Purification and properties of eIF-2 phosphatase

Eukaryotic initiation factor 2 (eIF-2) phosphatase has been purified 840-fold to apparent homogeneity from rabbit reticulocyte lysate. Native eIF-2 phosphatase has a Mr = 98,000, pI = 6.1, s20,w = 5.1, and a Stokes radius = 38 A. A subunit composition of one 60,000-dalton polypeptide and one 38,000-dalton polypeptide is indicated. The Km for [32P]eIF-2 is 30 microM and the Vmax = 1.1 nmol of phosphate released/min/microgram. The 38,000-dalton subunit of eIF-2 phosphatase does not co-electrophorese with the catalytic subunit of liver phosphorylase phosphatase, a type 1 protein phosphatase. The specificity of eIF-2 phosphatase for phosphorylation sites on th alpha- and beta-subunits of eIF-2 appears to be determined by the environment of the phosphatase and substrate. Both the alpha- and beta-subunits of [32P]eIF-2 are rapidly dephosphorylated by the purified phosphatase. In unfractionated lysate and in unfractionated lysate supplemented with an equivalent activity of the purified phosphatase, only the alpha-subunit of eIF-2 is dephosphorylated. This indicates other factors are present in the lysate which govern the dephosphorylation of eIF-2.     

Evidence that phosphorylation of eIF-2(alpha) prevents the eIF-2B-mediated dissociation of eIF-2 X GDP from the 60 S subunit of complete initiation complexes

Recent observations have indicated that eukaryotic initiation factor (eIF)-2 and GTP or GDP normally bind to 60 S ribosomal subunits in rabbit reticulocyte lysate and that when eIF-2 alpha is phosphorylated and polypeptide chain initiation is inhibited, eIF-2 X GDP accumulates on 60 S subunits due to impaired dissociation that is normally mediated by the reversing factor (eIF-2B). Current findings now indicate that inhibition due to phosphorylation of eIF-2 alpha is mediated, at least in part, by the inability to dissociate eIF-2 X GDP from the 60 S subunit of complete initiation complexes. At the onset of inhibition, there is an accumulation of Met-tRNA(f) and eIF-2 on the polysomes, despite a marked reduction in Met-tRNA(f) bound to 40 S subunits and Met-peptidyl-tRNA bound to the polysomes. This initial effect is not associated with the formation of "half-mers" (polysomes containing an extra unpaired 40 S subunit), and the 40 S X Met-tRNA(f) complexes, though reduced, still sediment at 43 S. When inhibition is maximal and the polysomes are largely disaggregated, there is an accumulation of 48 S complexes consisting of a 40 S subunit and Met-tRNA(f) bound to globin mRNA as well as small polysomal half-mers, such that residual protein synthesis occurs to about the same degree on "1 1/2"s and "2 1/2"s as on mono-, di-, and triribosomes. Exogenous eIF-2B increases protein synthesis on mono-, di-, and triribosomes and decreases that on half-mers. This is associated with reduced binding of Met-tRNA(f) and eIF-2 to ribosomal particles sedimenting at 80 S and greater and a shift from 48 S to 43 S complexes. These results suggest that eIF-2B must normally promote dissociation of eIF-2 X GDP from the 60 S subunit of complete initiation complexes before they can elongate but cannot when eIF-2 alpha is phosphorylated, resulting in the accumulation of these complexes, some of which dissociate into Met-tRNA(f) X 40 S X mRNA and 60 S X eIF-2 X GDP.     

Modulation of tRNA(iMet), eIF-2, and eIF-2B expression shows that GCN4 translation is inversely coupled to the level of eIF-2.GTP.Met-tRNA(iMet) ternary complexes.

