Mechanism of translational control by partial phosphorylation of the alpha subunit of eukaryotic initiation factor 2

Correlation between the distribution of the reversing factor and eukaryotic initiation factor 2 in heme-deficient or double-stranded RNA-inhibited reticulocyte lysates

The recycling of eukaryotic initiation factor eIF-2 requires the exchange of GDP for GTP, in a reaction catalyzed by the reversing factor (RF). Recent studies have suggested that a 60 S ribosomal subunit-bound eIF-2.GDP complex is an intermediate in protein chain initiation. We have monitored the distribution of RF in heme-deficient and dsRNA-inhibited lysates by immunoblot analysis of sucrose gradient fractions and have compared the distribution with that of eIF-2(alpha-32P). RF and eIF-2(alpha P) were both found to be tightly associated with 60 S and 80 S ribosomes, as their distribution did not change in gradients containing up to 0.1 M K+. The association of eIF-2(alpha-32P) and RF with 60 S and 80 S ribosomes was enhanced in the presence of F-, indicating the presence of an endogenous ribosome-associated phosphatase activity which is capable of dephosphorylating eIF-2(alpha P) in the absence of F-. These observations are consistent with the hypothesis that under physiologic conditions, RF interacts with the 60 S-bound eIF-2.GDP complex to promote the dissociation of GDP from eIF-2 and the release of eIF-2 from the 60 S subunit as a complex with RF.

Translational control by influenza virus: suppression of the kinase that phosphorylates the alpha subunit of initiation factor eIF-2 and selective translation of influenza viral mRNAs

Selective translation of influenza viral mRNAs occurs after influenza virus superinfection of cells infected with the VAI RNA-negative adenovirus mutant dl331 (M. G. Katze, Y.-T. Chen, and R. M. Krug, Cell 37:483-490, 1984). Cell extracts from these doubly infected cells catalyze the initiation of essentially only influenza viral protein synthesis, reproducing the in vivo situation. This selective translation is correlated with a 5- to 10-fold suppression of the dl331-induced kinase that phosphorylates the alpha subunit of eucaryotic initiation factor eIF-2. This strongly suggests that influenza virus encodes a gene product that, analogous to the adenoviral VAI RNA, prevents the shutdown of overall protein synthesis caused by an eIF-2 alpha kinase turned on by viral infection. Adenoviral mRNA translation was restored to the extract from the doubly infected cells by the addition of the guanine nucleotide exchange factor eIF-2B, which is responsible for the normal recycling of eIF-2 during protein synthesis. This indicates that the residual kinase in the doubly infected cells leads to a limitation in functional (nonsequestered) eIF-2B and hence functional (GTP-containing) eIF-2 and that under these conditions influenza viral mRNAs are selectively translated over adenoviral mRNAs. Addition of double-stranded RNA to the extracts from these cells restored the eIF-2 alpha kinase to a level approaching that seen in extracts from cells infected with dl331 alone and caused the inhibition of influenza viral mRNA translation. This suggests that the putative influenza viral gene product acts against the double-stranded RNA activation of the kinase and indicates that influenza viral mRNA translation is also linked to the level of functional eIF-2. Our results thus indicate that a limitation in functional eIF-2 which causes a nonspecific reduction in the rate of initiation of protein synthesis results in the preferential translation of the better mRNAs (influenza viral mRNAs) at the expense of the poorer mRNAs (adenoviral mRNAs).

