The Regulatory Ser262 of Microtubule-associated Protein Tau Is Phosphorylated by Phosphorylase Kinase*

(Received for publication, July 18, 1996, and in revised form, October 2, 1996)

Hemant K. Paudel Dagger

From the Bloomfield Center for Research in Aging, Lady Davis Institute for Medical Research, Sir Mortimer B. Davis-Jewish General Hospital and the Department of Neurology and Neurosurgery, McGill University, Montreal, Quebec, H3T 1E2, Canada

ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
FOOTNOTES
REFERENCES


ABSTRACT

Abnormally phosphorylated tau is the major component of paired helical filaments found in the brains of patients suffering from Alzheimer's disease. Therefore, the identification of kinases that phosphorylate tau is of considerable interest. A DEAE-Sepharose column resolved porcine brain extract into five tau kinase activity peaks. Among these peaks, two were completely inhibited by EGTA, indicating that these two activity peaks contained Ca2+-dependent tau kinases. One of the above two Ca2+-dependent tau kinase activity peaks also contained phosphorylase kinase activity. The tau kinase and phosphorylase kinase activities associated with this peak could not be separated from each other by Superose 12 gel filtration, hydroxylapatite, and calmodulin-agarose affinity chromatographies. Phosphorylase kinase, purified from rabbit skeletal muscle, phosphorylated tau to a stoichiometry of 2.1 mol of phosphate/mol of tau and converted tau to a species with a retarded mobility on SDS-polyacrylamide gel electrophoresis. The apparent Km and kcat values for tau phosphorylation by muscle phosphorylase kinase were 6.9 µM and 47.4 min-1, respectively. As a substrate of muscle phosphorylase kinase, phosphorylase was eight times better than tau. Sequence analyses of tryptic and thermolytic phosphopeptides derived from tau phosphorylated by muscle phosphorylase kinase revealed five phosphorylation sites, Ser237, Ser262, Ser285, Ser305, and Ser352. Among these sites, Ser262 was previously shown to be phosphorylated in human tau from fetal, adult, and Alzheimer's diseased brains (Seubert, P., Mawal-Dewan, M., Barbour, R., Jakes, R., Goedert, M., Johnson, G. V. W., Litersky, J. M., Schenk, D., Lieberburg, I., Trojanowski, J. Q., and Lee, V. M. Y. (1995) J. Biol. Chem. 270, 18917-18922); and its phosphorylation abolished tau's binding to microtubules (Drewes, G., Trinczek, B., Illenberger, S., Biernat, J., Schmitt-Ulms, G., Meyer, H. E., Mandelkow, E.-M., and Mandelkow, E. (1995) J. Biol. Chem. 270, 7679-7688). Slot-blot analysis using a monoclonal antibody against muscle phosphorylase kinase and an activity assay using phosphorylase revealed that phosphorylase kinase was present in microtubules extensively purified by repeated cycles of polymerization and depolymerization. Taken together, these results suggest that in neurons, phosphorylase kinase may be one of the kinases that participate in the phosphorylation of tau.


INTRODUCTION

Tau is a neuronal-specific microtubule-associated protein. It plays an important role in neuronal morphogenesis, the maintenance of axonal shape, and axonal transport through its ability to bind and regulate microtubule structure and dynamics (for reviews, see Refs. 1-3). There are six alternatively spliced isoforms of human tau ranging from 352 to 441 amino acids in length (4). The expression of these isoforms is developmentally regulated with only one isoform expressed in juvenile brain and all isoforms expressed in adult brain (2-4). These isoforms differ from each other by the presence of three or four tandem repeats of 31-32 amino acids (3). These repeats are rich in positively charged residues and are thought to interact with the negatively charged carboxyl-terminal region of tubulins exposed on the microtubule structure to result in a tau-microtubule binding (1, 5).

Tau is a phosphoprotein (6-8), and phosphorylation greatly reduces tau's affinity for microtubules (9-11). Since tau binding promotes microtubule assembly and enhances stability and stiffness of microtubules (12), phosphorylation of tau is likely to regulate microtubule structure, stiffness, and dynamics. In vivo, adult tau is phosphorylated on Thr181, Ser202, Thr231, Ser262, and Ser404 (all tau residues in this paper are numbered according to the longest isoform of human tau (4)) (6-8). Juvenile tau is phosphorylated on all the above sites plus Ser198, Ser199, Thr217, Ser235, Ser396, and Ser400 (6-8). Among all the above sites, Ser262 is the only site located within the microtubule-binding repeats of tau. Its phosphorylation was shown to drastically suppress tau's affinity for microtubules and was suggested to play a key role in the regulation of binding of tau to microtubules (10, 11).

Recently, tau became the focus of intense research due to its presence in paired helical filaments (PHFs)1. PHFs are the main structural component of neurofibrillary tangles (NFTs) which are a characteristic neuropathological lesion found in the brains of patients suffering from Alzheimer's disease (AD) (reviewed in Ref. 13). PHF-tau (tau associated with PHFs) is highly insoluble, displays retarded mobility on SDS-PAGE, and is abnormally phosphorylated (i.e. contains more phosphate than normal tau). Several studies indicate that it is the abnormal phosphorylation that is responsible for converting normal tau to PHF-tau (13-16).

PHF-tau is phosphorylated on 19 sites. These sites are Ser208, Ser210, Thr212, Ser214, Thr403, Ser409, Ser412, Ser413, and Ser422 plus the phosphorylation sites of juvenile tau mentioned above, except Thr181 (17). Among these sites nine contain a S/TP motif, recognized by various proline-directed kinases. These sites are Ser199, Ser202, Thr212, Thr217, Thr231, Ser235, Ser396, Ser404, and Ser422. Proline-directed kinases such as neuronal cdc2-like kinase (NCLK), Map-kinase, cdc2 kinase, and glycogen synthase kinase 3 efficiently phosphorylate tau on many of these sites in vitro (18-22). Of the 10 nonproline-directed sites of PHF-tau, Ser214 and Ser409 are phosphorylated by cAMP-dependent protein kinase (A kinase) (23). Two kinases, a 35/41-kDa (11) and a 116-kDa (10), were suggested to phosphorylate Ser262, a nonproline-directed site, but further characterization of these kinases has not been done. Protein kinase C (24) and calmodulin-dependent protein kinase II (Cam kinase II) (25) phosphorylate tau on Ser324 and Ser416, respectively, but these sites do not appear to be phosphorylated in normal (6) or PHF-tau (17). Kinases that phosphorylate nonproline-directed sites of tau, Ser198, Ser208, Ser210, Ser400, Thr403, Ser412, and Ser413 are yet to be identified.

For a complete elucidation of the cellular and molecular basis of PHF formation it is essential that all the kinases that phosphorylate tau be identified. In this study specific substrates and antibodies directed to various kinases were utilized to find out if a novel tau kinase could be detected in brain homogenates. Herein, it is reported that phosphorylase kinase, implicated in the phosphorylation of phosphorylase and regulation of glycogenolysis (26), copurifies with microtubules and phosphorylates tau in a Ca2+-dependent manner. In vitro, it phosphorylates tau on five sites including Ser262. Results presented in this paper suggest that phosphorylase kinase is one of the potential kinases involved in the phosphorylation of tau under normal and pathological conditions.


