We study mechanisms of transcriptional and translational regulation of the expression of amino acid–biosynthetic genes by nutrients in the yeast Saccharomyces cerevisiae. Translation of the transcriptional activator GCN4 is stimulated in amino acid–starved cells by a mechanism involving short open reading frames (uORFs) in the mRNA leader and phosphorylation of translation initiation factor 2 (eIF2), which, when bound to GTP, delivers initiator tRNAiMet to the 40S ribosome. Phosphorylation of eIF2 by the kinase GCN2 inhibits formation of the eIF2-GTP-tRNAiMet ternary complex (TC), reducing general protein synthesis but stimulating translation of GCN4. We are analyzing the physical and functional interactions of eIF2 with other initiation factors (eIF1, -1A, -3, and -5) and the 40S ribosome that promote TC recruitment and ribosomal scanning and recognition of AUG codons during general and GCN4-specific translation. We also study the mechanism of kinase GCN2 activation on the translation of ribosomes by uncharged tRNA (the starvation signal) and the GCN1-GCN20 complex. In addition, we are analyzing several co-activators required for transcriptional activation by GCN4. Our investigation of the co-activators involves the definition of the molecular program for recruitment of chromatin-remodeling enzymes and adaptor proteins that deliver TATA-binding protein, other general factors, and RNA polymerase II to the promoter at GCN4 target genes, thus stimulating transcription elongation.
Promotion of pre-initiation complex assembly and regulation of start codon selection by interactions of eukaryotic translation initiation factor 3 (eIF3) subunit NIP1/c with eIF1 and eIF5
Valasek,1 Nielsen, Fekete, Zhang, Hinnebusch
Assembly of the 80S translation initiation complex is a multistep process involving a large number of soluble eukaryotic initiation factors (eIFs). According to current models, the TC binds to the 40S ribosome with the help of eIF1, eIF1A, and eIF3. The 43S pre-initiation complex thus formed interacts with mRNA in a manner stimulated by eIF4F and poly(A)-binding protein. The resulting 48S complex scans the mRNA until the Met-tRNAiMet base-pairs with the AUG start codon. On AUG recognition, the GTPase-activating protein eIF5 stimulates GTP hydrolysis by eIF2 while the eIFs are ejected, and the 60S subunit joins with the 40S-Met-tRNAiMet-mRNA complex in a reaction stimulated by eIF5B. By generating mutations in eIF1, eIFA, eIF-3 and examining the consequences of the mutations on the rate of translation initiation, 43S/48S complex assembly, and GCN4 translational control in living cells, we are probing the relative importance of these factors in recruiting TC and mRNA to the 40S ribosome and in scanning and selecting AUG in vivo.
We showed previously that eIF3 contains six subunits and exists in a multifactor complex (MFC) with eIF1 and eIF5 and the eIF2-GTP-Met-tRNAiMet ternary complex. We demonstrated that MFC integrity is dependent on simultaneous interaction of the eIF5 C-terminal domain (CTD) with eIF1, the eIF3c/NIP1 N-terminal domain (NTD), and eIF2-beta as well as on interaction of the eIF3a/TIF32 CTD with eIF2-beta. Reductions in 40S binding of eIF3 and eIF2 and a strong decrease in translation initiation rates in vivo (Nielsen et al., 2004) resulted from the simultaneous disruption of both interactions that link eIF3 to eIF2 in the MFC when the tif5-7A mutation in eIF5 was combined with a dominant-negative truncation of the TIF32 CTD.