Formation of a translational inhibitor by interaction of phospholipid with the eukaryotic initiation factor 2

The polypeptide chain initiation factor 2 (eIF-2) binds phospholipid (PL) and becomes a potent inhibitor of translation in hemin-supplemented reticulocyte lysates. This binding is markedly reduced by prior treatment of eIF-2 with N-ethylmaleimide. Although PL is probably bound by all three eIF-2 subunits, our results suggest that the inhibitory molecule is produced by binding to the alpha subunit because, functionally, PL binding has the same effect on eIF-2 as alpha-subunit phosphorylation. This is suggested by the following findings. (i) Like translational inhibition due to heme deficiency, inhibition by small amounts of the eIF-2 X PL complex is prevented by small amounts of the GDP exchange factor (GEF). (ii) In the presence of Mg2+, the GEF-catalyzed formation of a ternary complex (eIF-2 X GTP X Met-tRNAi in which Met-tRNAi is the eukaryotic initiator methionyl tRNA) is inhibited by eIF-2 X PL just as well as by eIF-2 alpha-subunit phosphorylation. (iii) Also in the presence of Mg2+, GEF is unable to catalyze the exchange of free GDP with eIF-2 X PL-bound GDP, as it fails to catalyze the exchange of free GDP with GDP that is bound to alpha-subunit-phosphorylated eIF-2. These observations suggest that, like alpha-subunit-phosphorylated eIF-2, eIF-2 X PL traps GEF in a nondissociable eIF-2 X PL X GEF complex, whereby GEF is no longer able to catalyze ternary complex formation and initiation is inhibited.

Translational control by influenza virus: suppression of the kinase that phosphorylates the alpha subunit of initiation factor eIF-2 and selective translation of influenza viral mRNAs

Selective translation of influenza viral mRNAs occurs after influenza virus superinfection of cells infected with the VAI RNA-negative adenovirus mutant dl331 (M. G. Katze, Y.-T. Chen, and R. M. Krug, Cell 37:483-490, 1984). Cell extracts from these doubly infected cells catalyze the initiation of essentially only influenza viral protein synthesis, reproducing the in vivo situation. This selective translation is correlated with a 5- to 10-fold suppression of the dl331-induced kinase that phosphorylates the alpha subunit of eucaryotic initiation factor eIF-2. This strongly suggests that influenza virus encodes a gene product that, analogous to the adenoviral VAI RNA, prevents the shutdown of overall protein synthesis caused by an eIF-2 alpha kinase turned on by viral infection. Adenoviral mRNA translation was restored to the extract from the doubly infected cells by the addition of the guanine nucleotide exchange factor eIF-2B, which is responsible for the normal recycling of eIF-2 during protein synthesis. This indicates that the residual kinase in the doubly infected cells leads to a limitation in functional (nonsequestered) eIF-2B and hence functional (GTP-containing) eIF-2 and that under these conditions influenza viral mRNAs are selectively translated over adenoviral mRNAs. Addition of double-stranded RNA to the extracts from these cells restored the eIF-2 alpha kinase to a level approaching that seen in extracts from cells infected with dl331 alone and caused the inhibition of influenza viral mRNA translation. This suggests that the putative influenza viral gene product acts against the double-stranded RNA activation of the kinase and indicates that influenza viral mRNA translation is also linked to the level of functional eIF-2. Our results thus indicate that a limitation in functional eIF-2 which causes a nonspecific reduction in the rate of initiation of protein synthesis results in the preferential translation of the better mRNAs (influenza viral mRNAs) at the expense of the poorer mRNAs (adenoviral mRNAs).

Regulation of protein synthesis in virus-infected animal cells

This chapter summarizes the structural features that govern the translation of viral mRNAs: where the synthesis of a protein starts and ends, how many proteins can be produced from one mRNA, and how efficiently. It focuses on the interplay between viral and cellular mRNAs and the translational machinery. That interplay, together with the intrinsic structure of viral mRNAs, determines the patterns of translation in infected cells. It also points out some possibilities for translational regulation that can only be glimpsed at present, but are likely to come into focus in the future. The mechanism of selecting the initiation site for protein synthesis appears to follow a single formula. The translational machinery displays a certain flexibility that is exploited more frequently by viral than by cellular mRNAs. Although some of the parameters that determine efficiency have been identified, how efficiently a given mRNA will be translated cannot be predicted by summing the known parameters.