MATERIALS AND METHODS

Proteins

Tau protein was purified from lysates of Escherichia coli that overexpressed the longest isoform of human tau (htau 40) essentially as described (27) except the lysis buffer also contained 1% Triton X-100, and the clear supernatant obtained after the centrifugation of the bacterial lysate was passed through a Q-Sepharose column. Microtubules were purified from bovine brain extract by three cycles of temperature-induced polymerization and depolymerization of microtubules (20). Muscle phosphorylase kinase was purified from New Zealand White rabbit skeletal muscle through DEAE-cellulose chromatography (28). Phosphorylated muscle phosphorylase kinase was prepared by autophosphorylation of the kinase and resulted in the incorporation of 7.9 mol of phosphate/mol of kinase (29). The catalytic gamma  subunit of muscle phosphorylase kinase was prepared essentially as described previously (30). Phosphorylase was isolated from New Zealand White rabbit skeletal muscles using the method of Fisher and Krebs (31). Trypsin, A kinase, and thermolysin (protease type X) were from Sigma. Protein kinase C was purified from the rat brain homogenate (32). Bovine brain calmodulin was a generous gift from Dr. Jerry H. Wang (University of Calgary). Polyclonal antibodies against a synthetic peptides derived from amino-terminal region of NCLK (EKIGEGTYGVVYK) and carboxyl-terminal region of Map-kinase p43erk1 (residues 333-367) were prepared in the Medical Research Council, Signal Transduction Antibody Facility of the University of Calgary. The monoclonal antibodies, mAB 88 and mAB 979, against rabbit skeletal muscle phosphorylase kinase were kindly provided by Dr. Gerald M. Carlson, University of Tennessee, Memphis. The monoclonal antibody against protein kinase C was from Amersham Corp. (RPN 536).

Peptides

Synthetic peptide substrates of NCLK (KTPKKAKKPKTPKKAKKL) (20) and Map-kinase (APRTPGGRR) (33) were synthesized at the Medical Research Council, Signal Transduction Peptide Synthesis Core Facility of the University of Calgary. Kemptide (LRRASLG), Syntide 2 (PLARTLSVAGLPGKK), protein kinase C substrate peptide (KRTLRR), and A kinase inhibitory peptide (TTYADPIASGRTGRRNAIHD) were purchased from Sigma.

Protein and Peptide Concentrations

Concentrations of tau, phosphorylase kinase, calmodulin, and phosphorylase were based on their respective absorbance indices (28, 34-36). The concentration of phosphorylated phosphorylase kinase was determined by Bio-Rad protein assay using phosphorylase kinase as standard. Concentration of the catalytic gamma  subunit was determined as described previously (30). Concentrations of all the other proteins were determined by Bio-Rad protein assay using bovine serum albumin as standard. Concentrations of Kemptide, Syntide 2, protein kinase C substrate peptide, and protein kinase A inhibitory peptide were based on their dry weights. Concentrations of all other synthetic peptides were based on their amino acid analyses.

Kinase Assay

Unless otherwise stated, the phosphorylation of various substrates by various kinases was carried out as described previously (32). The final concentrations of assay components were 50 mM Tris-HCl (pH 8.0), 0.1 mM EDTA, 0.1 mM DTT, 0.3 mM CaCl2, 10 mM MgCl2, 0.5 mM [gamma -32P]ATP, 0.1 µM kinase, and 20 µM substrate protein or substrate peptide. When included, the concentration of EGTA was 1 mM. Reactions were initiated by adding a 10-µl aliquot of kinase into 90 µl of the phosphorylation mixture containing the rest of the components of the assay and carried out at 30 °C. After the indicated time points, aliquots were withdrawn and analyzed for the amount of radioactivity incorporated into the substrate by the filter paper assay (37). Phosphorylase kinase assays were carried out as above except the pH of the assay mixture was 8.2. The phosphorylation mixture of A kinase, in addition to all the above components of the assay, also contained 10 µM cAMP. The phosphorylation of tau by the catalytic gamma  subunit of phosphorylase kinase was carried out as described (30). After 15 min at room temperature, MgATP was added to initiate the reaction. The final concentrations of various components of the assay were 50 mM Hepes (pH 7.0), 0.1 mM EDTA, 0.1 mM DTT, 0.3 mM CaCl2, 0.5 mM [gamma -32P]ATP, 10 mM MgCl2, 200 mM urea, 25 µg/ml calmodulin, 30 µg/ml gamma  subunit, and 20 µM tau. After 30 min at 30 °C, aliquots were removed and subjected to SDS-PAGE followed by autoradiography to monitor the phosphorylation of tau.

SDS-PAGE, Immunoblot, and Slot Blot

SDS-PAGE was performed by the method of Laemmli (38). Immunoblots were carried out essentially as described (32). To perform slot-blot analysis, samples were applied to a nitrocellulose membrane in a slot-blot apparatus (Bio-Rad) and then the membrane was washed twice with TBS (25 mM Tris-HCl (pH 7.5) and 0.2 M NaCl). The membrane was removed from the slot-blot apparatus, washed once with TBS containing 0.2% Tween (TTBS), and blocked with 5% skim milk in TTBS for 30 min. The blocked membrane was probed with a primary antibody in 5% skim milk in TTBS for 16 h at 4 °C and developed with a secondary antibody conjugated to alkaline phosphatase as described (32).

Partial Purification of Phosphorylase Kinase from Porcine Brain

Unless otherwise stated all procedures were carried out at 4 °C. Fresh porcine brain (300 g) was homogenized with a blender in 450 ml of buffer A (50 mM beta -glycerophosphate (pH 7.0), 0.5 mM EDTA, 0.5 mM DTT, 0.1 mM EGTA, 10 mM NaF, and 1 mM phenylmethylsulfonyl fluoride) containing 1 µg/ml leupeptin, 1 µg/ml pepstatin, and 5 µg/ml benzamidine. The homogenate was centrifuged at 10,000 × g for 30 min. The supernatant was filtered through a cheesecloth and the filtrate centrifuged at 105 × g for 40 min. Resulting clear supernatant (400 ml) was loaded onto a DEAE Sepharose column (45 × 2.5 cm) previously equilibrated with buffer A. The column was washed with 3 column volumes of buffer A and eluted with 500 ml of a linear gradient (0-0.5 M) of NaCl in buffer A. The effluent fractions containing phosphorylase kinase activity were combined (~50 ml), concentrated to ~ 10 ml by aquacide III (Calbiochem), and passed through a Superose 12 gel filtration column (70 × 2 cm) previously equilibrated in buffer B (50 mM beta -glycerophosphate (pH 7.0), 0.5 mM EDTA, and 0. 5 mM DTT) containing 0.1 M NaCl. The effluent containing phosphorylase kinase activity was loaded onto a hydroxylapatite column (12 × 2.5 cm), pre-equilibrated in buffer B. The column was washed with 3 column volumes of buffer B and eluted with 100 ml of a linear gradient (0-0.5 M) of Na2HPO4 (pH 7.0) in buffer B. The eluent from the hydroxylapatite column containing phosphorylase kinase activity was adjusted to 0.3 mM in CaCl2 and loaded onto a ~1-ml size calmodulin-agarose (Sigma) affinity column previously equilibrated in buffer B containing 0.3 mM CaCl2 and 0.1 M NaCl. The column was washed with 20 ml of equilibration buffer, and the kinase activity was eluted with 1 mM EGTA in buffer B. Effluent fractions (1 ml each) were collected, and those containing phosphorylase kinase activity were stored at 4 °C.

Phosphopeptide Maps and Purification of Phosphopeptides

Tau was phosphorylated by muscle phosphorylase kinase as described above, except here the concentrations of muscle phosphorylase kinase were 0.15 µM. When brain phosphorylase kinase was used, phosphorylation was initiated by adding a 10-µl aliquot from Fig. 2C (fraction 8) to a vial containing 90 µl of reaction mixture that contained all the rest of the components for tau phosphorylation. After 5 h at 30 °C, the entire phosphorylation mixture was placed in a boiling water bath for 10 min to denature and precipitate phosphorylase kinase. The boiled sample was centrifuged in a bench-top centrifuge at full speed for 15 min, and the heat-stable, phosphorylated tau was recovered in the supernatant. The recovered phosphorylated tau was desalted on a Sephadex G25 column, lyophilized, redissolved in 200 µl of 50 mM (NH)4HCO3 (pH 8.0) containing 50 µg/ml trypsin and incubated at 37 °C for 16 h. After incubation, the sample was injected into a HPLC C18 reverse phase column previously equilibrated with 0.1% trifluoroacetic acid. The column was eluted with a linear gradient of 0-30% acetonitrile in 0.1% trifluoroacetic acid in 70 min. Fractions (0.5 ml each) were collected, and an aliquot from each fraction was counted in a liquid scintillation counter to determine the amount of radioactivity in each fraction. To purify phosphopeptides, each of the above fractions containing radioactivity was passed through a Sephadex G25 column (0.8 × 25 cm) pre-equilibrated in 0.1% trifluoroacetic acid to separate peptides based on their sizes. The effluent fractions (0.5 ml each) were collected, and each radioactive fraction was vacuum-dried, redissolved in 200 µl of 0.1% trifluoroacetic acid, and rechromatographed through a HPLC column as above, except the acetonitrile gradient was 0-40% in 50 min. Effluent fractions were collected, and those containing radioactivity were subjected to amino acid sequencing.


