12), respectively). Labelling reactions (50 .mu.l ) contained 50 mM KCl, 10 mM Tris-HCl (pH 8.8), 1.5 mM MgCl.sub.2, 0.1% Triton X-100, 1 U Taq DNA Polymerase, 20 .mu.M each dATP, dGTP, and dTTP, 2 .mu.M dCTP, 200 .mu.Ci of 6000 Ci/mmol [.alpha.-.sup.32 P]dCTP (NEN), 100 ng template DNA, and 0.5 .mu.M primer. Unincorporated [.alpha.-.sup.32 P]dCTP was removed by elution over a G50 Sephadex (Pharmacia) column. One set of transfers was hybridized with the sense strand probe, and the duplicate set was hybridized with the antisense strand probe. The final high stringency post-hybridization wash was in 0.2X SET and 0.1% SDS for 1 hour at 67.degree. C. Probe-positive plaques were purified and the cDNA or genomic inserts subcloned into pBluescript SK+ plasmid DNA (Stratagene) using standard techniques (Sambrook et al., Molecular Cloning: A laboratory Manual, 1989, Cold Spring Harbor Laboratory Press, which disclosure is hereby incorporated by reference).

6. PCR Amplification Of The cDNA 3' End:

Comparison of the DNA sequences of the isolated cDNA and the LCCH3 genomic DNA indicated that the former was truncated at its 3' end by approximately 500 bp due to the occurrence of an A-rich stretch of sequence in the coding region. In order to isolate the 3' end of the LCCH3 mRNA expeditiously, the pupal cDNA library was used as a template in PCR using primers selected on the basis of sequences from the LCCH3 genomic subclone. The downstream-directed primer (5'-TTTCGTGTTCGCAGCCCTCT-3' (SEQ.ID.NO.13)) was designed to be identical to the sense strand of the DNA coding for the -VFVFAAL- (SEQ. ID. NO. 29) amino acid sequence motif in the putative transmembrane domain M3. The upstream-directed primer (5'GGTCTAGAAGTCAATCTTTATTATGGAGG-3'(SEQ.ID.NO.14)) was designed to be identical to the complementary strand adjacent to a putative polyadenylation site located 106 bp downstream of the apparent termination signal, which was identified on the basis of the consensus polyadenylation signal (AATAAA) (SEQ. ID. NO. 30) followed by CA 14 bases downstream, with no other CA motifs nearby (Wahle et al., 1992, Annu. Rev. Biochem., vol. 61, pp. 419-440, which disclosure is hereby incorporated by reference). PCR reactions (50 .mu.l) included 50 ng cDNA template, 0.5.mu.M of each primer, 250 mM dNTP's, 50 mM KCl, 10 mMTris-HCl (pH 8.8), 1.5 mM MgCl.sub.2, 0.1% Triton X-100, 1 U Taq Polymerase (Perkin elmer Cetus) and 0.5 U Perfect Match (Stratagene). Amplications were performed using 35 cycles on a Techne PHC-1 thermal cycler using the following cycle program: denaturation for 1 minute at 94.degree. C., annealing for 1 minute at 56.degree. C., and extension for 0.5 minutes at 72.degree. C. DNA amplification products were detected by electrophoresis of 10 .mu.l of the reaction in 3% Nusieve GTG/1% Seakem ME (FMC) agarose gels, followed by ethidium bromide staining.

7. DNA Sequence Analysis:

DNA sequencing was performed using the dideoxynucleotide chain termination procedure as described by Sanger et al., 1977, cited elsewhere herein, which disclosure is hereby incorporated by reference, adapted for either .sup.35 S labelling with Sequenase and autoradiography or dye terminator labelling with Taq polymerase and automated sequencing using an Applied Biosystems Model 373 Automated DNA Sequencer. Templates for DNA sequencing were either: plasmid DNA isolated by the alkaline lysis procedure (Birnboim, 1983, cited elsewhere herein, which disclosure is hereby incorporated by reference), followed by PEG precipitation; single-stranded pBluescript DNA isolated by a standard method (Vieira et al., 1987, Method. Enzymol., vol. 153, pp. 3-11, which disclosure is hereby incorporated by reference) with modification as recommended by Stratagene for XL1-Blue cells; or linear DNA fragments synthesized by performing an additional 25 cycles of PCR on gel-purified PCR products, followed by purification in a Centicon-100 (Amicon) filtration device. DNA sequence comparisons were made using the Genetics Computer Group on BioVAX and MacVector (IBI) software.

