Portal de Periódicos da CAPES

Identification, Molecular Cloning, and Characterization of a Novel GABAA Receptor-associated Protein, GRIF-1

2002; Elsevier BV; Volume: 277; Issue: 33 Linguagem: Inglês

10.1074/jbc.m200438200

1083-351X

Mike Beck, Kieran Brickley, Helen Wilkinson, Seema Sharma, Miriam J. Smith, Paul L. Chazot, Simon Pollard, F. Anne Stephenson,

Neuropeptides and Animal Physiology

A novel 913-amino acid protein, γ-aminobutyric acid type A (GABAA) receptor interacting factor-1 (GRIF-1), has been cloned and identified as a GABAAreceptor-associated protein by virtue of its specific interaction with the GABAA receptor β2 subunit intracellular loop in a yeast two-hybrid assay. GRIF-1 has no homology with proteins of known function, but it is the rat orthologue of the human ALS2CR3/KIAA0549 gene. GRIF-1 is expressed as two alternative splice forms, GRIF-1a and a C-terminally truncated form, GRIF-1b. GRIF-1 mRNA has a wide distribution with a major transcript size of 6.2 kb. GRIF-1a protein is only expressed in excitable tissues, i.e. brain, heart, and skeletal muscle major immunoreactive bands ofMr ∼ 115 and 106 kDa and, in muscle and heart only, an additional 88-kDa species. When expressed in human embryonic kidney 293 cells, GRIF-1a yielded three immunoreactive bands withMr ∼ 115, 106, and 98 kDa. Co-expression of GRIF-1a and α1β2γ2 GABAA receptors in mammalian cells revealed some co-localization in the cell cytoplasm. Anti-FLAG-agarose specifically precipitated GRIF-1FLAG and GABAAreceptor β2 subunits from human embryonic kidney 293 cells co-transfected with GRIF-1aFLAG and β2 subunit clones. Further, immobilized GRIF-1-(8–633) specifically precipitated in vitro GABAA receptor α1 and β2 subunit immunoreactivities from detergent extracts of adult rat brain. The respective GABAA receptor β2 subunit/GRIF-1 binding domains were mapped using the yeast two-hybrid reporter gene assays. A possible role for GRIF-1 as a GABAA receptor β2 subunit trafficking factor is proposed. A novel 913-amino acid protein, γ-aminobutyric acid type A (GABAA) receptor interacting factor-1 (GRIF-1), has been cloned and identified as a GABAAreceptor-associated protein by virtue of its specific interaction with the GABAA receptor β2 subunit intracellular loop in a yeast two-hybrid assay. GRIF-1 has no homology with proteins of known function, but it is the rat orthologue of the human ALS2CR3/KIAA0549 gene. GRIF-1 is expressed as two alternative splice forms, GRIF-1a and a C-terminally truncated form, GRIF-1b. GRIF-1 mRNA has a wide distribution with a major transcript size of 6.2 kb. GRIF-1a protein is only expressed in excitable tissues, i.e. brain, heart, and skeletal muscle major immunoreactive bands ofMr ∼ 115 and 106 kDa and, in muscle and heart only, an additional 88-kDa species. When expressed in human embryonic kidney 293 cells, GRIF-1a yielded three immunoreactive bands withMr ∼ 115, 106, and 98 kDa. Co-expression of GRIF-1a and α1β2γ2 GABAA receptors in mammalian cells revealed some co-localization in the cell cytoplasm. Anti-FLAG-agarose specifically precipitated GRIF-1FLAG and GABAAreceptor β2 subunits from human embryonic kidney 293 cells co-transfected with GRIF-1aFLAG and β2 subunit clones. Further, immobilized GRIF-1-(8–633) specifically precipitated in vitro GABAA receptor α1 and β2 subunit immunoreactivities from detergent extracts of adult rat brain. The respective GABAA receptor β2 subunit/GRIF-1 binding domains were mapped using the yeast two-hybrid reporter gene assays. A possible role for GRIF-1 as a GABAA receptor β2 subunit trafficking factor is proposed. γ-aminobutyric acid amino acid(s) activation domain binding domain γ-aminobutyric acid type A receptor interacting factor-1 human embryonic kidney