FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2034304329> ?p ?o ?g. }

• W2034304329 endingPage "17596" @default.
• W2034304329 startingPage "17591" @default.
• W2034304329 abstract "When starved for a single amino acid, the budding yeast Saccharomyces cerevisiae activates the eukaryotic initiation factor 2α (eIF2α) kinase GCN2 in a GCN1-dependent manner. Phosphorylated eIF2α inhibits general translation but selectively derepresses the synthesis of the transcription factor GCN4, which leads to coordinated induction of genes involved in biosynthesis of various amino acids, a phenomenon called general control response. We recently demonstrated that this response requires binding of GCN1 to the GI domain occurring at the N terminus of GCN2 (Kubota, H., Sakaki, Y., and Ito, T. (2000)J. Biol. Chem. 275, 20243–20246). Here we provide the first evidence for the involvement of GCN1-GCN2 interaction in activation of GCN2 per se. We identified a C-terminal segment of GCN1 sufficient to bind the GI domain and used a novel dual bait two-hybrid method to identify mutations rendering GCN1 incapable of interacting with GCN2. The yeast bearing such an allele,gcn1-F2291L, fails to display derepression of GCN4translation and hence general control response, as does a GI domain mutant, gcn2-Y74A, defective in association with GCN1. Furthermore, we demonstrated that phosphorylation of eIF2α is impaired in both mutants. Since GCN2 is the sole eIF2α kinase in yeast, these findings indicate a critical role of GCN1-GCN2 interaction in activation of the kinase in vivo. When starved for a single amino acid, the budding yeast Saccharomyces cerevisiae activates the eukaryotic initiation factor 2α (eIF2α) kinase GCN2 in a GCN1-dependent manner. Phosphorylated eIF2α inhibits general translation but selectively derepresses the synthesis of the transcription factor GCN4, which leads to coordinated induction of genes involved in biosynthesis of various amino acids, a phenomenon called general control response. We recently demonstrated that this response requires binding of GCN1 to the GI domain occurring at the N terminus of GCN2 (Kubota, H., Sakaki, Y., and Ito, T. (2000)J. Biol. Chem. 275, 20243–20246). Here we provide the first evidence for the involvement of GCN1-GCN2 interaction in activation of GCN2 per se. We identified a C-terminal segment of GCN1 sufficient to bind the GI domain and used a novel dual bait two-hybrid method to identify mutations rendering GCN1 incapable of interacting with GCN2. The yeast bearing such an allele,gcn1-F2291L, fails to display derepression of GCN4translation and hence general control response, as does a GI domain mutant, gcn2-Y74A, defective in association with GCN1. Furthermore, we demonstrated that phosphorylation of eIF2α is impaired in both mutants. Since GCN2 is the sole eIF2α kinase in yeast, these findings indicate a critical role of GCN1-GCN2 interaction in activation of the kinase in vivo. eukaryotic initiation factor 2 open reading frame amino acid(s) polymerase chain reaction activation domain PC motif-containing region synthetic complete synthetic dextrose 3-aminotriazole 5-fluoro-orotic acid degenerate protein kinase domain Protein synthesis in eukaryotic cells is suppressed by stress-induced phosphorylation of eukaryotic initiation factor 2α (eIF2α)1 on a serine residue at position 51 (1). The phosphorylation converts eIF2α from the substrate to an inhibitor of eIF2B, the guanine nucleotide exchange factor of eIF2; phosphorylated eIF2-GDP forms a stable complex with eIF2B to hamper recycling of eIF2-GDP to eIF2-GTP (2Merrick W.C. Microbiol. Rev. 1992; 56: 291-315Crossref PubMed Google Scholar). Scarcity of eIF2-GTP accordingly decreases the level of the ternary complex composed of eIF2, GTP, and the charged initiator tRNA, a prerequisite for translational initiation, and hence leads to general suppression of protein synthesis. Thus, eIF2α kinases play pivotal roles in this famous translational control. Mammalian cells have four eIF2α kinases, each of which is activated in response to a distinct stress. Heme-regulated inhibitor is activated by hemin deprivation (3Clemens M.J. Translational Control. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY1996: 139-172Google Scholar); double-stranded RNA-dependent kinase is activated by double-stranded RNA (3Clemens M.J. Translational Control. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY1996: 139-172Google Scholar); RNA-dependent kinase-like endoplasmic reticulum kinase is activated by unfolded proteins (4Shi Y. Vattem K.M. Sood R. An J. Liang J. Stramm L. Wek R.C. Mol. Cell. Biol. 1998; 18: 7499-7509Crossref PubMed Google Scholar, 5Harding H.P. Zhang Y. Ron D. Nature. 1999; 397: 271-274Crossref PubMed Scopus (2558) Google Scholar); and GCN2 is activated by serum or amino acid starvation (6Berlanga J.J. Santoyo J. De Haro C. Eur. J. Biochem. 