G Thomas

The principal target of rapamycin-induced p70s6k inactivation is a novel phosphorylation site within a conserved hydrophobic domain.

The immunosuppressive agent rapamycin induces inactivation of p70s6k with no effect on other mitogen‐activated kinases. Here we have employed a combination of techniques, including mass spectrometry, to demonstrate that this effect is associated with selective dephosphorylation of three previously unidentified p70s6k phosphorylation sites: T229, T389 and S404. T229 resides at a conserved position in the catalytic domain, whose phosphorylation is essential for the activation of other mitogen‐induced kinases. However, the principal target of rapamycin‐induced p70s6k inactivation is T389, which is located in an unusual hydrophobic sequence outside the catalytic domain. Mutation of T389 to alanine ablates kinase activity, whereas mutation to glutamic acid confers constitutive kinase activity and rapamycin resistance. The importance of this site and its surrounding motif to kinase function is emphasized by its presence in a large number of protein kinases of the second messenger family and its conservation in putative p70s6k homologues from as distantly related organisms as yeast and plants.

The principal rapamycin-sensitive p70(s6k) phosphorylation sites, T-229 and T-389, are differentially regulated by rapamycin-insensitive kinase kinases.

Mitogen-induced activation of p70(s6k) is associated with the phosphorylation of specific sites which are negatively affected by the immunosuppressant rapamycin, the fungal metabolite wortmannin, and the methylxanthine SQ20006. Recent reports have focused on the role of the amino terminus of the p85(s6k) isoform in mediating kinase activity, with the observation that amino-terminal truncation mutants are activated in the presence of rapamycin while retaining their sensitivity to wortmannin. Here we show that the effects of previously described amino- and carboxy-terminal truncations on kinase activity are ultimately reflected in the phosphorylation state of the enzyme. Mutation of the principal rapamycin-targeted phosphorylation site, T-389, to an acidic residue generates a form of the kinase which is as resistant to wortmannin or SQ20006 as it is to rapamycin, consistent with the previous observation that T-389 was a common target of all three inhibitors. Truncation of the first 54 residues of the amino terminus blocks the serum-induced phosphorylation of three rapamycin-sensitive sites, T-229 in the activation loop and T-389 and S-404 in the linker region. This correlates with a severe reduction in the ability of the kinase to be activated by serum. However, loss of mitogen activation conferred by the removal of the amino terminus is reversed by additional truncation of the carboxy-terminal domain, with the resulting mutant demonstrating phosphorylation of the remaining two rapamycin-sensitive sites, T-229 and T-389. In this double-truncation mutant, phosphorylation of T-229 occurs in the basal state, whereas mitogen stimulation is required to induce acute upregulation of T-389 phosphorylation. The phosphorylation of both sites proceeds unimpaired in the presence of rapamycin, indicating that the kinases responsible for the phosphorylation of these sites are not inhibited by the macrolide. In contrast, activation of the double-truncation mutant is blocked in the presence of wortmannin or SQ20006, and these agents completely block the phosphorylation of T-389 while having only a marginal effect on T-229 phosphorylation. When the T-389 site is mutated to an acidic residue in the double-truncation background, the activation of the resulting mutant is insensitive to the wortmannin and SQ20006 block, but interestingly, the mutant is activated to a significantly greater level than a control in the presence of rapamycin. These data are consistent with the hypothesis that T-389 is the principal regulatory phosphorylation site, which, in combination with hyperphosphorylation of the autoinhibitory domain S/TP sites, is acutely regulated by external effectors, whereas T-229 phosphorylation is regulated primarily by internal mechanisms.

Dual requirement for a newly identified phosphorylation site in p70s6k.

The activation of p70s6k is associated with multiple phosphorylations at two sets of sites. The first set, S411, S418, T421, and S424, reside within the autoinhibitory domain, and each contains a hydrophobic residue at -2 and a proline at +1. The second set of sites, T229 (in the catalytic domain) and T389 and S404 (in the linker region), are rapamycin sensitive and flanked by bulky aromatic residues. Here we describe the identification and mutational analysis of three new phosphorylation sites, T367, S371, and T447, all of which have a recognition motif similar to that of the first set of sites. A mutation of T367 or T447 to either alanine or glutamic acid had no apparent effect on p70s6k activity, whereas similar mutations of S371 abolished kinase activity. Of these three sites and their surrounding motifs, only S371 is conserved in p70s6k homologs from Drosophila melanogaster, Arabidopsis thaliana, and Saccharomyces cerevisiae, as well as many members of the protein kinase C family. Serum stimulation increased S371 phosphorylation; unlike the situation for specific members of the protein kinase C family, where the homologous site is regulated by autophosphorylation, S371 phosphorylation is regulated by an external mechanism. Phosphopeptide analysis of S371 mutants further revealed that the loss of activity in these variants was paralleled by a block in serum-induced T389 phosphorylation, a phosphorylation site previously shown to be essential for kinase activity. Nevertheless, the substitution of an acidic residue at T389, which mimics phosphorylation at this site, did not rescue mutant p70s6k activity, indicating that S371 phosphorylation plays an independent role in regulating intrinsic kinase activity.

A stimulated S6 kinase from rat liver: identity with the mitogen activated S6 kinase of 3T3 cells.

A number of approaches were tested for their ability to induce S6 phosphorylation and S6 kinase activation in rat liver, including i.p. injection of insulin, sodium orthovanadate or cycloheximide, as well as refeeding starved animals. All treatments led to increased S6 phosphorylation and activation of the apparent same enzyme. The most potent activator of the S6 kinase in liver extracts was cycloheximide. Maximum activation was achieved in 20 min at 1 mg cycloheximide/100 g body weight, with half‐maximal activation in 10 min. Based on these findings a large‐scale kinase purification procedure was established involving seven steps of chromatography. Following the final step a major protein band of Mr 70,000 was revealed. The protein was purified 20,000‐fold, had a sp. act. of 640 nmol/min/mg of protein towards S6, autophosphorylated and was inactivated by phosphatase 2A. Peptide maps of autophosphorylated material were identical to those derived from the mitogen‐activated kinase of 3T3 cells.

Identification and early activation of a Xenopus laevis p70s6k following progesterone-induced meiotic maturation.

Employing oligonucleotide primers derived from the DNA sequence of rat p70s6k a homologous sequence was shown by polymerase chain reaction (PCR) to be present as a maternal transcript in stage IV‐VI Xenopus laevis oocytes. The sequence covered 665 bp of p70s6k and was 97% identical at the amino acid level. When used to probe a Northern blot of the poly(A)+ mRNA from stage VI oocytes, this sequence recognized four transcripts of 1.9, 2.5, 3.2 and 5.2 kb. Specific rat p70s6k antibodies immunoprecipitated active S6 kinase from stage VI oocytes but not unfertilized eggs. The basal level of activity was 3‐ to 5‐fold higher in primed versus non‐primed oocytes indicating that p70s6k activation was an early event in maturation. Consistent with this observation, progesterone induced a 10‐fold activation of the kinase in non‐primed oocytes within 1 h post‐induction at a time critical for early activation of protein synthesis. A much smaller, variable peak of activation was observed at 85% germinal vesicle breakdown (GVBD), which was dependent on the rate of maturation. Two members of the pp90rsk family, thought to be the sole S6 kinases present in X.laevis oocytes, exhibited distinct kinetics of activation. Finally, the S6 kinase activity present 1 h post‐progesterone stimulation was purified and shown to have a Mr of 70K.