A Monoclonal Antibody to Amyloid Precursor ... - Wiley Online Library

Journal of Neurochemistry Lippincott Williams & Wilkins, Inc., Philadelphia © 2000 International Society for Neurochemistry

A Monoclonal Antibody to Amyloid Precursor Protein Induces Neuronal Apoptosis Troy T. Rohn, Kathryn J. Ivins, *Ben A. Bahr, Carl W. Cotman, and David H. Cribbs Institute for Brain Aging and Dementia, Department of Neurology, University of California, Irvine, California, and *Department of Pharmaceutical Sciences and Neurosciences Program, University of Connecticut, Storrs, Connecticut, U.S.A.

precursor protein—22C11—Cortical neurons—Alzheimer’s disease. J. Neurochem. 74, 2331–2342 (2000).

Abstract: Although there is considerable evidence suggesting that altered metabolism of ␤-amyloid precursor protein (APP) and accumulation of its ␤-amyloid fragment are key features of Alzheimer’s disease (AD), the normal physiological function of APP remains elusive. We investigated the potential role of APP in neurons using the monoclonal antibody 22C11, which binds to the extracellular domain of the human, rat, or mouse APP. Exposure of cortical neurons to 22C11 induced morphological changes including neurite degeneration, nuclear condensation, and internucleosomal DNA cleavage that were consistent with neurons dying by apoptosis. Supporting a role for 22C11-mediated apoptosis occurring by binding to APP were data demonstrating that preincubation of 22C11 with either purified APP or a synthetic peptide (APP66 – 81) that contains the epitope for 22C11 significantly attenuated neuronal damage induced by 22C11. The specificity of 22C11 was further supported by data showing no apparent effects of either mouse IgG or the monoclonal antibody P2-1, which is specific for the aminoterminal end of human but not rat APP. In addition, biochemical features indicative of apoptosis were the formation of 120- and 150-kDa breakdown products of fodrin following treatment of cortical neurons with 22C11. Both the morphological and the biochemical changes induced by 22C11 were prevented following pretreatment of neurons with the general caspase inhibitor N-benzyloxycarbonyl-Val-Ala-Asp(O-methyl)-fluoromethyl ketone. Prior incubation of cortical neurons with GSH ethyl ester (GEE), a cell-permeable form of GSH, resulted in complete protection from the 22C11 insult, thus implicating an oxidative pathway in 22C11-mediated neuronal degeneration. This was further supported by the observation that prior treatment of neurons with buthionine sulfoximine, an inhibitor of ␥-glutamylcysteinyl synthetase, potentiated the toxic effects of 22C11. Finally, with use of compartmented cultures of hippocampal neurons, it was also demonstrated that selective application of 22C11 caused local neuritic degeneration that was prevented by the addition of GEE to the neuritic compartment. Thus, the binding of a monoclonal antibody to APP initially triggers neurite degeneration that is followed by caspasedependent apoptosis in neuronal cultures and illustrates a novel property of this protein in neurons that may contribute to the profound neuronal cell death associated with AD. Key Words: Apoptosis—Amyloid

In Alzheimer’s disease (AD), much attention has been paid to the possibility that senile plaques, which are composed of fibrillar deposits of the ␤-amyloid peptide, may be an early and critical feature of the disease. In spite of the extensive research on the importance of senile plaques in the pathology of AD, little is known of the normal functions of the ␤-amyloid precursor protein (APP) from which ␤-amyloid is derived. APP consists of 695–770 amino acids encoded by differentially spliced mRNAs transcribed from a single gene located on human chromosome 21 (Kang et al., 1987). The 695-amino acid APP is expressed preferentially in the brain and has a structure and membrane orientation that resemble those of a polypeptide hormone receptor (Nishimoto et al,. 1997). In this regard, it has been demonstrated that APP can associate with the GTP-binding protein Go (Nishimoto et al., 1993; Okamoto et al., 1995; Nishimoto, 1998; Brouillet et al., 1999). The exact function of APP in neurons is currently unknown; however, numerous studies have suggested that APP plays a role in promoting cell substrate adhesion as well as neurite outgrowth (for review, see Mattson, 1997). In familial AD (FAD), missense mutations in the 695-amino acid form of APP cosegregate with disease Resubmitted manuscript received January 27, 2000; accepted January 27, 2000. Address correspondence and reprint requests to Dr. D. H. Cribbs at Institute for Brain Aging and Dementia, Department of Neurology, University of California, Irvine, CA 92697-4540, U.S.A. E-mail: [email protected] Abbreviations used: AD, Alzheimer’s disease; anti-BDPc, calpain cleavage site-directed antibody to 150-kDa breakdown product of fodrin; APP, ␤-amyloid protein precursor; BDP, breakdown product; BSO, buthionine sulfoximine; DMEM, Dulbecco’s modified Eagle’s medium; FAD, familial Alzheimer’s disease; GEE, GSH ethyl ester; SST, staurosporine; Z-VAD-FMK, N-benzyloxycarbonyl-Val-AlaAsp(O-methyl)-fluoromethyl ketone.

2331

2332

T. T. ROHN ET AL.

phenotype in families with dominantly inherited AD (Karlinsky et al., 1992). Expression of these mutant forms of APP in COS cells results in DNA fragmentation and apoptosis mediated by Go (Okamoto et al., 1996; Yamatsuji et al., 1996a,b; Giambarella et al., 1997). These results raise an interesting hypothesis, namely, that proteolytic release of ␤-amyloid may leave behind a tonically active carboxyl-terminal portion of APP that could injure neurons through constitutive stimulation of a cell-signaling pathway involving Go. In the present study, we examined the role of APP in primary cultures of cortical neurons. Because the natural ligand of APP remains unknown, we determined the effects of cross-linking APP using 22C11, a monoclonal antibody raised against the extracellular domain of APP (Hilbich et al., 1993). Our laboratory has previously used 22C11 to demonstrate that ␤-amyloid causes increased expression and processing of APP in cortical neurons (Cribbs et al., 1995). Using this antibody to APP, we now show that exposure of cortical neurons to 22C11 resulted in neuronal toxicity by a mechanism involving apoptosis. In addition, we demonstrate that 22C11 can act directly on neurites, inducing local neurite degeneration. These results highlight a novel function for normal APP and suggest that inappropriate clustering of APP on the plasma membrane of neurons may be a contributing factor in the profound loss of neurons associated with AD (Bobinski et al., 1997). MATERIALS AND METHODS Materials All chemicals used were of the highest grade available. Poly-D-lysine, GSH, GSH ethyl ester (GEE), L-buthionine[S,R]-sulfoximine (BSO), mouse IgG1, and A23187 were all purchased from Sigma Chemical Co. (St. Louis, MO, U.S.A.). Anti-Alzheimer precursor protein A4 (clone 22C11) was from Boehringer Mannheim (Indianapolis, IN, U.S.A.). N-Benzyloxycarbonyl-Val-Ala-Asp(O-methyl)-fluoromethyl ketone (ZVAD-FMK) was from Enzyme Systems Products (Livermore, CA, U.S.A.). Calpeptin and staurosporine (SST) were from Calbiochem-Novabiochem (La Jolla, CA, U.S.A.). SYTO 11 was from Molecular Probes (Eugene, OR, U.S.A.). The ApopTag peroxidase kit was from Oncor (Gaithersburg, MD, U.S.A.). Trypan blue and BenchMark-prestained protein standards were purchased from GibcoBRL (Grand Island, NY, U.S.A.). Pertussis toxin was from RBI (Natick, MA, U.S.A.). Rabbit mammalian brain antispectrin antibody 922 was from Chemicon (Temecula, CA, U.S.A.). The epitope peptide APP66 – 81 for 22C11 (KEGILQYCQEVYPELQ) was synthesized and purified to ⬎97% purity by Multiple Peptide Systems (San Diego, CA, U.S.A.). The calpain cleaveage site-directed antibody to the 150-kDa breakdown product (BDP) of fodrin (anti-BDPc) was previously characterized (Bahr et al., 1995). Purified soluble APP and the monoclonal antihuman APP antibody (P2-1) were kind gifts from Dr. William Van Nostrand (State University of New York, Stony Brook, NY, U.S.A.).