1999; 265: 754-762Crossref PubMed Scopus (221) Google Scholar, 7Harding H.P. Novoa I. Zhang Y. Zeng H. Wek R. Schapira M. Ron D. Mol. Cell. 2000; 6: 1099-1108Abstract Full Text Full Text PDF PubMed Scopus (2431) Google Scholar). In contrast, the budding yeastSaccharomyces cerevisiae has the sole eIF2α kinase, GCN2, the founding member of this family. The yeast GCN2 is activated by starvation for amino acids, glucose deprivation, purine limitation, and impaired tRNA synthetase activity (8Yang R. Wek S.A. Wek R.C. Mol. Cell. Biol. 2000; 20: 2706-2717Crossref PubMed Scopus (144) Google Scholar, 9Hinnebusch A.G. Microbiol. Rev. 1988; 52: 248-273Crossref PubMed Google Scholar, 10Hinnebusch A.G. J. Biol. Chem. 1997; 272: 2161-21664Abstract Full Text Full Text PDF Scopus (434) Google Scholar). The gene for this kinase was originally identified in the studies of a response to amino acid starvation called general control of amino acid synthesis, and hence was termed GCN2 (generalcontrol nonderepressible 2). The molecular mechanism underlying general control response is currently considered as follows. When the budding yeast starves for a single particular amino acid, free tRNAs, which are not charged with amino acids, accumulate within the cells and bind to a bipartite domain composed of the histidyl tRNA synthetase-related domain and the C-terminal ribosome-binding domain of GCN2 (11Dong J. Qiu H. Garcia-Barrio M. Anderson J. Hinnebusch A.G. Mol. Cell. 2000; 6: 269-279Abstract Full Text Full Text PDF PubMed Scopus (333) Google Scholar). This bipartite domain forms an inhibitory interaction with the kinase domain, which is disrupted upon binding of tRNAs (11Dong J. Qiu H. Garcia-Barrio M. Anderson J. Hinnebusch A.G. Mol. Cell. 2000; 6: 269-279Abstract Full Text Full Text PDF PubMed Scopus (333) Google Scholar). In addition to uncharged tRNAs, which unmask the kinase domain, genetic evidence suggests that in vivo activation of GCN2 requires another gene, GCN1, encoding a protein bearing a region homologous to translation elongation factor 3 (12Marton M.J. Crouch D. Hinnebusch A.G. Mol. Cell. Biol. 1993; 13: 3541-3556Crossref PubMed Scopus (96) Google Scholar). GCN1 forms a stable complex with the ATP-binding cassette protein GCN20 and functions on an elongating ribosome (13Vazquez de Aldana C.R. Marton M.J. Hinnebusch A.G. EMBO J. 1995; 14: 3148-3199Crossref Scopus (125) Google Scholar, 14Marton M.J. Vazquez de Aldana C.R. Qiu H. Chalraburtty K. Hinnebusch A.G. Mol. Cell. Biol. 1997; 17: 4474-4489Crossref PubMed Google Scholar). GCN2 is activated by uncharged tRNAs in the presence of GCN1 and phosphorylates eIF2α to suppress protein synthesis via the mechanism described above. However, the mRNA encoding GCN4 is selectively translated by a unique mechanism, which depends on the four short open reading frames (ORFs) preceding the one for GCN4 (10Hinnebusch A.G. J. Biol. Chem. 1997; 272: 2161-21664Abstract Full Text Full Text PDF Scopus (434) Google Scholar). The transcription factor GCN4 induces the expression of genes involved in various amino acid synthetic pathways. In contrast to the action of uncharged tRNAs and the mechanism for derepressed translation of GCN4 mRNA, how GCN1 participates in the activation of GCN2 was poorly understood at the molecular level. Recently, we and others showed that a direct interaction between GCN1 and GCN2 is necessary for general control response, thereby providing the first insight into the underlying mechanism (15Kubota H. Sakaki Y. Ito T. J. Biol. Chem. 2000; 275: 20243-20246Abstract Full Text Full Text PDF PubMed Scopus (53) Google Scholar, 16Garcia-Barrio M. Dong J. Ufano S. Hinnebusch A.G. EMBO J. 2000; 17: 1887-1899Crossref Google Scholar). In this study, we determine the minimal essential regions of GCN1 and GCN2 for the complex formation and demonstrate that phosphorylation of eIF2α, translational derepression of GCN4 mRNA, and general control response are impaired in the gcn1 andgcn2 mutants defective in this interaction. These results provide the first direct evidence for a crucial role of GCN1-GCN2 interaction in the activation of the eIF2α kinase. The strains used in this study are summarized in Table I.Table IGenotypes of yeast strains used in this studyStrainGenotypeSourcePJ69–4AMAT a trp1–901 leu2–3,112 ura3–52 his3Δ200 gal4Δ gal80Δ LYS2::GAL1-HIS3 GAL2-ADE2 met2::GAL7-lacZRef. 37James P. Halladay J. Craig E.A. Genetics. 1996; 144: 1425-1436Crossref PubMed Google ScholarPJ69–4AΔMAT a trp1–901 leu2–3,112 ura3–52 his3Δ200 gal4Δ gal80Δ LYS2::GAL1-HIS3 GAL2-ADE2 met2::GAL7-lacZ lig4ΔK. OtaMavχMATα trp1–901 leu2–3,112 his3Δ200 ade2Δ gal4Δ gal80Δ SPAL10::URA3 LYS2::kanMX-Lexop-HIS3 GAL1::lacZK. OtaMB758–5BMAT a ura3A. FujitaJBY1MAT a ura3 gcn2-Y74ARef. 15Kubota H. Sakaki Y. Ito T. J. Biol. Chem. 2000; 275: 20243-20246Abstract Full Text Full Text PDF PubMed Scopus (53) Google ScholarJBY2MAT a ura3 GCN2-T7::URA3Ref. 15Kubota H. Sakaki Y. Ito T. J. Biol. Chem. 2000; 275: 20243-20246Abstract Full Text Full Text PDF PubMed Scopus (53) Google ScholarJBY3MAT a ura3 gcn2-Y74A-T7::URA3Ref. 15Kubota H. Sakaki Y. Ito T. J. Biol. Chem. 2000; 275: 20243-20246Abstract Full Text Full Text PDF PubMed Scopus (53) Google ScholarJBY4MAT a ura3 GCN2-T7This studyJBY5MAT a ura3 gcn2-Y74A-T7This studyJBZ1MAT a ura3 gcn1-F2291LThis studyJBZ2MAT a ura3 GCN1-T7::URA3This studyJBZ3MAT a ura3 gcn1-F2291L-T7::URA3This study Open table in a new tab The two-hybrid vectors, pGBK and pGAD424g, were described previously (17Ito T. Nakamura R. Sumimoto H. Takeshige K. Sakaki Y. FEBS Lett. 1996; 385: 229-232Crossref PubMed Scopus (51) Google Scholar,18Nakamura R. Sumimoto H. Mizuki K. Hata K. Ago T. Kitajima S. Takeshige K. Sakaki Y. Ito T. Eur. J. Biochem. 