huntingtin-associated protein intracellular loop neuron-specific γ,γ′-enolase rapid amplification of cDNA ends transmembrane region Tris-buffered saline group of overlapping clones γ-Aminobutyric acid type A (GABAA)1receptors, fast-acting ligand-gated chloride ion channels, are the major inhibitory neurotransmitter receptors in the mammalian central nervous system. There are multiple GABAA receptor subunit genes encoding the α1–6, β1–3, γ1–3, δ, ε, θ, and π subunits. These subunits co-assemble in different pentameric combinations throughout development to form functional receptors (reviewed in, e.g., Ref. 1Stephenson F.A. Stephenson F.A. Turner A.J. Amino Acid Neurotransmission. Portland Press, London1998: 65-92Google Scholar). The major GABAAreceptor subtype in adult brain is composed of α1, β2/3, and γ2 subunits with a probable stoichiometry of 2α1, 2β2/3, 1γ2. This subunit combination is expressed predominantly at the synapse, whereas other subunit combinations have different subcellular localizations, e.g. δ subunit-containing GABAAreceptors are localized extrasynaptically in cerebellar granule cells (2Nusser Z. Sieghart W. Somogyi P. J. Neurosci. 1998; 18: 1693-1703Crossref PubMed Google Scholar) and α2 subunit-containing receptors are found uniquely at axon initial segments in hippocampal pyramidal cells (3Nusser Z. Sieghart W. Benke D. Fritschy J.-M. Somogyi P. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 11939-11944Crossref PubMed Scopus (341) Google Scholar). By analogy with other neurotransmitter receptor systems, it is thought that GABAA receptor-associated proteins exist that determine receptor subcellular localization, mediate receptor clustering, and regulate receptor activity. For example, Sigel and colleagues (4Kannenberg K. Baur R. Sigel E. J. Neurochem. 1997; 68: 1352-1360Crossref PubMed Scopus (39) Google Scholar) reported co-purification of α1 subunit-containing GABAAreceptors and several proteins including actin and tubulin. Some of these proteins were identified as GABAA receptor-tubulin complex-associated proteins, thus showing a link between the receptors and the cytoskeleton. One of these proteins, GABAAreceptor-tubulin complex-associated protein 34, was shown to be a novel serine kinase with specificity for β3 subunits (5Kannenberg K. Schaerer M.T. Fuchs K. Sieghart W. Sigel E. J. Biol. Chem. 1999; 274: 21257-21264Abstract Full Text Full Text PDF PubMed Scopus (8) Google Scholar). More recently, the same group identified this protein as the mitochondrial, multifunctional protein, gC1q-R (6Schaerer M.T. Kannenberg K. Hunziker P. Baumann S. Sigel E. J. Biol. Chem. 2001; 276: 26597-266604Abstract Full Text Full Text PDF PubMed Scopus (23) Google Scholar). The glycine receptor-associated protein, gephyrin, has also been implicated in the synaptic clustering of GABAA receptors, although a direct association between the two proteins has not been shown (reviewed in Refs. 7Betz H. Nat. Neurosci. 1998; 1: 541-543Crossref PubMed Scopus (41) Google Scholar and 8Sassoe-Pognetto M. Fristchy J.-M. Eur. J. Neurosci. 2000; 12: 2205-2210Crossref PubMed Scopus (97) Google Scholar). Knock-out γ2 (−/−) mice have a parallel deficiency in both gephyrin and clustered post-synaptic GABAA receptors (9Essrich C. Lorez M. Benson J.A. Fritschy J.-M. Luscher B. Nat. Neurosci. 1998; 1: 563-571Crossref PubMed Scopus (726) Google Scholar). More recently, the yeast two-hybrid system has been used to identify GABAA receptor-associated proteins. Proteins identified via this route include GABARAP (GABAA receptor-associated protein), Plic-1 (GABAA receptor-associated ubiquitin-like protein), and microtubule-associated protein 1B (MAP-1B). MAP-1B was shown to link GABAC receptors to the cytoskeleton at retinal synapses (10Hanley J.G. Koulen P. Bedford F. Gordon-Weeks P.R. Moss S.J. Nature. 