The git5 Gß and git11 G Form an Atypical Gß Dimer Acting in the Fission Yeast Glucose/cAMP Pathway

Corresponding author: Charles S. Hoffman, Boston College, Biology Department, Higgins Hall 401B, Chestnut Hill, MA 02467., hoffmacs{at}bc.edu (E-mail)

Communicating editor: J. RINE

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Fission yeast adenylate cyclase, like mammalian adenylate cyclases, is regulated by a heterotrimeric G protein. The gpa2 G and git5 Gß are both required for glucose-triggered cAMP signaling. The git5 Gß is a unique member of the Gß family in that it lacks an amino-terminal coiled-coil domain shown to be essential for mammalian Gß folding and interaction with G subunits. Using a git5 bait in a two-hybrid screen, we identified the git11 G gene. Co-immunoprecipitation studies confirm the composition of this Gß dimer. Cells deleted for git11 are defective in glucose repression of both fbp1 transcription and sexual development, resembling cells lacking either the gpa2 G or the git5 Gß. Overexpression of the gpa2 G partially suppresses loss of either the git5 Gß or the git11 G, while mutational activation of the G fully suppresses loss of either Gß or G. Deletion of gpa2 (G), git5 (Gß), or git11 (G) confer quantitatively distinct effects on fbp1 repression, indicating that the gpa2 G subunit remains partially active in the absence of the Gß dimer and that the git5 Gß subunit remains partially active in the absence of the git11 G subunit. The addition of the CAAX box from the git11 G to the carboxy-terminus of the git5 Gß partially suppresses the loss of the G. Thus the G in this system is presumably required for localization of the Gß dimer but not for folding of the Gß subunit. In mammalian cells, the essential roles of the Gß amino-terminal coiled-coil domains and G partners in Gß folding may therefore reflect a mechanism used by cells that express multiple forms of both Gß and G subunits to regulate the composition and activity of its G proteins.

HETEROTRIMERIC G proteins, consisting of , ß, and subunits, relay external signals detected by ligand-activated seven-transmembrane receptors to a variety of effector molecules in eukaryotic cells (SIMON et al. 1991 ; NEER 1995 ). In the inactive state, the GDP-bound G subunit is associated with the Gß dimer to form the heterotrimer. Upon ligand binding, the receptor stimulates GDP release from G, allowing G to subsequently bind GTP. This nucleotide exchange activates the G protein by triggering a conformational change in G and its dissociation from the Gß dimer. In the activated state, the G subunit and the Gß dimer are free to regulate the activity of downstream effectors including adenylate cyclase, phospholipase C, mitogen-activated protein kinase (MAPK) cascades, and ion channels (GILMAN 1987 ; NEER 1995 ).

Genetic studies and genomic sequencing of the budding yeast Saccharomyces cerevisiae and the fission yeast Schizosaccharomyces pombe show that both organisms possess two G genes, but only one Gß gene. Due to the small size and weak conservation of G subunits, it is harder to identify the G genes in silico. In S. cerevisiae, the Gpa1 G, Ste4 Gß, and Ste18 G regulate the pheromone response pathway (DIETZEL and KURJAN 1987 ; WHITEWAY et al. 1989 ), with the Gß dimer activating a MAPK pathway (HIRSCH and CROSS 1992 ). The Gpa2 G, in conjunction with the G-protein-coupled receptor-like protein Gpr1, monitors glucose to activate adenylate cyclase (NAKAFUKU et al. 1988 ; COLOMBO et al. 1998 ; XUE et al. 1998 ; YUN et al. 1998 ). There does not appear to be a Gß acting in this signaling pathway. In S. pombe, the gpa1 G is a positive regulator of the pheromone response pathway (OBARA et al. 1991 ). As with S. cerevisiae Gpa2, gpa1 does not appear to associate with a Gß, although one study erroneously concluded that the git5/gpb1 Gß is a negative regulator of gpa1 (KIM et al. 1996 ). The S. pombe gpa2 G, in concert with the git5/gbp1 Gß and the G-protein-coupled receptor-like protein git3, activate adenylate cyclase in a glucose monitoring pathway (ISSHIKI et al. 1992 ; NOCERO et al. 1994 ; LANDRY et al. 2000 ; WELTON and HOFFMAN 2000 ).

The S. pombe gpa2 gene was also identified as git8 (git, glucose insensitive transcription) in a mutant screen for git genes required for glucose repression of transcription of the fbp1 gene that encodes the gluconeogenic enzyme fructose-1,6-bisphosphatase (HOFFMAN and WINSTON 1990 ; NOCERO et al. 1994 ). The gpa2/git8 gene, along with git1, git3, git5, git7, and git10, is required for adenylate cyclase (encoded by git2/cyr1) activation in response to glucose detection (HOFFMAN and WINSTON 1991 ; BYRNE and HOFFMAN 1993 ). The git5 gene, identical to gbp1, encodes a Gß subunit that positively regulates gpa2 (LANDRY et al. 2000 ). A git5 deletion confers the same phenotypes as a gpa2 deletion, including derepression of fbp1 transcription and starvation-independent conjugation and sporulation. Strains carrying a git5 point mutation or deletion display a defect in glucose-triggered cAMP signaling, although basal cAMP levels are unaffected (BYRNE and HOFFMAN 1993 ; LANDRY et al. 2000 ). In addition, the git5 deletion is partially suppressed by multicopy gpa2⁺ (LANDRY et al. 2000 ) and fully suppressed by an activated allele of gpa2, gpa2^R176H (WELTON and HOFFMAN 2000 ). Finally, as conjugation by git5 deletion strains remains pheromone dependent, git5 does not negatively regulate the gpa1-mediated pheromone response pathway (LANDRY et al. 2000 ).

Gß subunits comprise a highly conserved protein family whose structure includes an amino-terminal coiled-coil followed by a seven-bladed WD repeat ß-barrel (Fig 1; WALL et al. 1995 ; SONDEK et al. 1996 ). The git5 Gß is remarkable in that while it is 43% identical to members of the Gß family, it lacks the amino-terminal coiled-coil domain that includes 15 residues shown to form contacts with the G subunit in the mammalian Gß dimer (Fig 1; SONDEK et al. 1996 ; LANDRY et al. 2000 ). In mammalian systems, this domain appears to be essential for both Gß folding and assembly of the Gß dimer (GARRITSEN et al. 1993 ; WALL et al. 1995 ; GARCIA-HIGUERA et al. 1996 ; LAMBRIGHT et al. 1996 ; SONDEK et al. 1996 ; PELLEGRINO et al. 1997 ).