1998; 251: 583-589Crossref PubMed Scopus (72) Google Scholar). For high efficiency transformation, the protocol of Gietz and Schiestl (19Gietz R.D. Schiestl R.H. Methods Mol. Cell. Biol. 1995; 5: 255-269Google Scholar) was adopted except for the addition of 10% dimethyl sulfoxide prior to the heat shock step (20Ito T. Tashiro K. Muta S. Ozawa R. Chiba T. Nishizawa M. Yamamoto K. Kuhara S. Sakaki Y. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 1143-1147Crossref PubMed Scopus (667) Google Scholar). AGCN1 DNA fragment (nucleotides 6142–7146) was cloned between the segments encoding the GAL4 activation domain (AD) and the C-terminal 75-amino acid (aa) region of CDC24, which bears a PC motif and is called the PC motif-containing region (PCCR) (18Nakamura R. Sumimoto H. Mizuki K. Hata K. Ago T. Kitajima S. Takeshige K. Sakaki Y. Ito T. Eur. J. Biochem. 1998; 251: 583-589Crossref PubMed Scopus (72) Google Scholar). We subjected the GAL4 AD-GCN1-PCCR fragment to error-prone PCR in 50 μl of 1× PCR buffer containing 1 unit of Taq DNA polymerase, 200 μm dATP, 2 mm dCTP, 2 mm dGTP, 2 mm dTTP, 2 mm MgCl2, 5 pmol of each primer under the following thermal cycling: 94 °C for 3 min, followed by 30 cycles of 95 °C for 15 s, 55 °C for 30 s, and 72 °C for 2 min. The amplified products of error-prone PCR were cloned into pGAD424g by a transformation-associated recombination technique (21Oldenberg K., R. Vo K.T. Michaelis S. Paddon C. Nucleic Acids Res. 1997; 25: 451-452Crossref PubMed Scopus (417) Google Scholar) using PJ69–4AΔ (MAT a) as a host. Transformants were then mated with Mavχ (MATα) bearing pGBK-GCN2 and pHLZ-BEM1-PB1. The former plasmid encodes a hybrid protein between the GAL4 DNA-binding domain and GI domain of GCN2, and the latter one encodes a protein between LexA and the PB1 domain of BEM1 (aa 472–551), which specifically binds to PCCR of CDC24. 2T. Ito and H. Sumimoto, unpublished observation. Diploid cells were then plated onto synthetic complete medium (20Ito T. Tashiro K. Muta S. Ozawa R. Chiba T. Nishizawa M. Yamamoto K. Kuhara S. Sakaki Y. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 1143-1147Crossref PubMed Scopus (667) Google Scholar) lacking Trp, Leu, and His (SC−Trp−Leu−His) supplemented with 20 mm3-aminotriazole (3AT) and 0.2% 5-fluoro-orotic acid (5FOA). Thegcn1-F2291L DNA fragment (nucleotides 6538–7401) was cloned in pUC-URA3, a pUC13 derivative bearing the URA3 marker (15Kubota H. Sakaki Y. Ito T. J. Biol. Chem. 2000; 275: 20243-20246Abstract Full Text Full Text PDF PubMed Scopus (53) Google Scholar), to obtain pUC-URA3-gcn1-F2291L, which was subsequently linearized with Bglll, and introduced into the yeast MB758-5B cells. The Ura+ transformants were tested for successful targeted integration by diagnostic colony PCR. These clones were then selected for 5FOA resistance and examined for the desired allele replacement to obtain the strain JBZ1 (MAT a ura3 gcn1-F2291L). For epitope tagging of GCN1, we first constructed a pUC-URA3 derivative bearing a DNA fragment encoding the 3′-end portion of GCN1ORF with its flanking region (nucleotides 7705–8214) followed byAshbya gossypii TEF2 terminator, which is derived from thekanMX cassette (22Güldener U. Heck S. Fiedler T. Beinhauer J. Hegemann J.H. Nucleic Acids Res. 1996; 24: 2519-2524Crossref PubMed Scopus (1372) Google Scholar). Following the insertion of a T7 epitope-encoding sequence by an inverse PCR-mediated procedure to C-terminally tag GCN1, the plasmid was linearized and transformed into MB758-5B and JBZ1 to obtain the strains JBZ2 (MAT a ura3 GCN1-T7::URA3) and JBZ3 (MAT a ura3 gcn1-F2291L-T7::URA3), respectively. These strains were spotted onto agar plates of SC−Ura or SC−Ura−His plus 20 mm 3AT as described previously (15Kubota H. Sakaki Y. Ito T. J. Biol. Chem. 2000; 275: 20243-20246Abstract Full Text Full Text PDF PubMed Scopus (53) Google Scholar). The p180 reporter construct is a low copy number plasmid, which encodes an mRNA bearing the GCN4-lacZ fusion ORF preceded by the entire 5′ leader region of GCN4 mRNA containing four upstream ORFs to monitor the translational derepression (23Mueller P.P. Hinnebusch A.G. Cell. 1986; 45: 201-207Abstract Full Text PDF PubMed Scopus (360) Google Scholar). This plasmid was introduced into MB758-5B, JBZ1, JBY4, and JBY5 cells, and the transformants were grown to midlogarithmic phase and shifted to SC or synthetic dextrose (SD) medium (0.67% yeast nitrogen base without amino acids, 2% glucose) containing 10 or 20 mm3AT. Following the incubation at 30 °C for 3 h, cells were collected and subjected to β-galactosidase assay as described previously (18Nakamura R. Sumimoto H. Mizuki K. Hata K. Ago T. Kitajima S. Takeshige K. Sakaki Y. Ito T. Eur. J. Biochem. 