1999; 397: 66-69Crossref PubMed Scopus (120) Google Scholar). GABARAP is a 13.9-kDa microtubule-associated protein that binds specifically to the γ2 subunit and promotes the clustering of GABAA receptors expressed in Qt-6 quail fibroblasts (11Wang H. Bedford F.K. Brandon N.J. Moss S.J. Olsen R.W. Nature. 1999; 397: 69-72Crossref PubMed Scopus (491) Google Scholar, 12Chen L. Wang H. Vicini S. Olsen R.W. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 11557-11562Crossref PubMed Scopus (186) Google Scholar). Differences in functional properties were reported between the clustered and unclustered receptors (12Chen L. Wang H. Vicini S. Olsen R.W. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 11557-11562Crossref PubMed Scopus (186) Google Scholar). It has also recently been reported to mediate intracellular transport of GABAA receptors by virtue of its specific interaction withN-ethylmaleimide-sensitive factor (13Kittler J.T. Rostaing P. Schiavo G. Fritschy J.-M. Olsen R.W. Triller A. Moss S.J. Mol. Cell. Neurosci. 2001; 18: 13-25Crossref PubMed Scopus (198) Google Scholar). Plic-1 (formerly GRUB1) is a GABAA receptor-associated ubiquitin-like protein that binds the α1 subunit and plays a role in the stabilization of cell surface receptors (14Bedford F.K. Kittler J.T. Uren J.M. Thomas P. Smart T.G. Moss S.J. Eur. J. Neurosci. Suppl. 2000; 12 (016.14)Google Scholar, 15Bedford F.K. Kittler J.T. Muller E. Thomas P. Uren J.M. Merlo D. Wisden W. Triller A. Smart T.G. Moss S.J. Nat. Neurosci. 2001; 4: 908-916Crossref PubMed Scopus (204) Google Scholar). We have used the yeast two-hybrid system to identify GABAAreceptor-associated proteins. Using the GABAA receptor β2 intracellular loop (β2-IL) as a bait, we have discovered a novel protein GABAA receptor interacting factor (GRIF-1) that is expressed predominantly in excitable tissues. The characterization of this protein is reported in this paper. The IL of the GABAA receptor β2 subunit was selected as a bait in a yeast two-hybrid screen. The coding sequence of the human β2-IL (aa 301–426; Ref. 16Ymer S. Schofield P.R. Draguhn A. Werner P. Kohler M. Seeburg P.H. EMBO J. 1989; 9: 1665-1670Crossref Scopus (353) Google Scholar) was amplified by the polymerase chain reaction (PCR) utilizing primers that containedEcoRI and BamHI restriction enzyme sites. The product was subcloned into the EcoRI and BamHI sites of the bait vector, pAS2–1 (Matchmaker Two-Hybrid System 2, CLONTECH, Palo Alto, CA). Likewise, constructs encoding Gal4 DNA binding domain (BD) fusion proteins of the rat GABAA receptor α1 (aa 304–384), β1 (aa 293–426), β3 (aa 293–426), and γ2L (aa 317–411) subunit ILs and β2-TM3/IL/TM4 (aa 292–439) and β2-IL/TM4 (aa 272–439) in pAS2–1 were prepared. The authenticity of all constructs was verified by nucleotide sequencing utilizing ABI PRISM dye terminator chemistry on an ABI 310 Genetic Analyzer and immunoblotting of respective yeast protein extracts. All bait constructs tested negative for auto-activation of reporter gene activity in the yeast two-hybrid reporter strains, CG1945 and AH109. Yeast cultures were grown on standard solid or in liquid media using either YPAD (yeast, peptone, adenine, dextrose; 2% (w/v) peptone, 1% (w/v) yeast extract, 2% (w/v) glucose, 0.003% (w/v) adenine) or SD (synthetic dropout medium; 0.67% (w/v) yeast nitrogen base (BD PharMingen, Cowley, Oxford, UK), 2% (w/v) glucose, 1× dropout supplement (CLONTECH) as appropriate for the selection involved). All transformations were performed using the lithium acetate/polyethylene glycol method (17Geitz R.D. Schiestl R.H. Willems A.R. Woods R.A. Yeast. 