Recently, we used clustered alanine substitutions of 16 segments spanning the NIP1-NTD to investigate the physiological importance of the interactions of the NIP1/eIF3c NTD with eIF1 and eIF5 in the MFC. Mutations in several segments reduced the binding of recombinant eIF1 or eIF5 to the NIP1-NTD in vitro. Mutating a C-terminal segment of the NIP1-NTD increased use of UUG start codons at a his4 allele lacking the normal AUG initiation codon, conferring a Sui– (suppressor of initiation defect) phenotype. The NIP1-NTD mutation was lethal in cells expressing a mutant form of eIF5 (eIF5-G31R) that is hyperactive in stimulating GTP hydrolysis by the TC at AUG codons. eIF1 overexpression suppressed both effects of the NIP1 mutation, as did the Sui– phenotype conferred by eIF5-G31R. Mutations in two N-terminal segments of the NIP1-NTD suppressed the Sui– phenotypes produced by the eIF1-D83G and eIF5-G31R mutations. From these and other findings, we propose that the NIP1-NTD coordinates an interaction between eIF1 and eIF5 that enables eIF1 to inhibit the GTPase-activating function of eIF5 at non–AUG codons during the scanning process. The NIP1-NTD mutation with a Sui– phenotype would alter the interaction between eIF1 and eIF5 in a manner that increases eIF5 function in promoting GTP hydrolysis at non–AUG codons, whereas the NIP1-NTD mutations that suppress the Sui– phenotypes of the eIF1-D83G and eIF5-G31R mutations would overcome the elevated GTP hydrolysis conferred by these latter mutations. In further work, we found that two NIP1-NTD mutations derepressed GCN4 translation in a manner suppressed by overexpressing the TC, with both mutations leading to reduced 40S binding of TC and eIF5 in vivo. These last results indicate that the interactions of NIP1 with eIF1 and eIF5, which help anchor TC to eIF3 in the MFC, stimulate the rate of TC recruitment to 40S ribosomes in vivo. Thus, factor interactions within the MFC that are dependent on the NIP1-NTD of eIF3 are required for efficient assembly of 43S pre-initiation complexes and for stringent selection of AUG codons by the TC, eIF1, and eIF5 during ribosomal scanning in vivo (Valasek et al., 2004).
Nielsen KH, Szamecz B, Valasek L, Jivotovskaya A, Shin B, Hinnebusch AG. Functions of eIF3 downstream of 48S assembly impact AUG recognition and GCN4 translational control. EMBO J 2004;23:1166-1177.
Valasek L, Nielsen K, Zhang F, Fekete CA, Hinnebusch AG. Interactions of eukaryotic translation initiation factor 3 (eIF3) subunit NIP1/c with eIF1 and eIF5 promote preinitiation complex assembly and regulate start codon selection. Mol Cell Biol 2004;24:9437-9455.
Structural basis for autoinhibition and mutational activation of eIF2alpha protein kinase GCN2
Qiu, Hinnebusch; in collaboration with Burley
We showed previously that the wild-type GCN2 protein kinase (PK) domain is functionally inert when physically separated from the HisRS domain but can be activated by replacement of Arg794 with Gly in the PK domain (R794G). In collaboration with Stephen Burley’s group, we determined the crystal structures of the PK domains of wild-type and R794G mutant forms of GCN2 in both the apo state and bound to ATP or AMPPNP. The structures suggest that GCN2 autoinhibition results from stabilization of a closed conformation that restricts ATP binding to the active site. The R794G mutant shows increased flexibility in the hinge region connecting the N- and C-lobes, resulting from the loss of several interactions involving R794. This conformational change is associated with intradomain movement that enhances ATP binding and hydrolysis. We propose that intramolecular interactions following tRNA binding remodel the hinge region in a manner similar to the mechanism of enzyme activation elicited by the R794G mutation (Padyana et al., 2005).
Our GCN2 PK structures lack autophosphorylation of Thr882 in the activation loop, an event critical for PK function, and the observed configuration of active site residues suggests how autophosphorylation would facilitate kinase activation. The orientation of alpha helix C in the PK domain is incompatible with catalysis, allowing the invariant Glu643 residue to interact with Arg834 (the R of the conserved “HRD” motif in the catalytic loop) rather than forming the classical salt bridge with invariant Lys628 in the beta3 strand of the N-lobe. This interaction is likely responsible for the failure of Lys628 to orient the alpha- and beta-phosphate groups of ATP properly. By analogy with other “RD” kinases, autophosphorylation of Thr882 would enable the residue to displace Glu643 from Arg834 and allow Glu643 to form the crucial salt bridge with Lys628 that is needed for proper binding of ATP to the active site (Padyana et al., 2005).
Padyana AK, Qiu H, Roll-Mecak A, Hinnebusch AG, Burley SK. Structural basis for autoinhibition and mutational activation of eIF2alpha protein kinase GCN2. J Biol Chem 2005;280:29289-29299.
