Advertisement

Truncation of CDK5 Activator p35 Induces Intensive Phosphorylation of Ser202/Thr205 of Human Tau*

  • Mitsuko Hashiguchi
    Correspondence
    To whom correspondence should be addressed. Tel.: 81-3-3351-6141 (ext. 279/248); Fax: 81-3-5379-0658;
    Affiliations
    Department of Physiology, Tokyo Medical University, 6-1-1 Shinjuku, Shinjuku, Tokyo 160-8402, Japan and the
    Search for articles by this author
  • Taro Saito
    Affiliations
    Department of Biological Sciences, Graduate School of Science, Tokyo Metropolitan University, 1-1 Minami-ohsawa, Hachiohji, Tokyo 192-0397, Japan
    Search for articles by this author
  • Shin-ichi Hisanaga
    Affiliations
    Department of Biological Sciences, Graduate School of Science, Tokyo Metropolitan University, 1-1 Minami-ohsawa, Hachiohji, Tokyo 192-0397, Japan
    Search for articles by this author
  • Toshio Hashiguchi
    Affiliations
    Department of Physiology, Tokyo Medical University, 6-1-1 Shinjuku, Shinjuku, Tokyo 160-8402, Japan and the
    Search for articles by this author
  • Author Footnotes
    * The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Open AccessPublished:September 10, 2002DOI:https://doi.org/10.1074/jbc.M207426200
      Hyperphosphorylated tau is a major component of neurofibrillary tangles, one of the hallmarks of Alzheimer's disease. CDK5 is a kinase that phosphorylates the tau protein, and its endogenous activator, p35, regulates its activity. Recently, calpain was found to digest p35 to its truncated product, p25. Several lines of evidence suggest that p25-CDK5 has much more powerful kinase activity and that it may cause abnormal hyperphosphorylation of tau. In this study, we have examined the kinetic characteristics of in vitro phosphorylation of the longest isoform of human tau by CDK5 and its activators using recombinant proteins. Although the kinase activity of CDK5 in phosphorylating tau was significantly higher in the presence of p25, the affinity of CDK5 for tau was not different. Phosphopeptide mapping revealed enhanced phosphorylation of Ser202/Thr205 residues by p25-CDK5 (amino acid residues of tau are numbered according to the longest isoform of human tau). These results suggest that cleavage of p35 to p25 greatly enhances the kinase activity of CDK5 and increases the phosphorylation of Ser202/Thr205. Considering the fact that phosphorylation of Ser202/Thr205 antagonizes the tau-mediated nucleation of tubulin, p25-CDK5 may play a pivotal role in neuronal cell death in Alzheimer's disease.
      Aberrant phosphorylation of the tau protein is considered to play a decisive role in the pathogenesis of neurodegenerative disorders collectively called tauopathies (
      • Lee V.M.
      • Goedert M.
      • Trojanowski J.Q.
      ). Tau is a neuron-specific, microtubule-associated protein that plays major roles in the assembly and stabilization of microtubules. In the brains of patients with Alzheimer's disease, abnormally phosphorylated tau protein, paired helical filament (PHF)
      The abbreviations used are: PHF, paired helical filament; CDK5, cyclin-dependent kinase-5; PIPES, 1,4-piperazinediethanesulfonic acid; MOPS, 4-morpholinepropanesulfonic acid
      1The abbreviations used are: PHF, paired helical filament; CDK5, cyclin-dependent kinase-5; PIPES, 1,4-piperazinediethanesulfonic acid; MOPS, 4-morpholinepropanesulfonic acid
      tau, becomes dissociated from neuronal microtubules and accumulates in PHFs, the filamentous network generated by self-aggregation of hyperphosphorylated forms of tau. All six adult isoforms of tau are known to be hyperphosphorylated in PHFs (