Functional heterogeneity of GEF-free initiation factor 2 purified from suckling and adult rat brain

The functional behavior of initiation factor 2 was studied in purified preparations from the brains of suckling (4-12-day-old) and adult (60-day-old) rats. Adult eIF2 has lower GDP and GTP affinity than suckling eIF2, even in the presence of a large excess of GTP, whereas suckling eIF2 has a lower capacity to bind GTP. Since these two factors are free of guanine nucleotide exchange factor (GEF), and ribosomal fractions show an age-dependent difference in GEF activity, the observed functional heterogeneity may be due to a different ratio in eIF2 species (eIF2-GDP, eIF2(alpha P)).

Double-stranded RNA-dependent protein kinase and 2-5A system are both activated in interferon-treated, encephalomyocarditis virus-infected HeLa cells

Activation of the interferon-inducible, double-stranded RNA-dependent protein kinase was monitored in monolayer cultures of control and interferon-treated HeLa cells infected with encephalomyocarditis virus. The extent of phosphorylation in the intact cell of the alpha-subunit of eucaryotic protein synthesis initiation factor eIF2 by the kinase was determined for the first time in this type of system, using a two-dimensional immunoblot technique. Virus protein synthesis and the kinetics of activation of the ppp(A2'p)nA (n greater than or equal to 2) system were analyzed in parallel. Enhanced phosphorylation of eIF2-alpha was obvious at 9 h and increased by 12 h postinfection. ppp(A2'p)nA and ppp(A2'p)nA-mediated rRNA cleavage were observed from 6 h. No viral protein synthesis was detected in cells in which a general inhibition of protein synthesis developed with time. It can be concluded that both the kinase and ppp(A2'p)nA system are active in interferon-treated, encephalomyocarditis virus-infected HeLa cells.

Mechanism of activation of the heme-stabilized translational inhibitor of reticulocyte lysates by calcium ions and phospholipid

We have reported previously that calcium ions and phospholipid activate the heme-stabilized proinhibitor form (pro-HCI) of the heme-controlled translational inhibitor (HCI) in reticulocyte lysates and promote the first step of the reaction pro-HCI in equilibrium reversible HCI----irreversible HCI. This suggested the possible involvement of a Ca2+/phospholipid-dependent protein kinase (protein kinase C) in the activation. However, further investigation revealed, among other things, that polyunsaturated fatty acids (e.g., arachidonic acid) were as effective as Ca2+/phospholipid in promoting translational inhibition and phosphorylation of the alpha subunit of the chain-initiation factor eIF-2 and, moreover, HCI activation could be prevented or reversed in either case by NADPH-generating systems or by dithiols. Our results suggest that pro-HCI is activated by lipoperoxides produced in reticulocyte lysates from either phospholipid or polyunsaturated fatty acids; the presence of Ca2+ is required in the former but not in the latter case. The reversible activation of HCI by Ca2+ and phospholipid might suggest a possible modulatory role of Ca2+ in translational control.

Inhibition of binding to initiation complexes of nascent reovirus mRNA by double-stranded RNA-dependent protein kinase

A coupled, cell-free system for the transcription and translation of reovirus mRNA was developed. Activated reovirions were incubated with reticulocyte lysate and an appropriate energy mix. Active transcription was obtained, but protein synthesis was inhibited after a short lag even by low concentrations of reovirions. This inhibition was abolished by the addition of the kinase inhibitor 2-aminopurine. With this addition, the synthesis of viral proteins could be detected in reaction mixtures containing nuclease-treated reticulocyte lysate. The binding of nascent reovirus mRNA to 80S initiation complexes measured after 2 min of incubation was greatly inhibited, whereas the binding of cellular mRNA added to the same reaction mixtures for the next 2 min was not inhibited. The inhibition of reovirus mRNA binding could not be explained by the synthesis of defective templates, since most of the mRNA could be bound to 80S complexes after the addition of 2-aminopurine. These results indicate that the binding of nascent reovirus mRNA was preferentially inhibited by a protein kinase. Reovirions preincubated with reticulocyte lysate could phosphorylate initiation factor eIF-2. This phosphorylation was inhibited by the addition of high concentrations of double-stranded RNA, which are inhibitory for the eIF-2 kinase present in elevated levels in reticulocyte lysate and in interferon-treated cells. These results indicate that the translation of viral mRNA may be preferentially inhibited in interferon-treated cells by the eIF-2 kinase activated by viral transcriptional complexes containing double-stranded RNA.