PCR-based homology screening of Drosophila melanogaster genomic DNA with degenerate oligonucleotide primers was conducted (Gould et al., 1989, Proc. Natl. Acad. Sci. USA; , vol. 86, pp. 1934-9138; and Kamb et al., 1989, Proc. Natl. Acad. Sci. USA, vol. 86, pp. 4372-4376, which disclosures are hereby incorporated by reference) to amplify genomic sequences that encode the first seven amino acids of an octapeptide (TTVLTMTT)(SEQ.ID.NO.1) "signature motif," which is conserved in the M2 region of most of the characterized ligand-gated chloride channel genes. Using the single site PCR strategy, seven amplification products were obtained that ranged in size from approximately 100 to 300 bp from reactions employing the three target primer pools and either BglII- or BamHI-cut and anchor-adapted Drosophila melanogaster genomic DNA. Sequence analysis of the subcloned PCR products revealed that two of these sequence elements (designated LCCH1 and LCCH2) contained extended open reading frames having greater than 40% identity of their predicted amino acid sequences to the corresponding sequences encoded by vertebrate ligand-gated chloride channel subunit genes in this highly conserved region. The other five products had very short open reading frames or little similarity to ligand-gated chloride channel genes outside of the target sequence for PCR primer design.

A PCR homology probing strategy was conducted, with degenerate target primers specifying two defined sequence elements, in a search for Drosophila melanogaster genomic fragments with GABA receptor .beta. subunit-like sequences. This strategy yielded a single product (designated LCCH3) when TP1b was used as the downstream-directed target primer pool, but did not yield other products with either the TP1a or TP1c primer pools. LCCH3 contained an extended open reading frame, and the predicted amino acid sequence of this fragment exhibited greater than 40% identity to the corresponding sequences encoded by GABA receptor B subunit genes. A similar strategy was employed to search for GABA receptor .alpha. subunit-like sequences, based on the ATVNYFT (SEQ.ID.NO.31) amino acid sequence motif present in M3 of all vertebrate .alpha. subunits, but a specific amplification product using the .alpha. subunit-specific primer in combination with any of the three TP1 primer pools was not observed.

To further characterize the transcription units from which the LCCH1, LCCH2, and LCCH3 amplification products were derived, approximately 60,000 plaques were screened from a Drosophila melanogaster genomic library, and isolated clones identified by each probe. Southern analysis of restriction digests from one of each group of genomic clones identified the following fragments that were used in subsequent analysis: 1) a 2 kb BamHI/HindIII fragment of LCCH1; 2) a 1.7 kb EcoRI/BglII fragment of LCCH2; and 3) a 2.3 kb XbaI/EcoRI fragment of LCCH3. Sequencing of the regions flanking the original amplified segments of these DNAs provisionally identified the LCCH1, LCCH2, and LCCH3 transcription units as members of the ligand-gated chloride channel gene family of Drosophila melanogaster based on the conservation of several distinctive features of the amino acid sequences encoded by open reading frames of each genomic fragment. To illustrate these homology relationships, the deduced amino acid sequences of these clones in the putative M2-M3 regions are shown in FIG. 1 along with those of previously characterized representative ligand-gated chloride channel genes. With reference to FIG. 1, the expanded M2-M3 region is represented by the single letter amino acid code for putative Drosophila ligand-gated chloride channel homologues LCCH1, LCCH2, and LCCH3 and representative established gene family members. .alpha.1, .beta.1, .gamma.2, and .delta. are: bovine .alpha.1 and .beta.1 (Schofield et al., 1987, cited elsewhere herein, which disclosure is hereby incorporated by reference); and rat .gamma.2 and .delta. (Shivers et al., 1989, Neuron, vol. 3, pp. 327-337, which disclosure is hereby incorporated by reference) GABA receptor subunits. Lym .beta. is the Lymnaea GABA receptor .beta. subunit (Harvey et al., 1991, cited elsewhere herein, which disclosure is hereby incorporated by reference). Gl.alpha. is the rat .alpha.1 glycine receptor subunit (Grenningloh et al., 1987, Nature, vo. 328, pp. 215-220, which disclosure is hereby incorporated by reference). Amino acids that are conserved in all sequences are shown in boldface type. Dashes in the LCCH1 sequence identify the point of divergence of this fragment from the corresponding cDNA sequence (ffrench-Constant et al., 1991, cited elsewhere herein, which disclosure is hereby incorporated by reference). The DNA sequence obtained for the LCCH1 coding region was delimited by a Sau3AI subcloning site, which occurred just upstream of the conserved M2 sequence motif, and by an intron-exon splice junction near the downstream end of M3. Although only the putative M2-M3 regions of LCCH2 and LCCH3 are shown in FIG. 1, the open reading frames of both these clones extended beyond M2 upstream and beyond M3 downstream, as explained hereinafter and shown in FIG. 3.