To understand how phosphorylation of eukaryotic translation initiation factor (eIF)-2 alpha in Saccharomyces cerevisiae stimulates GCN4 mRNA translation while at the same time inhibiting general translation initiation, we examined the effects of altering the gene dosage of initiator tRNA(Met), eIF-2, and the guanine nucleotide exchange factor for eIF-2, eIF-2B. Overexpression of all three subunits of eIF-2 or all five subunits of eIF-2B suppressed the effects of eIF-2 alpha hyperphosphorylation on both GCN4-specific and general translation initiation. Consistent with eIF-2 functioning in translation as part of a ternary complex composed of eIF-2, GTP, and Met-tRNA(iMet), reduced gene dosage of initiator tRNA(Met) mimicked phosphorylation of eIF-2 alpha and stimulated GCN4 translation. In addition, overexpression of a combination of eIF-2 and tRNA(iMet) suppressed the growth-inhibitory effects of eIF-2 hyperphosphorylation more effectively than an increase in the level of either component of the ternary complex alone. These results provide in vivo evidence that phosphorylation of eIF-2 alpha reduces the activities of both eIF-2 and eIF-2B and that the eIF-2.GTP. Met-tRNA(iMet) ternary complex is the principal component limiting translation in cells when eIF-2 alpha is phosphorylated on serine 51. Analysis of eIF-2 alpha phosphorylation in the eIF-2-overexpressing strain also provides in vivo evidence that phosphorylated eIF-2 acts as a competitive inhibitor of eIF-2B rather than forming an excessively stable inactive complex. Finally, our results demonstrate that the concentration of eIF-2-GTP. Met-tRNA(iMet) ternary complexes is the cardinal parameter determining the site of reinitiation on GCN4 mRNA and support the idea that reinitiation at GCN4 is inversely related to the concentration of ternary complexes in the cell.     

Kinetics of dephosphorylation of eIF-2(alpha P) and reutilization of mRNA

Phosphorylation of the alpha subunit of eukaryotic initiation factor 2 (eIF-2) causes mRNA to accumulate in 48 S complexes containing Met-tRNAf and eIF-2(alpha P). When the eIF-2 alpha kinase is inhibited by 2-aminopurine, the mRNA is slowly transferred from 48 to 80 S initiation complexes after an initial lag. The cause of this lag was examined by investigating whether mRNA and Met-tRNAf dissociated from 48 S complexes before binding to 80 S. Both compounds were quantitatively transferred from 48 to 80 S complexes after addition of 2-aminopurine and the eIF-2(alpha P) bound to 48 S complexes was dephosphorylated after an initial lag more slowly than unbound eIF-2(alpha P), which was rapidly dephosphorylated. the eIF-2(alpha P) in isolated 48 S complexes was slowly dephosphorylated by partially purified lysate phosphatases, whereas free eIF-2(alpha P) was readily dephosphorylated. These results indicated that 48 S complexes could directly join to a 60 S ribosomal subunit after eIF-2(alpha P) dephosphorylation. The lag and slow kinetics of dephosphorylation of eIF-2(alpha P) bound to 48 S complexes accounted for the slow transfer of mRNA from 48 to 80 S complexes. Moreover, the mRNA bound to 48 S complexes was more susceptible to cleavage by an endonuclease than mRNA in polyribosomes, as shown by activating the (2'-5')oligo(A)-dependent endonuclease. This finding is discussed in view of the possible role of eIF-2 alpha kinase and endonuclease in the inhibition of viral mRNA translation in interferon-treated cells.     

Association of initiation factor eIF-4E in a cap binding protein complex (eIF-4F) is critical for and enhances phosphorylation by protein kinase C