View larger version (36K): In this window In a new window Download PPT slide   Figure 1. Protein modeling of the git5 Gß. (A) The git5 structure was determined by computer modeling of the git5 amino acid sequence (accession no. AAD09020) using the Swiss-Model program (PEITSCH 1996 ) and displayed via RasMol (SAYLE and MILNER-WHITE 1995 ). The amino terminus (N) of the git5 protein is indicated. (B) Model of the bovine Gß dimer as determined by SONDEK et al. 1996 . The Gß subunit, possessing an additional 36 amino-terminal residues relative to git5, is in blue and the G subunit is in red.

To test whether the S. pombe git5 Gß interacts with a G subunit, we conducted a two-hybrid screen to identify S. pombe proteins that physically interact with git5. One clone obtained from this screen encodes a recognizable G subunit possessing several lysine residues and a CAAX-box (CASEY 1994 ) at the carboxy-terminus. Co-immunoprecipitation studies in vivo confirm this interaction. Deletion of this gene, designated git11, confers phenotypes associated with a defect in glucose detection that, like a deletion of the git5 Gß gene, are partially suppressed by overexpression of the gpa2 G gene. These results identify git11 as the functional G partner of git5; thus the git5 Gß does not require an amino-terminal coiled-coil to assemble into a functional Gß dimer.

Additional characterization of the roles of the G, Gß, and G subunits in glucose repression of fbp1 transcription suggests that the git5-git11 Gß dimer is required for activation of the gpa2 G subunit and that, contrary to the data from mammalian studies, the git5 Gß retains some function in the absence of a G partner. Thus, the dependence upon the Gß amino-terminal coiled-coil and the G subunit for proper folding of mammalian Gß subunits (GARRITSEN et al. 1993 ; GARCIA-HIGUERA et al. 1996 ; PELLEGRINO et al. 1997 ) is not an intrinsic trait of Gß subunits. These features of mammalian G proteins not observed in S. pombe may reflect a mechanism employed by cells that express multiple forms of the both Gß and G subunits to tightly control the repertoire and activity of Gß dimers.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

S. pombe strains and growth media:
S. pombe strains used in this study are listed in Table 1. The fbp1::ura4⁺ and ura4::fbp1-lacZ reporters have been previously described (HOFFMAN and WINSTON 1990 ). Rich medium YEA (GUTZ et al. 1974 ) was supplemented with 2% casamino acids. Defined PM media (WATANABE et al. 1988 ) were supplemented with required nutrients at 75 mg/liter, except for leucine, which was at 150 mg/liter. SC solid medium containing 0.4 g/liter 5-fluoroorotic acid (5-FOA) and 8% glucose (HOFFMAN and WINSTON 1990 ) was used to determine 5-FOA sensitivity. Strains were grown at 30° unless otherwise indicated.

Recombinant DNA methodology:
All DNA manipulations were performed, unless otherwise stated, using reagents and protocols from New England Biolabs (Beverly, MA). Escherichia coli transformations were done using XL1-Blue electroporation competent cells (Stratagene, La Jolla, CA). The Expand High Fidelity PCR system (Roche Molecular Biochemicals, Indianapolis) was used for PCR reactions, according to the manufacturer's instructions. Oligonucleotides were obtained from Integrated DNA Technologies (Coralville, IA). S. pombe plasmid transformations were performed by overnight incubation in a polyethylene glycol-LiOAc-TE buffer as previously described (DAL SANTO et al. 1996 ).

Two-hybrid screening:
A two-hybrid screen was carried out to identify proteins that interact with the git5 Gß. The git5 bait was PCR amplified from pSL11 (LANDRY et al. 2000 ) using oligonucleotides 5-3PTH 5' GAAGATCTTGAAGCAGTCAACCTCCTAGAATCGA 3' and 5-5PTH 5' CCGGCCATGGAGGCCATGGATTCTGGGTCAAGAGTAAACGT 3'. The PCR product was digested with SfiI and BglII and ligated into SfiI- and BamHI-digested pAS2 (HARPER et al. 1993 ) to form plasmid pSL13 that expresses a Gal4 binding domain (GBD)-git5 fusion protein. Plasmid pSL13 was cotransformed with a S. pombe two-hybrid cDNA library in pACT [expressing fusions to the Gal4 activation domain (GAD); DURFEE et al. 1993] into YRG-2 competent yeast cells (Stratagene) according to the manufacturer's protocol. From 3.5 x 10⁵ transformants, 81 His⁺ candidates were screened for ß-galactosidase activity by Xgal filter lift assay (HOFFMAN and WINSTON 1990 ). Ten positive candidates were rescued into E. coli (HOFFMAN and WINSTON 1987 ), and the inserts were sequenced by Bioserve Biotechnologies, (Laurel, MD) using primer TH5-F1 5' CGTTTGGAATCACTACAGGG 3'. Of these 10 plasmids, only pSL20 displayed a bait-specific interaction. Plasmid pSL20 contains the entire git11 coding region with the exception of the start codon (see RESULTS). Specificity of the interaction was determined by cotransformation of bait plasmid pSL13 and prey plasmid pSL20 in appropriate combinations with pSE1112 (GBD-Snf1) or pSE1111 (GAD-Snf4; FIELDS and SONG 1989 ) followed by testing for growth on SC-Trp-Leu-His medium containing 25 mM 3-aminotriazole (3AT) and for ß-galactosidase production.

Construction of functional git11 clones:
Functional clones of the git11 gene were constructed as follows. Plasmid pSL24, expressing an HA-tagged form of git11 on a URA3⁺-based vector, was created by PCR amplifying git11 from plasmid pSL20 using oligonucleotides git11-HA-for 5' CTACTAGCTAGCATGGAAACAGAGGCTTTATTGAATG 3' (that restores the START codon) and git11-HA-rev 5' CGGGGTACCTTAGGAAATAGTACAGCATTTGGTAGTGGC 3'. The PCR product was gel purified, digested with NheI and KpnI, and ligated with NheI- and KpnI-treated plasmid pALU (CHANG et al. 1994 ). Plasmid pSL25 was created by replacing the 1.8-kb HindIII fragment containing the URA3⁺ selectable marker in pSL24 with a 2.2-kb HindIII fragment carrying the LEU2⁺ selectable marker from pARTCM (CHANG et al. 1994 ).