Fig. 2. Co-elution of phosphorylase kinase and tau kinase activities from various chromatographic columns. Fractions containing phosphorylase kinase activity from Fig. 1 were pooled and subjected to sequential column chromatographies, Superose 12 gel filtration, hydroxylapatite, and calmodulin-agarose affinity as described under "Materials and Methods." Aliquots from indicated effluent fractions were withdrawn and assayed for phosphorylase kinase and tau kinase activities as in Fig. 1. A, Superose 12 gel filtration chromatography; B, hydroxylapatite chromatography; C, calmodulin-agarose affinity chromatography. The size of each fraction in A, B, and C was 2, 2, and 1 ml, respectively.
[View Larger Version of this Image (25K GIF file)]


Phosphopeptide Sequencing

Phosphopeptides were sequence by a gas phase sequencer as described previously (20) and were carried out at the peptide sequencing facilities at The University of Calgary, and Department of Biochemistry and Microbiology, University of Victoria.


RESULTS

Tau Kinases in Brain Extracts

When an extract from fresh porcine brain was chromatographed through a DEAE-Sepharose column and various fractions assayed for tau phosphorylation activity, five prominent activity peaks, designated as 1, 2, 3, 4, and 5, eluted from the column (Fig. 1A). Immunoblot analysis, using a polyclonal antibody that recognized the amino-terminal region of NCLK, and an activity assay, using a synthetic peptide substrate directed to NCLK, revealed that NCLK did not bind to the column and was recovered in flow-through fractions as reported previously (39). Similarly, brain-specific Map-kinases, p43erk1 and p42erk2, were detected within fractions 50-80 by a polyclonal antibody that recognized the carboxyl termini of both p43erk1 and p44erk2. When these fractions were tested against a synthetic peptide substrate directed to these two Map-kinases, a very low phosphotransferase activity was observed (data not shown). These observations suggested that these two brain-specific Map-kinases were almost inactive in the above fractions. It is possible that the two Map-kinases may have been inactivated during the purification. It is also possible that these two Map-kinases in brain extract may require activation as was observed previously (21). In any case, these observations indicated that none of the peaks in Fig. 1A correspond to Map kinase or NCLK.


Fig. 1. DEAE-Sepharose chromatography of porcine brain extract. An extract from fresh porcine brain was chromatographed through a DEAE-Sepharose column as described under "Materials and Methods." Effluent fractions (5 ml each) were collected, and indicated fractions were assayed for indicated kinase activities using tau (A), syntide 2 (B), or phosphorylase (C) as the substrate. Activity assays were carried out for 30 min.
[View Larger Version of this Image (19K GIF file)]


When EGTA, a preferential chelator of Ca2+, was included in the assay mixture and the tau kinase activity in various fractions in Fig. 1A monitored, peaks 3 and 5 were completely suppressed (Fig. 1A, open circles). These observations indicated that peaks 3 and 5 correspond to Ca2+-dependent tau kinases.

When calmodulin, a Ca2+-binding protein that regulates the activities of several kinases (40), was included in the assay and tau phosphorylation activity of various fractions in Fig. 1A monitored, only peaks 3 and 5 were stimulated (data not shown). These results suggested that peaks 3 and 5 correspond to Ca2+- and calmodulin-dependent tau kinases. In brain, Cam kinase II is a Ca2+- and calmodulin-dependent kinase that phosphorylates tau in vitro (25). Previously syntide 2, a synthetic peptide derived from glycogen synthase, was shown to be an excellent substrate of Cam kinase II (41). Therefore, to find out if any of the above two Ca2+- and calmodulin-dependent tau kinase activity peaks in Fig. 1A corresponded to Cam kinase II, various fractions in Fig. 1A were examined for syntide 2 phosphorylation activity. As shown in Fig. 1B, three syntide 2 kinase activity peaks designated as a, b, and c eluted from the column. Peak c was the major and peaks a and b were the minor activity peaks (Fig. 1B). Peak a did not co-elute with any of the tau kinase activity peaks in Fig. 1A and was found to contain protein kinase C (see "Discussion"). Peak b was found to contain phosphorylase kinase activity (see below). Thus, peak c, the major syntide 2 kinase activity peak in Fig. 1B, is likely to be Cam kinase II. As shown in Fig. 1, the syntide 2 kinase activity peak c in Fig. 1B (Cam kinase II activity peak) and tau kinase activity peak 5 (Fig. 1A) co-eluted from the DEAE-Sepharose column. These observations and the fact that Cam kinase II phosphorylates tau in vitro (25) together suggested that either all or part of the tau kinase activity in peak 5 (Fig. 1A) is due to Cam kinase II.

As shown in Fig. 1, syntide 2 kinase activity peak b (Fig. 1B) and tau kinase activity peak 3 (Fig. 1A) co-eluted from the DEAE-Sepharose column (compare Fig. 1, A and B). These observations suggested that the tau kinase in peak 3 (Fig. 1A) may phosphorylate syntide 2. Previously, muscle phosphorylase kinase (a Ca2+- and calmodulin-dependent enzyme (26)) was reported to phosphorylate syntide 2 (42, 44). Since phosphorylase kinase is relatively abundant in brain (32, 43), various fractions in Fig. 1A were assayed for phosphorylase kinase activity using phosphorylase (a relatively specific substrate of phosphorylase kinase) as the substrate. Interestingly, phosphorylase kinase activity co-eluted with the tau kinase activity peak 3 (Fig. 1A) from the DEAE-Sepharose column (compare Fig. 1, A and C). To determine if the tau kinase activity in peak 3 (Fig. 1A) was due to phosphorylase kinase (or any other kinase), peak 3 fractions were combined, concentrated, and passed through a Superose 12 gel filtration column. As shown in Fig. 2A, the phosphorylase kinase activity eluted as a large species from the column, a characteristic feature of the muscle phosphorylase kinase which has a molecular size of 1.3 × 106 kDa (26, 28). When phosphorylase was replaced by tau as the substrate, only one tau kinase activity peak was observed (Fig. 2A). Furthermore, this tau kinase activity peak co-eluted with phosphorylase kinase activity peak from Superose 12 gel filtration column (Fig. 2A). When the activity assays in Fig. 2A were carried out in the presence of EGTA, both phosphorylase kinase and tau kinase activity peaks were completely suppressed (data not included). These observations suggested that there was only one tau kinase in peak 3 (Fig. 1A) whose size and requirement of Ca2+ for the activity was similar to muscle phosphorylase kinase.

To gain evidence in support of the notion that phosphorylase kinase is the Ca2+-dependent tau kinase in peak 3 (Fig. 1A), fractions containing phosphorylase kinase activity from Fig. 2A (fractions 14-20) were directly chromatographed through a hydroxylapatite column, and the effluent fractions were assayed for phosphorylase kinase and tau kinase activities. As shown in Fig. 2B, both tau kinase and phosphorylase kinase activities again co-eluted from the column. When EGTA was included in the assay, both activities were completely suppressed (data not included).