Cell culture Sprague–Dawley rat embryos (day 18 of gestation) were used to generate short-term cultures of cortical or hippocampal

J. Neurochem., Vol. 74, No. 6, 2000

neurons as previously described (Pike et al., 1993). The cortical neurons were kept in culture at 37°C with 5% CO2 in Dulbecco’s modified Eagle’s medium (DMEM; Gibco) supplemented with N2 components (Pike et al., 1993). For all morphological studies, cells were plated at a density of 1.5 ⫻ 105 cells/ml in poly-D-lysine-coated 24-well plates. For western blot analysis, a higher cell density was required (see below); therefore, cortical neurons were plated at a density of 3.75 ⫻ 105 cells/ml in poly-D-lysine-coated 24-well plates. All experiments were performed on day 4 of the neuronal cultures. Compartmented cultures of hippocampal neurons were prepared as previously described (Ivins et al., 1998). In brief, each chamber was constructed by affixing a 10 ⫻ 22-mm no. 1 coverslip to a hemisected Teflon ring to which a thin layer of silicon vacuum grease was applied to the bottom before placing chambers onto poly-D-lysine/laminin-2-coated six-well cluster dishes. Neuronal cultures were plated (1–2 ⫻ 105 cells/cm2) outside the chambers to allow neurites to be isolated and studied in a small volume of buffer within the interior of the chamber.

Treatment protocols

Before use, 22C11 (stock, 50 ␮g/ml) was extensively dialyzed against sterile DMEM. Both GEE and BSO were made up as 50 mM stock solutions in sterile, distilled H2O and filter sterilized before use. The calcium ionophore A23187 and SST were prepared as 5 mM stocks in ethanol and dimethyl sulfoxide, respectively. Z-VAD-FMK and calpeptin were prepared as 50 mM stocks in dimethyl sulfoxide. Cultures were used for experimentation on day 4 in vitro. To permit cellular loading, all drugs were added to cultures at least 1 h before toxic insult, with the exception of BSO and pertussis toxin (50 ␮g/ml), which were allowed to preincubate with neurons for 4 h. For experiments using synthetic APP66 – 81, 10 mM stock solutions were prepared in H2O. 22C11 (0.5–1 ␮g/ml) was mixed with either purified soluble APP or various concentrations of APP66 – 81 overnight at 4°C before application to neuronal cultures for 18 –24 h.

Cell death and viability Studies on 22C11-induced cell death were carried out by direct visual observation and trypan blue staining as previously described (Pike et al., 1991). In brief, after 18- to 24-h insult periods, cell survival was quantified by counts (three fields per well, two wells per condition, average of 50 cells per field). Live and dead cell counts were determined to control for the loss of neurons from cultures. For graphical presentation, raw data from a single representative experiment were converted to percentage cell viability.

Apoptotic assays Following treatment of cultured neurons with various insults, neurons were analyzed for changes in nuclear morphology with the membrane-permeable dye SYTO 11 Live Cell nucleic stain. Neurons were exposed to 0.5 ␮M SYTO 11 for 10 min and observed for nuclear condensation and fragmentation using a fluorescent microscope. The ApopTag peroxidase kit was used to generate random heteropolymers of digoxigenin-linked nucleotides on the 3⬘ ends using terminal deoxynucleotidyl transferase to detect 3⬘-hydroxy ends produced following DNA fragmentation during apoptosis. An antidigoxigenin antibody fragment conjugated to peroxidase was allowed to bind to labeled DNA fragments, and specific binding was detected using diaminobenzidine as the chromogenic substrate.

APP-MEDIATED PROGRAMMED CELL DEATH

2333

Western blot analysis For analysis of fodrin BDPs, cortical neurons, treated under various conditions, were extracted with 2⫻ sample buffer and subjected to sodium dodecyl sulfate electrophoresis on 7.5% polyacrylamide gels as previously described (Cribbs et al., 1995). In brief, transfers were incubated with either antispectrin 922 (1:200 dilution; Chemicon) or anti-BDPc (1:100 dilution) for 1 h, followed by horseradish peroxidase-conjugated goat anti-rabbit IgG secondary antibody (1:5,000) for 1 h. All incubations were at room temperature, and blots were developed using an enhanced chemiluminescence system (Amersham). Prestained molecular mass standards were used on all gels.

RESULTS Exposure of cortical neurons to the anti-APP antibody 22C11 induces significant morphological changes and cell death that are associated with an interaction between 22C11 and APP As neurons release soluble APP (Mattson, 1997), preliminary experiments were performed to determine if released APP might attenuate the effects of 22C11 on membrane-bound APP. In experiments where the culture medium bathing the neurons was replaced, the effects of 22C11 were much more pronounced (data not shown). These results support the conclusion that released soluble APP presumably attenuates the actions of 22C11 on neurons by binding to it and, in a sense, acting as a sink for 22C11. Because of this, in all experiments, the culture medium was removed and replaced with fresh DMEM-N2 prior to the addition of 22C11 to cortical neurons. To determine any possible effects of APP crosslinking by 22C11, we first examined neurons by phase contrast microscopy. As depicted in Fig. 1, neurons exposed to various concentrations of 22C11 had gross morphological changes including the complete loss of neurites and condensed translucent nuclei as compared with controls (Fig. 1C and D). The effect of 22C11 on neurons was concentration dependent, with beaded or fragmented neurites being detected following incubation of cortical neurons with 0.1 ␮g/ml 22C11 (Fig. 1B). Extensive neurite degeneration and condensed translucent nuclei were observed at concentrations of 0.5–1.0 ␮g/ml 22C11 (Fig. 1C and D). It is noteworthy that there was no apparent effect of 22C11 on cell adhesion, as incubation with 22C11 did not result in neurites or neurons lifting off the plate. However, the possibility that the effect of 22C11 on the neurons was due to disruption of an APP-mediated anchorage-dependent survival signal (Ruoslahti and Reed, 1994) and that it did not produce a detectable loss in cell attachment (Boudreau et al., 1995) led us to perform additional experiments to address this issue. We have performed experiments with factor XIa, a protease that has previously been shown to cleave the extracellular portion of APP and abolish the cell-adhesive properties of APP (Saporito-Irwin and Van Nostrand, 1995). Over a broad range of concentrations, factor XIa produced no measurable changes in neurite morphology or neuron viability (data not shown).

