Translational control of GCN4: an in vivo barometer of initiation-factor activity

Translation in plants--rules and exceptions

Translation processes in plants are very similar to those in other eukaryotic organisms and can in general be explained with the scanning model. Particularly among plant viruses, unconventional mRNAs are frequent, which use modulated translation processes for their expression: leaky scanning, translational stop codon readthrough or frameshifting, and transactivation by virus-encoded proteins are used to translate polycistronic mRNAs; leader and trailer sequences confer (cap-independent) efficient ribosome binding, usually in an end-dependent mechanism, but true internal ribosome entry may occur as well; in a ribosome shunt, sequences within an RNA can be bypassed by scanning ribosomes. Translation in plant cells is regulated under conditions of stress and during development, but the underlying molecular mechanisms have not yet been determined. Only a small number of plant mRNAs, whose structure suggests that they might require some unusual translation mechanisms, have been described.

Translational control of cellular and viral mRNAs

We are becoming increasingly aware of the role that translational control plays in regulating gene expression in plants. There are now many examples in which specific mechanisms have evolved at the translational level that directly impact the amount of protein produced from an mRNA. All regions of an mRNA, i.e., the 5' leader, the coding region, and the 3'-untranslated region, have the potential to influence translation. The 5'-terminal cap structure and the poly(A) tail at the 3' terminus serve as additional elements controlling translation. Many viral mRNAs have evolved alternatives to the cap and poly(A) tail that are functionally equivalent. Nevertheless, for both cellular and viral mRNAs, a co-dependent interaction between the terminal controlling elements appears to be the universal basis for efficient translation.

The plant translational apparatus

Protein synthesis in both eukaryotic and prokaryotic cells is a complex process requiring a large number of macromolecules: initiation factors, elongation factors, termination factors, ribosomes, mRNA, amino-acylsynthetases and tRNAs. This review focuses on our current knowledge of protein synthesis in higher plants.

eIF2B, the guanine nucleotide-exchange factor for eukaryotic initiation factor 2. Sequence conservation between the alpha, beta and delta subunits of eIF2B from mammals and yeast

The guanine nucleotide-exchange factor eIF2B mediates the exchange of GDP bound to translation initiation factor eIF2 for GTP. This exchange process is a key regulatory step for the control of translation initiation in eukaryotic organisms. To improve our understanding of the structure, function and regulation of eIF2B, we have obtained and sequenced cDNA species encoding all of its five subunits. Here we report the sequences of eIF2B beta and delta from rat. This paper focuses on sequence similarities between the alpha, beta and delta subunits of mammalian eIF2B. Earlier work showed that the amino acid sequences of the corresponding subunits of eIF2B in the yeast Saccharomyces cerevisiae (GCN3, GCD7 and GCD2) exhibit considerable similarity. We demonstrate that this is also true for the mammalian subunits. Moreover, alignment of the eIF2B alpha, beta and delta sequences from mammals and yeast, along with the sequence of the putative eIF2B alpha subunit from Caenorhabditis elegans and eIF2B delta from Schizosaccharomyces pombe shows that a large number of residues are identical or conserved between the C-terminal regions of all these sequences. This strong sequence conservation points to the likely functional importance of these residues. The implications of this are discussed in the light of results concerning the functions of the subunits of eIF2B in yeast and mammals. Our results also indicate that the large apparent differences in mobility on SDS/PAGE between eIF2B beta and delta subunits from rat and rabbit are not due to differences in their lengths but reflect differences in amino acid composition. We have also examined the relative expression of mRNA species encoding the alpha, beta, delta and epsilon subunits of eIF2B in a range of rat tissues by Northern blot analysis. As might be expected for mRNA species encoding subunits of a heterotrimeric protein, the ratios of expression levels of these subunits to one another did not vary between the different rat tissues examined (with the possible exception of liver). This represents the first analysis of the levels of expression of mRNA species encoding the different subunits of eIF2B.