Since muscle phosphorylase kinase is a calmodulin-dependent kinase and binds calmodulin in the presence of Ca2+ (26), fractions containing phosphorylase kinase activity from Fig. 2B (fractions 12-22) were pooled, adjusted to 0.3 mM in CaCl2, and loaded onto a calmodulin-agarose affinity column. As shown in Fig. 2C, both tau kinase and phosphorylase kinase activities bound to the column and both activities co-eluted from the column with EGTA. Thus, the Ca2+- and calmodulin-dependent tau kinase in peak 3 (Fig. 1A) and phosphorylase kinase could not be separated from each other by DEAE-Sepharose, Superose 12 gel filtration, hydroxylapatite, or calmodulin-agarose affinity chromatographies. These results indicated that the Ca2+- and calmodulin-dependent tau kinase in peak 3 (Fig. 1A) is either the same or very similar to phosphorylase kinase.

Phosphorylase Kinase Phosphorylates Tau

Consistent with the previous report (43), brain phosphorylase kinase was very unstable during purification. Further attempts to purify phosphorylase kinase from Fig. 2C resulted in a drastic loss of the kinase activity, and the enzyme remained nonhomogeneous. Therefore, homogeneous phosphorylase kinase was purified from rabbit skeletal muscle and incubated with tau in the presence of other components of the phosphorylation mixture. After various time points, the product was analyzed by SDS-PAGE followed by autoradiography. As shown in Fig. 3B, a progressive increase in the phosphorylation of tau with increasing time was observed (lanes 3-6). These results suggest that tau is phosphorylated by muscle phosphorylase kinase. This phosphorylated tau displayed a reduced mobility on a SDS-PAGE (Fig. 3A, lanes 3-6). Since tau can be phosphorylated by several kinases (18-25), the following experiments were performed to confirm that the phosphorylation of tau observed in Fig. 3 was due to the action of muscle phosphorylase kinase and not that of any other contaminating kinase in the muscle phosphorylase kinase preparation.


Fig. 3. Phosphorylation of tau by muscle phosphorylase kinase. Tau was incubated with muscle phosphorylase kinase in the presence of all the rest of the components of phosphorylation mixture. After various time points, samples were withdrawn and analyzed by SDS-PAGE (7.5%) and autoradiography. A, SDS-PAGE; B, autoradiography of A. Lane 1, muscle phosphorylase kinase control (0.5 µg) incubated with all the components of the phosphorylation mixture except tau for 90 min; lane 2, tau control (5 µg) incubated with all the components of phosphorylation mixture except muscle phosphorylase kinase for 90 min; lanes 3-6, tau (5 µg each) incubated with muscle phosphorylase kinase for 15, 30, 60, and 90 min, respectively.
[View Larger Version of this Image (83K GIF file)]


Since protein kinase C phosphorylates tau (24), muscle phosphorylase kinase preparations were examined for the presence of any contaminant protein kinase C. About 15 µg of purified muscle phosphorylase kinase from various preparations were immunoblotted with a monoclonal antibody against protein kinase C. Consistent with a previous report (32), this antibody detected 10 ng of protein kinase C in the control lane but completely failed to show any cross-reactive band in the lanes containing various muscle phosphorylase kinase preparations used in this study (data not shown).

To confirm that the phosphorylation of tau in Fig. 3 is not due to contaminant A kinase present in muscle phosphorylase kinase preparation, phosphorylation of tau by muscle phosphorylase kinase was carried out in the presence of 50 µg/ml A kinase inhibitory peptide. At this concentration, the A kinase inhibitory peptide completely inhibited the phosphorylation of tau by A kinase but failed to show any effect on the phosphorylation of tau by muscle phosphorylase kinase (data not shown).

Because muscle phosphorylase kinase is a Ca2+-dependent enzyme (26), phosphorylation of tau was carried out in the presence of EGTA. With increasing concentrations of EGTA, a progressive inhibition in the phosphorylation of tau by muscle phosphorylase kinase was observed (data not included).

Nonphosphorylated muscle phosphorylase kinase displays very little activity at neutral pH but is activated upon phosphorylation by A kinase or autophosphorylation (26). When tau was phosphorylated by autophosphorylated and nonphosphorylated muscle phosphorylase kinase under identical conditions at pH 6.8, autophosphorylated muscle phosphorylase kinase displayed 3.5 times higher activity than the nonphosphorylated counterpart (data not included).

To confirm that tau is phosphorylated by muscle phosphorylase kinase, tau was incubated with the purified catalytic gamma  subunit of muscle phosphorylase kinase in the presence of all the other components of phosphorylation mixture. When the phosphorylation mixture was analyzed by SDS-PAGE followed by autoradiography, tau was found to be phosphorylated by the gamma  subunit (data not shown). These results along with the observation that all the muscle phosphorylase kinase preparations used in this study phosphorylated tau in a similar manner indicate that muscle phosphorylase kinase indeed phosphorylates tau.

Tau was phosphorylated by muscle phosphorylase kinase for 6 h as described under "Materials and Methods" except that the concentration of muscle phosphorylase kinase was 0.3 µM. When the product was analyzed by the filter paper assay (37), it was determined that 2.1 mol of phosphate/mol of tau were incorporated.

The kinetic parameters of phosphorylation of phosphorylase and tau by muscle phosphorylase kinase are shown in Table I. The Km and kcat values for the phosphorylation of tau are 6.9 µM and 47.4 min-1, respectively. Thus, the Km and kcat values for the phosphorylation by muscle phosphorylase kinase of phosphorylase are 5.5 and 45 times higher than that of tau, respectively. These observations suggest that although tau binds to muscle phosphorylase kinase better than phosphorylase, the turnover rate of tau phosphorylation is very slow compared with that of phosphorylase. Overall, phosphorylase is 8 times better substrate than tau as indicated by kcat/Km values. Other substrates of muscle phosphorylase kinase such as glycogen synthase (45, 46), neurogranin, and GAP-43 (32) were reported to be 2, 13, and 44 times poorer than phosphorylase, respectively. Thus as a substrate of muscle phosphorylase kinase, tau appears to be 4 times poorer than glycogen synthase but ~2 and ~5 times better than neurogranin and GAP-43, respectively.

Table I.

Comparison of the kinetic parameters of phosphorylation of phosphorylase and tau by muscle phosphorylase kinase

Phosphorylation of phosphorylase and tau was carried out as described under "Materials and Methods" under identical conditions except the concentration of muscle phosphorylase kinase in phosphorylase phosphorylation mixture was 10 nM and increasing concentrations of each substrate was used in the assays. After 15 min at 30 °C, reactions were terminated, and the amount of phosphate incorporated into the each substrate was determined by filter paper assay (38). Kinetic parameters were calculated from the double-reciprocal plots of the data using linear regression analysis. The values are an average of two independent determinations.
Phosphorylase
Tau
Km kcat kcat/Km Km kcat kcat/Km

µm min-1 µm min-1
38.3 ± 3 2134 ± 59.1 55.7 6.9 ± 1.1 47.4 ± 3.3 6.8

Determination of Phosphorylation Sites

Tau (0.5 mg) was phosphorylated by purified muscle phosphorylase kinase. The phosphorylated tau was trypsinized, and the resulting tryptic peptides were separated by a HPLC C18 reverse phase column. As shown in Fig. 4A, five radioactive peaks eluted from the column. The radioactive fractions from each peak were pooled, and each pool was subjected to gel filtration through a Sephadex G 25 column followed by a HPLC reverse phase chromatography as described under "Materials and Methods." From each pool, one radioactive phosphopeptide was recovered. Thus, five tryptic phosphopeptides, designated as T1, T2, T3, T4, and T5, were purified from pools I, II, III, IV, and V, respectively.