Translational control by adenovirus: lack of virus-associated RNAI during adenovirus infection results in phosphorylation of initiation factor eIF-2 and inhibition of protein synthesis

The dl331 mutant of adenovirus serotype 5 fails to produce virus-associated (VA) RNAI, and cells infected with this mutant do not synthesize proteins efficiently at late times in infection. The translational defect occurs at the level of polypeptide chain initiation, and cell-free extracts prepared from dl331-infected cells exhibit the defect observed in vivo. Addition of either eukaryotic initiation factor 2 (eIF-2) or guanine nucleotide exchange factor (GEF) to these cell-free extracts restores translational activity, with GEF functioning more efficiently in this regard. These results suggest that cells infected with the dl331 mutant develop a translational block at the level of GEF-catalyzed guanine nucleotide exchange and that this block is most likely established through phosphorylation of the alpha subunit of eIF-2. In the present investigation we show that endogenous HeLa cell GEF activity is significantly reduced in cells infected with the dl331 mutant. Further, in contrast to cells infected with wild-type serotype 2 adenovirus, dl331-infected cells contain increased eIF-2 alpha kinase activity. These results indicate that VA RNAI plays a role in suppressing eIF-2 alpha kinase activity during adenovirus infection of HeLa cells.

Evaluation and significance of kinetic parameters governing function of protein synthesis initiation factors eIF-2 and eIF-2B

Published data dealing with the formation of the ternary complex eIF-2 X GTP X met-tRNAi involved in eukaryotic initiation have been evaluated to calculate the expected inhibition by GDP and the role of eIF-2B in limiting this inhibition. It is concluded that cellular levels of GDP are unlikely seriously to inhibit ternary complex formation if the reaction can proceed to equilibrium. However, derivation of 'on' and 'off' rates for the interaction of GTP and GDP with eIF-2 demonstrates that these are too slow in the absence of eIF-2B to support active protein synthesis, particularly if eIF-2 is released from ribosomes as eIF-2 X GDP. Whilst eIF-2 X GDP and eIF-2 X GTP appear to dissociate equally slowly, it is concluded that GDP binds to eIF-2 100-times faster than GTP. Addition of eIF-2B has the effect of raising k-1 for both GDP and GTP several hundred-fold and k+1 50- and 7000-fold, respectively. Thus, a kinetic block can be relieved even if there is no change in the thermodynamic state. Phosphorylation of the alpha-subunit of eIF-2 appears to affect only those parameters influenced by eIF-2B. The reported rescue of inhibited lysates by addition of 1 mM GTP is not by mass action but by some other mechanism. Consideration of the kinetic parameters favours the formation of a ternary complex of eIF-2 X eIF-2B X GDP en route to eIF-2 X GTP as opposed to displacement of GDP from eIF-2 X GDP by eIF-2B.

Protein phosphorylation and translational control in reticulocytes: activation of the heme-controlled translational inhibitor by calcium ions and phospholipid