git5 plasmid constructions:
Three plasmids expressing git5 Gß derivatives from the S. pombe nmt41 promoter were constructed by the insertion of PCR products into the pNMT41-TOPO vector (Invitrogen, Carlsbad, CA) according to the manufacturer's instructions. KpnI-linearized plasmid pSL12 (LANDRY et al. 2000 ) was used as the template DNA for the PCR reactions. The insert in plasmid pSL27, which expresses the wild-type git5 protein, was generated using primers git5-for-topo 5' ATGGATTCTGGGTCAAGAGTA 3' and git5-rev-topo 5' TTACCCTGACGAAGACCAGAGAC 3'. The insert in plasmid pSL28, which expresses a functional git5-V5 tagged protein [the V5 tag (SOUTHERN et al. 1991 ) is contributed by vector sequences], was generated using primers git5-for-topo 5' ATGGATTCTGGGTCAAGAGTA 3' and git5V5-rev 5' CCCTGACGAAGACCAGAGAC 3'. The insert in plasmid pSL29, which expresses the 305-amino-acid git5 Gß protein fused to the carboxy-terminal nine amino acids of the git11 G protein, was generated using primers git5-for-topo 5' ATGGATTCTGGGTCAAGAGTA 3' and git5-CAAX-rev 5' TTAGGAAATAGTACAGCATTTGGTAGTGGCCCCTGACGAAGACCAGAGAC-3'.

Deletion of the git11 gene:
Strains deleted for the git11 gene were constructed using the PCR-based gene targeting method of BAHLER et al. 1998 . Oligonucleotides git11-deltafor 5' TACTAGGTGAGCACAGACGGTAGGAAGTGCACGTAAGATGCTTAAACAACGTTCCACAAAACACGGATCCCCGGGTTAATTAA 3' and git11-deltarev 5' CAAGGCTATAATTTTACTTAACAGGCATTACTTATTGAAATTGTAGTTGATCGGTCCTTAAACAAGAATTCGAGCTCGTTTAAAC 3' were used to PCR amplify the kanMX6 cassette from pFA6a-GFP(S65T)-kanMX6 (WACH et al. 1997 ) such that the product was flanked with sequences from either side of the git11 open reading frame (ORF). The PCR product was used to transform strain FWP72 to G418 resistance. The git11 deletion was confirmed by PCR using oligonucleotides git11-test 5' CCAAGCAAAATCGCATCTA 3' and intKANtest 5' CATCCTATGGAACTGCCTCGG 3'.

Multicopy suppression analyses:
S. pombe strains FWP72 (wild type), CHP439 (gpa2), CP477 (git5), and SLP17 (git11) were transformed to Leu⁺ with plasmids expressing gpa2 (pRW7; expressing a myc-gpa2 fusion, R. M. WELTON and C. S. HOFFMAN, unpublished results), git5 (pSL26; a derivative of pSL11, LANDRY et al. 2000 , expressing a 6his-tagged git5 gene), or git11 (pSL25), as well as with the pART1 empty vector control (MCLEOD et al. 1987 ). ß-Galactosidase activity was determined from two independent transformants for each host and plasmid combination as previously described (NOCERO et al. 1994 ). The values given are the average specific activity ± standard error from three separate cultures of each transformant grown to exponential phase in PM medium containing 8% glucose (repressing conditions).

Mating assays:
Homothallic (h⁹⁰) strains CHP362 (wild type), CHP481 (gpa2), CHP486 (git5), CHP558 (git2), and SLP44 (git11) were grown to exponential phase in PM liquid medium (at 37° to inhibit conjugation) and diluted to 10⁶ cells/ml in PM liquid medium with or without 5 mM cAMP. Cultures were incubated 30 hr at 30° without shaking and photographed.

Co-immunoprecipitation studies:
S. pombe strain CHP463 (git5) was cotransformed to Leu⁺, Ura⁺ with either plasmids pSL28 (git5-V5) and pSL24 (HA-git11), pSL28 (git5-V5) and pALU (empty vector control), or pNMT41-TOPO (empty vector control) and pSL24 (HA-git11). Cultures were grown to exponential phase in PM-ura-leu (0.1% glucose, 3% glycerol), harvested, washed twice with chilled distilled water, and resuspended at a concentration of 1000:1 in chilled lysis buffer [50 mM HEPES, pH 7.6, 150 mM NaCl, 0.5% Triton X-100, 1 mM dithiothreitol, 2 mM phenylmethylsulfonyl fluoride, and Complete Protease Inhibitor Cocktail (Roche Molecular Biochemicals)]. Cell lysates were made by glass bead lysis, vortexing five times in a mini-beadbeater (BioSpec Products, Bartlesville, OK) at 4° for 1 min at maximum speed with 1-min intervals on ice. Protein extracts were clarified and quantitated with the bicinchonic acid (BCA) kit (Pierce Chemical Co., Rockford, IL). Co-immunoprecipitations were performed as described by CELENZA et al. 1989 with three modifications. The immunoprecipitation buffer contained 0.5 mg/ml of BSA instead of ovalbumin. The extracts were precleared with 0.5 µg of normal mouse IgG (Santa Cruz Biotechnology, Santa Cruz, CA) and Protein A beads (Sigma, St. Louis) for 1 hr. Immunoprecipitation of HA-git11 was carried out by incubating extracts with 1 µg of -HA (Roche Molecular Biochemicals) for 2 hr followed by the addition of Protein A beads for 1.5 hr. Whole-cell extracts and -HA precipitated proteins were resolved on a 15% SDS polyacrylamide gel and transferred to Immobilon P (Millipore Corp., Bedford, MA). The filter was probed with HRP-conjugated -HA (Roche Molecular Biochemicals) and HRP-conjugated -V5 (Invitrogen) antibodies and visualized with HRP-conjugated goat -mouse antibody and LumiGlo Chemiluminescent Substrate Kit (Kirkegaard & Perry Laboratories, Gaithersburg, MD).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

A two-hybrid screen identifies a potential G partner for the git5 Gß:
To investigate whether the git5 Gß interacts with a G partner, we conducted a two-hybrid screen for S. pombe genes whose products physically interact with git5. The bait in the screen was the S. cerevisiae GBD fused to the 305-residue git5 protein. Candidate plasmids from the S. pombe two-hybrid library, whose products interact with git5, trigger the expression of HIS3 and lacZ reporter genes in the host strain YRG-2, resulting in His⁺ (3AT-resistant) growth and the production of ß-galactosidase that causes colonies to stain blue in an Xgal filter lift. Plasmid pSL20, identified in this screen, displays a bait-specific interaction (Fig 2A), as indicated by the fact that only the combinations of the GBD-git5 bait and the pSL20-encoded GAD-git11 prey, or the control GBD-Snf1 bait and GAD-Snf4 prey (FIELDS and SONG 1989 ), produce 3AT-resistant, Xgal-blue transformants.