Regulation of GCN2 by GCN1 and YIH1
Sattlegger, Ashcraft,2 Hinnebusch; in collaboration with, Castilho, Link, Wek
Activation of GCN2 requires physical interaction between the GCN2-NTD and a C-terminal segment of the GCN1 protein. GCN1 interacts with ribosomes in cell extracts, but it was unknown whether such activity is crucial for GCN1’s ability to stimulate GCN2 function in starved cells. We isolated point mutations in two conserved, noncontiguous segments of GCN1, mutations that led to reduced polyribosome association with GCN1-GCN20 in living cells without reducing GCN1 expression or GCN1 interaction with GCN20. Simultaneous mutation of both segments produced a greater reduction in polyribosome binding by GCN1-GCN20 and a stronger decrease in eIF2 phosphorylation than mutation of only one segment. Our findings provide strong evidence that ribosome binding by GCN1 is required for the protein’s role as a positive regulator of GCN2. One mutation mapping to the GCN1 domain that is related in sequence to translation elongation factor 3 (eEF3) decreased GCN2 activation much more than it reduced ribosome binding by GCN1. Hence, the eEF3-like domain appears to have an effector function in GCN2 activation. This conclusion supports the model that an eEF3-related activity of GCN1 influences occupancy of the ribosomal decoding site by uncharged tRNA in starved cells.
The YIH1 protein has a domain related to the GCN2-NTD, and we found that, when overexpressed, YIH1 competes with GCN2 for GCN1 binding and dampens the induction of GCN4 and its target genes (Gcn– phenotype). GCN2 overexpression suppresses the Gcn– phenotype associated with YIH1 overexpression, and YIH1 overexpression partially suppresses the growth defect conferred by a constitutively active GCN2c allele. In vivo, overexpressed YIH1 binds to GCN1, reduces native GCN1/GCN2 complex formation, and suppresses eIF2alpha phosphorylation by GCN2. YIH1 interacts with the same GCN1 fragment that binds to the GCN2-NTD, and the YIH1-GCN1 interaction requires Arg-2259 in the GCN1 fragment both in vitro and in full-length GCN1 in vivo, as found previously for GCN2-GCN1 interaction (Sattlegger et al., 2004).
In collaboration with the groups of Beatriz Castilho and Ronald Wek, we showed recently that IMPACT, the mouse orthologue of YIH1, is also the functional counterpart of YIH1. In yeast, overexpression of IMPACT lowered both basal and amino acid starvation–induced levels of phosphorylated eIF2alpha, as we described for YIH1. Overexpression of IMPACT in mouse embryonic fibroblasts inhibited phosphorylation of eIF2alpha by GCN2 under leucine starvation conditions, abolishing expression of IMPACT’s downstream target genes ATF4 (CREB-2) and CHOP (GADD153). IMPACT bound to native mouse GCN1 as well as to the minimal yeast GCN1 segment required for interaction of GCN1 with GCN2 or YIH1. We detected IMPACT mainly in the brain, and the abundance of IMPACT correlated inversely with phosphorylated eIF2alpha levels in different brain areas. The results suggest that IMPACT ensures constant high levels of translation and low levels of ATF4 and CHOP in neuronal cells under amino acid–starvation conditions (Pereira et al., 2005).
Pereira CM, Sattlegger E, Jiang HY, Longo BM, Jaqueta CB, Hinnebusch AG, Wek RC, Mello LE, Castilho BA. IMPACT, a protein preferentially expressed in the mouse brain, binds GCN1 and inhibits GCN2 activation. J Biol Chem 2005;280:28316-28323.
Sattlegger E, Swanson MJ, Ashcraft E, Jennings J, Fekete R, Link AJ, Hinnebusch AG. YIH1 is an actin-binding protein that inhibits protein kinase GCN2 and impairs general amino acid control when overexpressed. J Biol Chem 2004;279:29952-29962.