The following elements, which are conserved in other members of this gene family, were found to be conserved across all three Drosophila melanogaster sequences (i.e., LCCH1, LCCH2, LCCH3): 1) invariant Arg and Giy residues in proper alignment in M2; 2) the first seven amino acids of the octapeptide signature motif; 3) the invariant Leu-Pro motif in the extracellular loop between M2 and M3; and 4) invariant Asp and Phe residues in proper alignment in M3. LCCH1 and LCCH3 contained the full octapeptide signature motif in M2, but LCCH2 showed a substitution of Phe for Thr in the eighth residue. Each Drosophila melanogaster fragment also contained other sequence elements that were conserved in comparison with some of the known ligand-gated chloride channel genes but were not invariant across the entire family. Hydropathy plots (not shown) of the predicted amino acid sequences deduced from the open reading frames of LCCH1, LCCH2, and LCCH3 were similar to those derived for the corresponding regions of other characterized ligand-gated chloride channel subunit genes, and showed highly hydrophobic segments corresponding to the inferred M2 and M3 regions separated by a short hydrophilic segment.

Paired comparisons of inferred amino acid sequences in the M2-M3 region (Table 2) showed that LCCH1 and LCCH3 were more similar to each other (68% amino acid identity) than either was to LCCH2 (46% and 49% amino acid identity, respectively).

In comparisons with other members of this gene family in the M2-M3 region, LCCH1 was most similar to the Lymnaea and rat GABA receptor .beta. subunits and the .alpha. subunit of the rat glycine receptor. LCCH3 exhibited a very high degree of sequence identity to the Lymnaea .beta. subunit (96%) and considerable sequence identity with the rat GABA receptor .beta. subunit (79%). In contrast, LCCH2 exhibited only 46% to 49% sequence identity in the region in pairwise comparisons with all other sequences.

The cytogenetic locations of the cloned LCCH1, LCCH2, and LCCH3 genomic DNAs were determined by hybridizing biotinylated probes prepared from each sequence to Drosophila melanogaster polytene chromosome squashes in situ. The LCCH1 probe labelled cytogenetic region 66F on the left arm of chromosome 2 (not shown), a finding consistent with the known location of Rdl (=LCCH1) to 66F (ffrench-Constant et al., 1991, cited elsewhere herein, which disclosure is hereby incorporated by reference). The LCCH2 probe labelled cytogenetic region 75A on the left arm of chromosome 3 (FIG. 2, top). The LCCH3 probe labelled cytogenetic region 13F on the X chromosome (FIG. 2, bottom).

The use of PCR-based homology probing to identify novel LCCHs elements in the Drosophila melanogaster genome, provisionally designated LCCH1, LCCH2, and LCCH3, permitted isolation of corresponding genomic DNA clones. Sequencing of a 2.3 kb XbaI/EcoRI restriction fragment of the LCCH3 genomic clone (data not shown) identified three open reading frames encompassing 987 bp, which encode amino acid sequences having high homology in regions corresponding to the four transmembrane domains (M1-M4) of GABA receptor .beta. subunits of vertebrates (Schofield et al., 1987, Nature, vol. 328, pp. 221-227; and Ymer et al., 1989, EMBO/J., vol. 6, pp. 1665-70, which disclosures are hereby incorporated by reference) and Lymnaea stagnalis (Harvey et al., 1991, cited elsewhere herein, which disclosure is hereby incorporated by reference).