FIG. 1. 22C11 induced concentration-dependent morphological changes in cortical neurons. Cortical neurons were treated for 18 –24 h with varying concentrations of 22C11 and then examined by phase microscopy. A: 0 ␮g/ml; B: 0.1 ␮g/ml; C: 0.5 ␮g/ml; D: 1.0 ␮g/ml. At concentrations of 0.5–1.0 ␮g/ml, 22C11 caused major changes in morphology, including neurite degeneration and soma shrinkage. Data are representative of three independent experiments from three separate neuronal preparations.

Exposure of cortical neurons to 22C11 also had a major effect on cell viability. Neurons treated with 0.1 ␮g/ml 22C11 exhibited a similar neurotoxicity profile as controls. However, significant cell death (⬎50% cell mortality) was observed at concentrations beginning at 1.0 ␮g/ml 22C11 within 24 h (Fig. 2), and nearly 100% cell death was observed at 2.0 ␮g/ml 22C11 (Fig. 2). Based on these results, we chose an intermediate value of 1.0 ␮g/ml 22C11, which resulted in significant neuronal damage and ⬃50% cell death, as a concentration for all further experiments, except where noted. Experiments were performed to determine the specificity of 22C11 actions on cortical neurons. No effect was observed if cortical neurons were treated with mouse IgG concentrations equivalent to those used with 22C11 that caused significant morphological changes and cell death (Fig. 3C). In addition, the mouse monoclonal antibody P2-1 (BioSource International, Camarillo, CA, U.S.A.), which is specific for the amino-terminal end of human but not rat APP, had no effect on neuronal morphology when used at equivalent concentrations to 22C11 (Fig. 3D). Experiments were also undertaken using purified soluble APP. As shown in Fig. 4A, 22C11 recognizes a single band in neuronal extracts corresponding to an identical molecular mass as that of purified soluble APP, an observation that is consistent with a


2334

T. T. ROHN ET AL.

FIG. 2. Exposure to 22C11 results in a concentration-dependence toxicity in cultured cortical neurons. Neurons were treated with various concentrations of 22C11 for 18 –24 h, and cell death was determined by morphological cell counts. Data show mean viability values (⫾SD) from a single representative experiment from three separate neuronal preparations.

specific reaction of 22C11 with APP. Furthermore, if 22C11 was premixed with soluble APP, 22C11-mediated damage was significantly attenuated (Fig. 4B). To further confirm the specificity of 22C11 on APP, we performed experiments using a synthetic peptide, APP66 – 81, that contains the epitope recognized by 22C11 (Hilbich et al., 1993). Treatment of cortical neurons with 22C11 alone induced characteristic neurite degeneration and cell death (Fig. 5B) that were completely prevented

FIG. 3. Specificity of 22C11-mediated neuronal damage in cultured cortical neurons. Cortical neurons were treated for 18 –24 h with various agents and then examined by phase microscopy. A: Control (DMEM only); B: 22C11 (1 ␮g/ml); C: IgG (1 ␮g/ml); D: monoclonal antibody P2-1 (2 ␮g/ml). Only 22C11 induced morphological changes including cell neurite degeneration and soma shrinkage, whereas IgG alone and the monoclonal antibody P2-1, which is specific for the amino-terminal end of human but not rat APP, had no apparent effects. Data are representative of three independent experiments.


FIG. 4. Premixing 22C11 with soluble APP attenuates 22C11mediated damage to cortical neurons. A: Purified soluble APP (5 ␮g) or cortical neuronal extracts (1 ⫻ 106 cells/ml) were separated by sodium dodecyl sulfate –polyacrylamide electrophoresis, transferred to nitrocellulose, and probed with 22C11 at 2.0 ␮g/ml. In cortical extracts, 22C11 recognizes a single band equivalent in molecular mass to that of purified soluble APP. B: Treatment of 22C11 with soluble APP in DMEM or with DMEM alone was carried out for 18 –24 h at 4°C. Following pretreatment, cortical neurons were incubated with DMEM alone (1), APP alone at 30 ␮g/ml (2), 22C11 alone at 0.5 ␮g/ml (3), or 22C11 premixed with 30 ␮g/ml APP (4). Data are representative of three independent experiments from two separate neuronal preparations.

if 22C11 was premixed with increasing concentrations of APP66 – 81 (Fig. 5C and D). This shows that APP66 – 81 competitively antagonizes the action of 22C11 in this system, consistent with the fact that APP66 – 81 is the epitope for 22C11, and provides strong evidence that


2335

(APLP2), because 22C11 has been shown to cross-react with APLP2 (Slunt et al., 1994).

FIG. 5. Premixing 22C11 with the epitope peptide APP66 – 81 prevents 22C11-mediated effects on cortical neurons. Treatment of 22C11 with the epitope peptide in DMEM or with DMEM alone was carried out for 18 –24 h at 4°C. Following pretreatment, cortical neurons were incubated with peptide alone at 500 ␮M (A), 22C11 alone at 1 ␮g/ml (B), or 22C11 premixed with either 50 ␮M peptide (C) or 500 ␮M peptide (D). Data are representative of three independent experiments from two separate neuronal preparations.

22C11 acts on APP to cause morphological changes and cell death. However, we cannot rule out the possibility that this antibody-induced effect may be at least partially initiated through binding to the APP-like protein

Morphological changes induced by 22C11 are mediated through an oxidative pathway Oxidative stress is thought to play a key role in the progression of many neurodegenerative diseases including AD. This oxidative hypothesis is supported not only by the presence of oxidative markers in AD brain (Smith et al., 1991; Pappolla et al., 1992) but also by research demonstrating that ␤-amyloid toxicity is associated with oxidative stress (Behl et al., 1992; Hensley et al., 1994; Pike et al., 1997). To determine a role for oxidative stress in the involvement of 22C11-mediated cell death, we examined the ability of the antioxidant GEE to prevent the morphological changes associated with 22C11. As shown in Fig. 6A, GEE by itself had no effect on the morphology of cortical neurons under these experimental conditions. Exposure of neurons to 22C11 resulted in the characteristic morphological changes, such as neurite degeneration and the appearance of condensed translucent nuclei (Fig. 6B), which were completely prevented following prior exposure of neurons to GEE (Fig. 6C). To further examine the role of oxidative stress in 22C11induced neurotoxicity, experiments were also undertaken using an inhibitor of ␥-glutamylcysteinyl synthetase, BSO. The rationale for such an experiment is that if GSH protects from 22C11-mediated damage, then prevention of GSH formation by inhibiting ␥-glutamylcysteinyl synthetase should potentiate the actions of 22C11. In these experiments, neurons were pretreated 4 h with BSO before the addition of 22C11. Interestingly, under these experimental conditions, there was no observable effect

FIG. 6. A role for oxidative stress in 22C11-mediated effects on cortical neurons. Cortical neurons were pretreated for 1 h with GEE (1 mM) (A and C) or 4 h with BSO (1 mM) (D and F), followed by the addition of 22C11 (1.0 ␮g/ml) for 18 h (B and C) or 6 h (E and F). Neurons pretreated with the antioxidant GEE were completely protected from the actions of 22C11 (C), whereas strong potentiation was observed in the presence of the glutathione synthase inhibitor BSO (F). Data are representative of three independent experiments from three different neuronal preparations.