Fig. 4. HPLC of tryptic digest of phosphorylated tau. Tryptic digest of tau, previously phosphorylated by muscle phosphorylase kinase, was chromatographed through a Delta Pak 5-µm 100-Å C18 reverse phase column (Millipore) using Waters HPLC system at a flow rate of 0.5 ml/min. The column was eluted with a linear gradient of 0-30% acetonitrile in 0.1% trifluoroacetic acid in 70 min. Effluent fractions (0.5 ml each) were collected, and 10 µl from each fraction was withdrawn and counted in a liquid scintillation counter to determine amount of radioactivity in each fraction. A, distribution of radioactivity in various fractions; B, HPLC profile of tryptic digest.
[View Larger Version of this Image (29K GIF file)]


The amino acid sequence of each phosphopeptide was determined by Edman degradation using a gas phase amino acid sequencer (20). As shown in Table II, the amino acid sequence of phosphopeptide T1 is DRVQXK. The 5th residue (indicated by X) must be the phosphorylated amino acid since its PTH-derivative could not be identified and the 5th cycle released very high levels of radioactivity. Based on these data and the published sequence of longest isoform of human tau (4), phosphopeptide T1 was concluded to extend from tau residues 348 to 353 and Ser352 identified as the phosphorylation site. Similarly, phosphopeptide T2 extends from 231 to 240 of human tau (4) with Ser237 being the phosphorylation site.

Table II.

Sequence determination of 32P-labeled tryptic and thermolytic peptides

The amino acid sequence of each phosphopeptide was determined as described under "Materials and Methods." A.A., indicates PTH-amino acid identified after each cycle; Yield, indicates pmol of PTH-amino acid released after each cycle; cpm, represents amount of radioactivity released in each cycle; X, represents the amino acid whose PTH-derivative could not be identified.
Cycle T1
T2
T3
T3-Th
T4
T5
A.A. Yield cpm A.A. Yield cpm A.A. Yield cpm A.A. Yield cpm A.A. Yield cpm A.A. Yield cpm

1 D 170 1500 T 165 355 I 624 325 I 102 42 K 450 364 H 79 94
2 R 264 660 P 478 135 G 522 170 G 123 62 L 435 950 V 123 151
3 V 128 520 P 365 85 S 148 6395 X 1084 D 330 915 P 93 200
4 Q 89 398 K 310 190 T 363 1095 T 67 600 L 418 420 G 80 356
5 X 7415 S 420 145 E 362 265 E 74 146 X 3217 G 114 275
6 K 50 5304 P 231 175 N 339 190 N 50 125 N 310 2473 G 123 300
7 X 5810 L 128 185 V 150 560 X 1571
8 S 290 1220 K 38 200 Q 85 895 V 34 1479
9 A 173 410 S 72 304 Q 38 222
10 K 128 290 K 87 620 I 11 210
11 V 10 173
12 Y 7 100
13 K 5 110

As shown in Table II, the amino acid sequence of phosphopeptide T3 is IGSTENLK. The 3rd cycle of this peptide released radioactivity indicating that the 3rd residue is phosphorylated. However, the 3rd cycle also released comparatively low yield of PTH-Ser. These results indicated that phosphopeptide T3 was contaminated with another peptide. Therefore, phosphopeptide T3 was digested with thermolysin and fractionated by a HPLC C18 reverse phase column as described in the legend to Fig. 5. As shown in Fig. 5B, one phosphopeptide peak eluted from the column. This peptide was designated as T3-Th, and 92% of the radioactivity injected into the column was recovered in this phosphopeptide. The amino acid sequence of this peptide was determined to be IGXTEN, and radioactivity was released during the 3rd cycle of Edman degradation (Table II). Based on these observations and the sequence of human tau (4), phosphopeptide T3-Th was determined to extend from residues 260 to 265 of human tau, and Ser262 is the phosphorylation site.


Fig. 5. Purification of phosphopeptides T3 and T3-Th. A, purification of phosphopeptide T3. Pool III from Fig. 4 was passed through a Sephadex G25 column, and the effluent fractions containing radioactivity were combined and lyophilized. The lyophilized sample was chromatographed through a HPLC reverse phase C18 column. All chromatographic conditions were the same as in Fig. 4 except the acetonitrile gradient was 0-40% in 50 min. B, purification of phosphopeptide T3-Th. Phosphopeptide T3 from A was lyophilized, redissolved in 200 µl of 50 mM (NH4)2HCO3 (pH 8.0) containing 50 µg/ml thermolysin, and incubated for 3 h at 37 °C. The incubated sample was fractionated through a HPLC column as in A.
[View Larger Version of this Image (27K GIF file)]


The amino acid sequence of phosphopeptide T4 is KLDLXNVQSK where the 5th residue is phosphorylated. This peptide extends residues 281-290 of human tau, and Ser285 is the phosphorylation site. Similarly, phosphopeptide T5 extends from residues 299 to 311 of human tau, and Ser305 is phosphorylated. Thus muscle phosphorylase kinase phosphorylates Ser237, Ser262, Ser285, Ser305, and Ser352 of human tau.

To find out if phosphorylase kinase from muscle and brain display any difference in site specificity in phosphorylating tau, two tau species (50 µg each), one phosphorylated by muscle phosphorylase kinase and another by phosphorylase kinase partially purified from brain homogenate (Fig. 2C), were prepared as described under "Materials and Methods." Both species were digested with trypsin under identical conditions. Each of the above digested species was injected into a HPLC C18 reverse phase column and phosphopeptide maps generated. Phosphopeptide maps of both the species contained five phosphopeptide peaks as in Fig. 4A, and both maps looked identical (Fig. 6). These observations indicate that phosphorylase kinase from muscle and brain phosphorylate the same five sites within tau.


Fig. 6. HPLC phosphopeptide maps of tau phosphorylated by phosphorylase kinase from muscle and brain. Two tau species, one phosphorylated by muscle phosphorylase kinase and another by brain phosphorylase kinase, were digested with trypsin under identical conditions, and the digests were analyzed by HPLC. All HPLC conditions were the same as in Fig. 4. A, phosphopeptide map of tau phosphorylated by muscle phosphorylase kinase. B, phosphopeptide map of tau phosphorylated by brain phosphorylase kinase.
[View Larger Version of this Image (21K GIF file)]


Phosphorylase Kinase Associates with Microtubules

Microtubules are generally purified from the brain homogenates by 2-3 cycles of temperature-induced polymerization and depolymerization of microtubules in the presence of 1 mM GTP (47). Purified microtubules contain tubulins and various microtubule-associated proteins that copurify with microtubules (1, 2, 47). To test if phosphorylase kinase copurifies with microtubules, an aliquot of the purified microtubule fraction was spotted onto a nitrocellulose membrane and subjected to a dot-blot analysis using a monoclonal antibody, mAB 88, against muscle phosphorylase kinase. As shown in Fig. 7A, the antibody did not cross-react with nonspecific proteins, bovine serum albumin, tau, and phosphorylase but displayed cross-reactivity with a brain extract, the muscle phosphorylase kinase control, and the microtubule fraction. When a similar experiment was carried out using mAB 979 (another monoclonal antibody against muscle phosphorylase kinase) identical results were obtained (data not shown). These observations suggested that, in brain, phosphorylase kinase is associated with microtubules.