In preliminary PCR-based screens of embryonic and pupal cDNA libraries, LCCH3-homologous sequences, identified by the presence of a predicted 134 bp amplification product, were detected only in the pupal library. A single 1.65 kb cDNA clone was isolated by screening approximately 850,000 plaques form this library using a hybridization probe derived form the M1-M3 transmembrane encoding region of the 2.3 kb XbaI/EcoRI genomic DNA restriction fragment. Sequencing of this clone revealed an open reading frame extending downstream from a putative translation initiation site near the 5' end to the 3' end of the clone. Comparison of this sequence to that of the 2.3 kb genomic subclone revealed that the cDNA clone was truncated at the 3' end in an adenosine-rich motif in the M3 domain, but was otherwise identical to two upstream exons identified on the genomic subclone. The availability of genomic DNA sequence, which identified both the cDNA truncation site and the 3' end of the transcription unit, permitted the design of PCR primers for the specific amplification of the 3' end of the cDNA from the pupal cDNA library. Thus, the complete coding sequence of LCCH3, comprising a single open reading frame of 1488 nucleotides (FIG. 4), was deduced from the 1.65 kb cDNA clone and an overlapping 615 bp amplification product having a nucleotide sequence identical to the downstream open reading frames identified in the LCCH3 genomic DNA, and corresponding to the 3' end of the gene. With reference to FIG. 4, the nucleotide sequence (SEQ.ID.NO.3) is numbered arbitrarily from the first ATG open reading frame. The amino acid sequence (SEQ.ID.NO.2) is numbered from the first residue of the predicted mature polypeptide. Amino acids in the putative signal sequence are given negative numbers. Conserved cysteine residues in the extracellular domain are marked with asterisks, and the four inferred hydrophobic transmembrane domains are underlined.

The LCCH3 cDNA encodes a 496-amino acid polypeptide (FIG. 4, SEQ.ID.NO.2). The hydropathy profile of the predicted gene product, illustrated by FIG. 5, shows five extended hydrophobic domains: one at the 5' end of the open reading frame forming a putative signal peptide sequence and four in the 3' half of the sequence corresponding to putative transmembrane domains. With reference to FIG. 5, the putative signal peptide sequence (SP) and transmembrane domains (M1-M4) are indicated. Both the relative position and the amino acid sequences of the four transmembrane domains are conserved with respect to other known members of the ligand-gated chloride channel gene family (FIG. 6).

With reference to FIG. 6, the alignment of the predicted amino acid sequence of the present isolated gene fragment (LCCH3, SEQ.ID.NO.2) with those of the Lymnaea stagnalis GABA receptor .beta.-like subunit (SEQ.ID.NO.24)(Harvey et al., 1991, cited elsewhere herein, which disclosure is hereby incorporated by reference), the rat GABA receptor .beta.1 subunit (SEQ.ID.NO.25) (Birnboim, H. C., 1983, cited elsewhere herein, which disclosure is hereby incorporated by reference), and the Drosophila melanogaster Rdl locus (SEQ.ID.NO.26)(ffrench-Constant et al., 1991, cited elsewhere herein, which disclosure is hereby incorporated by reference) using the GCG computer program is illustrated. All sequences are numbered from the first residue of the mature polypeptide, and putative signal sequence residues are given negative numbers. Amino acids that are identical in all four sequences are enclosed in boxes. The putative extracellular disulfide-bridged loop structure is underlined with a dotted line, and the four hydrophobic transmembrane domains are underlined with solid lines. The predicted mature gene product after signal peptide cleavage (von Heijne, G., 1986, Nucl. Acids Res., vol. 14, pp. 683-4690, which disclosure is hereby incorporated by reference) is a 476-amino acid polypeptide with a molecular weight of 54,442 daltons. The large N-terminal extracellular domain of this polypeptide contains a conserved 15-amino acid sequence element (amino acid residue 152-166) bounded by cysteine residues (Cys152 and Cys166) capable of forming the disulfide-bridged loop found in the extracellular domain of all members of the ligand-gated ion channel gene superfamily (Barnard et al., 1987, cited elsewhere herein, which disclosure is hereby incorporated by reference). The extracellular domain also contains two possible sites for N-linked glycosylation, Asn19 and Asn169, (Kornfeld et al., 1985, Annu. Rev. Biochem., vol. 54, pp. 931-964, which disclosure is hereby incorporated by reference). None of the putative intracellular domains of the predicted polypeptide contain a consensus recognition sequence (RRXSX) (SEQ.ID.NO.27) for phosphorylation by cAMP-dependent protein kinase, but the large intracellular domain contains two phosphorylation sites (Ser368 and Ser434) with alternative recognition sequences (Kemp et al., 1990, Trends Biochem. Sci., vol. 15, pp. 342-346, which disclosure is hereby incorporated by reference).