View larger version (24K): In this window In a new window Download PPT slide   Figure 2. Identification of the git11 G. (A) Two-hybrid screening strain YRG-2 was cotransformed with bait plasmids expressing either GBD-git5 or GBD-Snf1 and prey plasmids expressing either GAD-git11 or GAD-Snf4. Duplicate transformants were tested for growth in the absence or presence of 25 mM 3AT and for ß-galactosidase activity by Xgal filter lift. The results suggest that git5 and git11 physically interact, as do Snf1 and Snf4. (B) The amino acid sequence alignment of the predicted git11 protein and two human G subunits. The git11 sequence (accession no. CAA22118) was aligned with the human gamma 1 (accession no. Q08447) and human gamma 4 (accession no. NP_004476) protein sequences using the Clustal W (version 1.8) sequence alignment program (THOMPSON et al. 1994 ) and displayed using BOXSHADE. Identical residues are shaded in black with white letters, while conserved residues are shaded in gray with black letters.

DNA sequence analysis of the insert in pSL20 reveals a 71-codon open reading frame whose product bears key signatures of a G subunit. This gene is present on cosmid c215 (accession no. AL033534; open reading frame 4) from the S. pombe genome database (http://www.sanger.ac.uk/Projects/S_pombe/index.shtml), which carries a portion of chromosome 2. The sequence of the pSL20 cDNA insert, identical to base pairs 6367 to 6419 and 6509 to 7006 of cosmid c215, confirms the predicted splice junction. The genomic sequence indicates that only the start codon of this gene, designated git11, is missing from the cDNA clone. The 72-amino-acid git11 protein (accession no. CAA22118) is similar in length to mammalian G subunits and resembles the human G1 and G4 subunits (Fig 2B). Critical features shared by these proteins include lysine residues in the carboxy-terminal region of the proteins and a carboxy-terminal CAAX box, the site of prenylation needed to associate the Gß dimer with the peripheral membrane (CASEY 1994 ). Thus, git11 appears to be a viable candidate for the G partner to the git5 Gß.

Co-immunopreciptation of git5 Gß and git11 G:
We confirmed that the git5 and git11 proteins physically interact in S. pombe by co-immunoprecipitation. Protein extracts from cells expressing either a functional git5-V5 tagged Gß, a functional HA-git11 tagged G, or both tagged proteins were subjected to immunoprecipitation using anti-HA antibody. The precipitated proteins were examined by Western blot analysis for the presence of the git5-V5 Gß (see MATERIALS AND METHODS). Immunoblotting to detect the git5-V5 (Fig 3, top) or HA-git11 (Fig 3, bottom) proteins in whole-cell extracts revealed that the relative amount of git5-V5 was reduced in cells not overexpressing HA-git11 [ Fig 3, lanes 1 and 2; the host strain carries a git5 deletion while transcription of the endogenous git11 gene is very low (data not shown) compared to that of the adh-driven HA-git11 construct]. More importantly, the git5-V5 protein was detected only in the precipitated material from the extract from cells expressing HA-git11 (Fig 3, lanes 4–6). A similar interaction was seen using HA-git11 together with a git5-GFP tagged protein expressed from the git5 locus (data not shown). The git5-V5 tagged Gß expressed from plasmid pSL28 contains only the 305-codon git5 open reading frame fused to sequences encoding the V5 tag (SOUTHERN et al. 1991 ); thus the 305-amino-acid git5 Gß that lacks an amino-terminal coiled-coil domain physically interacts with the git11 G in vivo.

View larger version (39K): In this window In a new window Download PPT slide   Figure 3. Co-immunoprecipitation of the git5 Gß with the git11 G. Protein extracts were prepared from strain CHP463 (git5) transformed with either pSL28 (git5-V5) and pSL24 (HA-git11; lanes 1 and 4), pSL28 (git5-V5) and pALU (empty vector control; lanes 2 and 5), or pNMT41-TOPO (empty vector control) and pSL24 (HA-git11; lanes 3 and 6). Immunoblots were performed to detect git5-V5 (top) and HA-git11 (bottom) from whole-cell extracts (lanes 1–3). The HA-git11 protein was immunoprecipitated using -HA antibody (see MATERIALS AND METHODS) and the precipitated proteins were examined by immunoblot (lanes 4–6).

Deletion of the git11 gene confers phenotypes associated with defects in the glucose/cAMP pathway:
To test whether the putative git11 G acts in the glucose/cAMP pathway, we constructed git11 deletion strains and characterized them with regard to transcriptional regulation of the glucose-repressed fbp1 gene and regulation of conjugation and sporulation. Deletion of the git11 gene from strain FWP72 to create strain SLP17 causes a significant increase in ß-galactosidase expression from an integrated fbp1-lacZ reporter (HOFFMAN and WINSTON 1990 ; Table 2 and Table 3). SLP17 (git11) cells also display increased expression of an integrated fbp1-ura4⁺ reporter resulting in Ura⁺ and 5-FOA-sensitive growth similar to that of gpa2 and git5 strains, while FWP72 (git⁺) cells are Ura^- and 5-FOA resistant due to glucose repression of transcription from the fbp1 promoter (Fig 4; see empty vector control transformants). These growth phenotypes represent the criteria on which the original collection of git mutants, including gpa2/git8 and git5 mutants, was identified (HOFFMAN and WINSTON 1990 ). In addition and similar to our previous observations using strains carrying a point mutation or deletion of the git5 Gß gene (BYRNE and HOFFMAN 1993 ; LANDRY et al. 2000 ), SLP47 (git11) cells possess wild-type basal levels of intracellular cAMP but fail to increase intracellular cAMP in response to glucose (data not shown).

View larger version (58K): In this window In a new window Download PPT slide   Figure 4. Genetic analysis of gpa2, git5, and git11 deletions and multicopy expression on fbp1-ura4 expression. FWP72 (wild type), CHP439 (gpa2), CHP477(git5), and SLP17 (git11) strains were transformed with either an empty vector control (pARTCM; 6) or plasmids expressing gpa2 (pRW7), git5 (pSL26), or git11 (pSL25). Glucose repression of an integrated fbp1-ura4 reporter gene in these strains results in 5-FOA-resistant growth, while constitutive fbp1-ura4 expression results in 5-FOA sensitivity.