Phosphorylation by protein kinase C of the mRNA cap binding protein purified as part of a cap binding protein complex (eIF-4F) or as a single protein (eIF-4E), has been examined. Significant phosphorylation (up to 1 mol of phosphate/mol of p25 subunit) occurs only when the protein is part of the eIF-4F complex. With purified eIF-4E, using the same conditions, up to 0.1 mol of phosphate can be incorporated. Tryptic phosphopeptide maps show that the site phosphorylated in the Mr 25,000 subunit of eIF-4F (eIF-4F p25) is the same as that modified in purified eIF-4E. Kinetic measurements obtained from initial rates indicate that the Km values for eIF-4F and eIF-4E are similar, although the Vmax is 5-6 times higher for the complex. Dephosphorylation of eIF-4F p25, previously phosphorylated with protein kinase C, occurs in reticulocyte lysate with a half-life of 15-20 min, whereas little dephosphorylation is observed after 15 min with the purified phosphorylated eIF-4E. Phosphorylation of eIF-4F on the p220 and p25 subunits does not affect the stability of the complex as indicated by gel filtration on Sephacryl S-300. However, addition of non-phosphorylated eIF-4E to the phosphorylated complex results in the dissociation of the complex. These results suggest that interaction of p25 with other subunits in the complex greatly affects phosphorylation/dephosphorylation of p25. Since the rate of phosphorylation/dephosphorylation is significantly greater in the complex, regulation of the cap binding protein by phosphorylation appears to occur primarily on eIF-4F.     

Casein kinase II mediates multiple phosphorylation of Saccharomyces cerevisiae eIF-2 alpha (encoded by SUI2), which is required for optimal eIF-2 function in S. cerevisiae.

Previous studies have demonstrated that the alpha subunit of eukaryotic initiation factor 2 (eIF-2 alpha), encoded by the SUI2 gene in the yeast Saccharomyces cerevisiae, is phosphorylated at Ser-51 by the GCN2 kinase in response to general amino acid control. Here we describe that yeast eIF-2 alpha is a constitutively phosphorylated protein species that is multiply phosphorylated by a GCN2-independent mechanism. 32Pi labeling and isoelectric focusing analysis of a SUI2+ delta gcn2 strain identifies eIF-2 alpha as radiolabeled and a single isoelectric protein species. Treatment of SUI2+ delta gcn2 strain extracts with phosphatase results in the identification of three additional isoelectric forms of eIF-2 alpha that correspond to the stepwise removal of three phosphates from the protein. Mutational analysis of SUI2 coupled with biochemical analysis of eIF-2 alpha maps the sites to the carboxyl region of SUI2 that correspond to Ser residues at amino acid positions 292, 294, and 301 that compose consensus casein kinase II sequences. 32Pi labeling or isoelectric focusing analysis of eIF-2 alpha from conditional casein kinase II mutants indicated that phosphorylation of eIF-2 alpha is abolished or dephosphorylated forms of eIF-2 alpha are detected when these strains are grown at the restrictive growth conditions. Furthermore, yeast casein kinase II phosphorylates recombinant wild-type eIF-2 alpha protein in vitro but does not phosphorylate recombinant eIF-2 alpha that contains Ser-to-Ala mutations at all three consensus casein kinase II sequences. These data strongly support the conclusion that casein kinase II directly phosphorylates eIF-2 alpha at one or all of these Ser amino acids in vivo. Although substitution of SUI2 genes mutated at these sites for the wild-type gene have no obvious effect on cell growth, one test that we have used appears to demonstrate that the inability to phosphorylate these sites has a physiological consequence on eIF-2 function in S. cerevisiae. Haploid strains constructed to contain Ser-to-Ala mutations at the consensus casein kinase II sequences in SUI2 in combination with a mutated allele of either the GCN2, GCN3, or GCD7 gene have synthetic growth defects. These genetic data appear to indicate that the modifications that we describe at the carboxyl end of the eIF-2 alpha protein are required for optimal eIF-2 function in S. cerevisiae.     

