A distinct ER/IC gamma-secretase competes with the proteasome for cleavage of APP

The deposition of amyloid-beta peptides (Abeta) in senile plaques (SPs) is a central pathological feature of Alzheimer's disease (AD). Since SPs are composed predominantly of Abeta1-42, which is more amyloidogenic in vitro, the enzymes involved in generating Abeta1-42 may be particularly important to the pathogenesis of AD. In contrast to Abeta1-40, which is generated in the trans-Golgi network and other cytoplasmic organelles, intracellular Abeta1-42 is produced in the endoplasmic reticulum/intermediate compartment (ER/IC), where it accumulates in a stable insoluble pool. Since this pool of insoluble Abeta1-42 may play a critical role in AD amyloidogenesis, we sought to determine how the production of intracellular Abeta is regulated. Surprisingly, the production of insoluble intracellular Abeta1-42 was increased by a putative gamma-secretase inhibitor as well as by an inhibitor of the proteasome. We further demonstrate that this increased generation of Abeta1-42 in the ER/IC is due to a reduction in the turnover of Abeta-containing APP C-terminal fragments. We conclude that the proteasome is a novel site for degradation of ER/IC-generated APP fragments. Proteasome inhibitors may augment the availability of APP C-terminal fragments for gamma-secretase cleavage and thereby increase production of Abeta1-42 in the ER/IC. Based on the organelle-specific differences in the generation of Abeta by gamma-secretase, we conclude that intracellular ER/IC-generated Abeta1-42 and secreted Abeta1-40 are produced by different gamma-secretases. Further, the fact that a putative gamma-secretase inhibitor had opposite effects on the production of secreted and intracellular Abeta may have important implications for AD drug design.     

Caspase cleavage of members of the amyloid precursor family of proteins

The synapse loss and neuronal cell death characteristic of Alzheimer's disease (AD) are believed to result in large part from the neurotoxic effects of beta-amyloid peptide (Abeta), a 40-42 amino acid peptide(s) derived proteolytically from beta-amyloid precursor protein (APP). However, APP is also cleaved intracellularly to generate a second cytotoxic peptide, C31, and this cleavage event occurs in vivo as well as in vitro and preferentially in the brains of AD patients (Lu et al. 2000). Here we show that APPC31 is toxic to neurons in primary culture, and that like APP, the APP family members APLP1 and possibly APLP2 are cleaved by caspases at their C-termini. The carboxy-terminal peptide derived from caspase cleavage of APLP1 shows a degree of neurotoxicity comparable to APPC31. Our results suggest that even though APLP1 and APLP2 cannot generate Abeta, they may potentially contribute to the pathology of AD by generating peptide fragments whose toxicity is comparable to that of APPC31.     

Apolipoprotein E and beta-amyloid (1-42) regulation of glycogen synthase kinase-3beta

Glycogen synthase kinase-3beta (GSK-3beta) is implicated in regulating apoptosis and tau protein hyperphosphorylation in Alzheimer's disease (AD). We investigated the effects of two key AD molecules, namely apoE (E3 and E4 isoforms) and beta-amyloid (Abeta) 1-42 on GSK-3beta and its major upstream regulators, intracellular calcium and protein kinases C and B (PKC and PKB) in human SH-SY5Y neuroblastoma cells. ApoE3 induced a mild, transient, Ca2+-independent and early activation of GSK-3beta. ApoE4 effects were biphasic, with an early strong GSK-3beta activation that was partially dependent on extracellular Ca2+, followed by a GSK-3beta inactivation. ApoE4 also activated PKC-alpha and PKB possibly giving the subsequent GSK-3beta inhibition. Abeta(1-42) effects were also biphasic with a strong activation dependent partially on extracellular Ca2+ followed by an inactivation. Abeta(1-42) induced an early and potent activation of PKC-alpha and a late decrease of PKB activity. ApoE4 and Abeta(1-42) were more toxic than apoE3 as shown by MTT reduction assays and generation of activated caspase-3. ApoE4 and Abeta(1-42)-induced early activation of GSK-3beta could lead to apoptosis and tau hyperphosphorylation. A late inhibition of GSK-3beta through activation of upstream kinases likely compensates the effects of apoE4 and Abeta(1-42) on GSK-3beta, the unbalanced regulation of which may contribute to AD pathology.     