Both the nervous system and neuromuscular junctions of insects contain ligand-gated chloride channels (Lummis et al., 1990, cited elsewhere herein; and Sattelle, D. B., 1990, cited elsewhere herein, which disclosures are hereby incorporated by reference) that may be encoded by homologues of the ligand-gated chloride channel genes that have been described for vertebrates and molluscs. Despite the availability of cloned cDNAs of mammalian ligand-gated chloride channel subunit genes, there are no reports of the successful use of these cDNAs as probes for the isolation of homologous genes from insects. The existence of a short stretch of invariant amino acids (TTVLTMTT)(SEQ.ID.NO.1) in the second inferred transmembrane domain of all reported vertebrate ligand-gated chloride channels except for the GABA .rho. subunit (Cutting et al., 1991, Proc. Natl. Acad. Sci. USA, vol. 88, pp. 2673-2677, which disclosure is hereby incorporated by reference) led us to hypothesize that this octapeptide might be a conserved "signature motif" for the ligand-gated chloride channel gene family and therefore might be present in homologous insect ligand-gated chloride channels.

This hypothesis was tested by PCR-based homology probing, employing degenerate primer pools coding for the first seven amino acids of the signature motif. The existence in this gene family of only a single stretch of invariant amino acid sequence long enough for PCR primer design necessitated the use of a single site PCR approach (Roux et al., 1990, cited elsewhere herein, which disclosure is hereby incorporated by reference), which yielded a set of seven unique genomic fragments containing this motif. Within this set were two fragments (LCCH1 and LCCH2) with extensive open reading frames and considerable homology to established ligand-gated chloride channel genes. It is likely that other members of this sequence family exist in Drosophila melanogaster but were not identified in the present experiments because the single site PCR-based search was limited to those genes with BglII or BamHI restriction sites downstream of the signature motif and within range of PCR amplification under the specific reaction conditions that we employed. A more thorough single site search would include the use of other restriction sites in the digestion and modification of template DNA, upstream-directed primers based on the TTVLTMTT (SEQ.ID.NO.1) motif, and variations of the TTVLTMTT (SEQ.ID.NO.1) motif (Cutting et al., 1991, cited elsewhere herein, which disclosure is hereby incorporated by reference).

A supplementary approach was designed to probe specifically for GABA .beta. receptor subunit-like sequences. This strategy was based on the presence of a conserved CFVFVF (SEQ.ID.NO.28) motif in the third transmembrane domain of both vertebrate and molluscan GABA receptor .beta. subunits (Harvey et al., 1991, cited elsewhere herein, which disclosure is hereby incorporated by reference). When degenerate primers directed toward the CFVFVF (SEQ.ID.NO.28) motif were employed together with the available pools of primers for the octapeptide motif in conventional PCR amplifications, a single PCR product (LCCH3) was obtained with striking homology to the molluscan GABA receptor .beta. subunit. This finding provides evidence for a larger family of ligand-gated chloride channel sequence elements in the Drosophila melanogaster genome (identified in original homology probing strategy). However, a similar approach directed toward the isolation of GABA .alpha. receptor subunit-like sequences, based only on vertebrate sequence information, was not successful.