      • Luc B.
      • Thierry B.
      • Valrie B.S.
      • Andre D.
      • Patrick R.H.
      ).
      Several lines of evidence suggest that proline-directed serine/threonine kinases, together with protein phosphatases, play an important role in the hyperphosphorylation of tau (
      • Morishima-Kawashima M.
      • Hasegawa M.
      • Takio K.
      • Suzuki M.
      • Yoshida H.
      • Titani K.
      • Ihara Y.
      ). Candidate protein kinases include mitogen-activated protein kinase, glycogen synthase kinase-3β, and cyclin-dependent kinase-5 (CDK5). None of these enzymes can generate PHF-tau alone (
      • Zheng-Fischhofer Q.
      • Biernat J.
      • Mandelkow E.M.
      • Illenberger S.
      • Godemann R.
      • Mandelkow E.
      ).
      CDK5, initially known as brain proline-directed protein kinase or neuronal cdc2-like protein kinase, phosphorylates tau at a high stoichiometry (
      • Paudel H.K.
      • Lew J.
      • Ali Z.
      • Wang J.H.
      ,
      • Paudel H.K
      ). Unlike other tau kinases, the activity of CDK5 is regulated by endogenous regulatory proteins, p35 and p39. In this respect, CDK5 is distinctly different from other tau kinases (
      • Maccioni R.B.
      • Otth C.
      • Concha I.I
      • Munoz J.P.
      ).
      Recently, calpain, a calcium-dependent protease, was found to digest both p35 and p39 to their truncated products, p25 and p29, respectively (
      • Patrick G.N.
      • Zukerberg L.
      • Nikolic M., De La
      • Monte S.
      • Dikkes P.
      • Tsai L.H.
      ,
      • Kusakawa G.
      • Saito T.
      • Onuki R.
      • Ishiguro K.
      • Kishimoto T.
      • Hisanaga S.
      ). Furthermore, truncation of p35 to p25 and the subsequent formation of the p25-CDK5 complex result in neuronal cell death and hyperphosphorylation of tau (
      • Lee M.S.
      • Kwon Y.T., LI, M.
      • Peng J.
      • Friedlander R.M.
      • Tsai L.H.
      ). From these findings, the following hypothesis emerged. Initially, various stress signals activate calpain by recruiting intracellular Ca2+, and then the activated calpain digests p35 to p25. Finally, the p25-CDK5 complex hyperphosphorylates tau proteins, causing disruption of microtubule integrity and inevitable cell death (
      • Smith D.S.
      • Greer P.L.
      • Tsai L.H.
      ).
      Several criticisms of this hypothesis have appeared. Among them, the observed accumulation of p25 in the brains of patients with Alzheimer's disease may be accounted for by post-mortem degradation of p35 (
      • Taniguchi S.
      • Fujita Y.
      • Hayashi S.
      • Kakita A.
      • Takahashi H.
      • Murayama S.
      • Saido T.C.
      • Hisanaga S.
      • Iwatsubo T.
      • Hasegawa M.
      ). In fact, p35 is notoriously unstable. Most brain-derived p35 is lost or truncated to p25 during purification. Thus, the kinetic activity of tau phosphorylation by p35-CDK5 is only poorly understood.
      Lack of knowledge about the regulation of CDK5 by its activators, p35 and p25, severely limits our understanding of the possible pathogenesis induced by CDK5. We have examined the kinetic characteristics ofin vitro phosphorylation of the longest isoform of human tau by CDK5 using recombinant proteins. This study clarified not only the kinetics of overall phosphorylation, but also the details of site-specific phosphorylation of human tau by CDK5 complexes.