1998; 251: 583-589Crossref PubMed Scopus (72) Google Scholar). The yeast JBY2, JBY3, JBZ2, and JBZ3 cells were grown to midlogarithmic phase and shifted to YPAD (1% yeast extract, 2% peptone, 0.004% adenine sulfate, 2% glucose) or SD medium supplemented with 10 or 20 mm 3AT. Following incubation at 30 °C for 4 h, ∼107 cells were collected, resuspended in 1.0 ml of distilled H2O, and broken by the addition of 150 μl of breaking buffer (2 m NaOH, 2 mβ-mecaptoethanol). Following incubation on ice for 10 min, the lysate was neutralized by the addition of 130 μl of 60% trichloroacetic acid. Proteins were then collected by centrifugation at 15,000 ×g and washed twice with 700 μl of acetone. Protein extract equivalent to 106 cells was subjected to each lane of SDS-polyacrylamide gel electrophoresis and transferred to nitrocellulose filters. Filters were blocked by PBS-T (137 mm NaCl, 8.1 mmNa2HPO4, 2.68 mm KCl, 1.47 mm KH2PO4, 0.1% Tween 20, 4% nonfat dry milk). Phosphorylated eIF2α was visualized by using PBS-T containing an antibody that specifically recognizes eIF2α phosphorylated at serine 51 (BIO-SOURCE). We previously showed that the characteristic GI domain (aa 1–125) occurring at the N-terminal extremity of GCN2 directly binds to GCN1 (15Kubota H. Sakaki Y. Ito T. J. Biol. Chem. 2000; 275: 20243-20246Abstract Full Text Full Text PDF PubMed Scopus (53) Google Scholar). Others also demonstrated that GCN2 interacts with GCN1 via its N-terminal 272-aa region, which contains the GI domain followed by an acidic region (16Garcia-Barrio M. Dong J. Ufano S. Hinnebusch A.G. EMBO J. 2000; 17: 1887-1899Crossref Google Scholar). To delimit the minimal essential region to interact with GCN1, we prepared a series of truncated mutants for the N-terminal region of GCN2 and tested them for binding with GCN1 using the yeast two-hybrid system. All of the mutants bearing intact GI domain (aa 1–125) showed two-hybrid interactions with GCN1 (Fig.1). The yeast cells with the longest GCN2 hybrid protein (aa 1–598), which includes the GI domain, acidic region, and degenerate protein kinase domain (ψPK), showed higher β-galactosidase activity than those lacking ψPK (aa 1–272). Although many factors affecting two-hybrid interactions, including intracellular level of hybrid proteins and efficiency of nuclear transport, make it difficult to evaluate strength of binding by this method, the result described above is in accordance with the one obtained using an in vitro pull-down binding assay (16Garcia-Barrio M. Dong J. Ufano S. Hinnebusch A.G. EMBO J. 2000; 17: 1887-1899Crossref Google Scholar). Notably, removal of the acidic region did not diminish β-galactosidase activity any more, and we failed to detect any evidence for this region to bind GCN1 (Fig. 1). In contrast, further deletion from either end of the GI domain (aa 1–125) completely abolished the interaction (Fig. 1). We had also shown that substitution of conserved residues in the GI domain abrogated the interaction (15Kubota H. Sakaki Y. Ito T. J. Biol. Chem. 2000; 275: 20243-20246Abstract Full Text Full Text PDF PubMed Scopus (53) Google Scholar). These results indicated that the GI domain itself serves as the minimal essential region for the binding to GCN1 and that the acidic region has little if any role in the interaction. Two-hybrid screening using the GI domain as bait had revealed the C-terminal portion (aa 1925–2552) of GCN1 as its binding partner. This is coincident with the results recently reported by others, which identified the region spanning aa 2052–2428 as the sole GCN2-binding region of GCN1 (24Sattlegger E. Hinnebusch A.G. EMBO J. 2000; 19: 6622-6633Crossref PubMed Scopus (94) Google Scholar). To pinpoint the minimal binding region, we prepared variously truncated fragments around this region using PCR and tested their binding to the GI domain using the yeast two-hybrid system (Fig.2). Two-hybrid interactions were examined for growth on the medium lacking adenine and histidine. The segment spanning aa 2064–2382 of GCN1 supported the growth as efficiently as those by longer constructs when expressed as a DNA-binding domain fusion. However, we failed to detect the interaction in the opposite orientation for unknown reason. Further deletion either from its N- or C-terminal end completely abolished the interaction (Fig. 2). From these results, we concluded that the region spanning amino acids 2064–2382 is sufficient for binding to the GI domain of GCN2. We next intended to isolate gcn1 mutants defective in association with GCN2, because they would highlight critical residues for the recognition of the GI domain and because they can be used to examine the role of this interaction. For this purpose, we used a PCR-based random mutagenesis. However, since we had already pinpointed the minimal essential GCN1 region for GCN2 binding, we do not need nonsense mutations to truncate the protein anymore. To selectively obtain missense mutants, we developed a novel strategy described below (Fig. 3 A). We first modified the pGAD-GCN1-(2048–2382) plasmid so that the GAL4 AD-GCN1 fusion protein is further tailed with the PCCR of CDC24 (18Nakamura R. Sumimoto H. Mizuki K. Hata K. Ago T. Kitajima S. Takeshige K. Sakaki Y. Ito T. Eur. J. Biochem. 1998; 251: 583-589Crossref PubMed Scopus (72) Google Scholar), which specifically interacts with the PB1 domain occurring at the C-terminal end of BEM1.2 Following random mutagenesis to GCN1 by error-prone PCR, clones were selected for two-hybrid interaction with the PB1 domain, which guarantees that the hybrid protein retains the C-terminal PCCR and hence is not truncated within the GCN1 portion. From these untruncated populations, clones incapable of interacting with GCN2 were identified using the reverse two-hybrid selection based on URA3 reporter gene and 5FOA (25Vidal M. Brachmann R.K. Fattaey A. Harlow E. Boeke J. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 10315-10320Crossref PubMed Scopus (363) Google Scholar). Notably, this screening may be useful not only for the elimination of truncated proteins but also for the identification of those with missense mutations leading to protein instability, because such clones display weaker two-hybrid signals than the wild-type parental clone. In practice, we used a dual bait two-hybrid system to perform both selections simultaneously. We prepared a mutagenized GAL4 AD-GCN1-PCCR library in PJ69–4AΔ (MAT a). These cells were then mated with Mavχ (MATα) cells that bear two plasmids expressing GAL4 DNA-binding domain-GCN2 hybrid protein and LexA-PB1 hybrid protein. Note that the genome of Mavχ has URA3 andHIS3 reporter genes driven by GAL4 and LexA, respectively (Table I). The diploid cells formed by the mating were selected for resistance to both 3AT and 5FOA. The resistance to 3AT indicates that the HIS3 reporter gene is induced and that an interaction is occurred between the GCN1-hybrid protein and the PB1 domain. On the other hand, 5FOA resistance is an indicative of failed induction ofURA3 or impaired association of the GCN1-hybrid protein with GCN2. From the clones displaying resistance to both 3AT and 5FOA, we identified three single amino acid substitutions, namely F2291L, S2304P, and L2353P. While showing two-hybrid interactions with the PB1 domain comparable with that of the parental clone, these threegcn1 single point mutants all failed to interact with GCN2 (Fig. 3 B). These results suggest that the mutations affect the ability of GCN1 to interact with GCN2 but not the stability of hybrid proteins (see below). We also identified seven mutants, each bearing two amino acid substitutions, namely (L2303S,V2329D), (F2291S,V2376A), (K2317R,L2319P), (S2304P,L2356S), (F2291I,T2307N), (F2281L, Q2294R) and (F2299A,R2328D). Although we did not determine which of the two is critical for binding, it is intriguing to note that all of these mutations occurred in the C- terminal half of the pinpointed GCN1 segment, as do those in the three single point mutants (Fig. 4). To examine the role of GCN1-GCN2 interaction, we intended to generate a yeast strain bearing GCN1 incapable of interacting with GCN2. For this purpose, we chose F2291L substitution, because it occurs within a cluster of identically conserved amino acid residues among GCN1 from various species (Fig. 4). We also tagged GCN1 in this mutant and its parental strain with the T7-epitope at their C-terminal ends to facilitate detection by anti-T7 antibody. As shown in Fig. 5 A, comparable amounts of GCN1 were detected in wild type andgcn1-F2291L cells, thereby demonstrating that the mutation does not destabilize the full-length protein in vivo. We then examined these cells for sensitivity to 3AT, which is an inhibitor of HIS3, a typical GCN4 target, and has been used as an indicator of general control response. The mutant cells displayed remarkably higher 3AT-sensitivity than the parental strain (Fig. 5 B). A similar phenotype was reported for yeast cells bearing thegcn2-Y74A allele, which encodes GCN2 defective in interaction with GCN1 (15Kubota H. Sakaki Y. Ito T. J. Biol. Chem. 2000; 275: 20243-20246Abstract Full Text Full Text PDF PubMed Scopus (53) Google Scholar). Taken together, these results indicate that the interaction between GCN1 and GCN2 is necessary for general control of amino acid synthesis. The Gcn− phenotype described above suggests that the derepression of GCN4translation is impaired in both gcn1-F2291L andgcn2-Y74A mutants. We thus examined the translation ofGCN4 mRNA using a reporter construct bearinglacZ preceded by the characteristic GCN4 leader region, which is responsible for the derepression (23Mueller P.P. Hinnebusch A.G. Cell. 1986; 45: 201-207Abstract Full Text PDF PubMed Scopus (360) Google Scholar). The wild-type cells showed a remarkable induction of β-galactosidase activity under starved or derepressed conditions (Fig.6). In contrast, the induction was severely impaired in both gcn1-F2291L andgcn2-Y74A cells (Fig. 6). Thus, the interaction between GCN1 and GCN2 is required for efficient derepression of GCN4translation under amino acid-starved conditions. Finally, we intended to determine whether the eIF2α kinase is activated in these mutants upon amino acid starvation, because GCN2-independent mechanisms to derepressGCN4 translation are also possible (9Hinnebusch A.G. Microbiol. Rev. 1988; 52: 248-273Crossref PubMed Google Scholar, 10Hinnebusch A.G. J. Biol. Chem. 1997; 272: 2161-21664Abstract Full Text Full Text PDF Scopus (434) Google Scholar, 26Anderson J. Phan L. Cuesta R. Carlson B.A. Pak M. Asano K. Björk G.R. Tamame M. Hinnebusch A.G. Genes Dev. 1998; 12: 3650-3662Crossref PubMed Scopus (198) Google Scholar, 27Tavernarakis N. Alexandraki D. Liodis P. Tzamarias D. Thireos G. Gene ( Amst. ). 