The order of exposure of tau to signal transduction kinases alters the generation of "AD-like" phosphoepitopes.

1. The individual and sequential influence of protein kinase C (PKC), protein kinase A (PKA) and mitogen-activated protein kinase (MAP kinase) on human brain tau was examined. 2. A range of PKC concentrations generated certain phosphoepitopes common with paired helical filaments. These epitopes were masked by higher PKC concentrations, suggesting the presence of multiple tau phosphorylation sites for which PKC exhibited differing affinities and/or conformational alterations in tau induced by sequential PKC-mediated phosphorylation. 3. Prior phosphorylation by PKC enhanced the nature and extent of AD-like tau antigenicity generated by subsequent incubation with MAP kinase yet inhibited that generated by subsequent incubation with PKA. 4. Dephosphorylation of tau prior to incubation with kinases significantly altered the influence of individual and multiple kinase incubation on tau antigenicity in a site-specific manner, indicating that prior in situ phosphorylation events markedly influenced subsequent cell-free phosphorylation. 5. In addition to considerations of the potential impact of tau phosphorylation by individual kinases, these findings extend previous studies which indicate that tau antigenicity, and, presumably, its behavior in situ, is influenced by the sequential and convergent influences of multiple kinases.

A 26-30 kDa developmentally-regulated tau isoform localized within nuclei of mitotic human neuroblastoma cells.

Tau isoforms migrating at 46-68 and 97-115 kDa were prominent within heat-stable Triton-soluble material, and were present in lesser concentration with Triton-insoluble cytoskeletons, derived from undifferentiated SH-SY-5Y human neuroblastoma cells. Conversely, a 26-30 kDa tau isoform was enriched in the cytoskeleton and detected at relatively minor levels within cytosolic fractions. Pulse labeling with 35S-methionine indicated that this 26-30 kDa "small tau" did not represent a breakdown product of larger isoforms. Since the nucleus is retained within the Triton-insoluble cytoskeleton, additional cultures were fractionated onto sucrose to obtain purified nuclei. The vast majority of small tau was recovered within purified nuclei. Small tau was reactive with tau antibodies directed towards N-terminal, C-terminal and central epitopes, further confirming that this small isoform was not derived from proteolytic cleavage of larger tau isoforms. Small tau demonstrated alkaline phosphatase-sensitive reactivity with multiple phospho-dependent tau antibodies. Small tau was depleted within 3 days of retinoic acid-induced differentiation, suggesting that the putative function of this isoform may be obsolete following terminal differentiation of neurons.

Calcium influx into human neuroblastoma cells induces ALZ-50 immunoreactivity: involvement of calpain-mediated hydrolysis of protein kinase C.

Calcium influx into SH-SY5Y human neuroblastoma cells after ionophore treatment or transient permeabilization in calcium-containing medium increased ALZ-50 immunoreactivity markedly. This increase was prevented by inhibitors active against calpain or against protein kinase C (PKC), suggesting that both of these enzymes were required to mediate the effect of calcium influx on ALZ-50 immunoreactivity. Treatment with PKC activator TPA increased ALZ-50 immunoreactivity in the absence of calcium influx or after intracellular delivery of the specific calpain inhibitor calpastatin, indicating that the influence of PKC was downstream from that of calpain. Calcium influx also resulted in mu-calpain autolysis (one index of calpain activation) and the transient appearance of PKM (i.e., free PKC catalytic subunits, generated by calpain-mediated cleavage of the regulatory and catalytic PKC domains). Inhibition of calpain within intact cells resulted in a dramatic increase in steady-state levels of total tau (migrating at 46-52 kDa) but resulted in a relatively minor increase in 68-kDa ALZ-50-immunoreactive tau isoforms. Although calcium influx into intact cells resulted in accumulation of ALZ-50 immunoreactivity, total tau levels were, by contrast, rapidly depleted. Incubation of isolated fractions with calpain in the presence of calcium indicated that ALZ-50-immunoreactive tau isoforms were more resistant to calpain-mediated proteolysis than were non-ALZ-50 reactive tau isoforms. These data therefore indicate that calpain may regulate tau levels directly via proteolysis and indirectly through PKC activation. A consequence of the latter action is altered tau phosphorylation, perhaps involving one or more kinase cascades, and the preferential accumulation of ALZ-50-immunoreactive tau isoforms due to their relative resistance to degradation. These findings provide a basis for the possibility that disregulation of calcium homeostasis may contribute to the pathological levels of conversion of tau to A68 by hyperactivation of the calpain/PKC system.