2336

T. T. ROHN ET AL.

FIG. 7. Local neuritic degeneration by 22C11 is prevented by prior exposure to the antioxidant GEE. 22C11 (0.5 ␮g/ml) was added to the neuritic compartment of compartmented cultures of hippocampal neurons in the absence (C) or presence (D) of a 1-h pretreatment with 50 ␮M GEE. The addition of GEE had no effect on the morphology of the neuritic processes (B) compared with untreated neurons (A). Complete neuritic degeneration was observed following selective application of 22C11, which was significantly prevented by prior addition of GEE to the neuritic compartment. Data are representative of three independent experiments from three different neuronal preparations.

of BSO alone on neuronal morphology (Fig. 6D). To observe a possible potentiation of 22C11 damage by inhibition of ␥-glutamylcysteinyl synthetase, it was necessary to treat neurons with 22C11 for a time period (6 h) that induced observable but not maximal neuronal cell damage (Fig. 6E). Cortical neurons that were pretreated with BSO showed a significantly greater level of neurite degeneration as well as condensation and apparent fragmentation of nuclei (Fig. 6F) than neurons treated with 22C11 alone. Thus, pretreatment of cortical neurons with BSO greatly potentiated the effects of 22C11. Taken together, these results strongly support a role for oxidative stress in mediating the actions of 22C11. Local neurite degeneration occurred in cultured hippocampal neurons exposed to 22C11 Our laboratory has recently demonstrated that certain apoptotic insults, including ␤-amyloid and SST, may act directly on neuritic processes to promote local neurite degeneration by mechanisms common to apoptosis (Ivins et al., 1998). Using compartmented cultures of hippocampal neurons, we examined whether selective application of 22C11 causes local neurite degeneration and whether the antioxidant GEE could provide protection as it did in dissociated cortical cultures. Primary cultures of hippocampal neurons were plated outside fabricated chambers to allow neurites to be isolated and studied (see Materials and Methods). These chambers consist of a coverslip to separate two compartments: a


somal/proximal neurite compartment (or somal compartment) and a distal neurite compartment (or neuritic compartment) (Klostermann and Bonhoeffer, 1996; Ivins et al., 1998). Importantly, as previously demonstrated, leakage between the two compartments is negligible, thus allowing for neurites to be selectively exposed to 22C11 and/or GEE (Ivins et al., 1998). Figure 7 depicts the results of such an experiment where neurites were selectively exposed to 22C11. When neurites were exposed to 22C11, significant neuritic degeneration occurred that was characterized by neuritic defasiculation and membrane blebbing (Fig. 7C). To determine if antioxidant pretreatment could prevent 22C11-mediated neuritic degeneration as it did in Fig. 6, neurites were pretreated with GEE. GEE (50 ␮M) had no observable morphological effects when added to the neuritic chamber as compared with control neurites (Fig. 7A and B). When neurites were selectively pretreated with GEE, the effects of 22C11 were greatly diminished (Fig. 7D). Therefore, 22C11 may cause local neurite degeneration that is prevented by the addition of antioxidant GEE to the neuritic compartment. It should be noted that from these experiments, we were unable to unambiguously identify the cell bodies of neurites exposed to 22C11. However, preliminary results suggest that neurite degeneration is limited to the more distal portions of neurites, whereas the more proximal neurites and soma are unaffected (data not shown).


2337

FIG. 8. Neurons treated with 22C11 displayed changes in nuclear morphology consistent with apoptosis. A and B: Cortical cultures were treated with 22C11 (1.0 ␮g/ml) for 18 –24 h followed by in situ DNA end-labeling of neurons using the ApopTag kit to identify cortical neurons undergoing apoptosis. A few stained nuclei appear in control cultures (A). Neuronal cultures treated with 22C11 showed numerous stained nuclei and a few fragmented nuclei are also visible (B, arrowhead). C and D: Cortical neurons treated with 22C11 were examined for nuclear changes with the membrane-permeable nucleic acid stain SYTO 11. Nuclei from control neurons exhibited normal features in size and morphology (C) compared with 22C11-treated neurons, which showed numerous condensed and fragmented nuclei (D, arrowheads).

Treatment of neurons with 22C11 resulted in apoptosis Morphological features including changes in nuclear condensation and DNA fragmentation are major characteristics of apoptosis that may occur in certain neurodegenerative diseases such as AD (Lassmann et al., 1995; Thompson, 1995). To examine whether 22C11-mediated cell death occurred through apoptosis, experiments were undertaken using the technique of in situ end-labeling of DNA. Cultured neurons treated with 22C11 were screened using the ApopTag peroxidase method to determine whether or not DNA fragmentation had occurred. As shown in Fig. 8B, exposure of cortical neurons to 22C11 produced significantly more positively stained cells than control cultured neurons (Fig. 8A). Additional experiments using the membrane-permeable SYTO 11 nucleic acid stain were performed to monitor the nuclear morphology changes that occurred as a result of exposure of cortical neurons to 22C11. Untreated neurons exhibited large uniform staining (Fig. 8C). In

contrast, nuclei appeared condensed and fragmented following exposure of cortical neurons to 22C11 (Fig. 8D), supporting the results observed with the ApopTag peroxidase method. These results are similar to those previously obtained by our laboratory demonstrating that cross-linking of neural cell adhesion molecule or concanavalin A receptors results in DNA condensation and fragmentation (Cribbs et al., 1996; Azizeh et al., 1998) and supports the conclusion that 22C11-mediated neurotoxicity occurs through apoptosis. To further confirm a role for apoptosis in 22C11mediated neurotoxicity, experiments were performed using the broad caspase inhibitor Z-VAD-FMK. Caspases represent a class of aspartyl proteases that are crucial mediators for apoptosis (for review, see Nicholson and Thornberry, 1997). As depicted in Fig. 9A, Z-VADFMK alone had no effect on neuronal morphology. Extensive neurite degeneration occurred following exposure of cortical neurons to 22C11 alone (Fig. 9B) that was significantly prevented following prior exposure of

FIG. 9. Protection of 22C11-mediated morphological changes to cortical neurons by the broad caspase inhibitor Z-VAD-FMK. Cortical neurons were pretreated for 1 h in the presence of Z-VAD-FMK (150 ␮M) followed by the addition of 22C11 (1.0 ␮g/ml) for 18 –24 h and then examined for morphological changes by microscopy. A: Z-VADFMK alone; B: 22C11 alone; C: 22C11 ⫹ ZVAD-FMK. Z-VAD-FMK provided significant protection from neurite degeneration caused by 22C11. Data are representative of three independent experiments from three separate neuronal preparations.