Fig. 7. Presence of phosphorylase kinase in purified microtubule fraction. Microtubules purified by three cycles of temperature-induced polymerization and depolymerization were probed for the presence of phosphorylase kinase by a slot-blot analysis using a monoclonal antibody, mAB 88, against muscle phosphorylase kinase and the activity assay using phosphorylase (a specific substrate of phosphorylase kinase). A, slot-blot analysis. Indicated protein samples (10 µg each) were subjected to slot-blot analysis. B, SDS-PAGE of phosphorylase kinase assay. Phosphorylase was incubated in a vial with either brain homogenate or purified microtubule fraction in the presence of all the components of phosphorylation mixture. After 30 min at 30 °C, a 20-µl aliquot was withdrawn from each vial and subjected to SDS-PAGE and autoradiography. The final concentrations of various components of phosphorylation assay were 50 mM Tris-HCl (pH 8.2), 0.2 mM EGTA, 0.1 mM EDTA, 0.5 mM DTT, 0.5 mM CaCl2, 10 mM MgCl2, 0.2 mM [gamma -32P]ATP, 0.5 mg/ml phosphorylase, and 0.5 mg/ml protein from purified microtubule fraction or brain homogenate. Lane 1, phosphorylase control (10 µg) incubated with all the components of phosphorylation mixture except microtubule fraction or brain homogenate; lane 2, brain homogenate control incubated with all the components of phosphorylation mixture except phosphorylase; lane 3, microtubule fraction control incubated with all the components of phosphorylation mixture except phosphorylase; lane 4, phosphorylase (10 µg) incubated with microtubule fraction in the presence of the rest of the components of phosphorylation mixture; lane 5, phosphorylase (10 µg) incubated with brain homogenate in the presence of rest of the components of phosphorylation mixture. C, autoradiography of B.
[View Larger Version of this Image (47K GIF file)]


To confirm the suggestion that brain phosphorylase kinase associates and copurifies with microtubules, an aliquot from a purified microtubule fraction was transferred to a vial containing all the components of the phosphorylation mixture, including phosphorylase. After 30 min at 30 °C, samples were analyzed by SDS-PAGE followed by autoradiography for phosphorylase kinase activity. As shown in Fig. 7C, phosphorylase incubated with the microtubule fraction (lane 4) and brain extract (lane 5) is radiolabeled. There is no radioactivity in the phosphorylase control that was incubated with all the components of phosphorylation mixture except the microtubule fraction (lane 1). Similarly, no radioactive band comigrating with phosphorylase is present in the control microtubule fraction incubated with all the components of phosphorylation mixture except phosphorylase (lane 3). These observations indicate that a kinase capable of phosphorylating phosphorylase is associated with purified microtubules. Since phosphorylase kinase is the only enzyme so far known to phosphorylate phosphorylase (26, 49), it is concluded that the brain phosphorylase kinase associates with microtubules.


DISCUSSION

Elevated intracellular Ca2+ by glutamate or Ca2+ ionophores elicits the phosphorylation of tau in rat hippocampal (50) and human cortical (51) neurons. These observations suggest that increased Ca2+ level in neurons activates tau kinases. In vitro, two Ca2+-dependent kinases, protein kinase C and Cam kinase II, phosphorylate tau on Ser324 and Ser416, respectively (24, 25). Since both these sites are not phosphorylated in vivo (6-8, 17), protein kinase C and Cam kinase II may not directly phosphorylate tau in neurons.

Phosphorylase kinase is a Ca2+-dependent enzyme that regulates glycogenolysis by phosphorylating and activating phosphorylase (26). In muscle, where glycogen is abundant, phosphorylase kinase is expressed in high levels. In brain, however, despite a very low glycogen reserve, the activity ratio of phosphorylase kinase/phosphorylase is very high and is highest among all the tissues examined, including muscle (43). In vitro, phosphorylase kinase phosphorylates two neuronal proteins, GAP-43 and neurogranin (32). Phosphorylation of these two proteins is implicated in various neuronal functions, such as neurotransmitter release and the induction and maintenance of long-term potentiation (52), a form of neuronal plasticity that may be involved in the mammalian learning and memory. A 40-kDa protein prepared from the rat brain synaptic plasma membrane, whose phosphorylation is enhanced during electric stimulation of rat hippocampal tissue that induces long-term potentiation (53), is phosphorylated by phosphorylase kinase (54). Phosphorylase kinase is activated in the mouse brain cerebral cortex during seizures (55), anoxia (56), hypoglycemia (56), on electric stimulation of cervical vagus nerve (57) and treatment of the brain slices with neurotransmitters (58). These observations have raised the possibility that the brain phosphorylase kinase in addition to the phosphorylation of phosphorylase might be involved in the other neuronal functions.

In this study, a Ca2+-dependent tau kinase and phosphorylase kinase from a brain homogenate co-eluted throughout four different chromatographic procedures that operate by different principles (Figs. 1 and 2). Purified muscle phosphorylase kinase phosphorylated tau and converted it to a species with a retarded mobility on SDS-PAGE (Fig. 3A), a characteristic feature of highly phosphorylated tau (13). Among the five sites phosphorylated by muscle phosphorylase kinase in vitro, Ser262 is also phosphorylated in vivo (8, 17). Phosphorylase kinase is highly expressed in neurons (32, 43) and copurifies with microtubules purified from brain homogenates by temperature-induced repeated polymerization end depolymerization (Fig. 7). These observations, together, suggest that in neurons phosphorylase kinase participates in the phosphorylation of tau and regulation of microtubule structure and dynamics.

There are four microtubule-binding repeats in the longest isoform of human tau (3). These microtubule-binding repeats bind microtubules in a noncooperative manner, and repeat 1 has a 100-fold higher affinity for microtubules than repeats 2, 3, or 4 (5). It was suggested that tau binds to microtubules firmly through repeat 1 and weakly through repeats 2-4 (5). Interestingly, among all the phosphorylation sites of tau, Ser262 is the only site that is uniquely located within the first microtubule-binding repeat (17). Since tau is thought to bind to microtubules through electrostatic interaction between the positively charged microtubule-binding repeats and the negatively charged carboxyl region of tubulins (1, 5), the phosphorylation of Ser262 is very likely to interfere with the binding of tau to microtubules. Indeed, in vitro phosphorylation of tau on Ser262 alone strongly reduced the ability of tau to bind microtubules, whereas the phosphorylation of many Ser/Thr-Pro motif sites of tau showed moderate effects on the binding of tau to microtubules (10, 11). Furthermore, since tau binding makes microtubules dynamically stable, phosphorylation/dephosphorylation of tau residue Ser262 was suggested to play a critical role in the regulation of microtubule dynamics (10, 11)

In this study, phosphorylase kinase was found to phosphorylate Ser262 in vitro. It is possible that phosphorylase kinase phosphorylates tau on Ser262 in vivo also. Since Ca2+ activates phosphorylase kinase in brain (43), an increase in the intracellular Ca2+ would therefore activate phosphorylase kinase leading to phosphorylation of tau on Ser262 and destabilization of microtubules. Currently, it is not known if tau is phosphorylated on Ser262 in response to a Ca2+ signal in neurons. However, increased level of intracellular Ca2+ in cultured neurons leads to phosphorylation of tau (50, 51), altered neuronal cytoskeleton (59), and loss of microtubules (51).

PHF-tau that does not bind microtubules and thus is functionally inactive regains its ability to bind and promote microtubule assembly upon dephosphorylation (14-16). It was suggested that in AD brain, abnormal phosphorylation makes tau incapable of binding to microtubules leading to cytoskeletal dysfunction and neurodegeneration (13, 60). Furthermore, phosphorylation at Ser262 was implicated to play a major role in making PHF-tau functionally inactive (10, 11). The cause of the abnormal phosphorylation of tau in AD brain is not known, but an aberrant activation of the cellular signal transduction pathway leading to the activation of tau kinases was proposed (61, 62). It was suggested that the increased intracellular Ca2+ level in AD brain activates tau kinases leading to the hyperphosphorylation of tau (51, 61, 63). In this study, phosphorylase kinase, a Ca2+-dependent enzyme, was found to phosphorylate tau on Ser262, whose phosphorylation abolishes tau's binding to microtubules (10). Thus, phosphorylation of tau by phosphorylase kinase very likely prevents tau from binding to microtubules making these microtubules unstable, a pathological condition that has been observed in AD brain (60). It is not known if phosphorylase kinase is aberrantly activated in AD brain. However, muscle phosphorylase kinase is activated by diverse mechanisms including proteolysis by a Ca2+-dependent protease (26, 64) whose activity was found to be elevated in AD brain (65). Furthermore, studies have demonstrated that the brains of AD patients, which are under energy stress (66), contain increased levels of adenosine diphosphate (ADP), compared with normal controls (67). In vitro, ADP is a potent activator of muscle phosphorylase kinase and was suggested to be an allosteric activator of the kinase (68).