Requirement of a multiplicity of co-activators for pre-initiation complex assembly, and subunit requirements for Srb mediator recruitment, by GCN4 in vivo
Swanson,3 Qiu, Yoon,4 Sumibcay,5 Kim, Zhang, Hu, Hinnebusch
Transcriptional activation in eukaryotes typically involves sequence-specific DNA-binding proteins that bind upstream of promoters and recruit multisubunit co-activator complexes with the capacity to stimulate assembly of a pre-initiation complex (PIC) at the promoter. Some co-activators, including SWI/SNF and RSC, are ATP-dependent enzymes capable of remodeling the nucleosome structure of the promoter. Other co-activators, such as the SAGA complex, contain histone acetyltransferase (HAT) activities that facilitate chromatin remodeling or mark promoter nucleosomes as binding sites for other co-activators. A third class of co-activators, including Srb Mediator and TFIID, are physically associated with TATA-binding protein (TBP), other general transcription factors (GTFs), or RNA polymerase II (Pol II) and are thought to function as adaptors between the activator and transcriptional machinery that promote PIC assembly. Some co-activators, including SAGA, TFIID, and Mediator, are multifunctional complexes that perform both histone modifications and adaptor functions. Our previous studies showed that wild-type transcriptional activation by GCN4 is dependent on several co-activators, including SAGA, SWI/SNF, RSC, and Srb Mediator. We demonstrated that GCN4 can interact in vitro with each of these co-activators and that GCN4 recruits each of them to target promoters in vivo. We also established that all four co-activators are required for wild-type assembly of a pre-initiation complex as mutations in subunits of each co-activator reduce the recruitment of TBP and RNA polymerase II (Pol II) by GCN4 to the ARG1, ARG4, and SNZ1 promoters.
We recently showed that recruitment of SAGA, SWI/SNF, and Mediator to ARG1 occurs independently of the TATA element and pre-initiation complex formation, whereas efficient recruitment of all the general transcription factors requires the TATA box, suggesting that recruitment of the co-activators by GCN4 precedes PIC assembly. Supporting this prediction, kinetic analysis of co-activator binding at ARG1 revealed that recruitment of SAGA, SWI/SNF, and Mediator is nearly simultaneous with binding of GCN4 and is followed 5 to 10 minutes later by PIC assembly and transcription elongation by RNA polymerase II (Pol II) through the open reading frame of the gene. Despite the simultaneous recruitment of co-activators to ARG1, rapid recruitment of SWI/SNF depends on the HAT subunit of SAGA (GCN5), a non–HAT function of SAGA, and Mediator function. Mediator and the RSC complex, in turn, strongly stimulate SAGA recruitment. We observed such interdependencies in co-activator recruitment at the GCN4 target genes ARG4 and SNZ1. Recruitment of Mediator, by contrast, occurs independently of the other co-activators at ARG1 but requires SAGA at ARG4 and SNZ1. We found that Mediator and SAGA stimulate recruitment of the TBP, that all four co-activators enhance Pol II recruitment or promoter clearance following TBP binding, and that SWI/SNF and SAGA stimulate transcription elongation downstream from the promoter. Our findings reveal a program of co-activator recruitment and PIC assembly that distinguishes GCN4 from other yeast activators studied thus far (Govind et al., 2005; Qiu et al., 2005).
Govind CK, Yoon S, Qiu H, Govind S, Hinnebusch AG. Simultaneous recruitment of coactivators by Gcn4p stimulates multiple steps of transcription in vivo. Mol Cell Biol 2005;25:5626-5638.
Qiu H, Hu C, Zhang F, Hwang GJ, Swanson MJ, Boonchird C, Hinnebusch AG. Interdependent recruitment of SAGA and Srb mediator by transcriptional activator Gcn4p. Mol Cell Biol 2005;25:3461-3474.
1Leos Valasek, PhD, former Postdoctoral Fellow
2Emily Ashcraft, BS, former Predoctoral Fellow
3Mark Swanson, PhD, former Adjunct Scientist
4Sungpil Yoon, PhD, former Postdoctoral Fellow
5Laarni Sumibcay, BS, former Predoctoral Fellow
Collaborators
Stephen Burley, DPhil, MD, Structural GenomiX, Inc., San Diego, CA
Beatriz Castilho, PhD, Universidade Federal de São Paulo, São Paulo, Brazil
Andrew Link, PhD, Vanderbilt University, Nashville, TN
Ronald Wek, PhD, Indiana University School of Medicine, Indianapolis, IN
For further information, contact ahinnebusch@nih.gov.