Hyperphosphorylation of Tau and filopodial retraction following microinjection of protein kinase C catalytic subunits.

Limited proteolysis of protein kinase C (PKC) by calcium-activated proteolysis cleaves the regulatory and catalytic subunits of PKC, generating a free, constitutively activated kinase ("PKM") that, unlike the intact parent enzyme, is not calcium-dependent, and is not restricted to the plasma membrane. These latter properties leave open the possibility that PKM may have access to, and may therefore phosphorylate, substrates normally unavailable to intact PKC. We examined the potential involvement of such aberrant phosphorylation in certain aspects of the neurodegeneration accompanying Alzheimer's disease by microinjecting PKC and PKM, along with a rhodamine-conjugated dextran tracer, into undifferentiated NB2a/d1 mouse neuroblastoma cells. After 4 hr, cultures were fixed and processed for immunofluorescence with monoclonal antibodies (PHF-1, ALZ-50, Tau-1, AT8) directed against tau in various phosphorylation states followed by fluorescein-conjugated secondary antibodies. Microinjected cells were localized via co-injected rhodamine-conjugated dextran tracer under rhodamine illumination, after which antibody immunoreactivity was examined under fluorescein illumination. Microdensitometric analyses indicated that microinjection of PKC did not increase basal immunofluorescent intensities of the antibodies; by contrast, microinjection of PKM induced three- and twofold increases in PHF-1 and ALZ-50 levels, respectively. By contrast, no significant alteration was observed in AT8 and Tau-1 immunofluorescence following either PKC or PKM microinjection. Whereas undifferentiated NB2a/d1 cells typically elaborate short, filopodia-like neurites, phase-contrast microscopy revealed the absence of filopodia or neurites on PKM-injected cells, while a similar percentage of PKC-injected cells. Cell-free analyses confirmed the ability of PKC, in the presence of necessary co-factors, and PKM to increase PHF-1 and ALZ-50 immunoreactivity; no change was observed in AT8 or Tau-1 immunoreactivity. These findings underscore the possibility that an abnormal amplification in limited PKC proteolysis to generate PKM could, under certain pathological conditions, contribute to neuronal degeneration.

Phospholipids inhibit proteolysis of protein kinase C alpha by mM calcium-requiring calpain.

The alpha isoform of protein kinase C (PKC alpha) is rapidly hydrolyzed by mM Ca(2+)-requiring calpain (calcium-activated neutral proteinase) under cell-free conditions (Shea et al, 1994, FEBS Lett. 350:223). In the present study, we demonstrate that this hydrolysis is inhibited by phosphatidyl serine, diacylglycerol, phosphatidyl choline, phosphatidyl inositol, and phosphatidic acid. With the exception of phosphatidic acid, these phospholipids did not directly inhibit calpain activity as evidenced by degradation of [14C]azocasein, suggesting that the nature of inhibition of calpain-mediated PKC alpha degradation is due to an effect of phospholipids on PKC alpha conformation. These findings suggest that m calpain-mediated PKC alpha hydrolysis may be specifically minimized at the plasma membrane, and leave open the possibility that such a mechanism exists in situ. In addition, the unique inhibition of calpain activity by phosphatidic acid suggests the existence of a specific mechanism by which this phospholipid regulates PKC alpha activity.

Enhancement of neurite outgrowth following calpain inhibition is mediated by protein kinase C.