Cloning of cDNA for the gamma-subunit of mammalian translation initiation factor 2B, the guanine nucleotide-exchange factor for eukaryotic initiation factor 2

Peptide sequence data were obtained from rabbit protein synthesis initiation factor subunit eIF2B gamma. Searching the database of expressed sequence tags (dbEST) revealed nucleotide sequences potentially encoding human eIF2B gamma that contained peptides corresponding to those from the rabbit subunit. PCR primers were derived from these sequences and used to generate a probe. This was used to screen a rat skeletal muscle cDNA library, and a clone encoding rat eIF2B gamma was isolated. This cDNA gave a product in coupled transcription/translation that co-migrated with the gamma-subunit of purified eIF2B under SDS/PAGE. The sequence of this rat eIF2B gamma cDNA is reported. The protein sequence shows homology with that of yeast eIF2B gamma (the GCD1 gene product). We have also identified an open reading frame from the Caenorhabditis elegans genome project that probably encodes the gamma-subunit of C. elegans eIF2B. All these sequences show similarity to nucleotidyl- and acyltransferases, as previously reported for GCD1 [Koonin (1995) Protein Sci. 4, 1608-1617], and contain conserved motifs potentially involved in nucleotide binding. They also contain "I-patch' motifs: isoleucine-rich hexamer repeats that have been associated with the binding of acyl groups in bacterial acyltransferases. The roles of these motifs are discussed in relation to the known properties of eIF2B.

Dosage compensation and chromatin structure in Drosophila

The past year has brought significant advances in our understanding of the male-specific lethal (msl genes and dosage compensation in Drosophila. The molecular characterization of the msl-2 gene has, to a great extent, solved the question of how msl-mediated dosage compensation is restricted to males. Molecular analyses of the msl genes have substantiated the proposal that the MSL proteins function as a multimeric complex to mediate dosage compensation. The finding that MSL-2 protein has a RING finger and the demonstration that an insulator protein facilitates the dosage compensation of X-linked genes inserted into the autosomes have opened promising avenues to identify the cis-acting dosage-compensation determinants.

Multiple steps in the regulation of transcription-factor level and activity

This review focuses on the regulation of transcription factors, many of which are DNA-binding proteins that recognize cis-regulatory elements of target genes and are the most direct regulators of gene transcription. Transcription factors serve as integration centres of the different signal-transduction pathways affecting a given gene. It is obvious that the regulation of these regulators themselves is of crucial importance for differential gene expression during development and in terminally differentiated cells. Transcription factors can be regulated at two, principally different, levels, namely concentration and activity, each of which can be modulated in a variety of ways. The concentrations of transcription factors, as of intracellular proteins in general, may be regulated at any of the steps leading from DNA to protein, including transcription, RNA processing, mRNA degradation and translation. The activity of a transcription factor is often regulated by (de) phosphorylation, which may affect different functions, e.g. nuclear localization DNA binding and trans-activation. Ligand binding is another mode of transcription-factor activation. It is typical for the large super-family of nuclear hormone receptors. Heterodimerization between transcription factors adds another dimension to the regulatory diversity and signal integration. Finally, non-DNA-binding (accessory) factors may mediate a diverse range of functions, e.g. serving as a bridge between the transcription factor and the basal transcription machinery, stabilizing the DNA-binding complex or changing the specificity of the target sequence recognition. The present review presents an overview of different modes of transcription-factor regulation, each illustrated by typical examples.

Ribosome regulation by the nascent peptide

Studies of bacterial and eukaryotic systems have identified two-gene operons in which the translation product of the upstream gene influences translation of the downstream gene. The upstream gene, referred to as a leader (gene) in bacterial systems or an upstream open reading frame (uORF) in eukaryotes, encodes a peptide that interferes with a function(s) of its translating ribosome. The peptides are therefore cis-acting negative regulators of translation. The inhibitory peptides typically consist of fewer than 25 residues and function prior to emergence from the ribosome. A biological role for this class of translation inhibitor is demonstrated in translation attenuation, a form or regulation that controls the inducible translation of the chloramphenicol resistance genes cat and cmlA in bacteria. Induction of cat or cmlA requires ribosome stalling at a particular codon in the leader region of the mRNA. Stalling destabilizes an adjacent, downstream mRNA secondary structure that normally sequesters the ribosome-binding site for the cat or cmlA coding regions. Genetic studies indicate that the nascent, leader-encoded peptide is the selector of the site of ribosome stalling in leader mRNA by cis interference with translation. Synthetic leader peptides inhibit ribosomal peptidyltransferase in vitro, leading to the prediction that this activity is the basis for stall site selection. Recent studies have shown that the leader peptides are rRNA-binding peptides with targets at the peptidyl transferase center of 23S rRNA. uORFs associated with several eukaryotic genes inhibit downstream translation. When inhibition depends on the specific codon sequence of the uORF, it has been proposed that the uORF-encoded nascent peptide prevents ribosome release from the mRNA at the uORF stop codon. This sets up a blockade to ribosome scanning which minimizes downstream translation. Segments within large proteins also appear to regulate ribosome activity in cis, although in most of the known examples the active amino acid sequences function after their emergence from the ribosome, cis control of translation by the nascent peptide is gene specific; nearly all such regulatory peptides exert no obvious trans effects in cells. The in vitro biochemical activities of the cat/cmla leader peptides on ribosomes and rRNA suggest a mechanism through which the nascent peptide can modify ribosome behavior. Other cis-acting regulatory peptides may involve more complex ribosomal interactions.