The git11 allele was introduced into a homothallic (h⁹⁰) strain to determine the effect of this deletion on the regulation of sexual development. Homothallic strains undergo mating-type switching to produce mating partners in a purified population. Wild-type cells growing in a nutrient-rich medium show little or no mating (Fig 5), whereas strains defective in glucose signaling due to the loss of gpa2 (G), git5 (Gß), or git2 (adenylate cyclase) readily mate and sporulate to produce asci (Fig 5; MAEDA et al. 1990 ; ISSHIKI et al. 1992 ; LANDRY et al. 2000 ). The h⁹⁰ git11 cells also display starvation-independent sexual development. This unregulated conjugation and sporulation is suppressed in all four mutant strains by the addition of 5 mM cAMP to the growth medium (Fig 5), indicating that the defect in these mutant strains is in cAMP signaling.

View larger version (79K): In this window In a new window Download PPT slide   Figure 5. Starvation-independent sexual development in git2 (adenlyate cyclase), gpa2 (G), git5 (Gß), and git11 (G) homothallic strains. Cells of homothallic strains CHP362 (wild type), CHP481 (gpa2), CHP486 (git5), CHP558 (git2), and SLP44 (git11) were pregrown at 37° to inhibit conjugation and then grown overnight in glucose- and nitrogen-rich medium (in the presence or absence of 5 mM cAMP) at 30° before photographing.

The git5 Gß and git11 G genes display the same genetic relationship to the gpa2 gene:
We previously showed that multicopy gpa2⁺ partially suppresses a mutation or deletion of the git5 Gß gene, while multicopy git5⁺ has no effect on a gpa2 deletion (NOCERO et al. 1994 ; LANDRY et al. 2000 ). We have now extended this analysis to include the git11 gene. Overexpression of gpa2 (by expression of a functional myc-gpa2 fusion from the adh promoter) completely suppresses a gpa2 deletion and partially suppresses git5 and git11 deletions with regard to glucose repression of an fbp1-lacZ reporter. Multicopy git5 and multicopy git11 only affect fbp1-lacZ expression in git5 and git11 deletion strains, respectively (Table 2). The same suppression pattern is observed when examining the ability of these plasmids to restore glucose repression of an fbp1-ura4 reporter resulting in 5-FOA^R growth (Fig 4). Meanwhile, deletion of either git5 or git11 is completely suppressed by the gpa2^R176H "activated" allele that carries a mutation in the coding region of the GTPase domain of the G subunit (Table 3; ISSHIKI et al. 1992 ; WELTON and HOFFMAN 2000 ). We previously showed that the gpa2^R176H allele suppresses mutations in git5 (Gß) and git3 (putative G-protein-coupled glucose receptor), but not in git1, git7, or git10 (WELTON and HOFFMAN 2000 ). Thus, git5 and git11 display the same genetic relationship to gpa2, consistent with a model in which these genes encode the two subunits of the Gß dimer acting in the S. pombe glucose/cAMP pathway.

Deletion of the gpa2 G, git5 Gß, and git11 G genes produce quantitatively distinct effects on fbp1-lacZ expression:
To characterize the relative roles in cAMP signaling, we measured the effects of gpa2 (G), git5 (Gß), and git11 (G) deletions on fbp1-lacZ expression (Table 3). The effect of each deletion on fbp1-lacZ expression is distinguishable with the gpa2 deletion causing a 250-fold increase, the git5 deletion causing a >100-fold increase, and the git11 deletion causing a >30-fold increase (Table 3). These results suggest that the G subunit remains partially active in the absence of the Gß dimer. They also indicate that either the Gß subunit remains partially active in the absence of the G subunit or that git11 is not the only G partner for the git5 Gß.

To investigate whether the git5-git11 Gß carries out any additional role in glucose monitoring other than to facilitate the activation of the gpa2 G, we co-overexpressed git5 and git11 in gpa2 mutant strains. This overexpression failed to reduce fbp1-lacZ expression in a gpa2-60 (reduction in function allele; HOFFMAN and WINSTON 1990 ; NOCERO et al. 1994 ) strain (FWP175) grown under glucose-rich conditions, which possessed 458 ± 38 units of ß-galactosidase activity relative to 518 ± 14 units detected in a control transformant carrying empty vectors pART1 and pALU. It therefore appears that the only role of the Gß dimer is to regulate G activity.

Bypass of a git11 deletion by the addition of the git11 CAAX box to git5:
The quantitative difference between a git5 deletion (Gß) and a git11 deletion (G; Table 2 and Table 3) indicates that either the Gß is partially active in the absence of G or that git11 is not the only G partner for git5. The suggestion that a Gß subunit can remain partially functional in the absence of a G partner is unprecedented. This would imply that the Gß subunit can properly fold in the absence of G and may require G only for localization of Gß to the peripheral membrane. If true, the addition of a CAAX box to the git5 Gß subunit might partially or fully bypass the need for an intact git11 G subunit. However, if there are multiple G subunits, and Gß dimer formation is required for function, the addition of a CAAX box to the Gß subunit would most likely inhibit dimer formation and G-protein activity. To distinguish between these two hypotheses, we fused the coding region for the carboxy-terminal nine residues from git11 to the git5 ORF and tested the ability of this construct to suppress either or both git5 and git11 deletions. The git5-CAAX chimeric protein, expressed from the nmt41 promoter on plasmid pSL29, suppressed the 5-FOA^S growth (indicating a restoration of glucose repression of the fbp1-ura4 reporter gene) to strains carrying either a git5 or git11 single deletion or a git5 git11 double deletion (Fig 6). Thus the addition of the CAAX box increases the function of git5 in the absence of git11, since expression of the wild-type git5 protein from the same promoter on plasmid pSL27 only suppressed the growth defect in the git5 single deletion strain (Fig 6). A quantitative analysis of the effect of these plasmids on expression of the fbp1-lacZ reporter in these same transformants shows that the git5-CAAX protein is only partially better than wild-type git5 in restoring glucose repression to a git5 git11 double deletion strain (Table 4). However, this analysis is complicated by the fact that the git5-CAAX protein is less functional than the wild-type git5 protein in a strain that expresses a functional git11 protein (strain CHP477, Table 4).

View larger version (45K): In this window In a new window Download PPT slide   Figure 6. Functional testing of the git5-CAAX chimeric protein. Strains FWP72 (wild type), CHP477 (git5), SLP33 (git5 git11), and SLP17 (git11) were transformed with either pNMT41-TOPO (empty vector control), pSL27 (nmt41-driven git5+), or pSL29 (nmt41-driven git5-CAAX; git5 fused to the carboxy-terminal nine codons of git11). Purified transformants were pregrown on PM-leu (8% glucose) medium. Equal numbers of cells were spotted to either PM-leu (-5-FOA) or 5-FOA-containing (+5-FOA) medium and grown for 2 days before photographing. Glucose repression of an integrated fbp1-ura4 reporter gene in these strains results in 5-FOA-resistant growth, while constitutive fbp1-ura4 expression results in 5-FOA sensitivity.