      DISCUSSION

      This study determined the kinetic characteristics of in vitro phosphorylation of tau by CDK5 complexes using recombinant proteins. The p25-CDK5 complex was found to phosphorylate tau much faster than the p35-CDK5 complex. On the assumption that the time course of total phosphorylation can be approximated by a single exponential reaction, the time constant of the decline in phosphorylatable sites was determined, and the results suggest that p25 can accelerate the catalytic activity of CDK5 by ∼2.4-fold compared with p35.
      We determined the stoichiometry (moles of Pi/mol of tau) of tau phosphorylation by CDK5 and found that the p25-CDK5 complex can phosphorylate tau faster. Furthermore, it can incorporate more phosphate into tau compared with the p35-CDK5 complex.
      The stoichiometry of tau phosphorylation by neuronal cdc2-like protein kinase (brain-derived p25-CDK5 complex) was reported to be 3.8 mol of phosphate/mol of tau (
      • Paudel H.K.
      • Lew J.
      • Ali Z.
      • Wang J.H.
      ). Thus, the recombinant p25-CDK5 complex used in the present study is as potent as the naturally occurring CDK5 complex. Sironi et al. (
      • Sironi J.J.
      • Yen S.H.
      • Gondal J.A., Wu, Q.
      • Grundke-Iqbal I.
      • Iqbal K.
      ) studied in vitrophosphorylation of tau at Ser262 by three different kinases (calcium/calmodulin-dependent protein kinase II, protein kinase A, and phosphorylase kinase) and found that total Piincorporation into tau by any of the three enzymes saturated after 100 min of reaction. Upon saturation of overall phosphorylation, calcium/calmodulin-dependent protein kinase II, protein kinase A, and phosphorylase kinase incorporated 0.9–1.2 mol of phosphate/mol of tau. Purified phosphorylase kinase from rabbit skeletal muscle reportedly phosphorylated tau to a stoichiometry of 2.1 mol of phosphate/mol of tau (
      • Paudel H.K.
      ). Compared with these results, the stoichiometry of overall phosphorylation of tau by the p25-CDK5 complex is characteristically high. To evaluate the phosphorylation of tau by CDK5 quantitatively, kinetic parameters were determined using the Michaelis-Menten equation. The apparent K m andk cat values for tau phosphorylation by the p35-CDK5 complex were 33 μm and 2.6 min−1, respectively. The K m value for the p25-CDK5 complex was 27 μm and was unchanged from that for the p35-CDK5 complex (p < 0.05, n = 4). Thek cat value for the p25-CDK5 complex was 13 min−1 and was significantly larger than that for the p35-CDK5 (p < 0.05, n = 4). Thek cat/K m value for the p25-CDK5 complex was 6 times larger than that for the p35-CDK5 complex.
      The kinetic parameters of CDK5 complexes were also determined using histone H1 as a substrate. The observed K m value for the p35-CDK5 complex was in good agreement with previous results (
      • Sharma P.
      • Steinbach P.J.
      • Sharma M.
      • Amin N.D.
      • Barchi Jr., J.J.
      • Pant H.C.
      ). The difference in the K m values for the p25-CDK5 and p35-CDK5 complexes was not significant, whereas thek cat/K m value for the p25-CDK5 complex was significantly larger than that for the p25-CDK5 complex. Thus, the p25-CDK5 complex is a much more potent kinase, and the p25 regulatory unit accelerates the reactivity of CDK5 without changing its affinity for tau.
      For tau phosphorylation by phosphorylase kinase, theK m and k cat values were 6.9 μm and 47.4 min−1, respectively, indicating that phosphorylase kinase has a better affinity for tau and more efficient turnover of catalytic activity compared with CDK5 complexes (
      • Paudel H.K.
      ). But the stoichiometry of total phosphorylation of tau by the p25-CDK5 complex is larger than that for the other tau kinases, including the p35-CDK5 complex. As a tau kinase, p25-CDK5 could contribute to the high phosphorylation stoichiometry of PHF-tau.
      Our analysis of the total phosphorylation of tau revealed a difference in the catalytic activity of CDK5 complexes; site-specific phosphorylation must be clarified for a better understanding of CDK5. We have employed two independent approaches: phosphopeptide mapping of the tryptic digest of tau and Western blot analysis with phosphorylation-dependent anti-tau antibodies.
      By comparing the phosphopeptide maps for the early and saturated stages of phosphorylation, we found that 32Piincorporation into spot 1 was very slow. In contrast,32Pi incorporation into the other four spots was almost complete within 120 min, suggesting that the time course of overall phosphorylation of tau depends on the phosphorylation kinetics of Ser202, Ser235, and Ser404. We further examined this unreported characteristic of site-specific phosphorylation of Ser202 and Thr205.