Expression of vaccinia virus K3L protein in yeast inhibits eukaryotic initiation factor-2 kinase GCN2 and the general amino acid control pathway

Phosphorylation of the alpha subunit of eukaryotic initiation factor-2 (eIF-2) is a well characterized mechanism regulating protein synthesis. Viral and cellular proteins have been identified that regulate the activity of the eIF-2alpha kinases. The regulatory protein, K3L, from vaccinia virus is homologous to the amino terminus of eIF-2alpha and is thought to inhibit the activity of the double-stranded RNA-dependent kinase suppressing the antiviral mechanism mediated by this kinase. We investigated whether K3L can inhibit the activity of the yeast eIF-2alpha kinase GCN2. Expression of K3L protein in yeast reduced the level of eIF-2alpha phosphorylation by GCN2 and blocked the stimulation of the general amino acid control pathway in response to starvation conditions. Accompanying in vitro studies showed that recombinant K3L protein reduced GCN2 autophosphorylation and phosphorylation eIF-2alpha. In agreement with the hypothesis that K3L inhibits eIF-2alpha kinases by functioning as a pseudosubstrate, we observed that K3L directly interacted with the kinase catalytic domain of GCN2. Together, these results indicate that K3L is a specific inhibitor of eIF-2alpha kinases from mammals and yeast and suggest that the kinases contain common structural features important for recognition of their substrate eIF-2alpha.

Role of an upstream open reading frame in mediating arginine-specific translational control in Neurospora crassa

The Neurospora crassa arg-2 transcript contains an upstream open reading frame (uORF) specifying a 24-residue leader peptide and is subject to a novel form of negative translational regulation in response to arginine. The role of the arg-2 uORF in arginine-specific negative regulation was investigated by using translational fusions of wild-type and mutant arg-2 sequences to the Escherichia coli lacZ reporter gene specifying beta-galactosidase. The wild-type uORF conferred Arg-specific regulation on the reporter gene in N. crassa, but mutated or truncated uORFs did not, as determined by measurements of beta-galactosidase activity produced in N. crassa strains expressing arg-2-lacZ fusion genes. All effects on reporter gene expression were posttranscriptional, as determined by measurement of RNA levels. Both sequence-dependent and sequence-independent effects of uORFs were observed. Genes containing the wild-type uORF or a 21-codon mutated uORF showed reduced translation in comparison with that of a gene lacking a uORF. Both uORF-containing transcripts showed reduced association with polysomes relative to transcripts lacking a uORF, but only the transcript with the wild-type uORF showed a reduced average number of ribosomes associated with it in response to arginine addition. Direct translational fusions between uORF sequences and lacZ sequences indicated that the uORF is translated. Overlapping the uORF with the lacZ initiation codon indicated that ribosome reinitiation at a downstream start codon is not integral to uORF-mediated, Arg-specific translational regulation. These studies provide direct biochemical evidence for arg-2 uORF function in translational control.

mRNA sequences influencing translation and the selection of AUG initiator codons in the yeast Saccharomyces cerevisiae