In this study, a DEAE-Sepharose resolved five tau kinase activity peaks in the porcine brain homogenate (Fig. 1A). Among these, peaks 3 and 5 were identified as phosphorylase kinase and Cam kinase II, respectively. Peak 1 was inhibited by protein kinase A inhibitory peptide and phosphorylated kemptide, a relatively specific synthetic peptide substrate of protein kinase A (69, 70). These observations suggest that peak 1 corresponds to protein kinase A. Similarly, when tau was replaced by a protein kinase C synthetic peptide substrate, derived from epidermal growth factor receptor (71), and various fractions in Fig. 1A assayed, a single activity peak that did not co-elute with any of the peaks in Fig. 1A, but co-eluted with peak a syntide 2 kinase activity in Fig. 1B, was observed (data not shown). Since syntide 2 is phosphorylated by protein kinase C (42) and by peak a (Fig. 1B), peak a is likely to be protein kinase C. It should be noted that phospholipids that are required for the full activity of protein kinase C (71) were not included in the activity assays in this study. Therefore, it is possible that protein kinase C may have displayed a very low phosphotransferase activity against tau that could not have been detected in Fig. 1A. Nevertheless, these observations suggested that none of the activity peaks in Fig. 1A are likely to correspond to protein kinase C. Peaks 2 and 4 in Fig. 1A remained unidentified. Since both unidentified peaks are insensitive to Ca2+, they correspond to Ca2+-independent kinases. Among Ca2+-independent kinases, casein kinase I (72) and casein kinase II (73) were recently reported to phosphorylate tau in vitro. 35/41-kDa kinase (11) and 116-kDa kinase (10) activities in brain extract capable of phosphorylating tau were also reported. It is possible that some of these kinase are responsible for peaks 2 and 4 in Fig. 1A. It is also possible that one or both of these peaks contain novel tau kinases. Further studies are being carried out to identify tau kinases in these two peaks.


FOOTNOTES

*   This work was supported by Grant MT-13352 from Medical Research Council of Canada and Grant 96-07 from the Alzheimer's Association of Canada. 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.
Dagger    Recipient of Fraser, Monat and McPherson Scholarship from McGill University. To whom correspondence should be addressed: Bloomfield Center for Research in Aging, Lady Davis Institute for Medical Research, 3755 Cote Ste-Catherine, Montreal, Quebec, Canada H3T 1E2. Tel.: 514-340-8222 (ext. 4866); Fax: 514-340-8295.
1    The abbreviations used are: PHF, paired helical filament; AD, Alzheimer's disease; NCLK, neuronal cdc2-like protein kinase; PTH, phenylthiohydantoin; DTT, dithiothreitol; PAGE, polyacrylamide gel electrophoresis; Map, mitogen-activated protein; Cam, calmodulin; mAB, monoclonal antibody; HPLC, high performance liquid chromatography.