1996; 179: 271-277Crossref PubMed Scopus (8) Google Scholar, 28Qiu H. Hu C. Anderson J. Björk G.R. Sarkar S. Hopper A.K. Hinnebusch A.G. Mol. Cell. Biol. 2000; 20: 2505-2516Crossref PubMed Scopus (64) Google Scholar). The phosphorylated eIF2α in mutants and their parental strains were examined under rich or poor conditions using an antibody specific to eIF2α phosphorylated at Ser-51. While phosphorylated eIF2α was barely detected under rich or repressed conditions, substantial phosphorylation of eIF2α was readily observed in the wild-type cells subjected to amino acid starvation (Fig.7). In contrast, the induction of the phosphorylation was substantially impaired in the cells bearinggcn1-F2291L or gcn2-Y74A compared with their respective parental strains, although residual levels of phosphorylation were detected (Fig. 7). Since GCN2 is the sole eIF2α kinase in the budding yeast, these results indicate that the interaction with GCN1 is necessary for the GCN2 to be fully activated in amino acid-starved cells. GCN1 is required for the activation of GCN2 in the budding yeast under amino acid starvation; deletion or mutations of GCN1 were reported to abolish phosphorylation of eIF2α, which is mediated by the sole eIF2α kinase GCN2 (12Marton M.J. Crouch D. Hinnebusch A.G. Mol. Cell. Biol. 1993; 13: 3541-3556Crossref PubMed Scopus (96) Google Scholar, 14Marton M.J. Vazquez de Aldana C.R. Qiu H. Chalraburtty K. Hinnebusch A.G. Mol. Cell. Biol. 1997; 17: 4474-4489Crossref PubMed Google Scholar). It had, however, remained totally unknown how GCN1 activates GCN2 until we and others provided evidence for their direct association and its requirement for a general control response (15Kubota H. Sakaki Y. Ito T. J. Biol. Chem. 2000; 275: 20243-20246Abstract Full Text Full Text PDF PubMed Scopus (53) Google Scholar, 16Garcia-Barrio M. Dong J. Ufano S. Hinnebusch A.G. EMBO J. 2000; 17: 1887-1899Crossref Google Scholar). In this study, we determined the minimal essential regions for GCN1-GCN2 association and demonstrated, for the first time, that the interaction is critical to the activation of GCN2 itself, which leads to the selective derepression of GCN4translation and subsequent general control response. The minimal essential region of GCN2 to interact with GCN1 was mapped to its N-terminal 125 residues (Fig. 1), which we had designated as the GI domain (15Kubota H. Sakaki Y. Ito T. J. Biol. Chem. 2000; 275: 20243-20246Abstract Full Text Full Text PDF PubMed Scopus (53) Google Scholar). The N-terminal 272 residues involving the GI domain followed by an acidic region were reported necessary for the interaction by others (16Garcia-Barrio M. Dong J. Ufano S. Hinnebusch A.G. EMBO J. 2000; 17: 1887-1899Crossref Google Scholar). However, our data shown here and in a previous report (15Kubota H. Sakaki Y. Ito T. J. Biol. Chem. 2000; 275: 20243-20246Abstract Full Text Full Text PDF PubMed Scopus (53) Google Scholar) clearly indicate that the GI domain per se, but not the acidic region, serves as the core for the binding. As pointed out previously (15Kubota H. Sakaki Y. Ito T. J. Biol. Chem. 2000; 275: 20243-20246Abstract Full Text Full Text PDF PubMed Scopus (53) Google Scholar), the GI domain is found in various proteins other than GCN2. They include Impact (a product of an evolutionarily conserved gene that is genetically imprinted in mice) (29Hagiwara Y. Hirai M. Nishiyama K. Kanazawa I. Ueda T. Sakaki Y. Ito T. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 9249-9254Crossref PubMed Scopus (83) Google Scholar, 30Yamada Y. Hagiwara Y. Shiokawa K. Sakaki Y. Ito T. Biochem. Biophys. Res. Commun. 1999; 256: 162-169Crossref PubMed Scopus (13) Google Scholar, 31Okamura K. Hagiwara-Takeuchi Y. Li T. Vu T.H. Hirai M. Hattori M. Sakaki Y. Hoffman A.R. Ito T. Genome Res. 2000; 10: 1878-1889Crossref PubMed Scopus (66) Google Scholar), AO7 (a RING finger protein interacting with ubiquitin-conjugating enzymes) (32Lorick K.L. Jensen J.P. Fang S. Ong A.M. Hatakeyama S. Weissman A.M. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 11364-11369Crossref PubMed Scopus (947) Google Scholar), ARA54 (a coactivator of androgen receptor) (33Kang H.Y. Yeh S. Fujimoto N. Chang C. J. Biol. Chem. 1999; 274: 8570-8576Abstract Full Text Full Text PDF PubMed Scopus (195) Google Scholar), YDR152W (a yeast hypothetical protein), YLR419W (a member of the DEAH-box RNA helicase family), and so forth. It remains to be elucidated whether these GI domains also function in protein binding. We also found that amino acid residues 2064–2382 comprise the minimal essential region of GCN1 to recognize the GI domain of GCN2 (Fig. 2). In accordance with this result, the segment spanning residues 2052–2428 of GCN1 was recently reported responsible for binding to GCN2 (24Sattlegger E. Hinnebusch A.G. EMBO J. 2000; 19: 6622-6633Crossref PubMed Scopus (94) Google Scholar). In contrast to GCN2, the primary sequence of the GI domain-binding region of GCN1 lacks any apparently characteristic feature (Fig. 4). We thus took a random mutagenesis approach to identify critical residues for the recognition of GI domain. For this purpose, we developed a unique dual bait two-hybrid strategy, which allows one to selectively identify missense mutations leading to defective interaction (Fig. 3 A). Furthermore, this strategy would be also useful to eliminate mutants encoding unstable proteins, which can be identified as clones with weaker interaction between the C-terminally attached domain and its binding partner (i.e.PCCR and PB1). Therefore, it provides a versatile tool for fine analysis of protein-protein interactions. Using this strategy, we have so far identified three single amino acid substitutions, namely F2291L, S2304P, and L2353P, each of which abolishes interaction between GCN1 and GCN2 (Fig. 3 B). Others reported that GCN1(R2259A) also fails