The secondary structure and sequences influencing the expression and selection of the AUG initiator codon in the yeast Saccharomyces cerevisiae were investigated with two fused genes, which were composed of either the CYC7 or CYC1 leader regions, respectively, linked to the lacZ coding region. In addition, the strains contained the upf1-delta disruption, which stabilized mRNAs that had premature termination codons, resulting in wild-type levels. The following major conclusions were reached by measuring beta-galactosidase activities in yeast strains having integrated single copies of the fused genes with various alterations in the 89 and 38 nucleotide-long untranslated CYC7 and CYC1 leader regions, respectively. The leader region adjacent to the AUG initiator codon was dispensable, but the nucleotide preceding the AUG initiator at position -3 modified the efficiency of translation by less than twofold, exhibiting an order of preference A > G > C > U. Upstream out-of-frame AUG triplets diminished initiation at the normal site, from essentially complete inhibition to approximately 50% inhibition, depending on the position of the upstream AUG triplet and on the context (-3 position nucleotides) of the two AUG triplets. In this regard, complete inhibition occurred when the upstream and downstream AUG triplets were closer together, and when the upstream and downstream AUG triplets had, respectively, optimal and suboptimal contexts. Thus, leaky scanning occurs in yeast, similar to its occurrence in higher eukaryotes. In contrast, termination codons between two AUG triplets causes reinitiation at the downstream AUG in higher eukaryotes, but not generally in yeast. Our results and the results of others with GCN4 mRNA and its derivatives indicate that reinitiation is not a general phenomenon in yeast, and that special sequences are required.

Patterns of translational regulation in the mammalian testis

The translational activity of more than 40 different mRNAs in rodent testes has been analyzed by determining the proportions of inactive free-mRNPs and active polysomal mRNAs in sucrose gradients. These mRNAs can be sorted into several groups comprising mRNAs with similar patterns of translational activity in particular cell types. mRNAs in testicular somatic cells sediment primarily with polysomes, indicating that they are translated efficiently, whereas the vast majority of mRNAs in late meiotic and haploid spermatogenic cells display high levels of free-mRNAPs, indicative of a block to the initiation of translation. Protamine mRNAs exemplify a group of mRNAs that is transcribed in round spermatids, stored as free-mRNPs for several days, and translated in elongated spermatids after the cessation of transcription. The extent to which the free-mRNPs in primary spermatocytes and round spermatids are due to developmental changes in translational activity is unclear. mRNAs at these stages can often be detected earlier than the corresponding protein, implicating either a delay in translational activation or difficulties in detecting the protein. In contrast, sucrose gradients consistently indicate little difference in the proportions of various mRNAs in free-mRNPs in primary spermatocytes and round spermatids, implying that the proportions of translationally active mRNAs remain essentially constant. Since the levels of some mRNAs appear to greatly exceed the amount that is translated, the biological significance of some free-mRNPs in meiotic and early haploid cells in unclear. There are numerous examples of controls over the translation of individual mRNAs in meiotic and haploid cells; the proportions of various mRNAs in free-mRNPs range from virtually none to virtually all, and individual mRNAs are activated at specific stages in elongated spermatids. Existing evidence is contradictory whether the mRNAs in the protamine/transition protein gene family are repressed by mRNP proteins of sequestration.

Eukaryotic gene expression: metabolite control by amino acids

Our understanding of the metabolite control in mammalian cells lags far behind that in prokaryotes. This is particularly true for amino-acid-dependent gene expression. Few proteins have been identified for which synthesis is selectively regulated by amino-acid availability, and the mechanisms for control of transcription and translation in response to changes in amino-acid availability have not yet been elucidated. The intimate relationship between amino-acid supply and the fundamental cellular process of protein synthesis makes amino-acid-dependent control of gene expression particularly important. Future studies should provide important insight into amino-acid and other nutrient signaling pathways, and their impact on cellular growth and metabolism.

Essential amino acids regulate fatty acid synthase expression through an uncharged transfer RNA-dependent mechanism

To better understand the regulation of gene expression by amino acids, we studied the effects of these macronutrients on fatty acid synthase (FAS), an enzyme crucial for energy storage. When HepG2 cells were fed serum-free media selectively deficient in each amino acid, the omission of any single classic essential amino acid as well as Arg or His (essential in some rapidly growing cells) resulted in FAS mRNA levels that were about half of those in complete medium. Control message levels were unaffected and omission of nonessential amino acids did not alter FAS expression. FAS mRNA levels peaked 12-16 h after feeding complete and Ser (nonessential)-deficient media but did not increase in cells fed Lys (essential)-deficient medium. With Lys, FAS mRNA increased over the physiologic concentration range of 15-150 microM, and low concentrations of lysine decreased FAS but not apoB protein mass. Transcription inhibitors mimicked treatment with Lys-deficient media, and nuclear run-off assays showed that Lys-deficient media abolished FAS but not apoB transcription. After treatment with Lys-deficient media, the intracellular Lys pool was rapidly depleted in association with an increase of uncharged (deacylated) tRNA Lys from < 1 to 64% of available tRNA Lys. Even in the presence of the essential amino acid His, increasing the level of uncharged tRNA His with histidinol, a competitive inhibitor of the histidinyl-tRNA synthetase, blocked FAS expression. Tyrosinol treatment did not alter FAS mRNA levels. These results suggest that essential amino acids regulate FAS expression by altering uncharged tRNA levels, a novel mechanism for nutrient control of gene expression in mammalian cells.