REFERENCES

  1. Hirokawa, N. (1994) Curr. Opin. Cell Biol. 6, 74-81 [Medline] [Order article via Infotrieve]
  2. Matus, A. (1988) Annu. Rev. Neurosci. 11, 29-44 [CrossRef][Medline] [Order article via Infotrieve]
  3. Goedert, M., Crowther, R. A., and Garner, C. C. (1991) Trends Neurosci. 14, 193-199 [CrossRef][Medline] [Order article via Infotrieve]
  4. Goedert, M., Spillantini, M. G., Cairns, N. J., and Crowther, R. A. (1992) Neuron 8, 159-168 [Medline] [Order article via Infotrieve]
  5. Butner, K. A., and Kirschner, M. W. (1991) J. Cell Biol. 115, 717-730 [Abstract]
  6. Watanabe, A., Hasegawa, M., Suzuki, M., Takio, K., Morishima-Kawashima, M., Titani, K., Arai, T., Kosik, K. S., and Ihara, Y. (1993) J. Biol. Chem. 268, 25712-25717 [Abstract/Free Full Text]
  7. Matsuo, E. S., Shin, R.-W., Billingsley, M. L., DeVoorde, A. V., O'Connor, M., Trojanowski, J. Q., and Lee, V. M.-Y. (1994) Neuron 13, 989-1002 [Medline] [Order article via Infotrieve]
  8. Seubert, P., Mawal-Dewan, M., Barbour, R., Jakes, R., Goedert, M., Johnson, G. V. W., Litersky, J. M., Schenk, D., Lieberburg, I., Trojanowski, J. Q., and Lee, V. M.-Y. (1995) J. Biol. Chem. 270, 18917-18922 [Abstract/Free Full Text]
  9. Drechsel, D. N., Hyman, A. A., Cobb, M. H., and Kirschner, M. W. (1992) Mol. Biol. Cell 3, 1141-1154 [Abstract]
  10. Drewes, G., Trinczek, B., Illenberger, S., Biernat, J., Schmitt-Ulms, G., Meyer, H. E., Mandelkow, E.-M., and Mandelkow, E. (1995) J. Biol. Chem. 270, 7679-7688 [Abstract/Free Full Text]
  11. Biernat, J., Gustke, N., Drewes, G., Mandelkow, E.-M., and Mandelkow, E. (1993) Neuron 11, 153-163 [Medline] [Order article via Infotrieve]
  12. Matus, A. (1994) Trends Neurosci. 17, 19-22 [CrossRef][Medline] [Order article via Infotrieve]
  13. Goedert, M. (1993) Trends Neurosci. 16, 460-465 [CrossRef][Medline] [Order article via Infotrieve]
  14. Bramblett, G. T., Goedert, M., Jakes, R., Merrick, S. E., Trojanowski, J. Q., and Lee, V. M. Y. (1993) Neuron 10, 1089-1099 [Medline] [Order article via Infotrieve]
  15. Alonso, A. D. C., Zaidi, T., Grundke-Iqbal, I., and Iqbal, K. (1994) Proc. Natl. Acad. Sci. U. S. A. 91, 5562-5566 [Abstract]
  16. Wang, J.-Z., Gong, C.-X., Zaidi, T., Grundke-Iqbal, I., and Iqbal, K. (1995) J. Biol. Chem. 270, 4854-4860 [Abstract/Free Full Text]
  17. Morishima-Kawashima, M., Hasegawa, M., Takio, K., Suzuki, M., Yoshida, H., Titani, K., and Ihara, Y. (1995) J. Biol. Chem. 270, 823-829 [Abstract/Free Full Text]
  18. Vulliet, R., Halloram, S. M., Braun, R. K., Smith, A. J., and Lee, G. (1992) J. Biol. Chem. 267, 22570-22574 [Abstract/Free Full Text]
  19. Mawal-Dewan, M., Sen, P. C., Abdel-Ghany, M., Shalloway, D., and Racker, E. (1992) J. Biol. Chem. 267, 19705-19709 [Abstract/Free Full Text]
  20. Paudel, H. K., Lew, J., Ali, Z., and Wang, J. H. (1993) J. Biol. Chem. 268, 23512-23518 [Abstract/Free Full Text]
  21. Drewes, G., Lichtenberg-Kraag, B., Döring, F., Mandelkow, E.-M., Biernat, J., Goris, J., Dorée, M., and Mandelkow, E. (1992) EMBO J. 11, 2131-2138 [Abstract]
  22. Mandelkow, E.-M., Drewes, G., Biernat, J., Gustke, N., Lint, J. V., Vandenheede, J. R., and Mandelkow, E. (1992) FEBS Lett. 314, 315-321 [CrossRef][Medline] [Order article via Infotrieve]
  23. Scott, C. W., Spreen, R. C., Herman, J. L., Chow, F. P., Davison, M. D., Young, J., and Caputo, C. B. (1993) J. Biol. Chem. 268, 1166-1173 [Abstract/Free Full Text]
  24. Correas, I., Diaz-Nido, J., and Avila, J. (1992) J. Biol. Chem. 267, 15721-15728 [Abstract/Free Full Text]
  25. Steiner, B., Mandelkow, E.-M., Biernat, J., Gustke, N., Meyer, H. E., Schmidt, B., Mieskes, G., Soling, H. D., Drechsel, D., Mirschner, M. W., Goedert, M., and Mandelkow, E. (1990) EMBO J. 9, 3539-3544 [Abstract]
  26. Pickett-Gies, C. A., and Walsh, D. A. (1986) in The Enzymes (Boyer, P. D., and Krebs, E. G., eds), Vol. 17, pp. 395-459, Academic Press, Orlando, FL
  27. Crowther, R. A., Olesen, O. F., Smith, M. J., Jakes, R., and Goedert, M. (1994) FEBS Lett. 337, 135-138 [CrossRef][Medline] [Order article via Infotrieve]
  28. Cohen, P. (1973) Eur. J. Biochem. 34, 1-14 [Medline] [Order article via Infotrieve]
  29. Paudel, H. K., and Carlson, G. M. (1991) J. Biol. Chem. 266, 16524-16529 [Abstract/Free Full Text]
  30. Paudel, H. K., and Carlson, G. M. (1987) J. Biol. Chem. 262, 11912-11915 [Abstract/Free Full Text]
  31. Fischer, E. H., and Krebs, E. G. (1958) J. Biol. Chem. 231, 65-71 [Free Full Text]
  32. Paudel, H. K., Zwiers, H., and Wang, J. H. (1993) J. Biol. Chem. 268, 6207-6213 [Abstract/Free Full Text]
  33. Clark-Lewis, I., Sanghera, J. S., and Pelech, S. L. (1991) J. Biol. Chem. 266, 15180-15184 [Abstract/Free Full Text]
  34. Maulet, Y., and Cox, J. A. (1983) Biochemistry 22, 5680-5686 [Medline] [Order article via Infotrieve]
  35. Cleveland, D. W., Hwo, S. Y., and Kirschner, M. W. (1977) J. Mol. Biol. 116, 227-247 [Medline] [Order article via Infotrieve]
  36. Kastenschmidt, L. L., Kastenschmidt, J., and Helmreich, E. (1968) Biochemistry 7, 3590-3608 [Medline] [Order article via Infotrieve]
  37. Roskoski, R., Jr. (1983) Methods Enzymol. 99, 3-6 [Medline] [Order article via Infotrieve]
  38. Laemmli, U. K. (1970) Nature 227, 680-685 [Medline] [Order article via Infotrieve]
  39. Lew, J., Beaudette, K., Litwin, C. M. E., and Wang, J. H. (1992) J. Biol. Chem. 267, 13383-13390 [Abstract/Free Full Text]
  40. Klee, C. B., and Vanaman, T. C. (1982) Adv. Protein Chem. 35, 213-321 [Medline] [Order article via Infotrieve]
  41. Pearson, R. B., Woodgett, J. R., Cohen, P., and Krebs, E. G. (1985) J. Biol. Chem. 260, 14471-14476 [Abstract/Free Full Text]
  42. Hashimoto, Y., and Soderling, T. R. (1987) Arch. Biochem. Biophys. 252, 418-425 [Medline] [Order article via Infotrieve]
  43. Drummond, G. I., and Bellward, G. (1970) J. Neurochem. 17, 475-482 [Medline] [Order article via Infotrieve]
  44. Bollen, M., Kee, S. M., Graves, D. J., and Soderling, T. R. (1987) Arch. Biochem. Biophys. 254, 437-447 [Medline] [Order article via Infotrieve]
  45. Soderling, T. R., Srivastava, A. K., Bass, M. A., and Khatra, B. S. (1979) Proc. Natl. Acad. Sci. U. S. A. 76, 2536-2540 [Abstract]
  46. DePaoli-Roach, A. A., Roach, P. J., and Larner, J. (1979) J. Biol. Chem. 254, 4212-4219 [Medline] [Order article via Infotrieve]
  47. Vallee, R. B. (1986) Methods Enzymol. 134, 89-104 [Medline] [Order article via Infotrieve]
  48. Deleted in proofDeleted in proof
  49. Madsen, N. B. (1986) in The Enzymes (Boyer, P. D., and Krebs, E. G., eds), Vol. 17, pp. 365-394, Academic Press, Orlando, FL
  50. Mattson, M. P. (1990) Neuron 2, 105-117
  51. Mattson, M. P., Engle, M. G., and Rychlik, B. (1991) Mol. Chem. Neuropathol. 15, 117-142 [Medline] [Order article via Infotrieve]
  52. Coggins, P. J., and Zwiers, H. (1991) J. Neurochem. 56, 1095-1105 [Medline] [Order article via Infotrieve]
  53. Browning, M., Bennett, W., and Lynch, G. (1979) Nature 278, 273-275 [Medline] [Order article via Infotrieve]
  54. Browning, M., Dunwiddie, T., Nette, W., Gispen, W., and Lynch, G. (1979) Science 203, 60-62 [Medline] [Order article via Infotrieve]
  55. Folbergrova, J. (1977) Brain Res. 135, 337-346 [Medline] [Order article via Infotrieve]
  56. Breckenridge, B. McL., and Norman, J. H. (1965) J. Neurochem. 12, 51-57 [Medline] [Order article via Infotrieve]
  57. Landowne, D., and Ritchie, J. M. (1971) J. Physiol. (Lond.) 212, 503-517 [Medline] [Order article via Infotrieve]
  58. Edwards, C., Nahorski, S. R., and Roger, K. J. (1974) J. Neurochem. 22, 565-572 [Medline] [Order article via Infotrieve]
  59. Kater, S. B., Mattson, M. P., Cohan, C. S., and Connor, J. A. (1988) Trends Neurosci. 11, 315-321 [CrossRef][Medline] [Order article via Infotrieve]
  60. Lee, V. M.-Y., and Trojanowski, J. Q. (1992) Curr. Opin. Cell Biol. 2, 653-656
  61. Saitoh, T., Horsburgh, K., and Masliah, E. (1993) Ann. N. Y. Acad. Sci. 695, 34-41 [Abstract]
  62. Pelech, S. L. (1995) Neurobiol. Aging 16, 247-256 [CrossRef][Medline] [Order article via Infotrieve]
  63. Mattson, M. P. (1994) Ann. N. Y. Acad. Sci. 747, 50-76 [Abstract]
  64. Meyer, W. L., Fischer, E. H., and Krebs, E. G. (1964) Biochemistry 3, 1033-1039
  65. Saito, K. I., Elce, J. S., Hames, J. E., and Nixon, R. A. (1993) Proc. Natl. Acad. Sci. U. S. A. 90, 2628-2632 [Abstract]
  66. Beal, M. F. (1992) Ann. Neurol. 31, 119-130 [Medline] [Order article via Infotrieve]
  67. Pettegrew, J. W., Panchalingam, K., Klunk, W. E., McClure, R. J., and Muenz, L. R. (1994) Neurobiol. Aging 15, 117-132 [Medline] [Order article via Infotrieve]
  68. Cheng, A., Fitzgerald, T. J., and Carlson, G. M. (1985) J. Biol. Chem. 260, 2535-2542 [Abstract]
  69. Maller, J. L., Kemp, B. E., and Krebs, E. G. (1978) Proc. Natl. Acad. Sci. U. S. A. 75, 248-251 [Abstract]
  70. Pearson, R. B., and Kemp, B. E. (1991) Methods Enzymol. 200, 62-81 [Medline] [Order article via Infotrieve]
  71. Heasley, L. E., and Johnson, G. L. (1989) J. Biol. Chem. 264, 8646-8652 [Abstract/Free Full Text]
  72. Singh, T. J., Grundke-Iqbal, I., and Iqbal, K. (1995) J. Neurochem. 64, 1420-1423 [Medline] [Order article via Infotrieve]
  73. Greenwood, J. A., Scott, C. W., Spreen, R. C., Caputo, C. B., and Johnson, G. V. W. (1994) J. Biol. Chem. 269, 4373-4380 [Abstract/Free Full Text]

©1997 by The American Society for Biochemistry and Molecular Biology, Inc.