View this table: In this window In a new window   Table 4. Partial bypass of the loss of the git11 G by a git5-CAAX chimeric protein

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The cloning and characterization of the S. pombe git11 gene suggest that its product is the G subunit of the gpa2-git5-git11 heterotrimeric G protein responsible for adenylate cyclase activation in response to glucose detection. While git11 is a conventional G subunit, its git5 Gß partner lacks the amino-terminal coiled-coil domain that has been suggested by both structural and functional studies to be critical to the assembly of the Gß dimer. Our previous observation that the S. pombe git5 Gß lacks this domain (LANDRY et al. 2000 ) brought this system into conflict with one of two well-established paradigms of G-protein research. The first is that G proteins of this type, as opposed to the Ras-like monomeric G proteins, are heterotrimeric. This principle is already challenged by the fact that the S. cerevisiae Gpa2 and S. pombe gpa1 G subunits appear to function in the absence of a Gß dimer. Thus it seemed equally plausible that a heterodimeric G protein lacking a G subunit could exist. The second paradigm is that the Gß amino-terminal coiled-coil is essential for both the folding of the Gß subunit and its assembly with G (GARRITSEN et al. 1993 ; GARCIA-HIGUERA et al. 1996 ; PELLEGRINO et al. 1997 ). As we have demonstrated here, the interaction between the git5 Gß and the git11 G indicates that the WD repeat ß-barrel of the Gß subunit can be sufficient to allow Gß dimer assembly.

The functional relationship of the G-protein subunits in the S. pombe glucose/cAMP pathway is clearly different from that of the S. cerevisiae pheromone response pathway in which the Gß dimer activates a downstream MAPK pathway. Due to this functional relationship of the G-protein subunits in S. cerevisiae, mutations in the GPA1 G gene confer the opposite phenotypes to those associated with mutations in the STE4 Gß gene or the STE18 G gene (DIETZEL and KURJAN 1987 ; WHITEWAY et al. 1989 ; HIRSCH and CROSS 1992 ). In the S. pombe glucose/cAMP pathway, the gpa2 (G), git5 (Gß), and git11 (G) genes function cooperatively, as evidenced by the similar defects in fbp1 transcriptional regulation (Fig 4, Table 3) and nutrient regulation of sexual development (Fig 5) observed in gpa2, git5, and git11 mutants. These results, along with the multicopy suppression studies (Table 2), indicate that the gpa2 G is the key activator of adenylate cyclase in response to glucose detection and that the git5-git11 Gß is a positive regulator of G. Consistent with this model, the loss of the gpa2 G has a greater effect on glucose repression of fbp1 transcription than does the loss of the git5 Gß or the git11 G (Table 3). Conversely, mutational activation of gpa2 fully suppresses the loss of either git5 or git11 (Table 3). For proper glucose signaling to occur, the Gß dimer may be required to promote an efficient interaction between the G subunit and git3, the likely glucose receptor (WELTON and HOFFMAN 2000 ). A similar role has been observed for the S. cerevisiae Ste4-Ste18 Gß in coupling of the Gpa1 G to the Ste2 pheromone receptor (BLUMER and THORNER 1990 ). As overexpression of both the Gß and the G subunit has no effect on fbp1-lacZ expression in a gpa2 mutant strain, we conclude that the Gß dimer does not have any G-independent role in glucose monitoring.

Deletion of the git5 Gß gene confers a two- to fourfold greater increase in fbp1-lacZ expression than does the deletion of the git11 G gene (Table 2 Table 3 Table 4); therefore it appears that the git5 Gß retains some function in the absence of its G partner. Consistent with this suggestion, the addition of a CAAX box to the carboxy-terminus of the git5 Gß increases the function of Gß in cells lacking G (Fig 6, Table 4). As mentioned above, this result does not support the alternative hypothesis that S. pombe expresses multiple G subunits. Thus the git11 G appears to act to localize the git5 Gß to the peripheral membrane but is not required for proper folding of the Gß subunit. While the git5-CAAX protein clearly bypasses the need for a git11 G as judged by the 5-FOA growth test (Fig 6), results from ß-galactosidase assays measuring the effect of these constructs on expression of the fbp1-lacZ reporter (Table 4) are not as convincing. We have observed similar discrepancies between 5-FOA growth results and ß-galactosidase activity while studying suppression of mutations affecting the protein kinase A pathway by multicopy pyp1 (DAL SANTO et al. 1996 ; only a partial reduction in ß-galactosidase activity was observed) and sck1 (JIN et al. 1995 ; no reduction in ß-galactosidase activity was observed). We believe that these discrepancies represent heterogeneity in the population of plasmid-containing cells. If a subpopulation of cells establishes glucose repression of the fbp1-ura4 reporter due to a particular level of expression of the git5-CAAX fusion protein, these cells will grow on the 5-FOA medium to produce a 5-FOA^R patch. However, if the majority of the cells fail to establish glucose repression of the fbp1-lacZ reporter due to over- or underexpression of the git5-CAAX fusion protein, the ß-galactosidase activity in the overall culture will show little or no change from control cultures.

Our results stand in striking contrast to data from studies showing that mammalian Gß subunits are unable to fold in the absence of G partners (GARCIA-HIGUERA et al. 1996 ) and that the Gß amino-terminal coiled-coil is essential for dimer assembly (GARRITSEN et al. 1993 ). The apparent conflict between those studies and ours may point to an inherent difference in the biology of cells that express a single Gß and G subunit vs. ones that express multiple forms of each subunit. In mammalian cells, the requirement of the Gß amino-terminal coiled-coil for Gß association and thus the proper folding of the Gß subunit may allow cells to tightly regulate the combinations of Gß dimers assembled. If mammalian Gß subunits could fold into ß-barrels in the absence of G subunits, it might reduce the stringency of the dimer interaction and allow the subsequent assembly of a broader range of Gß dimers. Alternatively, these monomeric Gß subunits might either promote or interfere with signaling in pathways other than the ones in which they normally act. These issues do not arise in S. pombe, which expresses a single species of Gß and G subunit. Even so, we detect lower levels of the Gß subunit in extracts from cells overexpressing the tagged git5-V5 Gß subunit alone relative to those of cells overexpressing both git5-V5 and HA-git11 G (Fig 3, lanes 1 and 2). Thus, stability of the git5 Gß subunit may be partially dependent upon assembly of a Gß dimer, showing that the behavior of the fission yeast Gß is not wholly unlike that of the mammalian dimer. In the same vein, it is possible that some as yet untested mammalian Gß subunits may behave more like the git5 Gß with respect to folding in the absence of a G partner. The continued study of the structure and function of G proteins may therefore identify a broad spectrum of structural tendencies rather than a few strictly followed rules.