2338

T. T. ROHN ET AL. neurons to Z-VAD-FMK (Fig. 9C). These results suggest a role for caspases in mediating the actions of 22C11 and provide further support that apoptosis is the primary pathway of 22C11-mediated neuronal damage. It is noteworthy that unlike the results using the antioxidant GEE, Z-VAD-FMK did not provide complete protection from 22C11-mediated changes in neuronal morphology (compare Fig. 9 and Fig. 6A–C).

FIG. 10. Biochemical characterization of caspase and calpain cleavage of fodrin following exposure of cortical neurons to 22C11. A: The 120-kDa and 150-kDa BDPs of fodrin produced following exposure of cortical neurons to 22C11 are prevented by the caspase inhibitor Z-VAD-FMK. Cortical neurons were pretreated for 1 h with Z-VAD-FMK (150 ␮M) followed by the addition of either SST (1 ␮M) or 22C11 (1.0 ␮g/ml) for 24 h. Following treatment, neuronal extracts were separated by sodium dodecyl sulfate–polyacrylamide electrophoresis, transferred to nitrocellulose, and probed with the antispectrin antibody 922. SST and 22C11 increased the formation of both the 120-kDa and the 150-kDa BDPs of fodrin, which was prevented in the additional presence of Z-VAD-FMK. B: 22C11-mediated generation of the 150-kDa BDP of fodrin is prevented by the calpain inhibitor calpeptin. Cortical neurons were pretreated for 1 h with calpeptin, followed by the addition of either the calcium ionophore A23187 for an additional 2 h (1) or 22C11 for 18 –24 h (2). Following treatment protocol, western blot analysis was performed. Neuronal extracts were immunoblotted with a calpain-


Biochemical characterization of caspase cleavage of fodrin following exposure of cortical neurons to 22C11 Data presented in Fig. 9 support a role for caspases in 22C11-mediated neurotoxicity. A prominent target for caspases is nonerythroid brain spectrin (fodrin), a major component of the membrane cytoskeleton in neurons (Goodman and Zagon, 1986; Goodman et al., 1987). During the initiation phase of apoptosis, fodrin is one of the first substrates to be cleaved by caspases (Martin et al., 1995; Cryns et al., 1996), resulting in the formation of stable 120-kDa and 150-kDa BDPs of fodrin (Nath et al., 1996; Wang et al., 1998). We examined whether exposure of cortical neurons to 22C11 resulted in a similar cleavage of fodrin and compared the cleavage products with those produced by SST, a classic apoptotic insult (Jacobson et al., 1993). Figure 10A depicts western blot analysis of neuronal extracts treated under various conditions and probed with the anti-␣-spectrin antibody 922. Even under control conditions, limited generation of a 150-kDa but not 120-kDa BDP of fodrin is observed (Fig. 10A, Control). Treatment of cortical neurons with SST results in an increase in the formation of the 150kDa BDP and the generation of a 120-kDa BDP of fodrin (Fig. 10A, SST). A similar but larger increase in the generation of both the 120- and the 150-kDa BDPs of fodrin (Fig. 10A, 22C11) was observed following treatment of neurons with 22C11. If, however, cortical neurons were pretreated with the caspase inhibitor Z-VADFMK generation of both the 120-kDa and the 150-kDa BDPs of fodrin was completely prevented (Fig. 10A, SST ⫹ Z-VAD and 22C11 ⫹ Z-VAD). These results support the hypothesis that 22C11 leads to the activation of cellular caspases that are capable of targeting certain cytoplasmic proteins including fodrin. Fodrin not only is a target for certain caspases but is also an especially sensitive target for calpain (Martin et al., 1995). Calpain cleavage of fodrin generates a

mediated cleavage site-directed antibody (anti-BDPc), which recognizes the 150-kDa BDP of fodrin but not full-length fodrin. C: Prevention of 22C11-mediated generation of both the 120kDa and the 150-kDa BDPs of fodrin by Z-VAD-FMK but not calpeptin is shown. Cortical neurons were pretreated for 1 h with either calpeptin or Z-VAD-FMK, as indicated, followed by the addition of 22C11 for 18 –24 h. Following treatment, western blot analysis was performed where neuronal extracts were immunoblotted with the antispectrin antibody 922. Data are representative of three independent experiments from three separate neuronal preparations.

APP-MEDIATED PROGRAMMED CELL DEATH 150-kDa BDP that has been used as a marker for pathogenic calpain activation (Lee et al., 1991; del Cerro et al., 1994; Bahr et al., 1995). In addition, it is now clear that both calpain and certain caspases may be activated following an apoptotic insult (Nath et al., 1996). Our results showing an increase in both 120-kDa and 150-kDa BDPs of fodrin suggest that, in addition to the activation of certain caspases, calpain may also be activated and contribute to the cleavage of fodrin following treatment of cortical neurons with 22C11. To examine this possibility, experiments were undertaken using a calpain-mediated cleavage site-directed antibody (anti-BDPc) that recognizes the 150-kDa BDP of fodrin but not full-length fodrin (Bahr et al., 1995). Figure 10B depicts the result of an experiment in which cortical neurons were treated under various conditions and neuronal extracts were then analyzed by western blot analysis using anti-BDPc. As shown in Fig. 10B(1), exposure of cortical neurons to the calcium ionophore A23187 generated a large 150-kDa BDP of fodrin, which was completely prevented following prior exposure of neurons to the calpain inhibitor calpeptin. Treatment of cortical neurons with 22C11 also resulted in a significant increase in the formation of the 150-kDa BDP of fodrin [Fig. 10B(2), 22C11]. In addition, prior treatment with calpeptin greatly attenuated the generation of this band [Fig. 10B(2), ⫹ Calpeptin]. Thus, these results support the conclusion that the generation of the 150-kDa BDP of fodrin following treatment of cortical neurons with 22C11 is a result of activation of calpain. Interestingly, even though calpeptin prevented to a large degree the formation of the 150-kDa BDP of fodrin, it did not prevent neurite degeneration or cell death associated with either A23187 or 22C11 (data not shown). A final experiment was undertaken to try to further establish the relative roles of caspases and calpain in 22C11-mediated neurotoxicity and apoptosis. Neurons were pretreated and either Z-VAD-FMK or calpeptin, followed by exposure to 22C11, and neuronal extracts were then analyzed by western blot analysis using the antispectrin 922 antibody. As shown in Fig. 10C, 22C11 caused an increase in both the 120-kDa and the 150-kDa BDPs of fodrin, and both bands were blocked by prior exposure of neurons to Z-VAD-FMK (Fig. 10C, 22C11 and ⫹ Z-VAD). In contrast, prior exposure of neurons to calpeptin prevented the formation of the 150-kDa but not the 120-kDa BDP of fodrin (Fig. 10C, 22C11 and ⫹ Calpeptin). Taken together, the results presented in Fig. 10 and the fact that only Z-VAD-FMK but not calpeptin provided protection from 22C11-induced morphological changes including neurite degeneration and nuclear condensation suggest a primary role for caspases but not calpain in 22C11-mediated apoptosis. DISCUSSION The major finding of the present study is that putative cross-linking of APP by the monoclonal antibody 22C11

