Criteria for establishment of the biological significance of ribosomal protein phosphorylation

Regulation of an epitope-tagged recombinant Rsk-1 S6 kinase by phorbol ester and erk/MAP kinase

Phorbol ester tumor promoters (TPA) activate the endogenous erk/MAP kinases and Rsk S6 kinases but not the p70S6 kinase in COS cells. DNA sequences encoding the rat Rsk-1 S6 kinase (homologous to Xenopus rsk alpha), modified by insertion of a peptide epitope at the polypeptide aminoterminus, were expressed transiently in COS cells. TPA stimulates the 40S and peptide kinase activity of the recombinant epitope-tagged Rsk-1, as well as the extent of Rsk-1 autophosphorylation in vitro (32P-Ser >> 32P-Thr). Indications that the conformation of the recombinant Rsk-1 polypeptide is substantially changed after activation by TPA in situ include a retarded mobility of the Rsk-1 polypeptide on SDS-PAGE and the appearance of new 32P-peptides during autophosphorylation in vitro. All these features of the TPA-activated Rsk-1 S6 kinase are abolished by dephosphorylation of the kinase in vitro with Ser/Thr phosphatase-2A. TPA increases 32P incorporation into recombinant Rsk-1 by 2-3-fold (32P-Ser >> 32P-Thr). Peptide mapping exhibits a single major 32P-peptide in Rsk-1 isolated from unstimulated cells and 10-12 additional 32P peptides after TPA treatment in situ. Phosphorylation of basal or phosphatase-2A-treated recombinant Rsk-1 in vitro with erk2/MAP kinase increases Rsk-1 40S kinase, peptide kinase, and autophosphorylating activity, retards migration of Rsk-1 polypeptides on SDS-PAGE, and generates new sites of Rsk-1 autophosphorylation in vitro. By contrast, TPA-activated Rsk-1 is not altered in these properties by autophosphorylation in vitro. By contrast, TPA-activated Rsk-1 is not altered in these properties by phosphorylation in vitro with erk2/MAP kinase. Activation of Rsk-1 in situ with TPA diminishes by over 90% the extent of Rsk-1 phosphorylation achieved in vitro by erk2/MAP kinase, as compared to the parallel phosphorylation of a phosphatase-2A-treated Rsk-1; basal Rsk-1 is intermediate. Peptide maps of phosphatase-2A-treated Rsk-1 after phosphorylation in vitro with erk2/MAP kinase exhibit 32P-peptides that comigrate with nearly all of the 32P-peptides present in TPA-activated-32P Rsk-1 labeled in situ, plus several 32P-peptides characteristic of Rsk-1 autophosphorylation in vitro.(ABSTRACT TRUNCATED AT 400 WORDS)

Rapamycin-induced inhibition of the 70-kilodalton S6 protein kinase

The immunosuppressant rapamycin inhibited proliferation of the H4IIEC hepatoma cell line. Rapamycin, but not its structural analog FK506, also inhibited the basal and insulin-stimulated activity of the p70 ribosomal protein S6 kinase. By contrast, insulin stimulation of the p85 Rsk S6 kinase and mitogen-activated protein (MAP) kinase activity were unaffected by drug. Rapamycin treatment of COS cells transfected with recombinant p70 S6 kinase completely inhibited the appearance of the hyperphosphorylated form of p70 S6 kinase concomitant with the inhibition of enzyme activity toward 40S subunits. Thus, rapamycin inhibits a signal transduction element that is necessary for the activation of p70 S6 kinase and mitogenesis but unnecessary for activation of p85 Rsk S6 kinase or MAP kinase.