      We have made the following observations. (a) The p35-CDK5 complex promoted 32Pi incorporation into spot 1 very slowly; and (b) the p25-CDK5 complex promoted32Pi incorporation into spot 1 more rapidly, but only after 2 h. These results raise the possibility of sequential phosphorylation of Thr205 after Ser202.
      To investigate 32Pi incorporation, we conducted phosphopeptide mapping after having occluded possible phosphorylation sites with unlabeled ATP. Newly incorporated32Pi was visualized as an intense spot on the map. After occlusion with the p25-CDK5 complex, p25-CDK5 promoted32Pi incorporation into spot 1 only.32Pi incorporation into any of the major spots did not happen. These findings suggest that phosphorylation of Ser202 was already saturated and that only Thr205 was slowly phosphorylated over the extended time of incubation. In contrast, the p35-CDK5 complex failed to occlude both spots 1 and 2, as suggested by the intense signal on the two spots. The occlusion experiment also confirmed that slow phosphorylation by CDK5 complexes happens only at Ser202and Thr205.
      To clarify further the time dependence of phosphorylation of Ser202 and Thr205, Western blot analysis of tau was carried out. The three phosphorylation-dependent anti-tau antibodies have distinct epitopes. AT8 reacts with tau only when multiple sites around Ser202, including Ser199, Ser202, and Thr205, are phosphorylated. Single phosphorylation of any of the residues is not enough for AT8 reactivity (
      • Goedert M.
      • Jakes R.
      • Vanmechelen E.
      ,
      • Preuss U.
      • Doring F.
      • Illenberger S.
      • Mandelkow E.M.
      ). Thus, AT8 is useful in detecting phosphorylation of Ser202/Thr205 for proline-directed kinases, including CDK5 complexes. In contrast, Tau-1 requires an absolutely non-phosphorylated epitope around Ser202 (
      • Szendrei G.I.
      • Lee V.M.
      • Otvos Jr., L.
      ). In the present study, Tau-1 reacted with recombinant tau quite well. However, phosphorylation of Ser202 or Thr205 caused loss of Tau-1 reactivity. The pS202 antibody is raised against phosphorylated Ser202 of human tau.
      AT8 reactivity was latent, whereas the Tau-1 signal started to decline without delay, suggesting that Ser202 and Thr205 cannot be phosphorylated simultaneously. Either Ser202 or Thr205 must be phosphorylated first. The phosphorylated Ser202 signal (immunoreactivity of the pS202 antibody) started to develop as soon as the Tau-1 signal disappeared. Although phosphorylation of Thr205 as a first step of sequential phosphorylation cannot be ruled out, the initial phosphorylation of Ser202 and the subsequent phosphorylation of Thr205 seem to be the principal route to double phosphorylation.
      Many of the pathologic epitopes on hyperphosphorylated tau in the brains of Alzheimer's disease patients were thought to be generated by sequential phosphorylation by different tau kinases. For example, phosphorylation of a specific amino acid residue by CDK5 is a prerequisite for the subsequent kinase action of glycogen synthase kinase-3β (
      • Sengupta A., Wu, Q.
      • Grundke-Iqbal K.
      • Singh T.
      ,
      • Alvarrez A.
      • Toro R.
      • Cacere A.
      • Maccioni R.B.
      ). Phosphorylation of Thr212 by glycogen synthase kinase-3β is known to facilitate protein kinase A action on Ser214 (
      • Zheng-Fischhofer Q.
      • Biernat J.
      • Mandelkow E.M.
      • Illenberger S.
      • Godemann R.
      • Mandelkow E.
      ). Such hierarchy among tau kinases (
      • Jicha G.A.
      • O'Donnell A.
      • Weaver C.
      • Angeletti R.
      • Davies P.
      ) makes it complicated to clarify the mechanism of abnormal phosphorylation of tau.
      We have shown evidence for the sequential phosphorylation of Ser202 and Thr205 of tau by the p25-CDK5 complex. Phosphorylation of Thr205 occurred only after Ser202 was phosphorylated. As a result, phosphorylation of Thr205 was extremely slow compared with phosphorylation of other sites such as Ser235 and Ser404. In addition, the p35-CDK5 complex had weak or negligible kinase action on Thr205.
      Our findings strongly suggest that cleavage of p35 to p25 regulates not only the overall kinase activity of CDK5, but also the sequential phosphorylation of Ser202 and Thr205. The association of CDK5 with a particular activator unit, p35 or p25, provides a novel mechanism of controlling kinase characteristics.