1995; 11: 355-360Crossref PubMed Scopus (1711) Google Scholar).Saccharomyces cerevisiae strain CG1945 (MATa, ura3–52, his3–200, ade2–101, lys2–801, trp1–901, leu2–3, 112, gal4–542, gal80–538, cyhr2, LYS2::GAL1UAS-GAL1TATA-HIS3, URA3::GAL417-mer (3x)-CYC1TATA-lacZ) was used to screen a rat brain cDNA library (in vector, pGAD10; CLONTECH) with pAS2–1β2-IL as a bait. Resulting colonies were assessed for reporter gene activation by initial nutritional selection (growth on SD −His) and subsequentlacZ activity assay. lacZ activity was determined by filter lift analysis, including appropriate positive and negative interaction controls. Clones encoding putative interacting proteins were isolated and analyzed following standard procedures (18Bartel P.L. Fields S. The Yeast Two-hybrid System. Oxford University Press, Oxford1997Google Scholar). Their association with the GABAA receptor β2-IL was verified by reconstitution assays with a selection of different bait constructs introduced by both yeast mating assays and repeated co-transformation. β-Galactosidase reporter gene activity was determined by liquid culture assays using o-nitrophenyl β-d-galactopyranoside as substrate according to protocols provided by CLONTECH. Briefly, double transformants of yeast strain AH109 expressing the appropriate GAL4 BD and AD fusions were grown overnight in selective medium followed by expansion to mid-log phase in YPAD medium. Cells were harvested, washed once with Z-buffer (0.1 m sodium phosphate, pH 7.0, 1 mmKCl, 1 mm MgSO4), and resuspended in 0.2 of the original volume in Z-buffer. For each sample, 100 μl of cell suspension was subjected to three freeze-thaw cycles in liquid nitrogen, and then 0.96 ml of Z-buffer containing 38 mmβ-mercaptoethanol and 0.16 mmo-nitrophenyl β-d-galactopyranoside was added. The mixture was incubated at 30 °C for 10 min to overnight depending on the time taken for color development. Reactions were stopped by the addition of 1 m Na2CO3 (0.4 ml), cellular debris removed by centrifugation, and OD at λ = 420 nm measured against Z-buffer blanks. Values of β-galactosidase activity were normalized to the cell density of the expansion culture and then expressed as a percentage of wild-type Gal4p. The full-length GRIF-1 cDNA was obtained by a combination of cDNA library screening and rapid amplification of cDNA ends (RACE). The clone resulting from the initial yeast two-hybrid screen was labeled with α-32P by the random primer method (19Sambrook J. Russell D.W. Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY2001Google Scholar). It was used to screen a rat brain cDNA library by colony hybridization. DNA from positive clones were sequenced and used for subsequent screenings. In addition, 5′- and 3′-RACE was performed on rat brain cDNA using the Marathon cDNA amplification system (CLONTECH). RACE products were TA-subcloned into pCR2.1 (Invitrogen, Groningen, The Netherlands) for further analysis. Cloned fragments and RACE PCR products were sequenced as described. The GRIF-1 contig was assembled using Lasergene software (DNAstar Inc., Madison, WI). The assembled cDNA and deduced amino acid sequence were subjected to analysis by a variety of predictive algorithms. BLAST searches were conducted against GenBankTM and EMBL data bases. Procrustes (20Gelfand M.S. Mironov A.A. Pevzner P.A. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 9061-9066Crossref PubMed Scopus (247) Google Scholar) analysis and a selection of hits obtained by BLAST searches against the human subset of GenBankTM and EMBL sequences were used to assemble the cDNA sequence of the human homologue of GRIF-1. Sequence alignments were performed with CLUSTAL. Structure predictions used a variety of publicly available resources such as psort (Ref. 21Nakai K. Horton P. Trends Biochem. Sci. 1999; 24: 34-35Abstract Full Text Full Text PDF PubMed Scopus (1838) Google Scholar; psort.nibb.ac.jp) and SMART (Ref. 22Schultz J. Copley R.R. Doerks T. Ponting C.P. Bork P. Nucleic Acids Res. 2000; 28: 231-234Crossref PubMed Scopus (1056) Google Scholar; smart.embl-heidelberg.de). GRIF-1 cDNA fragments were radiolabeled with α-32P by a standard random priming method and used to probe a rat multiple tissue Northern blot (CLONTECH). Hybridization was carried out using ExpressHyb solution according to the manufacturer's instructions. After final washing (0.1× SSC, 0.1% (w/v) SDS, 50 °C), the blot was exposed to imaging plates for 20 h. Plates were read with a PhosphorImager (Amersham Biosciences, Aylesbury, Bucks., UK) and analyzed using ImageQuant software. A set of PCR reactions with primers spanning the whole GRIF-1 coding region was employed to screen for alternative spliced isoforms in rat brain. Amplification was usually performed in a total volume of 25 μl containing 1× PCR buffer (10 mm Tris-HCl, pH 9.0, 50 mm KCl, 0.1% (v/v) Triton X-100; Promega Corp., Madison, WI), 1.5 mm MgCl2, 0.5 mm each dNTP, 0.2 μm each primer, 0.5 unit of Taq DNA polymerase, and 0.5 μl of cDNA (CLONTECH). Reactions were cycled through a profile consisting of 5 min of initial denaturation at 94 °C, followed by 30 cycles consisting each of 25 s at 94 °C, 25 s at 55 °C, and 1 min at 72 °C, followed by a final extension step of 7 min at 72 °C on a Hybaid OmniGene thermal cycler. PCR products were separated by flatbed agarose gel electrophoresis, visualized by ethidium bromide staining, and analyzed using the EDAS120 imaging system (Eastman Kodak Co.). A fusion protein was used to generate anti-GRIF-1 polyclonal antibodies in rabbits. For the fusion protein, the DNA encoding GRIF-1-(8–633), i.e. the original library clone, was subcloned into the bacterial expression vector, pTrcHisB (Invitrogen). BL 21 Escherichia coli transformed with the recombinant pTrcHisBGRIF-1-(8–633) were cultured overnight, diluted 1:50 with LB and grown at 37 °C until mid-log phase (OD λ = 600 nm= 0.6). Expression of the poly(His)-GRIF-1-(8–633) fusion protein was induced by the addition of 1 mmisopropyl-β-d-thiogalactopyranoside for 3–5 h at 30 °C. The E. coli culture was centrifuged at 5000 × g for 10 min at 4 °C, the supernatant discarded, and the pellet resuspended in lysis buffer, which was 20 mmsodium phosphate, pH 7.8, 6 m guanidine hydrochloride, 500 mm sodium chloride (10 ml) at 37 °C. The lysate was stirred gently at room temperature for 10 min followed by sonication (three 5-s pulses at high intensity) on ice. The cleared suspension was centrifuged at 3000 × g for 15 min at 4 °C and the poly(His)-GRIF-1-(8–633) fusion protein purified from the supernatant by Ni2+ affinity chromatography using ProBond columns exactly as per the manufacturer's instructions (Invitrogen) followed by preparative sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). The purified fusion protein, Mr ∼ 70 kDa, was used to generate anti-GRIF-1-(8–633) polyclonal antibodies in rabbits. Anti-GRIF-1-(8–633) antibodies were affinity-purified by GRIF-1-(8–633)-Ni2+-agarose affinity chromatography prior to use (23Harlow E. Lane D. Using Antibodies: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY1999Google Scholar). SDS-PAGE was carried out using 10% polyacrylamide slab gels under reducing conditions. Samples were prepared using the chloroform/methanol method of protein precipitation, and immunoblotting was carried out as previously described using affinity-purified anti-GABAA receptor α1 subunit (24Pollard S. Thompson C.L. Stephenson F.A. J. Biol. Chem. 1995; 270: 21285-21290Abstract Full Text Full Text PDF PubMed Scopus (63) Google Scholar), anti-β2 subunit (25Pollard S. Stephenson F.A. Biochem. Soc. Trans. 