Cloning and expression of two human p70 S6 kinase polypeptides differing only at their amino termini

Two classes of human cDNA encoding the insulin/mitogen-activated p70 S6 kinase have been isolated; the two classes differ only in the 5' region, such that the longer polypeptide (p70 S6 kinase alpha I; calculated Mr 58,946) consists of 525 amino acids, of which the last 502 residues are identical in sequence to the entire polypeptides encoded by the second cDNA (p70 S6 kinase alpha II; calculated Mr 56,153). Both p70 S6 kinase polypeptides predicted by these cDNAs are present in p70 S6 kinase purified from rat liver, and each is thus expressed in vivo. Moreover, both polypeptides are expressed from a single mRNA transcribed from the (longer) p70 S6 kinase alpha I cDNA through the utilization of different translational start sites. Although the two p70 S6 kinase polypeptides differ by only 23 amino acid residues, the slightly longer alpha I polypeptide exhibits anomalously slow mobility on sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), migrating at an apparent Mr of 90,000 probably because of the presence of six consecutive Arg residues immediately following the initiator methionine. Transient expression of p70 alpha I and alpha II S6 kinase cDNA in COS cells results in a 2.5- to 4-fold increase in overall S6 kinase activity. Upon immunoblotting, the recombinant p70 polypeptides appear as a closely spaced ladder of four to five bands between 65 and 70 kDa (alpha II) and 85 and 90 kDa (alpha I). Transfection with the alpha II cDNA yields only the smaller set of bands, while transfection with the alpha I cDNA generates both sets of bands. Mutation of Met-24 in the alpha I cDNA to Leu or Thr suppresses synthesis of the alpha II polypeptides. Only the p70 alpha I and alpha II polypeptides of slowest mobility on SDS-PAGE comigrate with the 70- and 90-kDa proteins observed in purified rat liver S6 kinase. Moreover, it is the recombinant p70 polypeptides of slowest mobility that coelute with S6 kinase activity on anion-exchange chromatography. The slower mobility and higher enzymatic activity of these p70 proteins is due to Ser/Thr phosphorylation, inasmuch as treatment with phosphatase 2A inactivates kinase activity and increases the mobility of the bands on SDS-PAGE in an okadaic acid-sensitive manner. Thus, the recombinant p70 S6 kinase undergoes multiple phosphorylation and partial activation in COS cells. Acquisition of S6 protein kinase catalytic function, however, is apparently restricted to the most extensively phosphorylated recombinant polypeptides.

Cloning and expression of two human p70 S6 kinase polypeptides differing only at their amino termini

Two classes of human cDNA encoding the insulin/mitogen-activated p70 S6 kinase have been isolated; the two classes differ only in the 5' region, such that the longer polypeptide (p70 S6 kinase alpha I; calculated Mr 58,946) consists of 525 amino acids, of which the last 502 residues are identical in sequence to the entire polypeptides encoded by the second cDNA (p70 S6 kinase alpha II; calculated Mr 56,153). Both p70 S6 kinase polypeptides predicted by these cDNAs are present in p70 S6 kinase purified from rat liver, and each is thus expressed in vivo. Moreover, both polypeptides are expressed from a single mRNA transcribed from the (longer) p70 S6 kinase alpha I cDNA through the utilization of different translational start sites. Although the two p70 S6 kinase polypeptides differ by only 23 amino acid residues, the slightly longer alpha I polypeptide exhibits anomalously slow mobility on sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), migrating at an apparent Mr of 90,000 probably because of the presence of six consecutive Arg residues immediately following the initiator methionine. Transient expression of p70 alpha I and alpha II S6 kinase cDNA in COS cells results in a 2.5- to 4-fold increase in overall S6 kinase activity. Upon immunoblotting, the recombinant p70 polypeptides appear as a closely spaced ladder of four to five bands between 65 and 70 kDa (alpha II) and 85 and 90 kDa (alpha I). Transfection with the alpha II cDNA yields only the smaller set of bands, while transfection with the alpha I cDNA generates both sets of bands. Mutation of Met-24 in the alpha I cDNA to Leu or Thr suppresses synthesis of the alpha II polypeptides. Only the p70 alpha I and alpha II polypeptides of slowest mobility on SDS-PAGE comigrate with the 70- and 90-kDa proteins observed in purified rat liver S6 kinase. Moreover, it is the recombinant p70 polypeptides of slowest mobility that coelute with S6 kinase activity on anion-exchange chromatography. The slower mobility and higher enzymatic activity of these p70 proteins is due to Ser/Thr phosphorylation, inasmuch as treatment with phosphatase 2A inactivates kinase activity and increases the mobility of the bands on SDS-PAGE in an okadaic acid-sensitive manner. Thus, the recombinant p70 S6 kinase undergoes multiple phosphorylation and partial activation in COS cells. Acquisition of S6 protein kinase catalytic function, however, is apparently restricted to the most extensively phosphorylated recombinant polypeptides.