Specific phosphorylation of pig liver initiation factor eIF-2 by the N-ethylmaleimide-treated hemin-controlled translational inhibitor

The specific phosphorylation of pig liver initiation factor 2(eIF-2) by the N-ethylmaleimide (NEM)-treated hemin-controlled translational inhibitor (HCI) from rabbit reticulocytes was investigated. The inhibitor phosphorylated the serine residue of the alpha subunit of eIF-2 (eIF-2 alpha) and 1 mol of phosphate was incorporated into 1 mol of eIF-2 alpha by the inhibitor on maximal phosphorylation, even when eIF-2 was pretreated with alkaline phosphatase prior to phosphorylation. The 32P-labeled eIF-2 alpha was subjected to tryptic digestion and the tryptic digest was analyzed by two-dimensional peptide mapping on a cellulose thin-layer sheet. After 94 h digestion, the autoradiograph of the peptide map showed a single 32P-labeled band with a molecular weight of approximately 1,200. These findings suggest that one specific serine residue of pig liver eIF-2 alpha was phosphorylated by the NEM-treated HCI.     

eIF-2B and the exchange of guanine nucleotides bound to eIF-2

Available data for the formation of the ternary complex eIF-2 X GTP X methionyl-tRNAi involved in eukaryotic initiation and of the inhibition of ternary complex formation by GDP have been examined with a view to determining the mechanism by which eIF-2B facilitates nucleotide exchange. Two mechanisms have been considered--first a displacement reaction in which eIF-2B displaces GDP and GTP in a manner analogous to a "ping-pong" enzyme mechanism, and secondly the possibility that binding of eIF-2B to eIF-2 nucleotide complexes enhances the rate of nucleotide exchange without itself inducing nucleotide displacement. Comparison has been made between the properties of eIF-2 and eIF-2B and of the bacterial elongation factors Tu and Ts. It seems most probable that, as previously suggested by others for Ts, eIF-2B effectively catalyses an exchange reaction through a "ping-pong" type mechanism. Possible explanations of data suggesting otherwise are put forward. Both eIF-2 and bacterial Tu are complex allosteric proteins subject to a variety of influences which in the case of eIF-2 include phosphorylation of the alpha subunit. This phosphorylation appears to change the equilibria in the reaction mechanism such that the transferred entity (eIF-2) becomes firmly bound to the catalyst (eIF-2B). Minimum rate constants for the formation of eIF-2 X eIF-2B from eIF-2 X GDP and eIF-2 X GTP and reverse reactions are derived. These values suggest that the initiation factors are likely to have to operate in a restricted environment if rates of protein synthesis seen in vivo are to be sustained.     

Evidence that the primary effect of phosphorylation of eukaryotic initiation factor 2(alpha) in rabbit reticulocyte lysate is inhibition of the release of eukaryotic initiation factor-2.GDP from 60 S ribosomal subunits

The phosphorylation of eukaryotic initiation factor (eIF) 2 alpha that occurs when rabbit reticulocyte lysate is incubated in the absence of hemin or with poly(I.C) causes inhibition of polypeptide chain initiation by preventing a separate factor (termed RF) from promoting the exchange of GTP for GDP on eIF-2. When lysate was incubated in the presence of hemin and [14C] eIF-2 or [alpha-32P]GTP, we observed binding of eIF-2 and GDP or GTP to 60 S ribosomal subunits that was slightly greater than that bound to 40 S subunits and little binding to 80 S ribosomes. When incubation was in the absence of hemin or in the presence of hemin plus 0.1 microgram/ml poly(I.C), eIF-2 and GDP binding to 60 S subunits was increased 1.5- to 2-fold, that bound to 80 S ribosomes was almost as great as that bound to 60 S subunits, and that bound to 40 S subunits was unchanged. Our data indicate that about 40% of the eIF-2 that becomes bound to 60 S subunits and 80 S ribosomes in the absence of hemin or with poly(I.C) is eIF-2(alpha-P) and suggest that the eIF-2 and GDP bound is probably in the form of a binary complex. The accumulation of eIF-2.GDP on 60 S subunits occurs before binding of Met-tRNAf to 40 S subunits becomes reduced and before protein synthesis becomes inhibited. The rate of turnover of GDP (presumably eIF-2.GDP) on 60 S subunits and 80 S ribosomes in the absence of hemin is reduced to less than 10% the control rate, because the dissociation of eIF-2.GDP is inhibited. Additional RF increases the turnover of eIF-2.GDP on 60 S subunits and 80 S ribosomes to near the control rate by promoting dissociation of eIF-2.GDP but not eIF-2(alpha-P).GDP. Our findings suggest that eIF-2.GTP binding to and eIF-2.GDP release from 60 S subunits may normally occur and serve to promote subunit joining. The phosphorylation of eIF-2 alpha inhibits polypeptide chain initiation by preventing dissociation of eIF-2.GDP from either free 60 S subunits (thus inhibiting subunit joining directly) or the 60 S subunit component of an 80 S initiation complex (thereby blocking elongation and resulting in the dissociation of the 80 S complex).     