The predicted amino acid sequences of the three Drosophila melanogaster genomic fragments was compared to assess the degrees of relatedness among these fragments and other genes in this family (Table 2). The comparisons were restricted to the presumptive M2 and M3 domains because sequence data with open reading frames corresponding to these regions were available for all three Drosophila melanogaster sequence elements. The M2 and M3 domains contain amino acid sequence elements that are conserved across all known members of the gene family as well as elements characteristic of each GABA receptor subunit class. In this region, ligand-gated chloride channel gene family members all have greater than 40% amino acid identity to each other but share less than 20% identity with more distantly related members of the ligand-gated ion channel gene superfamily (Schofield et al., 1987, cited elsewhere herein, which disclosure is hereby incorporated by reference).

LCCH1 shares 100% amino acid and nucleotide sequence identity in the M2-M3 region with the Drosophila melanogaster cDNA from the Rdl locus (ffrench-Constant et al., 1991, cited elsewhere herein, which disclosure is hereby incorporated by reference). Comparison of the sequences of LCCH1 and the homologous cDNA identified an apparent exon/intron splice junction at Glu.sup. 353 of the cDNA. This finding is in agreement with a recent report describing the intron-exon splice junction locations for the Rdl locus (ffrench-Constant et al., 1992, cited elsewhere herein, which disclosure is hereby incorporated by reference). The identification of Rdl (=LCCH1) as a GABA receptor subunit is based on both the inferred structure and the functional expression of the Rdl gene product (ffrench-Constant et al., 1991; ffrench-Constant et al., 1993, Nature, vol. 363, pp. 449-451, which disclosure is hereby incorporated by reference).

In contrast to LCCH1, both LCCH2 and LCCH3 represent previously undescribed genes in Drosophila melanogaster. Comparisons of amino acid sequence in the M2-M3 region between LCCH2 and established ligand-gated chloride channel genes showed that LCCH2 contains the invariant landmarks of this gene family with the notable exception that the octapeptide motif is truncated by the replacement of the terminal Thr with Phe. The predicted amino acid sequence of LCCH2 in this region is equally divergent from all other known sequences in the gene family. This result suggests that LCCH2 may represent a novel class of GABA receptor subunit or may encode a subunit of a ligand-gated chloride channel that is gated by an agonist other than GABA. Such genes are implicated to exist by the physiological evidence for distinct glutamate-, taurine-, and histamine-gated chloride channels in insects (Lummis et al., 1990, cited elsewhere herein, which disclosure is hereby incorporated by reference).

Results of in situ hybridization experiments show that LCCH2 is located at cytogenetic region 75A. A Drosophila melanogaster genomic DNA clone isolated with a rat glycine receptor .alpha. subunit cDNA probe has also been reported to map onto this region, but there is no reported mutation mapping to this cytogenetic interval that might provide insight into the physiological function of the LCCH2 gene product (Lindsley et al., 1992, The genome of Drosophila melanogaster, Academic Press, New York, which disclosure is hereby incorporated by reference).

LCCH3 exhibited 96% amino acid sequence identity to the established molluscan GABA receptor .beta. subunit (Harvey et al., 1991, cited elsewhere herein, which disclosure is hereby incorporated by reference) in the putative M2 and M3 domains, a degree of sequence identity comparable to that shared between subunit isomorphs in vertebrates. Moreover, sequence analysis of the entire 501-nucleotide open reading frame of the genomic subclone corresponding to LCCH3 showed a high degree of sequence identity between LCCH3 and the molluscan GABA receptor .beta. subunit upstream from the region of comparison through M1 and into the presumed extracellular domain (data not shown). Thus, LCCH3 appears to represent the homologue in Drosophila melanogaster of the molluscan GABA receptor .beta. subunit. This conclusion is supported herein by the determination of the complete LCCH3 cDNA sequence, which shows a much higher degree of similarity both within transmembrane domains M1 -M3 and overall to the rat GABA receptor .beta. subunit than to any other vertebrate GABA receptor subunit group. LCCH3 also exhibits a high degree of predicted amino acid sequence identity to .beta. subunit-like cDNA fragment previously isolated from Drosophila melanogaster (Sattelle et al., 1991, In Transmitter amino acid receptors: Structures, transduction and models for drug development (Edited by E. A. Barnard and E. Costa) pp. 273-291, Thieme Medical Publishers, New York, which disclosure is hereby incorporated by reference). However, the lack of either a published nucleotide sequence or a cytogenetic location for that cDNA fragment precludes a determination of whether it and LCCH3 are identical.