to associate with GCN2 (24Sattlegger E. Hinnebusch A.G. EMBO J. 2000; 19: 6622-6633Crossref PubMed Scopus (94) Google Scholar). Intriguingly, all of these mutations occur in the C-terminal half of the GCN2-binding region. In addition, 14 amino acid substitutions in the seven mutants that we identified as bearing two mutations were also mapped to this portion (Fig. 4). These results suggest that the C-terminal half functions as the interaction surface with the GI domain. The apparent lack of N-terminal mutants may indicate that the N-terminal half is not involved in target recognition and rather plays a role in structural integrity of the domain. Then, most mutations in this region lead to disordered and unstable proteins and hence would be excluded by our screening strategy as discussed above. More exhaustive screening to fully uncover the important residues is necessary to obtain deeper insight into the mechanism for recognition of the GI domain. It would also be informative to compare the sequence of GCN1 with those of the binding partners for other GI domains, which remain to be identified. We and others showed that yeast cells defective in GCN1-GCN2 interaction display 3AT-sensitive growth, indicative of impaired induction of HIS3, which is under the regulation of the transcription factor GCN4 (15Kubota H. Sakaki Y. Ito T. J. Biol. Chem. 2000; 275: 20243-20246Abstract Full Text Full Text PDF PubMed Scopus (53) Google Scholar, 16Garcia-Barrio M. Dong J. Ufano S. Hinnebusch A.G. EMBO J. 2000; 17: 1887-1899Crossref Google Scholar, 24Sattlegger E. Hinnebusch A.G. EMBO J. 2000; 19: 6622-6633Crossref PubMed Scopus (94) Google Scholar) (Fig. 5). Indeed, the derepression of GCN4 translation was impaired in these mutants (Fig. 6). However, note that the derepression of GCN4 translation is not always caused by the activation of GCN2. Mutations in subunits of eIF2 and eIF2B can mimic the effect of phosphorylation of eIF2α as exemplified by several GCD genes (9Hinnebusch A.G. Microbiol. Rev. 1988; 52: 248-273Crossref PubMed Google Scholar, 10Hinnebusch A.G. J. Biol. Chem. 1997; 272: 2161-21664Abstract Full Text Full Text PDF Scopus (434) Google Scholar). The reduction in amounts (10Hinnebusch A.G. J. Biol. Chem. 1997; 272: 2161-21664Abstract Full Text Full Text PDF Scopus (434) Google Scholar) or impaired base modification (26Anderson J. Phan L. Cuesta R. Carlson B.A. Pak M. Asano K. Björk G.R. Tamame M. Hinnebusch A.G. Genes Dev. 1998; 12: 3650-3662Crossref PubMed Scopus (198) Google Scholar) of initiator tRNA also induces a similar phenotype. Other studies demonstrated that GCN2-independent derepression of GCN4 translation is elicited by overexpression of NME1 (27Tavernarakis N. Alexandraki D. Liodis P. Tzamarias D. Thireos G. Gene ( Amst. ). 1996; 179: 271-277Crossref PubMed Scopus (8) Google Scholar), encoding the RNA component of RNase MRP (34Schmitt M.E. Clayton D.A. Genes Dev. 1992; 6: 1975-1985Crossref PubMed Scopus (93) Google Scholar), or PUS4, encoding the tRNA pseudouridine 55 synthase (28Qiu H. Hu C. Anderson J. Björk G.R. Sarkar S. Hopper A.K. Hinnebusch A.G. Mol. Cell. Biol. 2000; 20: 2505-2516Crossref PubMed Scopus (64) Google Scholar). We thus directly examined the phosphorylation of eIF2α in these mutants and demonstrated its considerable impairment (Fig. 7); the Gcn− phenotype of mutants defective in GCN1-GCN2 interaction is associated with impaired phosphorylation of eIF2α. Therefore, these mutants fail to fully activate GCN2, because it is the sole eIF2α kinase of the budding yeast. Residual phosphorylation of eIF2α in these mutants (Fig. 7) indicates that weakened GCN1-GCN2 interactions can still support minimal activation of the kinase. Notably, GCN1 and GCN2 are anchored onto ribosome through their respective ribosome-binding domains (14Marton M.J. Vazquez de Aldana C.R. Qiu H. Chalraburtty K. Hinnebusch A.G. Mol. Cell. Biol. 1997; 17: 4474-4489Crossref PubMed Google Scholar, 24Sattlegger E. Hinnebusch A.G. EMBO J. 2000; 19: 6622-6633Crossref PubMed Scopus (94) Google Scholar, 35Ramirez M. Wek R.C. Hinnebusch A.G. Mol. Cell. Biol. 1991; 11: 3027-3036Crossref PubMed Scopus (123) Google Scholar,36Zhu S. Wek R.C. J. Biol. Chem. 1998; 273: 1808-1814Abstract Full Text Full Text PDF PubMed Scopus (53) Google Scholar), presumably, in such close proximity that they can interact. It is thus conceivable that the two proteins occasionally take a configuration leading to activation of the kinase although the interaction is considerably impaired by the mutations. In this context, it is intriguing to note that overexpression of GCN2 can suppress the Gcn− phenotype of gcn1-R2259A, also defective in binding to GCN2, but not that of gcn1-ΔD, which encodes GCN1 totally lacking the GCN2-binding region (24Sattlegger E. Hinnebusch A.G. EMBO J. 2000; 19: 6622-6633Crossref PubMed Scopus (94) Google Scholar). Based on these observations, we conclude that the GI domain-mediated association of GCN2 to GCN1 is necessary for the full activation of GCN2 kinase in vivo upon amino acid starvation and hence for efficient derepression of GCN4 translation, leading to general control response. We thank R. C. Wek (Indiana University) for the generous gift of plasmid p180 and H. Sumimoto (Kyushu University) for critical reading of the manuscript and encouragement." @default.