Regulation of protein synthesis in eukaryotes. Eukaryotic initiation factor eIF-2 and eukaryotic recycling factor eRF from neuroblastoma cells

eIF-2 purified from neuroblastoma cells consists of three subunits, which appear to be of molecular weight identical to those of the subunits of rabbit reticulocyte eIF-2. A protein fraction has been isolated from neuroblastoma cells with characteristics similar to eRF from reticulocytes: stimulation of amino acid incorporation in a hemin-deprived reticulocyte lysate, the removal of GDP from eIF-2-GDP complexes, a 4-5-fold stimulatory effect in a two-step reaction measuring 40 S preinitiation complex formation and a 3-3.5-fold stimulation in the methionyl-puromycin synthesis. In the methionyl-puromycin-synthesizing system phosphorylated eIF-2 is not responsive to the addition of this fraction from neuroblastoma cells. The protein fraction contains eRF which seems to be similar to the eRF isolated from Ehrlich ascites tumor cells and somewhat distinct from the reticulocyte factor. Incubation of neuroblastoma cell lysate in the presence of [gamma-32P]ATP results in the phosphorylation of a protein of Mr 36 000, migrating on SDS-polyacrylamide gels to the position of eIF-2 alpha. This protein is also phosphorylated in vitro by HRI from reticulocytes. These results may reflect a common underlying principle for the quantitative regulation of protein synthesis in eukaryotic cells.     

Two phosphorylation sites on eIF-2 alpha

Protein synthesis in mammalian cells can be regulated through phosphorylation/dephosphorylation of the alpha subunit of initiation factor 2, eIF-2. Two specific kinases have been identified that apparently phosphorylate the same site(s). Controversy exists as to whether serine-48 is a phosphorylation site in addition to serine-51. A recent publication is discussed that, in this author's view, answers the question of the phosphorylation sites. It is suggested that phosphorylation proceeds sequentially with serine-51 being the first and serine-48 the second phosphorylation site. Phosphorylation of both sites is required for inhibition of protein synthesis.     

Phosphorylation of translational initiation factor 3 (eIF-3) by cyclic AMP-regulated protein kinase.

Translational initiation factor 3 (eIF-3) is phosphorylated by the cyclic AMP-regulated protein kinases from rabbit reticulocytes. eIF-3 is a large molecular weight complex which facilitates binding of the ternary complex containing met tRNA_f, GTP and initiation factor 2 to 40S ribosomal subunits. A single polypeptide with a molecular weight of 130,000 is modified. The phosphorylation is dependent upon the presence of cyclic AMP and is inhibited by the inhibitor protein diagnostic for cyclic AMP-regulated protein kinase. Assuming a molecular weight of 700,000 for eIF-3, one mole of phosphate is incorporated per mole of eIF-3. Thus the phosphorylation of two interacting components of the protein synthesizing system, 40S ribosomal subunits and eIF-3, is controlled by cyclic AMP.     