2339

leads to neuronal toxicity through apoptosis. Since the initial characterization of 22C11 (Weidemann et al., 1989; Hilbich et al., 1993), this antibody has been extensively used to follow the expression and processing of APP in a number of cell types (Jung et al., 1984; Cribbs et al., 1995; Rohan de Silva et al., 1997). Although the recognition potency is not high, antibody absorption (Milward et al., 1992) and immunoprecipitation (Nishimoto et al., 1993) indicate that 22C11 does recognize a native form of APP. The epitope region for 22C11 has been determined to be located in the amino-terminal region in the ectoplasmic Cys-containing domain to APP66 – 81 (Hilbich et al., 1993). Based on other studies, it is likely that this monoclonal antibody, which belongs to the divalent IgG isotype, activates APP by facilitating its dimerization (Okamoto et al., 1995). Other studies using neural cell adhesion molecule antibodies, ␤-amyloid, and concanavalin A have supported the hypothesis that membrane receptor cross-linking can be a mechanism that triggers apoptosis in neurons (Pike et al., 1993; Cribbs et al., 1996; Azizeh et al., 1998). Cross-linking of APP may in turn initiate downstream signal transduction cascades that elicit apoptosis, as is the case for the Fas ligand (Itoh et al., 1991). Several reports have suggested that the actions of APP may be mediated through an interaction with the GTP-binding protein Go (Nishimoto et al., 1993; Okamoto et al., 1995; Brouillet et al., 1999). Okamoto et al. (1995), using phospholipid vesicles consisting of baculovirally made APP and brain trimeric Go, demonstrated that 22C11 increased guanosine-5⬘-O-(3-thiotriphosphate) binding and the turnover number of GTPase of Go. These same authors have also previously reported that expression of three FAD mutants (FAD-APPs), but not normal APP, causes COS cells to undergo apoptosis by activating Go (Okamoto et al., 1996; Yamatsuji et al., 1996a; Giambarella et al., 1997; Nishimoto et al., 1997). Their results suggest that FAD-APPs are constitutively active Golinked receptors that may lead to abnormal signaling and cellular injury. However, to date, no evidence is available suggesting a coupling of Go with APP in neurons. In any event, their data are suggestive of an interesting hypothesis whereby the proteolytic release of ␤-amyloid fragments may produce a tonically active APP that could injure neurons through constitutive stimulation. Our results, using cultured cortical neurons, suggest that APP may act as a cell surface receptor and, when activated by 22C11, may initiate a death signal in neuronal cultures. Our results demonstrating that antibodyinduced activation of APP leads to apoptosis in cortical neurons extend the observations of Nishimoto et al. (1993) and support their hypothesis that APP may act as a cell surface receptor. Moreover, inappropriate activation of wild-type APP, perhaps by aggregates of ␤-amyloid, on neurons represents a plausible mechanism that may contribute to the profound loss of neurons associated with AD. Recently, it has been demonstrated that APP is directly cleaved by caspases during apoptosis,


2340

T. T. ROHN ET AL.

resulting in elevated ␤-amyloid peptide formation (Gervais et al., 1999). If ␤-amyloid is capable of binding to and activating APP, this may produce a cyclic positive loop where ␤-amyloid activation of APP leads to apoptosis and further production of ␤-amyloid caspase cleavage, all of which may contribute to the profound neuron loss associated with AD. At the present time, we do not know how activation of APP in neurons leads to neuronal apoptosis. There is, however, a large body of literature demonstrating that APP can play a role in promoting cell substrate adhesion (for review, see Mattson, 1997). An alternative interpretation of our results is that by binding to the extracellular domain of APP, 22C11 disrupts an APP-mediated anchorage-dependent survival signal, thereby initiating apoptosis (Ruoslahti and Reed, 1994). However, the antigenic epitope for 22C11 is not within any of the putative adhesion domains identified in APP. Moreover, a number of reports have shown that disruption of neuronal adhesion is dependent on using antibodies with specificity for the adhesion domains of APP and generally requires the neurons to be growing on specific extracellular matrix molecules (Breen et al., 1991; Chen and Yankner, 1991; Coulson et al., 1997). In addition, several reports have indicated that neurons growing on polylysine, the substrate used in the current experiments, are not affected by antibodies to adhesion domains in APP (Breen et al., 1991; Coulson et al., 1997). Finally, the experiments with protease factor XIa, previously shown to cleave the extracellular portion of APP and abolish the cell-adhesive properties to APP (Saporito-Irwin and Van Nostrand, 1995), did not induce apoptosis in our cultures. We have also investigated the possibility that the link between the activation of APP and eventual apoptosis in neurons may involve a pertussis toxin-sensitive G protein. In some cases, we have observed protection from the morphological changes associated with 22C11, including neurite degeneration, by pretreating cortical neurons for 4 h with 200 ng/ml pertussis toxin. However, we have been unable to reliably reproduce this observation. Whether this relates to an inability of pertussis toxin to completely inactivate Go in our model system or some other factors is unknown to us at this time. Further investigation is necessary to clarify this question. Previous studies have shown that, in addition to being expressed and processed in neuronal cell bodies, APP may also be expressed on neuritic processes (Kawarabayashi et al., 1993; Cribbs et al., 1995). In a recent study, our laboratory demonstrated that apoptotic insults can act directly on neurites in experiments using compartmented cultures of hippocampal neurons (Ivins et al., 1998). Both ␤-amyloid and concanavalin A, when selectively exposed to neurites only, caused neurite degeneration, extracellular exposure of phosphatidylserine, and caspase activation, all hallmarks of apoptosis. These results suggest that distal neuritic processes contain all the necessary machinery to respond to apoptotic microenvironments not shared by their cell bodies. In the


present study, the possibility that 22C11, in addition to promoting apoptosis in dissociated cultures, may also lead to local neuritic degeneration via activation of an apoptotic program in neurites was investigated. The addition of 22C11 to the neuritic compartment resulted in selective neurite degeneration that was similar to that previously observed with ␤-amyloid or concanavalin A (Ivins et al., 1998). Of interest was the observation that if neurites were pretreated with the antioxidant GEE, protection from neurite degeneration occurred. Preliminary results suggest that only the distal neuritic processes and not the more proximal portions of the same neurites and cell bodies are affected by the selective addition of 22C11. Thus, the same processes involved in mediating apoptosis in the soma of neurons, such as oxidative stress, are also present in the microenvironment of neurites. However, because we are unable at this time to identify the cell bodies of the neurites exposed to 22C11 unambiguously, the possibility that a signal is transported retrogradely to the nucleus that subsequently directs the neurite to degenerate must be considered. However, the fact that 22C11-induced neurite degeneration can be inhibited by selective application of certain antioxidants to the neuritic chamber supports the role for a local mechanism in mediating the actions of 22C11. In conclusion, we have demonstrated that exposure of cortical neurons to an anti-APP antibody leads to neuronal cell death that was characterized by neurite degeneration, cell shrinkage, condensation of nuclear chromatin and internucleosomal DNA cleavage, oxidative stress, caspase activation, and proteolytic cleavage of fodrin, all characteristics of apoptosis. In addition, it was demonstrated that selective APP activation on neuritic processes resulted in neurite degeneration that was protected by local application of the antioxidant GEE. These results reveal a novel function for APP in neurons and suggest that unwarranted activation of APP may lead to apoptosis that could contribute to the loss of neurons associated with AD. Acknowledgment: The authors thank Serena Wong for her excellent technical assistance. We thank Dr. William Van Nostrand for supplying us with purified soluble APP and monoclonal antibody P2-1. This work was supported in part by NIA grants 5T32AG00096-16 and 5R01AG13007.