Characterization of the in vivo phosphorylation sites of the mRNA.cap-binding complex proteins eukaryotic initiation factor-4E and p20 in Saccharomyces cerevisiae

Eukaryotic translation is believed to be regulated via the phosphorylation of specific eukaryotic initiation factors (eIFs), including one of the cap-binding complex proteins, eIF-4E. We show that in the yeast Saccharomyces cerevisiae, both eIF-4E and another cap-binding complex protein, p20, are phosphoproteins. The major sites of phosphorylation of yeast eIF-4E are found to be located in the N-terminal region of its sequence (Ser2 and Ser15) and are thus in a different part of the protein from the main phosphorylation sites (Ser53 and Ser209) proposed previously for mammalian eIF-4E. The most likely sites of p20 phosphorylation are at Ser91 and/or Ser154. All of the major sites in the two yeast proteins are phosphorylated by casein kinase II in vitro. Casein kinase II phosphorylation of cap-complex proteins should therefore be considered as potentially involved in the control of yeast protein synthesis. Mutagenesis experiments revealed that yeast eIF-4E activity is not dependent on the presence of Ser2 or Ser15. On the other hand, we observed variations in the amount of (phosphorylated) p20 associated with the cap-binding complex as a function of cell growth conditions. Our results suggest that interactions of yeast eIF-4E with other phosphorylatable proteins, such as p20, could play a pivotal role in translational control.

Protein synthesis in eukaryotic organisms: new insights into the function of translation initiation factor eIF-3

The pathway for initiation of protein synthesis in eukaryotic cells has been defined and refined over the last 25 years using purified components and in vitro reconstituted systems. More recently, powerful genetic analysis in yeast has proved useful in unraveling aspects of translation inherently more difficult to address by strictly biochemical approaches. One area in particular is the functional analysis of multi-subunit protein factors, termed eukaryotic initiation factors (eIFs), that play an essential role in translation initiation. eIF-3, the most structurally complex of the eIFs, has until recently eluded this approach. The identification of the yeast GCD10 gene as the structural gene for the zeta subunit of yeast eIF-3(1) and the analysis of mutant phenotypes has opened the door to the genetic dissection of the eIF-3 protein complex.

Sequences 5' of the first upstream open reading frame in GCN4 mRNA are required for efficient translational reinitiation

Translation of yeast GCN4 mRNA occurs by a reinitiation mechanism that is modulated by amino acid levels in the cell. Ribosomes which translate the first of four upstream open reading frames (uORFs) in the mRNA leader resume scanning and can reinitiate downstream. Under non-starvation conditions reinitiation occurs at one of the remaining three uORFs and GCN4 is repressed. Under starvation conditions, in contrast, ribosomes bypass the uORFs and reinitiate at GCN4 instead. The high frequency of reinitiation following uORF1 translation depends on an adequate distance to the next start codon and particular sequences surrounding the uORF1 stop codon. We present evidence that sequences 5' to uORF1 also strongly enhance reinitiation. First, reinitiation was severely inhibited when uORF1 was transplanted into the position of uORF4, even though the native sequence environment of the uORF1 stop codon was maintained, and this effect could not be accounted for by the decreased uORF1-GCN4 spacing. Second, insertions and deletions in the leader preceding uORF1 greatly reduced reinitiation at GCN4. Sequences 5' to uORF1 may influence the probability of ribosome release following peptide termination at uORF1. Alternatively, they may facilitate rebinding of an initiation factor required for reinitiation prior to resumption of the scanning process.