Proteolytic antibody light chains alter beta-amyloid aggregation and prevent cytotoxicity

Beta-amyloid (Abeta), a peptide generated by proteolytic cleavage of the amyloid precursor protein (APP), is a major constituent of the neuritic plaques associated with Alzheimer's disease (AD). Up-regulation of alpha-secretase, which can hydrolyze Abeta between Lys16 and Leu17, has been proposed as a potential therapeutic strategy in the treatment of AD. Previously, we identified two light-chain antibody fragments that had proteolytic activity against Abeta, one with alpha-secretase-like activity and one with carboxypeptidase-like activity. Here we show that cleavage of Abeta40 by hk14, the light-chain antibody having carboxypeptidase-like activity, alters aggregation of Abeta and neutralizes any cytotoxic effects of the peptide. Cleavage of Abeta40 with c23.5, the light chain having alpha-secretase-like cleavage, substantially increases the aggregation rate of Abeta; however, it does not show any corresponding increase in cytotoxicity. The increase in aggregation resulting from hydrolysis by c23.5 can be attributed to the decreased solubility of the hydrolyzed products relative to the parent Abeta40, primarily the Abeta17-40 fragment. These results demonstrate that antibody fragment mediated proteolytic degradation of Abeta peptide can be a potential therapeutic route to control Abeta aggregation and toxicity in vivo. Our results also suggest that increasing alpha-secretase activity as a therapeutic route must be approached with some caution because this can lead to a substantial increase in aggregation.     

Cortical and leptomeningeal cerebrovascular amyloid and white matter pathology in Alzheimer's disease

Alzheimer's disease (AD) is characterized by neurofibrillary tangles and by the accumulation of beta-amyloid (Abeta) peptides in senile plaques and in the walls of cortical and leptomeningeal arteries as cerebral amyloid angiopathy (CAA). There also is a significant increase of interstitial fluid (ISF) in cerebral white matter (WM), the pathological basis of which is largely unknown. We hypothesized that the accumulation of ISF in dilated periarterial spaces of the WM in AD correlates with the severity of CAA, with the total Abeta load in the cortex and with Apo E genotype. A total of 24 AD brains and 17 nondemented age-matched control brains were examined. CAA was seen in vessels isolated from brain by using EDTA-SDS lysis stained by Thioflavin-S. Total Abeta in gray matter and WM was quantified by immunoassay, ApoE genotyping by PCR, and dilatation of perivascular spaces in the WM was assessed by quantitative histology. The study showed that the frequency and severity of dilatation of perivascular spaces in the WM in AD were significantly greater than in controls (P< 0.001) and correlated with Abeta load in the cortex, with the severity of CAA, and with ApoE epsilon4 genotype. The results of this study suggest that dilation of perivascular spaces and failure of drainage of ISF from the WM in AD may be associated with the deposition of Abeta in the perivascular fluid drainage pathways of cortical and leptomeningeal arteries. This failure of fluid drainage has implications for therapeutic strategies to treat Alzheimer's disease.     

Amyloidogenic function of the Alzheimer's disease-associated presenilin 1 in the absence of endoproteolysis.

Alzheimer's disease (AD) is characterized by the invariant accumulation of senile plaques predominantly composed of the pathologically relevant 42-amino acid amyloid beta-peptide (Abeta42). The presenilin (PS) proteins play a key role in Abeta generation. FAD-associated mutations in PS1 and PS2 enhance the production of Abeta42, and PS1 is required for physiological Abeta production, since a gene knockout of PS1 and dominant negative mutations of PS1 abolish Abeta generation. PS proteins undergo endoproteolytic processing, and current evidence indicates that fragment formation may be required for the amyloidogenic function of PS. We have now determined the sequence requirements for endoproteolysis of PS1. Mutagenizing amino acids at the previously determined major cleavage site (amino acid 298) had no effect on PS1 endoproteolysis. In contrast, mutations or deletions at the additional cleavage site around amino acid 292 blocked endoproteolysis. The uncleavable PS1 derivatives accumulated as full-length proteins and replaced the endogenous PS1 proteins. In contrast to the previously described aspartate mutations within transmembrane domains 6 and 7, the uncleaved PS1 variants do not act as dominant negative inhibitors of Abeta production. Moreover, when a FAD-associated mutation (M146L) was combined with a mutation blocking endoproteolysis, Abeta42 production still reached pathological levels. These data therefore demonstrate that endoproteolysis of presenilins is not an absolute prerequisite for the amyloidogenic function of PS1. These data also show that accumulation of the PS1 holoprotein is not associated with the pathological activity of PS1 mutations as suggested previously.     