1997; 25: 547SCrossref PubMed Scopus (3) Google Scholar), affinity-purified anti-GRIF-1-(8–633), and anti-neuron-specific γ,γ′-enolase (NSE) antibodies in conjunction with the enhanced chemiluminescence (ECL) Plus Western blotting detection system (Amersham Biosciences) for the detection of immunoreactive species. Anti-rabbit and mouse immunoglobulin horseradish peroxidase-linked whole antibodies (Amersham Biosciences) were used at a final dilution of 1:2000. Where applicable, immunoblots were quantified by densitometry using a Personal Densitometer and ImageQuant (AmershamBiosciences) in the linear range of the film. Tissues from rat forebrain, cerebellum, liver, kidney, heart, and spleen were each homogenized in nine volumes of homogenizing buffer (20 mm Tris-HCl, pH 7.4, 0.25 m sucrose, containing benzamidine (1 μg/ml), bacitracin (1 μg/ml), soybean trypsin inhibitor (1 μg/ml), chicken egg trypsin inhibitor (1 μg/ml), and phenylmethylsulfonyl fluoride (1 mm)) at 4 °C using a hand-held glass-glass homogenizer. Skeletal and cardiac muscle was removed, minced using scissors, and homogenized using an Ultra-Turrax (six 10-s pulses at medium speed) followed by homogenization as above. Homogenates were centrifuged at 600 ×g for 10 min at 4 °C. The supernatants were collected and the P1 nuclear pellets re-homogenized in three volumes of homogenizing buffer and re-centrifuged at 600 × g for 10 min at 4 °C. Both supernatants were combined and centrifuged at 100,000 × g for 40 min at 4 °C to yield the soluble (S) and P2 membrane fractions. The P1 and P2 pellets were re-suspended in homogenizing buffer and stored at −20 °C until use. The cDNA encoding the full-length GRIF-1a was obtained by PCR from rat brain cDNA and subcloned into both the mammalian expression vectors, pCIS and pCMV-Tag4 (Stratagene, La Jolla, CA). The latter yielded a C-terminal FLAG-tagged GRIF-1. HEK 293 cells were cultured and transfected in 250-ml flasks using the calcium phosphate method as described previously (26Hawkins L.M. Chazot P.L. Stephenson F.A. J. Biol. Chem. 1999; 274: 27211-27218Abstract Full Text Full Text PDF PubMed Scopus (105) Google Scholar). Cells were transfected with pCIS GRIF-1a or pCMV-Tag4GRIF-1a alone (10 μg of DNA) or with the GABAA receptor clones pCDM8α1 (bovine), pCISβ2 (rat), pCDM8γ2L (bovine), and pCISGRIF-1 or pCMV-TagGRIF-1 in a 1:1:1:1 ratio with a total of 10 μg of DNA. Cells were harvested 24–48 h following transfection and analyzed by immunoblotting. For immunocytochemical studies, transfections were carried out by the calcium phosphate method on HEK 293 cells adhered to poly-l-lysine-coated cover slips. Transfected cells were cultured for 40 h, the cover slips washed three times for 5 min each at 20 °C with 20 mm Tris-HCl, pH 7.25, 145 mm NaCl (TBS), followed by fixation with 4% (w/v) paraformaldehyde in TBS for 10 min at 20 °C. Cells were washed three times for 5 min each with TBS containing 0.15% (v/v) Triton X-100 followed by three 5-min washes with TBS containing 20% (v/v) goat serum, 2% (w/v) bovine serum albumin, and 0.1% (w/v)dl-lysine. Transfected cells were incubated for 2 h at 20 °C with the primary antibody diluted appropriately in TBS containing 10% (v/v) goat serum, 1% (w/v) bovine serum albumin, and 0.1% (w/v) dl-lysine followed by five 5-min washes with TBS at 20 °C. Incubation with the secondary antibody, either goat anti-rabbit Ig-Alexa Fluor 594 or goat anti-mouse Ig-AlexaFluor 488 (Molecular Probes, Eugene, OR) diluted 1:150 in TBS. Finally, cells were washed five times for 5 min each time with TBS, mounted in UV-free mounting medium (H.D. Supplies, Muratech Scientific, Aylesbury, Bucks., UK), and viewed with a Leica TCSSP confocal microscope. HEK 293 cells were co-transfected with pCMV-Tag4GRIF-1a + pCISβ2; pCISβ2 + pCIS, pCMV-Tag4GRIF-1a + pCDM8α1, or pCMV-Tag4GRIF-1a + pCIS (1:1 ratio with 10 μg of total DNA). Cells were harvested 48 h after transfection and solubilized for 1 h at 4 °C with 10 mm HEPES, 150 mm NaCl, 5 mm EDTA, 5 mm EGTA, 1% (v/v) Triton X-100, and protease inhibitors as described above under "Subcellular Fractionation of Rat Tissues." The solubilized material was collected by centrifugation at 100,000 × g, diluted 1:3 in 10 mm HEPES, 50 mm NaCl, 16 mm KCl, 1 mm EDTA + protease inhibitors, and incubated (8 ml) with anti-FLAG M2-agarose (30 μl; Sigma, Poole, Dorset, UK) for 1 h at 37 °C. Samples were centrifuged for 20 s at 3000 ×g and the pellet washed with two 0.5-ml (17 volumes) amounts of 10 mm HEPES, 75 mm NaCl, 12.5 mmKCl, 2 mm EDTA, 1.25 mm EGTA + protease inhibitors, 0.25% (v/v) Triton X-100, followed by 17 volumes of the same buffer but with a Triton X-100 concentration of 1% (v/v). The pellets were collected by centrifugation as above and analyzed by immunoblotting. BL 21 cells were transformed with either pTrcHisBGRIF-1-(8–633) or the empty, pTrcHisB expression vector and the respective bacterial cell lysates prepared as detailed under "Generation of Anti-GRIF-1 Antibodies." ProBond resin (Invitrogen) was washed three times with seven volumes of distilled H2O followed by seven volumes of denaturing binding buffer, which was 20 mm sodium phosphate, pH 7.8, 8 m urea, 500 mm NaCl. The resin (25 μl) was incubated with either the pTrcHis-GRIF-(8–633) lysate, the pTrcHis lysate or lysis buffer (1 ml) for 10 min at 20 °C. Each resin was then washed two times with 14-volumes native binding buffer, which was 20 mm sodium phosphate, pH 7.8, 500 mm NaCl. A detergent-solubilized rat forebrain P2 extract was prepared by re-suspension of a 100,000 × gP2 pellet in 10 mm HEPES, pH 7.5, 50 mm KCl, 1 mm EDTA, 1% (v/v) Triton X-100, benzamidine HCl (1 μg/ml), bacitracin (1 μg/ml), soybean trypsin inhibitor (1 μg/ml), chicken egg trypsin inhibitor (1 μg/ml), and phenylmethylsulfonyl fluoride (1 mm) to a concentration of 3 mg of protein. Solubilization was carried out for 1 h at 4 °C and the detergent extract cleared by centrifugation at 100,000 ×g for 40 min at 4 °C. The derivatized resins were incubated with the solubilized P2 membrane proteins diluted 1:3 with 10 mm HEPES, pH 7.5, 17 mm KCl, with the protease inhibitors as above to give final concentrations of 0.25% (v/v) Triton X-100 and 25 mm KCl (1 ml of 1 mg/ml protein) overnight at 4 °C. The supernatant was removed by centrifugation of the resins at 200 × g for 10 s at 4 °C. Each resin was washed once with 40 volumes of 20 mm sodium phosphate, pH 7.8, 500 mm NaCl., 0.05% (v/v) Triton X-100. The bound proteins were extracted with SDS-PAGE sample buffer (50 μl), heated for 5 min at 100 °C, and the supernatants analyzed by quantitative immunoblotting using affinity-purified anti-β2 fragment 381–395 (25Pollard S. Stephenson F.A. Biochem. Soc. Trans. 1997; 25: 547SCrossref PubMed Scopus (3) Google Scholar), anti-α1 fragment 413–429 (27Duggan M.J. Stephenson F.A. J. Neurochem. 1989; 53: 132-139Crossref PubMed Scopus (27) Google Scholar), and anti-NSE antibodies (Affiniti Research Products Ltd., Mamhead, Exeter, UK). The results were analyzed by paired Student's t test. A series of GRIF-1 fragments were constructed by either standard subcloning techniques from the isolated library clone, C3, or PCR amplification. These were subcloned into pGAD10, pACT2, or pGADT7 (CLONTECH) to yield Gal4 AD fusion proteins. Similarly, fragments of the β2-IL (aa 303–427, 303–370, 355–427, 371–427, 303–394, 324–394, and 324–427) were fused to the GAL4 BD in pAS2–1. The yeast strain AH109 (MATa, trp1–901, leu2–3, 