Herpes simplex virus type-1-induced stimulation of ribosomal protein S6 phosphorylation is inhibited in neomycin-treated human epidermoid carcinoma 2 cells and in ras-transformed cells

Neomycin, an inhibitor of inositol phospholipid turnover, prevents Herpes-simplex-virus-type-1 (HSV-1)-induced stimulation of ribosomal protein S6 phosphorylation, but does not impair the S6 phosphorylation induced by serum. Long-term treatment with phorbol 12-myristate 13-acetate, which down-regulates protein kinase C activity, does not inhibit virus-induced S6 phosphorylation. In ras-transformed cells, S6 phosphorylation is not stimulated after HSV-1 infection. These results suggest that activation of the inositol phospholipid pathway is involved in the HSV-1-induced stimulation of S6 phosphorylation. However, protein kinase C activation does not appear to be necessary for HSV-1-induced S6 phosphorylation.

Molecular structure of a major insulin/mitogen-activated 70-kDa S6 protein kinase

The molecular structure of a rat hepatoma 70-kDa insulin/mitogen-stimulated S6 protein kinase, obtained by molecular cloning, is compared to that of a rat homolog of the 85-kDa Xenopus S6 protein kinase alpha; both kinases were cloned from H4 hepatoma cDNA libraries. The 70-kDa S6 kinase (calculated molecular mass of 59,186 Da) exhibits a single catalytic domain that is most closely related in amino acid sequence (56% identity) to the amino-terminal, kinase C-like domain of the rat p85 S6 kinase (calculated molecular mass of 82,695 Da); strong similarity extends through a further 67 residues carboxyl-terminal to the catalytic domain (40% identity), corresponding to a region also conserved among the kinase C family. Outside of this segment of approximately 330 amino acids, the structures of the p70 and p85 S6 kinases diverge substantially. The p70 S6 kinase is known to be activated through serine/threonine phosphorylation by unidentified insulin/mitogen-activated protein kinases. A model for the regulation of p70 S6 protein kinase activity is proposed wherein the low activity of the unphosphorylated enzyme results from the binding of a basic, inhibitory pseudosubstrate site (located carboxyl-terminal to the extended catalytic domain) to an acidic substrate binding region (located amino-terminal to the catalytic domain); substrate binding is thereby prevented. S6 kinase activation requires displacement of this inhibitory segment, which is proposed to occur consequent to its multiple phosphorylation. The putative autoinhibitory segment contains several serine and threonine residues, each followed directly by a proline residue. This motif may prevent autophosphorylation but permit transphosphorylation; two of these serine residues reside in a maturation promoting factor (MPF)/cdc-2 consensus motif. Thus, hormonal regulation of S6 kinase may involve the action of MPF/cdc-2 or protein kinases with related substrate specificity.

Ribosome and protein synthesis modifications after infection of human epidermoid carcinoma cells with herpes simplex virus type 1

Modifications of ribosomes have been investigated in human epidermoid carcinoma-2 cells at different stages of herpes simplex virus type 1 infection. Very early in infection, there is an increase in ribosomal protein S6 phosphorylation even in the absence of serum. The same result is obtained in the presence of actinomycin D. At early infection time, ribosomal proteins S2, S3a and Sa are newly phosphorylated. At early and early-late times, three phosphorylated non-ribosomal proteins (v1, v2 and v3) are differently associated temporally to ribosomes. Analyses of proteins extracted from 40S subunits, 80S ribosomes and polysomes show that v1 and v2 are distributed differently among the different ribosomal populations. S6 phosphopeptides were found to be identical after serum stimulation and after viral infection. In every case phosphoserine and phosphothreonine were identified in S6. Only phosphoserine was found in other phosphorylated proteins. Our results indicate that herpes simplex virus type 1 is able to modify pre-existing ribosomes: (i) by stimulating a pre-existing kinase for S6 phosphorylation even in the absence of serum and of viral genome expression; (ii) by inducing new specific kinase activity(ies); and (iii) by association of new, phosphorylated proteins to ribosomes. These ribosomal modifications are correlated with changes in protein synthesis, as shown by two-dimensional electrophoretic analyses of newly synthesized 35S-labelled proteins.