Insulin promoted decrease in the phosphorylation of protein synthesis initiation factor eIF-2

Insulin stimulates cellular protein synthesis in calf chondrocytes in suspension culture. This enhanced synthetic activity is seen in association with a decrease in phosphorylation of the α subunit of protein synthesis initiation factor eIF-2. [³²P] associated with the α subunit is reduced approximately 50% by insulin treatment of chondrocytes incubated in [³²P] containing media. Identical or closely located amino acids in the eIF-2 α subunit are phosphorylated by the chondrocyte kinase(s) and the rabbit reticulocyte hemin regulated kinase as indicated by comparative peptide fragment analysis of [³²P] labeled α subunits.     

Phosphorylation of initiation factor eIF-2 alpha, binding of mRNA to 48 S complexes, and its reutilization in initiation of protein synthesis

The formation of 80 S initiation complexes containing labeled viral mRNA was drastically inhibited when mRNA binding assays were carried out with reticulocyte lysate preincubated with double-stranded RNA (dsRNA). When the assays were analyzed by centrifugation on sucrose gradients, the mRNA incubated with lysate pretreated with dsRNA sedimented as a 48 S complex. Met-tRNA, GDP, and phosphorylated initiation factor eIF-2(alpha P) were shown to co-sediment with the 48 S complex. Therefore, the formation of this complex was attributed to the phosphorylation of eIF-2 alpha by a dsRNA-activated protein kinase. These observations suggested that mRNA could bind to a 40 S ribosomal subunit containing Met-tRNAf, GDP, and eIF-2(alpha P), but the joining of a 60 S ribosomal subunit was inhibited. When the 48 S complex was isolated and incubated with lysate without added dsRNA, the mRNA could form 80 S initiation complexes. The shift of mRNA from 48 S to 80 S complexes was also observed when the eIF-2 alpha kinase activity was inhibited by the addition of 2-aminopurine. This shift was quite slow, however, when compared to the rate of binding of free mRNA to 80 S initiation complexes. The 2-aminopurine was effective in reversing the inhibition of protein synthesis by dsRNA and in maintaining a linear rate of protein synthesis for 3 h in lysates. Without added 2-aminopurine, protein synthesis was inhibited after 90 min even in lysates supplemented with hemin and eIF-2(alpha P) was detected in these lysates. This finding indicated that eIF-2 alpha phosphorylation could be in part responsible for limiting the duration of protein synthesis in mammalian cell-free systems.     

Expression of mutant eukaryotic initiation factor 2 alpha subunit (eIF-2 alpha) reduces inhibition of guanine nucleotide exchange activity of eIF-2B mediated by eIF-2 alpha phosphorylation.

The inhibition of protein synthesis that occurs upon phosphorylation of the alpha subunit of eukaryotic initiation factor 2 (eIF-2 alpha) at serine 51 correlates with reduced guanine nucleotide exchange activity of eIF-2B in vivo and inhibition of eIF-2B activity in vitro, although it is not known if phosphorylation is the cause of the reduced eIF-2B activity in vivo. To characterize the importance of eIF-2 alpha phosphorylation in the regulation of eIF-2B activity, we studied the overexpression of mutant eIF-2 alpha subunits in which serine 48 or 51 was replaced by an alanine (48A or 51A mutant). Previous studies demonstrated that the 51A mutant was resistant to phosphorylation, whereas the 48A mutant was a substrate for phosphorylation. Additionally, expression of either mutant partially protected Chinese hamster ovary (CHO) cells from the inhibition of protein synthesis in response to heat shock treatment (P. Murtha-Riel, M. V. Davies, J. B. Scherer, S. Y. Choi, J. W. B. Hershey, and R. J. Kaufman, J. Biol. Chem. 268:12946-12951, 1993). In this study, we show that eIF-2B activity was inhibited in parental CHO cell extracts upon addition of purified reticulocyte heme-regulated inhibitor (HRI), an eIF-2 alpha kinase that phosphorylates Ser-51. Preincubation with purified HRI also reduced the eIF-2B activity in extracts from cells overexpressing wild-type eIF-2 alpha. In contrast, the eIF-2B activity was not readily inhibited in extracts from cells overexpressing either the eIF-2 alpha 48A or 51A mutant. In addition, eIF-2B activity was decreased in extracts prepared from heat-shocked cells overexpressing wild-type eIF-2 alpha, whereas the decrease in eIF-2B activity was less in heat-shocked cells overexpressing either mutant 48A or mutant 51A. While the phosphorylation at serine 51 in eIF-2 alpha impairs the eIF-2B activity, we propose that serine 48 acts to maintain a high affinity between phosphorylated eIF-2 alpha and eIF-2B, thereby inactivating eIF-2B activity. These findings support the hypothesis that phosphorylation of eIF-2 alpha inhibits protein synthesis directly through reducing eIF-2B activity and emphasize the importance of both serine 48 and serine 51 in the interaction with eIF-2B and regulation of eIF-2B activity.     