112, ura3–52, his3–200, gal4Δ, gal80D, LYS2::GAL1UAS-GAL1TATA-HIS3, GAL2UAS-GAL2TATA-ADE2, URA3::MEL1UAS-MEL1TATA-lacZ; CLONTECH) was transformed with pair-wise combinations of β2-ILBD and GRIF-1AD fusions. The interaction between pairs of proteins was assayed by nutritional selection on SD −4 (−His, −Trp, −Ade, −Leu) plates as well as β-galactosidase and α-galactosidase (Mel1) activity. α-Galactosidase activity was determined by overlaying the plates with X-α Gal-top agar (0.4% (w/v) low melting point agarose, 0.001% (w/v) X-α Gal (Melford Laboratories, Cambridge, UK) in Z-buffer. Coloration usually occurred within 3–6 h. The GABAAreceptor β subunits have been reported to determine the subcellular localization of αβγ receptors (28Connolly C.N. Wooltorton J.R.A. Smart T.G. Moss S.J. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 9899-9904Crossref PubMed Scopus (120) Google Scholar). The β2 subunit is the most abundantly expressed of the β isoforms in adult brain (see, e.g., Refs. 29Li M. De Blas A.L. J. Biol. Chem. 1997; 272: 16564-16569Abstract Full Text Full Text PDF PubMed Scopus (61) Google Scholar and 30Sur C. Wafford K.A. Reynolds D.S. Hadingham K.L. Bromidge F. Macaulay A. Collinson N. O'Meara G. Howell O. Newman R. Myers J. Atack J.R. Dawson G.R. McKernan R.M. Whiting P.J. Rosahl T.W. J. Neurosci. 2001; 21: 3409-3418Crossref PubMed Google Scholar). Therefore, the GABAAreceptor β2-IL was selected as a bait to identify β subunit interacting proteins that may function as targeting or clustering molecules. A yeast two-hybrid cDNA library screen of 1.5 × 106 independent transformants with the GABAAreceptor β2-IL as the bait yielded initially 36 positively growing clones of which one, C3, showed lacZ reporter gene activity. The plasmid isolated contained a 1.8-kb insert with an in-frame fusion encoding 626 aa. Re-transformation of this clone into yeast strain, AH109, with a variety of different GABAA receptor subunit IL bait constructs, i.e. α1, β1, β3, and γ2-ILs, showed that C3 interacted specifically with the β2-IL (Table I). Further, C3 did not yield reporter gene activity when alternative β2-IL constructs that incorporated either both TM3 and TM4, i.e. β2-TM3/ILTM4, or TM4 only, i.e. β2-IL/TM4, were used as baits (Table I). Quantification of lacZ reporter gene activity revealed a relative strength of the C3/β2-IL interaction of ∼1.5% relative to 100% of the pCL-1 (Gal4p wild-type) positive control (Table I).Table IA summary of the specificity of the GRIF-1/GABAA receptor subunit interactions in the yeast two-hybrid systemGal4 binding domain fusion constructActivation domain constructHIS3, ADE2, lacZ, MEL1 reporter gene activitieslacZ activity%pAS2–1α1-ILpGAD10GRIF-1-(8–633)−NDpAS2–1β1-ILpGAD10GRIF-1-(8–633)−NDpAS2–1β2-ILpGAD10GRIF-1-(8–633)+1.5 ± 0.4 (n = 3)pAS2–1β3-ILpGAD10GRIF-1-(8–633)−NDpAS2–1β2-TM3/IL/TM4pGAD10GRIF-1-(8–633)−NDpAS2–1β2-IL/TM4pGAD10GRIF-1-(8–633)−NDpAS2–1γ2-ILpGAD10GRIF-1-(8–633)−NDpAS2–1β2-ILpGADT7GRIF-1-(1–913)+1.7 ± 0.4 (n = 3)pAS2–1β2-ILpGADT7KIAA0549-(1–483)+1.2 ± 0.4 (n = 3)pAS2–1β2-ILpGAD10KIAA0549-(446–914)−NDpGAD10GRIF-1-(8–633)−NDpVA3–1pGAD10GRIF-1-(8–633)−NDpVA3–1PTD1–1+++100PCL-1+++100 Open table in a new tab Nucleotide sequencing revealed that C3 contained an open reading frame of novel sequence fused in-frame to the GAL4 activation domain. The full-length cDNA of the protein was obtained by cDNA library screening using the C3 probe in conjunction with 5′- and 3′-RACE. The assembled cDNA sequence (3641 bp) contains an open reading frame of 2739 bp predicting a protein of 913 aa residues, a calculated molecular mass of 102 kDa, and an acidic pI = 5.14. Within the protein, the original two-hybrid interacting fragment (C3) extends from aa 8 to 633. We have called the protein, GRIF-1, forGABAAreceptorinteracting factor. The nucleotid