Phosphorylation of the Saccharomyces cerevisiae equivalent of ribosomal protein S6 has no detectable effect on growth

The phosphorylation of mammalian ribosomal protein S6 is affected by a variety of agents, including growth factors and tumor promoters, as well as by expressed oncogenes. Its potential role in the regulation of protein synthesis has been the object of much study. We have developed strains of Saccharomyces cerevisiae in which the phosphorylatable serines of the equivalent ribosomal protein (S10) were converted to alanines by site-directed mutagenesis. The S10 of such cells is not phosphorylated. Comparison of these cells with the parental cells, whose genomes differ by only six nucleotides, revealed no differences in the lag phase or logarithmic phase of a growth cycle, in growth on different carbon sources, in sporulation, or in sensitivity to heat shock. We conclude that in S. cerevisiae the phosphorylation of ribosomal protein S10 may play no role in regulating the synthesis of proteins. This conclusion leads one to ask whether certain protein phosphorylations are simply the adventitious, if easily observable, result of the imperfect specificity of one or another protein kinase.

Phosphorylation of the Saccharomyces cerevisiae equivalent of ribosomal protein S6 has no detectable effect on growth

The phosphorylation of mammalian ribosomal protein S6 is affected by a variety of agents, including growth factors and tumor promoters, as well as by expressed oncogenes. Its potential role in the regulation of protein synthesis has been the object of much study. We have developed strains of Saccharomyces cerevisiae in which the phosphorylatable serines of the equivalent ribosomal protein (S10) were converted to alanines by site-directed mutagenesis. The S10 of such cells is not phosphorylated. Comparison of these cells with the parental cells, whose genomes differ by only six nucleotides, revealed no differences in the lag phase or logarithmic phase of a growth cycle, in growth on different carbon sources, in sporulation, or in sensitivity to heat shock. We conclude that in S. cerevisiae the phosphorylation of ribosomal protein S10 may play no role in regulating the synthesis of proteins. This conclusion leads one to ask whether certain protein phosphorylations are simply the adventitious, if easily observable, result of the imperfect specificity of one or another protein kinase.

Starvation-induced alterations of ribosomal protein phosphorylation in Bombyx mori L. Evidence for different phosphorylation kinetics in free and membrane-bound ribosomes

In the posterior silk gland of Bombyx mori, ribosomal protein S1, homologous to S6 in mammals, is partially phosphorylated in a normally fed animal. Before the first meal of the fifth larval instar, S1 is completely dephosphorylated. Likewise, starvation induces rapid dephosphorylation of the protein in both free and membrane-bound ribosomes. Upon refeeding after 48 h of starvation, S1 becomes phosphorylated again, first on membrane-bound ribosomes, then on free ribosomes, with a lag time of about 3 h. Following 48 h of refeeding, the most highly phosphorylated form of S1 predominates in both populations of ribosomes. These variations in phosphorylation are correlated with the level of protein synthesis in the posterior silk gland, 70% of the ribosomes occurring in polysomes upon feeding and only 30% upon starvation [Prudhomme, J.-C. & Couble, P. (1979) Biochimie (Paris) 61, 215-227]. After in vivo 32P labelling, the phosphopeptides of S1 from free and membrane-bound ribosomes were found to be identical and phosphoserine (only) was found in each S1. These results suggest the involvement of S1 phosphorylation in the regulation of protein synthesis at the translational level and the existence of at least two different pathways controlling this phosphorylation: one for the free ribosomes, the other for the membrane-bound ribosomes.