Suppression of Abeta deposition in brain by peripheral administration of Fab fragments of anti-seed antibody

Assembly and deposition of amyloid beta-protein (Abeta) in the brain is a fundamental process of Alzheimer's disease (AD). We previously hypothesized that GM1 ganglioside-bound Abeta (GAbeta) is an endogenous seed for Abeta assembly in brain. Recently, we have succeeded in generation of a monoclonal antibody specific to GAbeta. Notably, this antibody, 4396C, per se substantially inhibits Abeta assembly in vitro. Here we report that the peripheral administration of Fab fragments of 4396C into transgenic mice expressing a mutant amyloid precursor protein gene, following the conjugation of the protein transduction domain of the Tat protein, markedly suppressed Abeta deposition in the brain. This result further supports our previous hypothesis and also provides a new insight into develop AD therapy through targeting seed Abeta in the brain, which selectively inhibits the initial step of the pathological process of AD.     

APP transgenic mice Tg2576 accumulate Abeta peptides that are distinct from the chemically modified and insoluble peptides deposited in Alzheimer's disease senile plaques.

The amyloid (Abeta) peptides generated in Hsiao's APP Tg2576 transgenic (Tg) mice are physically and chemically distinct from those characteristic of Alzheimer's disease (AD). Transgenic mouse Abeta peptides were purified using sequential size-exclusion and reverse-phase chromatographic systems and subjected to amino acid sequencing and mass spectrometry analyses. The mouse Abeta peptides lacked the extensive N-terminal degradations, posttranslational modifications, and cross-linkages abundant in the stable Abeta peptide deposits observed in AD. Truncated Abeta molecules appear to be generated in vivo by hydrolysis at multiple sites rather than by post-mortem C-terminal degradation. In contrast to AD amyloid cores, the Tg mice peptides were soluble in Tris-SDS-EDTA solutions, revealing both monomeric and SDS-stable oligomeric species of Abeta. In contrast to our report on Novartis Pharma APP23 Tg mice [Kuo et al. (2001) J. Biol. Chem. 276, 12991], which maintain high levels of soluble Abeta early on with later development of extensive vascular amyloid, Tg2576 mice exhibited an age-related elevation of soluble Abeta with relatively limited vascular amyloid deposition. The transgenic mouse levels of carboxy-terminal (CT) APP fragments were nearly 10-fold greater than those of human brains, and this condition may contribute to the unique pathology observed in these animals. Immunization of transgenic mice may act to prevent the pathological effects of betaAPP overproduction by binding CT molecules or halting their processing to toxic forms, in addition to having any effects on Abeta itself. Thus, differences in disease evolution and biochemistry must be considered when using transgenic animals to evaluate drugs or therapeutic interventions intended to reduce the Abeta burden in Alzheimer's disease.     

Amyloid precursor protein and amyloid beta peptide in human platelets. Role of cyclooxygenase and protein kinase C

The main component of Alzheimer's disease (AD) senile plaques is amyloid-beta peptide (Abeta), a proteolytic fragment of the amyloid precursor protein (APP). Platelets contain both APP and Abeta and may contribute to the perivascular amyloid deposition seen in AD. However, no data are available concerning the biochemical mechanism(s) involved in their formation and release by these cells. We found that human platelets released APP and Abeta following activation with collagen or arachidonic acid. Inhibition of platelet cyclooxygenase (COX) reduced APP but not Abeta release following those stimuli. In contrast, activation of platelets by thrombin and calcium ionophore caused release of both APP and Abeta in a COX-independent fashion. Ex vivo studies showed that, despite suppression of COX activity, administration of aspirin did not modify Abeta or APP levels in serum or plasma, suggesting that this enzyme plays only a minor role in vivo. We examined the regulation of APP cleavage and release from activated platelets and found that cleavage requires protein kinase C (PKC) activity and is regulated by the intracellular second messengers phosphatidylinositol 2-phosphate and Ca(2+). Our data provide the first evidence that in human platelets COX is a minor component of APP secretion whereas PKC plays a major role in the secretory cleavage of APP. By contrast, Abeta release may represent secretion of preformed peptide and is totally independent of both COX and PKC activity.     