	ACKNOWLEDGMENTS

We give special thanks to Eva Neer, of blessed memory, for her advice and interest over the past several years. We thank Stephen Elledge for the S. pombe two-hybrid library and advice on screening, Tom Chappell for advice on use of the pNMT41-TOPO cloning vector, Eric Chang for plasmids pALU and pARTCM, and Rob Welton for strains and plasmids. We also thank David Burgess for the use of his microscope and Brad Shuster for advice and assistance with the microscopy. This work was supported by National Institutes of Health grant GM-46226 to C.S.H. and by a Clare Boothe Luce graduate fellowship to S.L.

Manuscript received October 23, 2000; Accepted for publication December 4, 2000.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

BAHLER, J., J. Q. WU, M. S. LONGTINE, N. G. SHAH, A. MCKENZIE, III et al., 1998 Heterologous modules for efficient and versatile PCR-based gene targeting in Schizosaccharomyces pombe. Yeast 14: 943–951.

BLUMER, K. J. and J. THORNER, 1990  Beta and gamma subunits of a yeast guanine nucleotide-binding protein are not essential for membrane association of the alpha subunit but are required for receptor coupling. Proc. Natl. Acad. Sci. USA 87:4363-4367[Abstract/Free Full Text].

BYRNE, S. M. and C. S. HOFFMAN, 1993  Six git genes encode a glucose-induced adenylate cyclase activation pathway in the fission yeast Schizosaccharomyces pombe. J. Cell Sci. 105:1095-1100[Abstract].

CASEY, P. J., 1994  Lipid modifications of G proteins. Curr. Opin. Cell Biol. 6:219-225[Medline].

CELENZA, J. L., F. J. ENG, and M. CARLSON, 1989  Molecular analysis of the SNF4 gene of Saccharomyces cerevisiae: evidence for physical association of the SNF4 protein with the SNF1 protein kinase. Mol. Cell. Biol. 9:5045-5054[Abstract/Free Full Text].

CHANG, E. C., M. BARR, Y. WANG, V. JUNG, and H. P. XU et al., 1994  Cooperative interaction of S. pombe proteins required for mating and morphogenesis. Cell 79:131-141[Medline].

COLOMBO, S., P. MA, L. CAUWENBERG, J. WINDERICKX, and M. CRAUWELS et al., 1998  Involvement of distinct G-proteins, Gpa2 and Ras, in glucose- and intracellular acidification-induced cAMP signalling in the yeast Saccharomyces cerevisiae. EMBO J. 17:3326-3341[Medline].

DAL SANTO, P., B. BLANCHARD, and C. S. HOFFMAN, 1996  The Schizosaccharomyces pombe pyp1 protein tyrosine phosphatase negatively regulates nutrient monitoring pathways. J. Cell Sci. 109:1919-1925[Abstract].

DIETZEL, C. and J. KURJAN, 1987  The yeast SCG1 gene: a G alpha-like protein implicated in the a- and alpha-factor response pathway. Cell 50:1001-1010[Medline].

DURFEE, T., K. BECHERER, P. L. CHEN, S. H. YEH, and Y. YANG et al., 1993  The retinoblastoma protein associates with the protein phosphatase type 1 catalytic subunit. Genes Dev. 7:555-569[Abstract/Free Full Text].

FIELDS, S. and O. SONG, 1989  A novel genetic system to detect protein-protein interactions. Nature 340:245-246[Medline].

GARCIA-HIGUERA, I., J. FENOGLIO, Y. LI, C. LEWIS, and M. P. PANCHENKO et al., 1996  Folding of proteins with WD-repeats: comparison of six members of the WD-repeat superfamily to the G protein beta subunit. Biochemistry 35:13985-13994[Medline].

GARRITSEN, A., P. J. VAN GALEN, and W. F. SIMONDS, 1993  The N-terminal coiled-coil domain of beta is essential for gamma association: a model for G-protein beta gamma subunit interaction. Proc. Natl. Acad. Sci. USA 90:7706-7710[Abstract/Free Full Text].

GILMAN, A. G., 1987  G proteins: transducers of receptor-generated signals. Annu. Rev. Biochem. 56:615-649[Medline].

GUTZ, H., H. HESLOT, U. LEUPOLD and N. LOPRIENO, 1974 Schizosaccharomyces pombe, pp. 395–446 in Handbook of Genetics, edited by R. C. KING. Plenum Press, New York.

HARPER, J. W., G. R. ADAMI, N. WEI, K. KEYOMARSI, and S. J. ELLEDGE, 1993  The p21 Cdk-interacting protein Cip1 is a potent inhibitor of G1 cyclin-dependent kinases. Cell 75:805-816[Medline].

HIRSCH, J. P. and F. R. CROSS, 1992  Pheromone response in yeast. Bioessays 14:367-373[Medline].

HOFFMAN, C. S. and F. WINSTON, 1987  A ten-minute DNA preparation from yeast efficiently releases autonomous plasmids for transformation of Escherichia coli. Gene 57:267-272[Medline].

HOFFMAN, C. S. and F. WINSTON, 1990  Isolation and characterization of mutants constitutive for expression of the fbp1 gene of Schizosaccharomyces pombe. Genetics 124:807-816[Abstract].

HOFFMAN, C. S. and F. WINSTON, 1991  Glucose repression of transcription of the Schizosaccharomyces pombe fbp1 gene occurs by a cAMP signaling pathway. Genes Dev. 5:561-571[Abstract/Free Full Text].

ISSHIKI, T., N. MOCHIZUKI, T. MAEDA, and M. YAMAMOTO, 1992  Characterization of a fission yeast gene, gpa2, that encodes a G alpha subunit involved in the monitoring of nutrition. Genes Dev. 6:2455-2462[Abstract/Free Full Text].

JIN, M., M. FUJITA, B. M. CULLEY, E. APOLINARIO, and M. YAMAMOTO et al., 1995  sck1, a high copy number suppressor of defects in the cAMP-dependent protein kinase pathway in fission yeast, encodes a protein homologous to the Saccharomyces cerevisiae SCH9 kinase. Genetics 140:457-467[Abstract].