Site-specific dephosphorylation of smooth muscle myosin light chain kinase by protein phosphatases 1 and 2A.

Smooth muscle myosin light chain kinase is phosphorylated at two sites (A and B) by different protein kinases. Phosphorylation at site A increases the concentration of Ca2+/calmodulin required for kinase activation. Diphosphorylated myosin light chain kinase was used to determine the site-specificity of several forms of protein serine/threonine phosphatase. These phosphatases readily dephosphorylated myosin light chain kinase in vitro and displayed differing specificities for the two phosphorylation sites. Type 2A protein phosphatase specifically dephosphorylated site A, and binding of Ca2+/calmodulin to the kinase had no effect on dephosphorylation. The purified catalytic subunit of type 1 protein phosphatase dephosphorylated both sites in the absence of Ca2+/calmodulin but only dephosphorylated site A in the presence of Ca2+/calmodulin. A protein phosphatase fraction was prepared from smooth muscle actomyosin by extraction with 80 mM MgCl2. On the basis of sensitivity to okadaic acid and inhibitor 2, this activity was composed of multiple protein phosphatases including type 1 activity. This phosphatase fraction dephosphorylated both sites in the absence of Ca2+/calmodulin. However, dephosphorylation of both sites A and B was completely blocked in the presence of Ca2+/calmodulin. These results indicate that two phosphorylation sites of myosin light chain kinase are dephosphorylated by multiple protein serine/threonine phosphatases with unique catalytic specificities.     

Comparative analysis of phosphorylation of translational initiation and elongation factors by seven protein kinases

Four initiation factors (eIF-2, -3, -4B, and -4F), previously shown to be phosphorylated in vivo, are each phosphorylated to a significant extent in vitro (greater than 0.3 mol of phosphate/mol of factor) by at least three different protein kinases. An S6 kinase from liver, an active form of protease-activated kinase II which modifies the same sites on S6 as those phosphorylated in vivo in response to mitogens, phosphorylates the beta subunit of eIF-2, eIF-3 (p120-p130), eIF-4B, and eIF-4F (p220). The Ca2+, phospholipid-dependent protein kinase phosphorylates eIF-2 beta, eIF-3 (p170, p120-p130), eIF-4B, and eIF-4F (p220, p25). The cAMP-dependent protein kinase significantly modifies eIF-4B and, to a lesser extent, eIF-3 (p130). Casein kinase I incorporates phosphate only into eIF-4B, but to a limited extent. Casein kinase II phosphorylates eIF-2 beta, eIF-3 (p170, p120), and eIF-4B, while protease-activated kinase I modifies eIF-3 (p170, p120-p130), eIF-4B, and eIF-4F (p220). The mitogen-stimulated S6 kinase from 3T3-L1 cells, activated in response to insulin, does not phosphorylate any of the initiation factors. There is no significant incorporation of phosphate into eIF-2 alpha or -gamma, eIF-4A, eIF-4C, eIF-4D, EF-1, or EF-2 by any of the protein kinases examined. Phosphopeptide mapping of tryptic digests of the phosphorylated subunits shows that the individual protein kinases modify different sites. The sites phosphorylated in vitro reflect those modified in vivo as shown with eIF-4F in concomitant studies with reticulocytes treated with tumor-promoting phorbol ester (Morley, S.J., and Traugh, J. A. J. Biol. Chem., in press). Thus, we have identified multipotential protein kinases which modify four initiation factors phosphorylated in vivo and have shown that phosphorylation of these translational components can be coordinately regulated.     