Changes in phosphoprotein pattern in Schizosaccharomyces pombe

A variety of evidence suggests that protein phosphorylation (pp) may be important in cell-cycle control. Phosphorylated proteins from S. pombe have been examined for phosphorylation changes under several conditions: known triggers of the division control (low nitrogen, low phosphate), cell size mutants (WEE1 and CDC2 alleles) and cell cycle mutants (CDC2, CDC10, CDC17, CDC25 alleles). Three major phosphorylated proteins (pp38, pp45 and pp54) showed the greatest response to nutritional shifts. The changes in the phosphorylated states of these proteins correlated with growth rate. Some phosphorylations (e.g. pp53) occurred transiently following a stimulus to cell division suggesting a possible involvement with the division mechanism. An allele-specific alteration of charge was noted for pp45 suggesting that this protein is the product of the CDC2 gene. The wee1-6 phosphoprotein pattern is similar to wild-type indicating that this mutant cell line accurately senses its nutritional environment and that the mutation likely affects the transfer of this information to the division control. Cells blocked by various temperature-sensitive cell cycle mutants did not show an alteration of phosphoprotein pattern.

Ribosome-associated cyclic nucleotide-independent protein kinase of Artemia salina cryptobiotic gastrulae

An extra-ribosomal cAMP-independent protein kinase from cryptobiotic embryos of Artemia salina has been purified to near homogeneity by gel filtration on Bio-Gel A-0.5 m, ion-exchange chromatography on DEAE-cellulose and phosphocellulose P11 and affinity chromatography on casein-Sepharose 4B and ATP-agarose. The enzymatic activity has a broad optimum at pH 7-8. Maximal activity is obtained in the presence of 5-6 mM MgCl2. The activity is inhibited by Mn2+, Ca2+ and K+. The enzyme has an Mr of 127 000, utilizes both ATP and GTP as phosphoryl donors and is completely inhibited by heparin and poly(L-glutamic acid). According to its properties, the enzyme can be classified as a casein kinase type II. Although the enzyme is associated with ribosomes, ribosomal proteins are not among the main substrates. The kinase is able to phosphorylate both the alpha and the beta subunits of initiation factor eIF2 using ATP or GTP as phosphoryl donors. The function of phosphorylation in the initiation of protein synthesis is discussed.

Phosphorylation of ribosomal and ribosome-associated proteins in isoproterenol-induced cardiac hypertrophy

Cardiac hypertrophy in adult rabbits was induced by subcutaneous injection of isoproterenol. The rate of [3H]leucine incorporation into acid insoluble material was increased and the extent of [32P]phosphate incorporation into several ribosomal proteins was altered. Specifically, a ribosomal protein with a molecular weight of 32,000 from the 40S ribosomal subunit showed a five-fold increase in phosphate incorporation in the hypertrophic heart whereas a protein with a molecular weight of 28,000 from the 60S subunit showed a four-fold decrease. Phosphorylation of ribosome-associated proteins, which could be removed from ribosomes with 0.72 M KCl, was also changed in the hypertrophic hearts. Six major phosphoproteins (with molecular weights 62,000, 49,000, 36,000, 30,000, 20,000 and 12,000) were detected in both the normal and the hypertrophic hearts. Phosphorylation of the 62 K and the 49 K protein was increased by two- and three-fold, respectively, in the hypertrophic hearts, whereas phosphorylation of the 36 K and the 30 K protein decreased by two-fold. The level of phosphorylation of the 20 K and the 12 K protein was not significantly changed in hypertrophic hearts.