APP substitutions V715F and L720P alter PS1 conformation and differentially affect Abeta and AICD generation

The 37-43 amino acid Abeta peptide is the principal component of beta-amyloid deposits in Alzheimer's disease (AD) brain, and is derived by serial proteolysis of the amyloid precursor protein (APP) by beta- and gamma-secretase. gamma-Secretase also cleaves APP at Val50 in the Abeta numbering (epsilon cleavage), resulting in the release of a fragment called APP intracellular domain (AICD). The aim of this study was to determine whether amino acid substitutions in the APP transmembrane domain differentially affect Abeta and AICD generation. We found that the APPV715F substitution, which has been previously shown to dramatically decrease Abeta40 and Abeta42 while increasing Abeta38 levels, does not affect in vitro generation of AICD. Furthermore, we found that the APPL720P substitution, which has been previously shown to prevent in vitro generation of AICD, completely prevents Abeta generation. Using a fluorescence resonance energy transfer (FRET) method, we next found that both the APPV715F and APPL720P substitutions significantly increase the distance between the N- and C-terminus of presenilin 1 (PS1), which has been proposed to contain the catalytic site of gamma-secretase. In conclusion, both APPV715F and APPL720P change PS1 conformation with differential effects on Abeta and AICD production.     

Effects of peptides derived from BACE1 catalytic domain on APP processing

One of the hallmarks of Alzheimer's disease (AD) is the deposition of beta-amyloid (Abeta) peptides in neuritic plaques. Abeta peptides are derived from sequential cleavage of amyloid precursor protein (APP) by beta- and gamma-secretases. beta-APP cleaving enzyme-1 (BACE1) has been shown to be the major beta-secretase and is a primary therapeutic target for AD. We report here novel BACE1 inhibitory peptidomimetics, which are derived from catalytic domains of BACE1 themselves, instead of APP cleavage sites and are structurally modified by myristoylation in N-terminus for efficient cell permeability. The peptides not only inhibited the formation of APPbeta (a soluble N-terminal fragment of APP cleaved by beta-secretase), but also significantly reduced Abeta40 production. Our results suggest a new approach for identifying inhibitory agents for the treatment of AD.     

Molecular and cellular mechanisms for Alzheimer's disease: understanding APP metabolism

Alzheimer's disease (AD) is the most common neurodegenerative disease associated with aging. One important pathologic feature of AD is the formation of extracellular senile plaques in the brain, whose major components are small peptides called beta-amyloid (Abeta) that are derived from beta-amyloid precursor protein (APP) through sequential cleavages by beta-secretase and gamma-secretase. Because of the critical role of Abeta in the pathogenesis of AD, unraveling the cellular and molecular events underlying APP/Abeta metabolism has been and remains, of paramount importance to AD research. In this article we will focus on the regulation of APP metabolism leading to Abeta generation. We will review current knowledge of the secretases (alpha-, beta-, and gamma-secretases) involved in APP processing and various molecular and cellular mechanisms underlying intracellular trafficking of APP, which is a highly regulated process and whose disturbance has direct impacts on the production of Abeta.     