Translational regulation in response to changes in amino acid availability in Neurospora crassa

We examined the regulation of Neurospora crassa arg-2 and cpc-1 in response to amino acid availability.arg-2 encodes the small subunit of arginine-specific carbamoyl phosphate synthetase; it is subject to unique negative regulation by Arg and is positively regulated in response to limitation for many different amino acids through a mechanism known as cross-pathway control. cpc-1 specifies a transcriptional activator important for crosspathway control. Expression of these genes was compared with that of the cytochrome oxidase subunit V gene, cox-5. Analyses of mRNA levels, polypeptide pulse-labeling results, and the distribution of mRNA in polysomes indicated that Arg-specific negative regulation of arg-2 affected the levels of both arg-2 mRNA and arg-2 mRNA translation. Negative translational effects on arg-2 and positive translational effects on cpc-1 were apparent soon after cells were provided with exogenous Arg. In cells limited for His, increased expression of arg-2 and cpc-1, and decreased expression of cox-5, also had translational and transcriptional components. The arg-2 and cpc-1 transcripts contain upstream open reading frames (uORFs), as do their Saccharomyces cerevisiae homologs CPA1 and GCN4. We examined the regulation of arg-2-lacZ reporter genes containing or lacking the uORF start codon; the capacity for arg-2 uORF translation appeared critical for controlling gene expression.

Stress-induced transcriptional activation

Living cells, both prokaryotic and eukaryotic, employ specific sensory and signalling systems to obtain and transmit information from their environment in order to adjust cellular metabolism, growth, and development to environmental alterations. Among external factors that trigger such molecular communications are nutrients, ions, drugs and other compounds, and physical parameters such as temperature and pressure. One could consider stress imposed on cells as any disturbance of the normal growth condition and even as any deviation from optimal growth circumstances. It may be worthwhile to distinguish specific and general stress circumstances. Reasoning from this angle, the extensively studied response to heat stress on the one hand is a specific response of cells challenged with supra-optimal temperatures. This response makes use of the sophisticated chaperoning mechanisms playing a role during normal protein folding and turnover. The response is aimed primarily at protection and repair of cellular components and partly at acquisition of heat tolerance. In addition, heat stress conditions induce a general response, in common with other metabolically adverse circumstances leading to physiological perturbations, such as oxidative stress or osmostress. Furthermore, it is obvious that limitation of essential nutrients, such as glucose or amino acids for yeasts, leads to such a metabolic response. The purpose of the general response may be to promote rapid recovery from the stressful condition and resumption of normal growth. This review focuses on the changes in gene expression that occur when cells are challenged by stress, with major emphasis on the transcription factors involved, their cognate promoter elements, and the modulation of their activity upon stress signal transduction. With respect to heat shock-induced changes, a wealth of information on both prokaryotic and eukaryotic organisms, including yeasts, is available. As far as the concept of the general (metabolic) stress response is concerned, major attention will be paid to Saccharomyces cerevisiae.

Multidomain organization of eukaryotic guanine nucleotide exchange translation initiation factor eIF-2B subunits revealed by analysis of conserved sequence motifs

Computer-assisted analysis of amino acid sequences using methods for database screening with individual sequences and with multiple alignment blocks reveals a complex multidomain organization of yeast proteins GCD6 and GCD1, and mammalian homolog of GCD6-subunits of the eukaryotic translation initiation factor eIF-2B involved in GDP/GTP exchange on eIF-2. It is shown that these proteins contain a putative nucleotide-binding domain related to a variety of nucleotidyltransferases, most of which are involved in nucleoside diphosphate-sugar formation in bacteria. Three conserved motifs, one of which appears to be a variant of the phosphate-binding site (P-loop) and another that may be considered a specific version of the Mg(2+)-binding site of NTP-utilizing enzymes, were identified in the nucleotidyltransferase-related domain. Together with the third unique motif adjacent to the the P-loop, these motifs comprise the signature of a new superfamily of nucleotide-binding domains. A domain consisting of hexapeptide amino acid repeats with a periodic distribution of bulky hydrophobic residues (isoleucine patch), which previously have been identified in bacterial acetyltransferases, is located toward the C-terminus from the nucleotidyltransferase-related domain. Finally, at the very C-termini of GCD6, eIF-2B epsilon, and two other eukaryotic translation initiation factors, eIF-4 gamma and eIF-5, there is a previously undetected, conserved domain. It is hypothesized that the nucleotidyltransferase-related domain is directly involved in the GDP/GTP exchange, whereas the C-terminal conserved domain may be involved in the interaction of eIF-2B, eIF-4 gamma, and eIF-5 with eIF-2.