On the role of energy metabolism in Neurospora circadian clock function

Neurospora crassa (bdA) mycelia were kept in liquid culture. Without rhythmic conidiation the levels of adenine nucleotides undergo circadian changes in constant darkness. Maxima occur 12-17 hr and 33-35 hr after initiation of the rhythm, i.e., at CT 0-6 hr. Pulses of metabolic inhibitors such as vanadate (Na3Vo4), molybdate (Na2MoO4 : 2 H2O), N-ethylmaleimide (NEM), azide (NaN3), cyanide (NaCN) and oligomycin phase shift the circadian conidiation rhythm of Neurospora crassa. Maximal advance phase shifts are observed at about CT 6 with all inhibitors. Pulses of N,N'dicyclohexylcarbodiimide (DCCD) and light phase shift the conidiation rhythm following a phase response curve different from those of the other agents (maximal advance at about CT 18-24). The phase shifts with DCCD and light are significantly larger in the wild type compared to the mitochrondrial mutant poky. Such differences are not found in PRCs of the protein synthesis inhibitor cycloheximide. [31P] NMR spectra of wild type Neurospora crassa and the clock mutants frq 1 and frq 7 which differ in their circadian period lengths did not reveal differences in the concentrations of adenine nucleotides, pyridine nucleotides or sugar phosphates. Starvation causes drastic changes of the levels of adenine nucleotides, phosphate and mobile polyphosphate without effecting phase or period length of the circadian rhythm.

Phosphorylation of ribosomal protein S6 in the aquatic fungus Blastocladiella emersonii

The changes in the degree of phosphorylation of ribosomal protein S6 during the life cycle of the aquatic fungus Blastocladiella emersonii were analyzed by two-dimensional gel electrophoresis. Three phosphorylated derivatives of S6 are present throughout the entire life cycle. However, under certain germination conditions, more highly phosphorylated derivatives of S6 appear. Nonetheless, the resumption of protein synthesis that occurs during germination is not dependent on those highly phosphorylated derivatives of S6. The pattern and sites of phosphorylation of S6 labelled in vivo with [32P]orthophosphate have been compared with those of 40S ribosomal subunit labelled in vitro by partially purified protein kinases. Three major phosphopeptides were found in S6 isolated from the zoospore, while six phosphopeptides were found after zoospore germination (in germling cells). The phosphopeptide patterns of S6 phosphorylated by the cAMP-dependent protein kinase and by casein kinases I and II were completely distinct. Only the cAMP-dependent protein kinase gives rise to a phosphopeptide found in 32P-labelled cells, indicating that one of sites phosphorylated in vivo is also phosphorylated in vitro by the cAMP-dependent protein kinase.

Identification and properties of the 38 000-Mr poly(A)-binding protein of non-polysomal messenger ribonucleoproteins of cryptobiotic gastrulae of Artemia salina

The Mr-38 000 poly(A)-binding protein interacts with synthetic and natural RNA. A sequence-independent stoichiometry of one protein per 8 - 12 nucleotides is measured by filter binding and sucrose gradient centrifugation. Specificity for the poly(A) sequence is demonstrated from poly(A)/RNA mixing experiments. The poly(A)-binding protein has been identified as the helix-destabilizing protein HD40[Marvil, D. K., Nowak, L. and Szer, W. (1980) J. Biol. Chem. 255, 6466 - 6472] and is characterized by the existence of at least seven ionic species with a pI ranging from 9.2 to 6.6. Acidic ionic species are generated by phosphorylation with mRNP-associated protein kinase. Different ionic species are present on free mRNP and ribosomes-mRNP preinitiation complexes. The poly(A)-binding protein affects mRNA translation and (A)4 polyadenylation. The multifunctionality of the protein is discussed.

Studies of the kinetics and ionic requirements for the phosphorylation of ribosomal protein S6 after fertilization of Arbacia punctulata eggs

Fertilization of the eggs of the sea urchin Arbacia punctulata is followed by the phosphorylation of ribosomal protein S6. The increase in phosphorylation starts at the same time that protein synthesis begins to increase, and leads to the appearance of mono-, di-, and triphosphorylated S6 derivatives. Essentially all the S6 is phosphorylated by first cleavage. This phosphorylation requires the occurrence of both the normal Ca2+ transient and the consequent Na+-H+ exchange. Protein synthesis can be partially activated by an increase in intracellular pH brought about by weak bases, but this neither causes S6 phosphorylation, nor the inactivation of the specific S6 phosphatase present in unfertilized Arbacia eggs.