• W2034304329 created "2016-06-24" @default.
• W2034304329 creator A5037367171 @default.
• W2034304329 creator A5043778187 @default.
• W2034304329 creator A5056531353 @default.
• W2034304329 creator A5068543044 @default.
• W2034304329 date "2001-05-01" @default.
• W2034304329 modified "2023-10-17" @default.
• W2034304329 title "Budding Yeast GCN1 Binds the GI Domain to Activate the eIF2α Kinase GCN2" @default.
• W2034304329 cites W1517699507 @default.
• W2034304329 cites W1560235081 @default.
• W2034304329 cites W1564571700 @default.
• W2034304329 cites W1769175751 @default.
• W2034304329 cites W1850569493 @default.
• W2034304329 cites W1903792504 @default.
• W2034304329 cites W1969034370 @default.
• W2034304329 cites W1971679867 @default.
• W2034304329 cites W1986627314 @default.
• W2034304329 cites W1988202801 @default.
• W2034304329 cites W1998334193 @default.
• W2034304329 cites W2003574018 @default.
• W2034304329 cites W2013317097 @default.
• W2034304329 cites W2014194720 @default.
• W2034304329 cites W2043522659 @default.
• W2034304329 cites W2052093902 @default.
• W2034304329 cites W2066825851 @default.
• W2034304329 cites W2076955269 @default.
• W2034304329 cites W2085387146 @default.
• W2034304329 cites W2111096302 @default.
• W2034304329 cites W2113263393 @default.
• W2034304329 cites W2120736516 @default.
• W2034304329 cites W2121401282 @default.
• W2034304329 cites W2127848114 @default.
• W2034304329 cites W2138047093 @default.
• W2034304329 cites W2158031669 @default.
• W2034304329 cites W2163285372 @default.
• W2034304329 cites W2165055107 @default.
• W2034304329 cites W2166558964 @default.
• W2034304329 cites W2166661434 @default.
• W2034304329 cites W2167768067 @default.
• W2034304329 cites W2170598671 @default.
• W2034304329 cites W2100424580 @default.
• W2034304329 doi "https://doi.org/10.1074/jbc.m011793200" @default.
• W2034304329 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/11350982" @default.
• W2034304329 hasPublicationYear "2001" @default.
• W2034304329 type Work @default.
• W2034304329 sameAs 2034304329 @default.
• W2034304329 citedByCount "57" @default.
• W2034304329 countsByYear W20343043292012 @default.
• W2034304329 countsByYear W20343043292013 @default.
• W2034304329 countsByYear W20343043292014 @default.
• W2034304329 countsByYear W20343043292016 @default.
• W2034304329 countsByYear W20343043292017 @default.
• W2034304329 countsByYear W20343043292018 @default.
• W2034304329 countsByYear W20343043292019 @default.
• W2034304329 countsByYear W20343043292020 @default.
• W2034304329 countsByYear W20343043292021 @default.
• W2034304329 countsByYear W20343043292022 @default.
• W2034304329 countsByYear W20343043292023 @default.
• W2034304329 crossrefType "journal-article" @default.
• W2034304329 hasAuthorship W2034304329A5037367171 @default.
• W2034304329 hasAuthorship W2034304329A5043778187 @default.
• W2034304329 hasAuthorship W2034304329A5056531353 @default.
• W2034304329 hasAuthorship W2034304329A5068543044 @default.
• W2034304329 hasBestOaLocation W20343043291 @default.
• W2034304329 hasConcept C134306372 @default.
• W2034304329 hasConcept C185592680 @default.
• W2034304329 hasConcept C198410946 @default.
• W2034304329 hasConcept C2777576037 @default.
• W2034304329 hasConcept C2779222958 @default.
• W2034304329 hasConcept C2992884146 @default.
• W2034304329 hasConcept C33923547 @default.
• W2034304329 hasConcept C36503486 @default.
• W2034304329 hasConcept C55493867 @default.
• W2034304329 hasConcept C86803240 @default.
• W2034304329 hasConcept C95444343 @default.
• W2034304329 hasConceptScore W2034304329C134306372 @default.
• W2034304329 hasConceptScore W2034304329C185592680 @default.
• W2034304329 hasConceptScore W2034304329C198410946 @default.
• W2034304329 hasConceptScore W2034304329C2777576037 @default.
• W2034304329 hasConceptScore W2034304329C2779222958 @default.
• W2034304329 hasConceptScore W2034304329C2992884146 @default.
• W2034304329 hasConceptScore W2034304329C33923547 @default.
• W2034304329 hasConceptScore W2034304329C36503486 @default.
• W2034304329 hasConceptScore W2034304329C55493867 @default.
• W2034304329 hasConceptScore W2034304329C86803240 @default.
• W2034304329 hasConceptScore W2034304329C95444343 @default.
• W2034304329 hasIssue "20" @default.
• W2034304329 hasLocation W20343043291 @default.
• W2034304329 hasOpenAccess W2034304329 @default.
• W2034304329 hasPrimaryLocation W20343043291 @default.
• W2034304329 hasRelatedWork W1963983419 @default.
• W2034304329 hasRelatedWork W1982584572 @default.
• W2034304329 hasRelatedWork W2001703407 @default.
• W2034304329 hasRelatedWork W2010133910 @default.
• W2034304329 hasRelatedWork W2207860391 @default.
• W2034304329 hasRelatedWork W2397070560 @default.
• W2034304329 hasRelatedWork W2528336156 @default.

• next