Selective Dephosphorylation of the Subunits of Skeletal Muscle Calcium Channels by Purified Phosphoprotein Phosphatases

Abstract: Multiple sites on the α1 and β subunits of purified skeletal muscle calcium channels are phosphorylated by cyclic AMP-dependent protein kinase, resulting in three different tryptic phosphopeptides derived from each subunit. Phosphoprotein phosphatases dephosphorylated these sites selectively. Phosphoprotein phosphatase 1 (PP1) and phosphoprotein phosphatase 2A (PP2A) dephosphorylated both α1 and β subunits at similar rates, whereas calcineurin dephosphorylated β subunits preferentially. PP1 dephosphorylated phosphopeptides 1 and 2 of the α1 subunit more rapidly than phosphopeptide 3. In contrast, PP2A dephosphorylated phosphopeptide 3 of the α1 subunit preferentially. All three phosphoprotein phosphatases preferentially dephosphorylated phosphopeptide 1 of the β subunit and dephosphorylated phosphopeptides 2 and 3 more slowly. Mn²⁺ increased the rate and extent of dephosphorylation of all sites by calcineurin so that >80% dephosphorylation of both α1 and β sub-units was obtained. The results demonstrate selective dephosphorylation of different phosphorylation sites on the α1 and β subunits of skeletal muscle calcium channels by the three principal serine/threonine phosphoprotein phosphatases.     

Mutations in the GCD7 subunit of yeast guanine nucleotide exchange factor eIF-2B overcome the inhibitory effects of phosphorylated eIF-2 on translation initiation.

Phosphorylation of the alpha subunit of eukaryotic translation initiation factor 2 (eIF-2 alpha) impairs translation initiation by inhibiting the guanine nucleotide exchange factor for eIF-2, known as eIF-2B. In Saccharomyces cerevisiae, phosphorylation of eIF-2 alpha by the protein kinase GCN2 specifically stimulates translation of GCN4 mRNA in addition to reducing general protein synthesis. We isolated mutations in several unlinked genes that suppress the growth-inhibitory effect of eIF-2 alpha phosphorylation catalyzed by mutationally activated forms of GCN2. These suppressor mutations, affecting eIF-2 alpha and the essential subunits of eIF-2B encoded by GCD7 and GCD2, do not reduce the level of eIF-2 alpha phosphorylation in cells expressing the activated GCN2c kinase. Four GCD7 suppressors were shown to reduce the derepression of GCN4 translation in cells containing wild-type GCN2 under starvation conditions or in GCN2c strains. A fifth GCD7 allele, constructed in vitro by combining two of the GCD7 suppressors mutations, completely impaired the derepression of GCN4 translation, a phenotype characteristic of deletions in GCN1, GCN2, or GCN3. This double GCD7 mutation also completely suppressed the lethal effect of expressing the mammalian eIF-2 alpha kinase dsRNA-PK in yeast cells, showing that the translational machinery had been rendered completely insensitive to phosphorylated eIF-2. None of the GCD7 mutations had any detrimental effect on cell growth under nonstarvation conditions, suggesting that recycling of eIF-2 occurs efficiently in the suppressor strains. We propose that GCD7 and GCD2 play important roles in the regulatory interaction between eIF-2 and eIF-2B and that the suppressor mutations we isolated in these genes decrease the susceptibility of eIF-2B to the inhibitory effects of phosphorylated eIF-2 without impairing the essential catalytic function of eIF-2B in translation initiation.