We examined the interdependence of calpain and protein kinase C (PKC) activities on neurite outgrowth in SH-SY-5Y human neuroblastoma cells. SH-SY-5Y cells elaborated neurites when deprived of serum or after a specific thrombin inhibitor, hirudin, was added to serum-containing medium. The extent of neurite outgrowth under these conditions was enhanced by treatment of cells with the cell-permeant cysteine protease inhibitors N-acetyl-leucyl-leucyl-norleucinal ("C1") and calpeptin or by the phospholipid-mediated intracellular delivery of either a recombinant peptide corresponding to a conserved inhibitory sequence of human calpastatin or a neutralizing anti-calpain antisera. Calpain inhibition in intact cells was confirmed by immunoblot analysis showing inhibition of calpain autolysis and reduced proteolysis of the known calpain substrates fodrin and microtubule-associated protein 1. The above inhibitory peptides and antiserum did not induce neurites in medium containing serum but lacking hirudin, suggesting that increased surface protein adhesiveness is a prerequisite for enhancement of neurite outgrowth by calpain inhibition. Treatment of cells with the PKC inhibitor H7, staurosporine, or sphingosine induced neurite outgrowth independently of serum concentration. Because calpain is thought to regulate PKC activity, we examined this potential interrelationship during neurite outgrowth. Simultaneous treatment with calpain and PKC inhibitors did not produce additive or synergistic effects on neurite outgrowth. PKC activation by 2-O-tetradecanoylphorbol 13-acetate (TPA) prevented and reversed both neurite initiation by serum deprivation and its enhancement by calpain inhibitors. Treatment of cells with the calpain inhibitor C1 retarded PKC down-regulation following TPA treatment. Cell-free analyses demonstrated the relative specificity of various protease and kinase inhibitors for calpain and PKC and confirmed the ability of millimolar calcium-requiring calpain to cleave the SH-SY-5Y PKC regulatory subunit from the catalytic subunit, yielding a free catalytic subunit (protein kinase M). These findings suggest that the influence of PKC on neurite outgrowth is downstream from that of surface adhesiveness and calpain activity.

Calcium influx recruits an additional class of kinases to hyperphosphorylate tau.

SH-SY-5Y human neuroblastoma cells were treated with combinations of the kinase inhibitors HA-1004, W-7 and H-7 and calcium ionophore A23187. Microdensitometric analyses revealed that, in the absence of ionophore-mediated calcium influx, PHF-1 levels were reduced by approximately half in cultures treated with HA-1004 or W-7, but were not reduced by H-7. By contrast, the doubling in PHF-1 immunoreactivity that resulted following ionophore treatment was prevented by all three inhibitors. These analyses demonstrate the recruitment of an additional kinase or kinases in tau phosphorylation following calcium influx, and underscore the possibility that de novo hyperactivation of calcium-dependent kinases may be involved in the early events that propagate PHF formation.

Proteolysis of protein kinase C: mM and microM calcium-requiring calpains have different abilities to generate, and degrade the free catalytic subunit, protein kinase M.

Limited proteolysis of protein kinase C (PKC) by calpain under cell free conditions cleaves the regulatory and catalytic PKC subunits, generating a free, co-factor independent catalytic subunit, termed PKM. In the present study, we demonstrate distinct differences in the rate, nature, and lipid-sensitivity of PKC and PKM proteolysis by microM and mM calcium-requiring calpain isozymes (mu calpain or m calpain, respectively). PKC is a preferred substrate for m calpain; not even a 100-fold increase in mu calpain was capable of degrading PKC as fast as in calpain. PKM was generated by both m and mu calpains, but was itself rapidly degraded by m calpain and therefore was only transiently detectable. By contrast, PKM was formed but not degraded by mu calpain, and persisted in the presence of mu calpain long after all PKC had been degraded. Phosphatidyl serine (PS) inhibited PKC hydrolysis by m calpain yet enhanced PKC hydrolysis by mu calpain. The ability of either calpain isoenzyme to degrade [14C]azocasein was unaffected by PS, suggesting that the influence of PS was on PKC conformation. These findings point towards distinct roles for mu and m calpain in PKC regulation.