REFERENCES Azizeh B. Y., Cribbs D. H., Kreng V. M., and Cotman C. W. (1998) Cross-linking of NCAM receptors on neurons induces programmed cell death. Brain Res. 796, 20 –26. Bahr B. A., Tiriveedhi S., Park G. Y., and Lynch G. (1995) Induction of calpain-mediated spectrin fragments by pathogenic treatments in long-term hippocampal slices. J. Pharmacol. Exp. Ther. 273, 902–908. Behl C., Davis J., Cole G. M., and Schubert D. (1992) Vitamin E protects nerve cells from amyloid beta protein toxicity. Biochem. Biophys. Res. Commun. 186, 944 –950.

APP-MEDIATED PROGRAMMED CELL DEATH Bobinski M., Wegiel J., Tarnawski M., Bobinski M., Reisberg B., de Leon M. J., Miller D. C., and Wisniewski H. M. (1997) Relationships between regional neuronal loss and neurofibrillary changes in the hippocampal formation and duration and severity of Alzheimer disease. J. Neuropathol. Exp. Neurol. 56, 414 – 420. Boudreau N., Sympson C. J., Werb Z., and Bissell M. J. (1995) Suppression of ICE and apoptosis in mammary epithelial cells by extracellular matrix. Science 267, 891– 893. Breen K. C., Bruce M., and Anderton B. H. (1991) Beta amyloid precursor protein mediates neuronal cell-cell and cell-surface adhesion. J. Neurosci. Res. 28, 90 –100. Brouillet E., Trembleau A., Galanaud D., Volovitch M., Bouillot C., Valenza C., Prochiantz A., and Allinquant B. (1999) The amyloid precursor protein interacts with G(o) heterotrimeric protein within a cell compartment specialized in signal transduction. J. Neurosci. 19, 1717–1727. Chen M. and Yankner B. A. (1991) An antibody to ␤-amyloid and the amyloid precursor protein inhibits cell-substratum adhesion in many mammalian cell types. Neurosci. Lett. 125, 223–226. Coulson E. J., Barrett G. L., Storey E., Bartlett P. F., Beyreuther K., and Masters C. L. (1997) Down-regulation of the amyloid protein precursor of Alzheimer’s disease by antisense oligonucleotides reduces neuronal adhesion to specific substrata. Brain Res. 770, 72– 80. Cribbs D. H., Davis-Salinas J., Cotman C. W., and Van Nostrand W. E. (1995) A␤ induces increased expression and processing of amyloid precursor protein in cortical neurons. Alzheimer Res. 1, 197– 200. Cribbs D. H., Kreng V. M., Anderson A. J., and Cotman C. W. (1996) Crosslinking of membrane glycoproteins by concanavalin A induces apoptosis in cortical neurons. Neuroscience 75, 173–185. Cryns V. L., Bergeron L., Zhu H., Li H., and Yuan J. (1996) Specific cleavage of alpha-fodrin during Fas- and tumor necrosis factorinduced apoptosis is mediated by an interleukin-1 beta-converting enzyme/Ced-3 protease distinct from the poly(ADP-ribose) polymerase protease. J. Biol. Chem. 271, 31277–31282. del Cerro S., Arai A., Kessler M., Bahr B. A., Vanderklish P., Rivera S., and Lynch G. (1994) Stimulation of NMDA receptors activates calpain in cultured hippocampal slices. Neurosci. Lett. 167, 149 – 152. Gervais F. G., Xu D., Robertson G. S., Vaillancourt J. P., Zhu Y., Huang J., LeBlanc A., Smith D., Rigby M., Shearman M. S., Clarke E. E., Zheng H., Van Der Ploeg L. H., Ruffolo S. C., Thornberry N. A., Xanthoudakis S., Zamboni R. J., Roy S., and Nicholson D. W. (1999) Involvement of caspases in proteolytic cleavage of Alzheimer’s amyloid-beta precursor protein and amyloidogenic A beta peptide formation. Cell 97, 395– 406. Giambarella U., Yamatsuji T., Okamoto T., Matsui T., Ikezu T., Murayama Y., Levine M. A., Katz A., Gautam N., and Nishimoto I. (1997) G protein betagamma complex-mediated apoptosis by familial Alzheimer’s disease mutant of APP. EMBO J. 16, 4897– 4907. Goodman S. R. and Zagon I. S. (1986) The neural cell spectrin skeleton: a review. Am. J. Physiol. 250, C347–C360. Goodman S. R., Zagon I. S., and Riederer B. M. (1987) Spectrin isoforms in mammalian brain. Brain Res. Bull. 18, 787–792. Hensley K., Carney J. M., Mattson M. P., Aksenova M., Harris M., Wu J. F., Floyd R. A., and Butterfield D. A. (1994) A model for beta-amyloid aggregation and neurotoxicity based on free radical generation by the peptide: relevance to Alzheimer disease. Proc. Natl. Acad. Sci. USA 91, 3270 –3274. Hilbich C., Monning U., Grund C., Masters C. L., and Beyreuther K. (1993) Amyloid-like properties of peptides flanking the epitope of amyloid precursor protein-specific monoclonal antibody 22C11. J. Biol. Chem. 268, 26571–26577. Itoh N., Yonehara S., Ishii A., Yonehara M., Mizushima S.-I., Sameshima M., Hase A., Seto Y., and Nagata S. (1991) The polypeptide encoded by the cDNA for human cell surface antigen Fas can mediate apoptosis. Cell 66, 233–243.