KIM, D. U., S. K. PARK, K. S. CHUNG, M. U. CHOI, and H. S. YOO, 1996  The G protein beta subunit Gpb1 of Schizosaccharomyces pombe is a negative regulator of sexual development. Mol. Gen. Genet. 252:20-32[Medline].

LAMBRIGHT, D. G., J. SONDEK, A. BOHM, N. P. SKIBA, and H. E. HAMM et al., 1996  The 2.0 A crystal structure of a heterotrimeric G protein. Nature 379:311-319[Medline].

LANDRY, S., M. PETTIT, A. APOLINARIO, and C. S. HOFFMAN, 2000  The fission yeast git5 gene encodes a Gß subunit required for glucose-triggered adenylate cyclase activation. Genetics 154:1463-1471[Abstract/Free Full Text].

MAEDA, T., N. MOCHIZUKI, and M. YAMAMOTO, 1990  Adenylyl cyclase is dispensable for vegetative cell growth in the fission yeast Schizosaccharomyces pombe. Proc. Natl. Acad. Sci. USA 87:7814-7818[Abstract/Free Full Text].

MCLEOD, M., M. STEIN, and D. BEACH, 1987  The product of the mei3⁺ gene, expressed under control of the mating-type locus, induces meiosis and sporulation in fission yeast. EMBO J. 6:729-736[Medline].

NAKAFUKU, M., T. OBARA, K. KAIBUCHI, I. MIYAJIMA, and A. MIYAJIMA et al., 1988  Isolation of a second yeast Saccharomyces cerevisiae gene (GPA2) coding for guanine nucleotide-binding regulatory protein: studies on its structure and possible functions. Proc. Natl. Acad. Sci. USA 85:1374-1378[Abstract/Free Full Text].

NEER, E. J., 1995  Heterotrimeric G proteins: organizers of transmembrane signals. Cell 80:249-257[Medline].

NOCERO, M., T. ISSHIKI, M. YAMAMOTO, and C. S. HOFFMAN, 1994  Glucose repression of fbp1 transcription of Schizosaccharomyces pombe is partially regulated by adenylate cyclase activation by a G protein alpha subunit encoded by gpa2 (git8). Genetics 138:39-45[Abstract].

OBARA, T., M. NAKAFUKU, M. YAMAMOTO, and Y. KAZIRO, 1991  Isolation and characterization of a gene encoding a G-protein alpha subunit from Schizosaccharomyces pombe: involvement in mating and sporulation pathways. Proc. Natl. Acad. Sci. USA 88:5877-5881[Abstract/Free Full Text].

PEITSCH, M. C., 1996  ProMod and Swiss-Model: Internet-based tools for automated comparative protein modelling. Biochem. Soc. Trans. 24:274-279[Medline].

PELLEGRINO, S., S. ZHANG, A. GARRITSEN, and W. F. SIMONDS, 1997  The coiled-coil region of the G protein beta subunit: mutational analysis of G and effector interactions. J. Biol. Chem. 272:25360-25366[Abstract/Free Full Text].

SAYLE, R. A. and E. J. MILNER-WHITE, 1995  RASMOL: biomolecular graphics for all. Trends Biochem. Sci. 20:374[Medline].

SIMON, M. I., M. P. STRATHMANN, and N. GAUTAM, 1991  Diversity of G proteins in signal transduction. Science 252:802-808[Abstract/Free Full Text].

SONDEK, J., A. BOHM, D. G. LAMBRIGHT, H. E. HAMM, and P. B. SIGLER, 1996  Crystal structure of a G-protein beta gamma dimer at 2.1A resolution. Nature 379:369-374[Medline].

SOUTHERN, J. A., D. F. YOUNG, F. HEANEY, W. K. BAUMGARTNER, and R. E. RANDALL, 1991  Identification of an epitope on the P and V proteins of simian virus 5 that distinguishes between two isolates with different biological characteristics. J. Gen. Virol. 72:1551-1557[Abstract/Free Full Text].

THOMPSON, J. D., D. G. HIGGINS, and T. J. GIBSON, 1994  CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22:4673-4680[Abstract/Free Full Text].

WACH, A., A. BRACHAT, C. ALBERTI-SEGUI, C. REBISCHUNG, and P. PHILIPPSEN, 1997  Heterologous HIS3 marker and GFP reporter modules for PCR-targeting in Saccharomyces cerevisiae. Yeast 13:1065-1075[Medline].

WALL, M. A., D. E. COLEMAN, E. LEE, J. A. INIGUEZ-LLUHI, and B. A. POSNER et al., 1995  The structure of the G protein heterotrimer Gi alpha 1 beta 1 gamma 2. Cell 83:1047-1058[Medline].

WATANABE, Y., Y. LINO, K. FURUHATA, C. SHIMODA, and M. YAMAMOTO, 1988  The S. pombe mei2 gene encoding a crucial molecule for commitment to meiosis is under the regulation of cAMP. EMBO J. 7:761-767[Medline].

WELTON, R. M. and C. S. HOFFMAN, 2000  Glucose monitoring in fission yeast via the gpa2 G, the git5 Gß, and the git3 putative glucose receptor. Genetics 156:513-521[Abstract/Free Full Text].

WHITEWAY, M., L. HOUGAN, D. DIGNARD, D. Y. THOMAS, and L. BELL et al., 1989  The STE4 and STE18 genes of yeast encode potential beta and gamma subunits of the mating factor receptor-coupled G protein. Cell 56:467-477[Medline].

XUE, Y., M. BATLLE, and J. P. HIRSCH, 1998  GPR1 encodes a putative G protein-coupled receptor that associates with the Gpa2p Galpha subunit and functions in a Ras-independent pathway. EMBO J. 17:1996-2007[Medline].

YUN, C. W., H. TAMAKI, R. NAKAYAMA, K. YAMAMOTO, and H. KUMAGAI, 1998  Gpr1p, a putative G-protein coupled receptor, regulates glucose-dependent cellular cAMP level in yeast Saccharomyces cerevisiae. Biochem. Biophys. Res. Commun. 252:29-33[Medline].

This article has been cited by other articles:

				M. A. Alaamery and C. S. Hoffman Schizosaccharomyces pombe Hsp90/Git10 Is Required for Glucose/cAMP Signaling Genetics, April 1, 2008; 178(4): 1927 - 1936. [Abstract] [Full Text] [PDF]

C. S. Hoffman Propping Up Our Knowledge of G Protein Si