The localization of LCCH3 to cytogenetic region 13F may provide additional insight into the function of the product of this gene. Region 13F also contains the slrp (slow receptor potential) locus, identified on the basis of mutants exhibiting altered electroretinogram potentials (Pak 1975) and behavioral deficits that include general hypoactivity, ether-induced leg shaking, and cold-induced paralysis (Homyk, T. Jr., 1977, Genetics, vol. 87, pp. 105-128; Homyk et al., 1977, Genetics, vol. 87, pp. 95-104; and Homyk et al., 1989, J. Neurogenet., vol. 5, pp. 37-48, which disclosures are hereby incorporated by reference). The various manifestations of the slrp phenotype, which involve or implicate abnormal transmission at neuronal or neuromuscular synapses (Homyk et al., 1989, cited elsewhere herein, which disclosure is hereby incorporated by reference), are consistent with the existence of a neurotransmitter receptor gene at this locus.

In Harvey et al., 1991, cited elsewhere herein, which disclosure is hereby incorporated by reference, the authors found that the positions of six intron-exon splice junctions in the Lymnaea stagnalis GABA receptor .beta. subunit gene were conserved in position when compared to splice junctions in vertebrate GABA receptor .beta. and .delta. subunits that have been characterized at the level of genomic organization. Moreover, this conservation did not extend to glycine receptor subunit genes or other, more distantly related members of the ligand-gated ion channel gene superfamily (Sommer et al., 1990, DNA Cell Biol., vol. 9, pp. 561-568; and Kirkness et al., 1991, Genomics, vol. 10, pp. 985-995, which disclosures are hereby incorporated by reference). These observations led the authors in Harvey et al., to conclude that the structural organization of GABA receptor genes was established prior to the evolutionary divergence of molluscs and chordates and prior to the duplication and divergence events that established the different GA3A receptor subtypes.

It was determined that each of the three Drosophila melanogaster sequences examined lacked at least one of the six conserved intron-exon splice junctions (Harvey et al., 1991, cited elsewhere herein, which disclosure is hereby incorporated by reference, and FIG. 3). Rdl (=LCCH1) lacked both the fourth (upstream of M1) and fifth (within M2) conserved splice junctions, and it also contained a nonconserved splice junction in M3 (ffrench-Constant et al., 1992, cited elsewhere herein, which disclosure is hereby incorporated by reference). LCCH2 lacked the fifth conserved splice junction in M2 and also contained uninterrupted coding sequence extending downstream well beyond the sixth conserved splice junction (within the M3-M4 intracellular loop). However, the splice junction at the N-terminus of the LCCH2 open reading frame was conserved in both position and amino acid sequence context. LCCH3 contained splice junctions at the termini of the open reading frame that appear to be conserved with respect to vertebrate and molluscan genes. The splice junction at the 5' end of the open reading frame was conserved both in position relative to M1 and amino acid sequence context. The splice junction at the 3' end of the LCCH3 open reading frame was approximately conserved in position relative to M3 but occurred in a region of low amino acid sequence conservation. On the basis of our results and those of others (ffrench-Constant et al., 1992, Neurochem., vol. 59, pp. 1562-1565, which disclosure is hereby incorporated by reference), it is concluded that there is considerable but not absolute conservation of intron location among structurally-divergent members of the ligand-gated chloride channel gene family of Drosophila melanogaster.

The gene product of LCCH3 exhibited all of the conserved features commonly found in members of this gene family: a structural organization encompassing a large N-terminal extracellular domain, four hydrophobic domains capable of forming transmembrane helices, and a large intracellular domain lying between transmembrane domains M3 and M4; the conserved octapeptide (-TTVLTMTT-)(SEQ.ID.NO.1) "signature motif" for this gene family located in M2; and the conserved cysteine residues and associated sequence elements capable of forming the disulfide-bridged loop structure in the extracellular domain (Burt et al., 1991, cited elsewhere herein; Langosch et al., 1990, Eur. J. Biochem., vol. 194, pp. 1-8; and Barnard et al., 1987, cited elsewhere herein, which disclosures are hereby incorporated by reference).

Within the ligand-gated chloride channel gene family, LCCH3 is most similar to members of the GABA receptor .beta. subunit sequence subfamily (Table 3).