2341

Ivins K. J., Bui E. T. N., and Cotman C. W. (1998) ␤-Amyloid induces local neurite degeneration in cultured hippocampal neurons: evidence for neuritic apoptosis. Neurobiol. Dis. 5, 365–378. Jacobson M. D., Burne J. F., King M. P., Miyashita T., Reed J. C., and Raff M. C. (1993) Bcl-2 blocks apoptosis in cells lacking mitochondrial DNA. Nature 361, 365–369. Jung G., Helm R. M., Carraway C. A., and Carraway K. L. (1984) Mechanism of concanavalin A-induced anchorage of the major cell surface glycoproteins to the submembrane cytoskeleton in 13762 ascites mammary adenocarcinoma cells. J. Cell Biol. 98, 179 –187. Kang J., Lemaire H. G., Unterbeck A., Salbaum J. M., Master C. L., Grezeschik K. H., Multhaup G., Beyruther K., and Mu¨ller-Hill B. (1987) The precursor of Alzheimer’s disease amyloid A4 protein resembles a cell-surface receptor. Nature 325, 733–736. Karlinsky H., Vaula G., Haines J. L., Ridgley J., Bergeron C., Mortilla M., Tupler R. G., Percy M. E., Robitaille Y., Noldy N. E., et al. (1992) Molecular and prospective phenotypic characterization of a pedigree with familial Alzheimer’s disease and a missense mutation in codon 717 of the beta-amyloid precursor protein gene [Erratum in Neurology (1992) 42, 1662]. Neurology 42, 1445– 1453. Kawarabayashi T., Shoji M., Yamaguchi H., Tanaka M., Harigaya Y., Ishiguro K., and Hirai S. (1993) Amyloid beta protein precursor accumulates in swollen neurites throughout rat brain with aging. Neurosci. Lett. 153, 73–76. Klostermann S. and Bonhoeffer F. (1996) Investigations of signaling pathways in axon growth and guidance. Perspect. Dev. Neurobiol. 4, 237–252. Lassmann H., Bancher C., Breitschopf H., Wegiel J., Bobinski M., Jellinger K., and Wisniewski H. M. (1995) Cell death in Alzheimer’s disease evaluated by DNA fragmentation in situ. Acta Neuropathol. (Berl.) 89, 35– 41. Lee K. S., Frank S., Vanderklish P., Arai A., and Lynch G. (1991) Inhibition of proteolysis protects hippocampal neurons from ischemia. Proc. Nat. Acad. Sci. USA 88, 7233–7237. Martin S. J., O’Brien G. A., Nishioka W. K., McGahon A. J., Mahboubi A., Saido T. C., and Green D. R. (1995) Proteolysis of fodrin (non-erythroid spectrin) during apoptosis. J. Biol. Chem. 270, 6425– 6428. Mattson M. P. (1997) Cellular actions of beta-amyloid precursor protein and its soluble and fibrillogenic derivatives. Physiol. Rev. 77, 1081–1132. Milward E. A., Papadopoulos R., Fuller S. J., Moir R. D., Small D., Beyreuther K., and Masters C. L. (1992) The amyloid protein precursor of Alzheimer’s disease is a mediator of the effects of nerve growth factor on neurite outgrowth. Neuron 9, 129 –137. Nath R., Raser K. J., Stafford D., Hajimohammadreza I., Posner A., Allen H., Talanian R. V., Yuen P., Gilbertsen R. B., and Wang K. K. (1996) Non-erythroid alpha-spectrin breakdown by calpain and interleukin 1 beta-converting-enzyme-like protease(s) in apoptotic cells: contributory roles of both protease families in neuronal apoptosis. Biochem. J. 319, 683– 690. Nicholson D. W. and Thornberry N. A. (1997) Caspases: killer proteases. Trends Biochem. Sci. 22, 299 –306. Nishimoto I. (1998) A new paradigm for neurotoxicity by FAD mutants of ␤APP: a signaling abnormality. Neurobiol. Aging 19, S33–S38. Nishimoto I., Okamoto T., Matsuura Y., Takahashi S., Okamoto T., Murayama Y., and Ogata E. (1993) Alzheimer amyloid protein precursor complexes with brain GTP-binding protein G(o). Nature 362, 75–79. Nishimoto I., Okamoto T., Giambarella U., and Iwatsubo T. (1997) Apoptosis in neurodegenerative diseases. Adv. Pharmacol. 41, 337–368. Okamoto T., Takeda S., Murayama Y., Ogata E., and Nishimoto I. (1995) Ligand-dependent G protein coupling function of amyloid transmembrane precursor. J. Biol. Chem. 270, 4205– 4208. Okamoto T., Takeda S., Giambarella U., Murayama Y., Matsui T., Katada T., Matsuura Y., and Nishimoto I. (1996) Intrinsic signal-


2342

T. T. ROHN ET AL.

ing function of APP as a novel target of three V642 mutations linked to familial Alzheimer’s disease. EMBO J. 15, 3769 –3777. Pappolla M. A., Omar R. A., Kim K. S., and Robakis N. K. (1992) Immunohistochemical evidence of antioxidant stress in Alzheimer’s disease. Am. J. Pathol. 140, 621– 628. Pike C. J., Walencewicz A. J., Glabe C. G., and Cotman C. W. (1991) In vitro aging of beta-amyloid protein causes peptide aggregation and neurotoxicity. Brain Res. 563, 311–314. Pike C. J., Burdick D., Walencewicz A. J., Glabe C. G., and Cotman C. W. (1993) Neurodegeneration induced by beta-amyloid peptides in vitro: the role of peptide assembly state. J. Neurosci. 13, 1676 –1687. Pike C. J., Ramezan-Arab N., and Cotman C. W. (1997) ␤-Amyloid neurotoxicity in vitro: evidence of oxidative stress but not protection by antioxidants. J. Neurochem. 69, 1601–1611. Rohan de Silva H. A., Jen A., Wickenden C., Jen L. S., Wilkinson S. L., and Patel A. J. (1997) Cell-specific expression of beta-amyloid precursor protein isoform mRNAs and proteins in neurons and astrocytes. Brain Res. Mol. Brain Res. 47, 147–156. Ruoslahti E. and Reed J. C. (1994) Anchorage dependence, integrins, and apoptosis. Cell 77, 477– 478. Saporito-Irwin S. M. and Van Nostrand W. E. (1995) Coagulation factor XIa cleaves the RHDS sequence and abolishes the cell adhesive properties of the amyloid beta-protein. J. Biol. Chem. 270, 26265–26269. Slunt H. H., Thinakaran G., Von Koch C., Lo A. C., Tanzi R. E., and Sisodia S. S. (1994) Expression of a ubiquitous, cross-reactive


homologue of the mouse beta-amyloid precursor protein (APP). J. Biol. Chem. 269, 2637–2644. Smith C. D., Carney J. M., Starke R. P., Oliver C. N., Stadtman E. R., Floyd R. A., and Markesbery W. R. (1991) Excess brain protein oxidation and enzyme dysfunction in normal aging and in Alzheimer disease. Proc. Natl. Acad. Sci. USA 88, 10540 – 10543. Thompson C. B. (1995) Apoptosis in the pathogenesis and treatment of disease. Science 267, 1456 –1462. Wang K. K., Posmantur R., Nath R., McGinnis K., Whitton M., Talanian R. V., Glantz S. B., and Morrow J. S. (1998) Simultaneous degradation of alphaII- and betaII-spectrin by caspase 3 (CPP32) in apoptotic cells. J. Biol. Chem. 273, 22490 –22497. Weidemann A., Ko¨nig G., Bunke D., Fischer P., Salbaum J. M., Masters C. L., and Beyreuther K. (1989) Identification, biogenesis, and localization of precursors of Alzheimer’s disease A4 amyloid protein. Cell 57, 115–126. Yamatsuji T., Matsui T., Okamoto T., Komatsuzaki K., Takeda S., Fukumoto H., Iwatsubo T., Suzuki N., Asami-Odaka A., Ireland S., Kinane T. B., Giambarella U., and Nishimoto I. (1996a) G protein-mediated neuronal DNA fragmentation induced by familial Alzheimer’s disease-associated mutants of APP. Science 272, 1349 –1352. Yamatsuji T., Okamoto T., Takeda S., Murayama Y., Tanaka N., and Nishimoto I. (1996b) Expression of V642 APP mutant causes cellular apoptosis as Alzheimer trait-linked phenotype. EMBO J. 15, 498 –509.