      Acknowledgments

      We appreciate the initial help of Dr. H. K. Paudel. We thank Dr. L. H. Tsai for providing human CDK5 cDNA and Dr. M. Goedert for the gift of human tau cDNA.

      References

        • Lee V.M.
        • Goedert M.
        • Trojanowski J.Q.
        Annu. Rev. Neurosci. 2001; 24: 1121-1159
        • Luc B.
        • Thierry B.
        • Valrie B.S.
        • Andre D.
        • Patrick R.H.
        Brain Res. Rev. 2000; 33: 95-130
        • Morishima-Kawashima M.
        • Hasegawa M.
        • Takio K.
        • Suzuki M.
        • Yoshida H.
        • Titani K.
        • Ihara Y.
        J. Biol. Chem. 1995; 270: 823-829
        • Zheng-Fischhofer Q.
        • Biernat J.
        • Mandelkow E.M.
        • Illenberger S.
        • Godemann R.
        • Mandelkow E.
        Eur. J. Biochem. 1998; 252: 542-552
        • Paudel H.K.
        • Lew J.
        • Ali Z.
        • Wang J.H.
        J. Biol. Chem. 1993; 268: 23512-23518
        • Paudel H.K
        J. Biol. Chem. 1997; 272: 28328-28334
        • Maccioni R.B.
        • Otth C.
        • Concha I.I
        • Munoz J.P.
        Eur. J. Biochem. 2001; 268: 1518-1527
        • Patrick G.N.
        • Zukerberg L.
        • Nikolic M., De La
        • Monte S.
        • Dikkes P.
        • Tsai L.H.
        Nature. 1999; 402: 615-622
        • Kusakawa G.
        • Saito T.
        • Onuki R.
        • Ishiguro K.
        • Kishimoto T.
        • Hisanaga S.
        J. Biol. Chem. 2000; 275: 17166-17172
        • Lee M.S.
        • Kwon Y.T., LI, M.
        • Peng J.
        • Friedlander R.M.
        • Tsai L.H.
        Nature. 2000; 405: 360-364
        • Smith D.S.
        • Greer P.L.
        • Tsai L.H.
        Cell Growth & Differ. 2001; 12: 277-283
        • Taniguchi S.
        • Fujita Y.
        • Hayashi S.
        • Kakita A.
        • Takahashi H.
        • Murayama S.
        • Saido T.C.
        • Hisanaga S.
        • Iwatsubo T.
        • Hasegawa M.
        FEBS Lett. 2001; 489: 46-50
        • Sobue K.
        • Agarwal-Mawal A., Li, W.
        • Sun W.
        • Miura Y.
        • Paudel H.K.
        J. Biol. Chem. 2000; 275: 16673-16680
        • Hashiguchi M.
        • Sobue K.
        • Paudel H.K.
        J. Biol. Chem. 2000; 275: 25247-25254
        • Boyle W.J.
        • van der Geer P.
        • Hunter T.
        Methods Enzymol. 1991; 201: 110-149
        • Sironi J.J.
        • Yen S.H.
        • Gondal J.A., Wu, Q.
        • Grundke-Iqbal I.
        • Iqbal K.
        FEBS Lett. 1998; 436: 471-475
        • Paudel H.K.
        J. Biol. Chem. 1997; 272: 1777-1785
        • Sharma P.
        • Steinbach P.J.
        • Sharma M.
        • Amin N.D.
        • Barchi Jr., J.J.
        • Pant H.C.
        J. Biol. Chem. 1999; 274: 9600-9606
        • Goedert M.
        • Jakes R.
        • Vanmechelen E.
        Neurosci. Lett. 1995; 189: 167-169
        • Preuss U.
        • Doring F.
        • Illenberger S.
        • Mandelkow E.M.
        Mol. Biol. Cell. 1995; 6: 1397-1410
        • Szendrei G.I.
        • Lee V.M.
        • Otvos Jr., L.
        J. Neurosci. Res. 1993; 34: 243-249
        • Sengupta A., Wu, Q.
        • Grundke-Iqbal K.
        • Singh T.
        Mol. Cell. Biochem. 1997; 167: 99-105
        • Alvarrez A.
        • Toro R.
        • Cacere A.
        • Maccioni R.B.
        FEBS Lett. 1999; 459: 421-426
        • Jicha G.A.
        • O'Donnell A.
        • Weaver C.
        • Angeletti R.
        • Davies P.
        J. Neurochem. 1999; 72: 214-224

      Linked Article