GCD10, a translational repressor of GCN4, is the RNA-binding subunit of eukaryotic translation initiation factor-3

GCN4 mRNA is translated by a reinitiation mechanism involving four short upstream open reading frames (uORFs) in its leader sequence. Decreasing the activity of eukaryotic initiation factor-2 (eIF-2) by phosphorylation inhibits general translation in yeast but stimulates GCN4 expression by allowing ribosomes to scan past the uORFs and reinitiate at GCN4 instead. GCD10 was first identified genetically as a translational repressor of GCN4. We show here that GCD10 is an essential protein of 54.6 kD that is required in vivo for the initiation of total protein synthesis. GCD10 binds RNA in vitro and we present strong biochemical evidence that it is identical to the RNA-binding subunit of yeast initiation factor-3 (eIF-3). eIF-3 is a multisubunit complex that stimulates translation initiation in vitro at several different steps. We suggest that gcd10 mutations decrease the ability of eIF-3 to stimulate binding of eIF-2/GTP/Met-tRNA(iMet) ternary complexes to small ribosomal subunits in vivo. This would explain why mutations in eIF-3 mimic eIF-2 alpha phosphorylation in allowing ribosomes to bypass the uORFs and reinitiate at GCN4. Our results indicate that GCN4 expression provides a sensitive in vivo assay for the function of eIF-3 in initiation complex formation.

Translational regulation: versatile mechanisms for metabolic and developmental control

It has become clear that many vital metabolic circuits and early developmental programs are regulated translationally. Until recently, the mechanisms underlying most of these observations were poorly understood. The past year has witnessed several important advances in the understanding of how the translational apparatus is controlled by different regulatory mechanisms.

PKR: a new name and new roles

The double-stranded RNA (dsRNA)-activated protein kinase, now called PKR, was first discovered by virtue of its ability to phosphorylate translation initiation factor eIF-2 and inhibit its activity. Recent studies have shown that expression of inactive mutants of PKR in cultured cells causes them to acquire characteristics typical of transformed cells. These and other findings indicate that PKR plays a role in the normal control of cell growth and differentiation. It seems likely that, in addition to eIF-2, PKR has other substrates including the protein I-kappa B, which regulates the transcription of certain genes. Indeed, it now seems likely that PKR mediates the regulation of selected genes by dsRNA.

Cytoplasmic mRNA-protein interactions in eukaryotic gene expression

Post-transcriptional mechanisms contribute in many important ways to the overall control and regulation of gene expression, and in doing so employ a veritable army of proteins that bind a wide range of targets in messenger RNA (mRNA). The full range of these RNA-protein interactions is only just beginning to emerge, and much remains to be learned about the mechanisms underlying the rapidly increasing number of regulatory systems now being described.

The relationship between eukaryotic translation and mRNA stability. A short upstream open reading frame strongly inhibits translational initiation and greatly accelerates mRNA degradation in the yeast Saccharomyces cerevisiae

A new strategy was developed to study the relationship between the translation and degradation of a specific mRNA in the yeast Saccharomyces cerevisiae. A series of 5'-untranslated regions (UTR) was combined with the cat gene from the bacterial transposon Tn9, allowing us to test the influence of upstream open reading frames (uORFs) on translation and mRNA stability. The 5'-UTR sequences were designed so that the minimum possible sequence alteration, a single nucleotide substitution, could be used to create a 7-codon ORF upstream of the cat gene. The uORF was translated efficiently, but at the same time inhibited translation of the cat ORF and destabilized the cat mRNA. Investigations of various derivatives of the 5'-UTR indicated that cat translation was primarily attributable to leaky scanning of ribosomes past the uORF rather than to reinitiation. Therefore, these data directly demonstrate destabilization of a specific mRNA linked to changes in translational initiation on the same transcript. In contrast to the previously proposed nonsense-mediated mRNA decay pathway, destabilization was not triggered by premature translational termination in the main ORF and was not discernibly dependent upon a reinitiation-driven mechanism. This suggests the existence of an as yet not described pathway of translation-linked mRNA degradation.