Over its entire coding region, LCCH3 exhibits 56% sequence identity with the L. stagnalis GABA receptor .beta. subunit (Harvey et al., 1991, cited elsewhere herein, which disclosure is hereby incorporated by reference) and 47% identity with the bovine GABA receptor .beta.1 subunit (Ymer et al., 1989, cited elsewhere herein, which disclosure is hereby incorporated by reference). Within the highly conserved core region of the gene product encompassing transmembrane domains M1 through M3, LCCH3 exhibits 98% amino acid identity with the L. stagnalis GABA receptor .beta. subunit (Harvey et al., 1991, cited elsewhere herein, which disclosure is hereby incorporated by reference) and 77% identity with the bovine GABA receptor .beta.1 subunit (Ymer et al., 1989, cited elsewhere herein, which disclosure is hereby incorporated by reference). The high sequence conservation between LCCH3 and the L. stagnalis GABA receptor .beta. subunit in the M1-M3 core region strongly suggests from structural grounds that the isolated gene fragment of the present invention LCCH3 is the homologue in D. melanogaster of the L. stagnalis subunit (Zaman et al., 1992, FEBS Lett., vol. 307, pp. 351-354, which disclosure is hereby incorporated by reference).

Despite the overall similarity of LCCH3 to other members of the GABA receptor .beta. subunit sequence subfamily, it lacks the consensus sequence for a cAmp-dependent phosphorylation site that is found in the M3-M4 intracellular loop of all other known vertebrate and invertebrate .beta. subunits (Burt et al., 1991, cited elsewhere herein; and Harvey et al., cited elsewhere herein, which disclosures are hereby incorporated by reference). Recent studies have shown that this site in the rat .beta.1 subunit is a substrate for phosphorylation by both cAMP-dependent protein kinase and protein kinase C (Moss et al., 1992, J. Biol. Chem., vol. 267, pp. 14470-14476, which disclosure is hereby incorporated by reference) and that phosphorylation by cAMP-dependent protein kinase modulates the responses to GABA of heteromultimeric (.alpha.1/.beta.1 or .alpha.1/.beta.1/.gamma.2) receptors expressed in transfected cells Moss et al., 1992, Science, vol. 257, pp. 661-665, which disclosure is hereby incorporated by reference). The findings of the present invention nevertheless identify two possible phosphorylation sites with alternative recognition cAMP-dependent protein kinase. Alternatively, these receptors in Drosophila melanogaster might be modulated by phosphorylation on other, as yet unidentified subunits or by mechanisms other than cAMP-dependent phosphorylation.

The LCCH3 gene product is less similar to the other known GABA receptor-like sequence from Drosophila melanogaster, the Rdl gene product (ffrench-Constant et al., 1991, cited elsewhere herein, which disclosure is hereby incorporated by reference), than to members of the .beta. subunit sequence subfamily (FIG. 6 and Table 3). The inferred amino acid sequence of LCCH3 possesses only 26% identity overall to Rdl, and the degree of identity in the M1-M3 core region is only 64%. The existence in Drosophila melanogaster of two divergent members of this sequence family raises the possibility that the two gene products could form subunits of a heteromultimeric receptor. Based on the properties of the homomultimeric and chimeric receptors formed by the L. stagnalis subunit in Xenopus oocytes (Harvey et al., 1991, and Zaman et al., 1992, FEBS Lett., vol. 307, pp. 351-354, cited elsewhere herein, which disclosures are hereby incorporated by reference), it is likely that the LCCH3 gene product would need to be co-expressed with another subunit to form a receptor that exhibits the pharmacological properties of GABA receptors in insect tissues. The linkage between the Rdl gene product and GABA receptor function is based on the ability of a mutation at the Rdl locus to confer resistance to picrotoxinin and cyclodiene insecticides, which act as blockers of the GABA-gated chloride channel (ffrench-Constant et al., 1991, cited elsewhere herein, Bloomquist et al., 1991, Pestic. Sci., vol. 32, pp. 463-469, which disclosures are hereby incorporated by reference).

Although the invention has been described in detail for the purpose of illustration, it is understood that such detail is solely for that purpose, and variations can be made therein by those skilled in the art without departing from the spirit and scope of the invention which is described by the following claims.