The synthesis of globin, the major protein synthesized by reticulocytes, requires the presence of heme, the prosthetic group of hemoglobin. The absence of heme leads to the activation of a nucleotide-independent protein kinase that phosphorylates the alpha subunit of the chain initiation factor eIF-2. This modification interferes with the catalytic function of eIF-2 in protein synthesis initiation. Recent progress in our understanding of the molecular mechanism of this inhibition is briefly reviewed. The same phosphorylation is catalyzed by a different enzyme (DAI) which, while constitutive in reticulocytes, is induced by interferon in other cells. This enzyme is activated by low concentrations of double-stranded RNA in conjunction with ATP. The mechanisms of activation of these enzymes are still poorly understood. HCI is believed to form an inactive complex with heme and become active when the heme is removed by hemoglobin formation. The proinhibitor form of HCI (proHCI) is unstable in vitro and, even in the presence of heme, is irreversibly inactivated by SH-binding reagents, alkaline pH, slightly elevated temperatures, or high hydrostatic pressure. In hemin-supplemented reticulocyte lysates proHCI can also be reversibly activated by oxidized glutathione (GSSG) or NADPH depletion as well as by polyunsaturated fatty acids and by Ca2+-phospholipid. The mechanism of activation of HCI by GSSG has not been clarified although it appears to involve oxidation of proHCI SH groups to disulfides. Like activation by GSSG, the activation of HCI by polyunsaturated fatty acids and by Ca2+-phospholipid also appears to be largely due to oxidation of some of the enzyme's SH groups. There thus appear to be two fully independent mechanisms of HCI activation in reticulocyte lysates, one involving heme deficiency, the other involving oxidation of proHCI SH groups. The latter, but not the former, can be prevented or reversed by NADPH generators or dithiols. ProHCI appears to be maintained in the reduced, inactive state by a system involving NADPH, thioredoxin, and thioredoxin reductase.

Regulation of protein synthesis in eukaryotes. Mode of action of eRF, an eIF-2-recycling factor from rabbit reticulocytes involved in GDP/GTP exchange

The rate of initiation of protein synthesis appears to be controlled at the level of recycling of eIF-2. In this process a new factor, designated eRF, plays an important role. The factor has been purified from the post-ribosomal supernatant and has been called formerly anti-HRI and anti-inhibitor [Amesz, H., Goumans, H., Haubrich-Morree, Th., Voorma, H.O., and Benne, R. (1979) Eur. J. Biochem. 98, 513-520]. Its effect on the initiation of protein synthesis has been studied in several assays: a small but distinct effect is found in the assay for the formation of a ternary complex between eIF-2, GTP and Met-tRNA; a 4-5-fold stimulation is obtained in assays for 40S preinitiation complex formation and in the methionyl-puromycin reaction. In the latter assay a catalytic use of eIF-2 occurs provided that eRF is present. eRF forms a complex with eIF-2 which results in a decrease of the affinity of eIF-2 for GDP, giving it the properties of a GDP/GTP exchange factor. The model stresses the catalytic use of eIF-2 in initiation provided that conditions are met for GDP/GTP exchange by a transient complex formation between eIF-2 and eRF. On the other hand, it is shown that phosphorylation of eIF-2 by the hemin-regulated inhibitor (HRI) abolishes the recycling of eIF-2, by the formation of another stable complex comprising eIF-2 alpha P, GDP and eRF.

Distribution of reversing factor in reticulocyte lysates during active protein synthesis and on inhibition by heme deprivation or double-stranded RNA

We have recently shown a direct correlation between protein synthetic activity and the function of reversing factor (RF) as a catalyst of GDP-GTP exchange in whole reticulocyte lysates under normal conditions and on inhibition of protein synthesis by heme deficiency, double-stranded RNA, or oxidized glutathione. In this paper we report that RF is detectable as a nonribosomal complex with eukaryotic initiation factor 2 phosphorylated in its alpha subunit [eIF-2(alpha P)] in whole lysates inhibited by heme deprivation or by double-stranded RNA. The complex contains no unphosphorylated eIF-2 alpha, and the GDP present is freely dissociable. All nonribosomal eIF-2(alpha P) is complexed with RF in fully inhibited lysates; we have not detected free eIF-2(alpha P). RF in this [RF X eIF-2(alpha P)] complex is unavailable to catalyze the release of GDP from eIF-2-GDP. Dephosphorylation of eIF-2(alpha P) present in nonribosomal fractions releases active RF, which is able to carry out its normal guanine nucleotide exchange function.