Degradation of beta-amyloid by proteolytic antibody light chains

Deposition of beta-amyloid (Abeta) is considered an important early event in the pathogenesis of Alzheimer's disease (AD). Clearance of Abeta thus represents a potential therapeutic approach. Antibody-mediated clearance of Abeta by vaccination inhibited and cleared Abeta deposition in animal models; however, inflammatory side effects were observed in humans. An alternative potentially noninflammatory approach to facilitate clearance is to proteolytically cleave Abeta. We screened 12 proteolytic recombinant antibody fragments for potential alpha-secretase activity, a naturally occurring enzyme that cleaves between the Lys16 and Leu17 residues of the amyloid precursor protein (APP). We utilized the synthetic alpha-secretase substrate, benzyloxycarbonyl-l-lysine o-nitrophenyl ester (Z-lys-o-Np) as a preliminary screen for alpha-secretase activity. Two antibody light chain fragments that hydrolyzed Z-lys-o-Np were identified. Abeta hydrolysis was studied using mass spectrometry to identify the cleavage patterns of the antibodies. The recombinant antibody light chain antibody fragment, c23.5, showed alpha-secretase-like activity, producing the 1-16 and 17-40 amino acid fragments of Abeta. The second light chain antibody fragment, hk14, demonstrated carboxypeptidase-like activity, cleaving sequentially from the carboxyl terminal of Abeta. These antibody light chains provide a novel route toward engineering efficient therapeutic antibodies capable of cleaving Abeta in vivo.     

Different tau epitopes define Abeta42-mediated tau insolubility

Alzheimer's disease (AD) is characterized by extracellular beta-amyloid (Abeta(42))-containing plaques and intracellular neurofibrillary tangles. The latter are composed of hyperphosphorylated filamentous aggregates of the microtubule-associated protein tau. Previously we demonstrated pathological interactions between these two histopathological hallmarks using human SH-SY5Y neuroblastoma cells overexpressing wild-type and mutant forms of human tau. Exposure to pre-aggregated forms of Abeta(42) caused both the formation of AD-like tau-containing filaments and a decreased solubility of tau, both of which were prevented by mutating the S422 phospho-epitope of tau. Here, we expressed additional tau mutants in SH-SY5Y cells to assess the role of phosphorylation and cleavage sites of tau in tau aggregation. We found that the Abeta(42)-mediated decrease in tau solubility depends on the interplay of distinct phospho-epitopes of tau and not only on phosphorylation of the S422 epitope.     

Zebrafish (Danio rerio) presenilin promotes aberrant amyloid beta-peptide production and requires a critical aspartate residue for its function in amyloidogenesis.

Alzheimer's disease (AD) is characterized by the invariable accumulation of senile plaques composed of amyloid beta-peptide (Abeta). Mutations in three genes are known to cause familial Alzheimer's disease (FAD). The mutations occur in the genes encoding the beta-amyloid precursor protein (betaAPP) and presenilin (PS1) and PS2 and cause the increased secretion of the pathologically relevant 42 amino acid Abeta42. We have now cloned the zebrafish (Danio rerio) PS1 homologue (zf-PS1) to study its function in amyloidogenesis and to prove the critical requirement of an unusual aspartate residue within the seventh putative transmembrane domain. In situ hybridization and reverse PCR reveal that zf-PS1 is maternally inherited and ubiquitously expressed during embryogenesis, suggesting an essential housekeeping function. zf-PS1 is proteolytically processed to produce a C-terminal fragment (CTF) of approximately 24 kDa similar to human PS proteins. Surprisingly, wt zf-PS1 promotes aberrant Abeta42 secretion like FAD associated human PS1 mutations. The unexpected pathologic activity of wt zf-PS1 may be due to several amino acid exchanges at positions where FAD-associated mutations have been observed. The amyloidogenic function of zf-PS1 depends on the conserved aspartate residue 374 within the seventh putative transmembrane domain. Mutagenizing this critical aspartate residue abolishes endoproteolysis of zf-PS1 and inhibits Abeta secretion in human cells. Inhibition of Abeta secretion is accompanied by the accumulation of C-terminal fragments of betaAPP, suggesting a defect in gamma-secretase activity. These data provide further evidence that PS proteins are directly involved in the proteolytic cleavage of betaAPP and demonstrate that this function is evolutionarily conserved.     

Anti-amyloid activity of neprilysin in plaque-bearing mouse models of Alzheimer's disease

Abnormally high concentrations of beta-amyloid peptide (Abeta) and amyloid plaque formation in Alzheimer's disease (AD) may be caused either by increased generation or by decreased degradation of Abeta. Therefore, activation of mechanisms that lower brain Abeta levels is considered valuable for AD therapy. Neuronal upregulation of neprilysin (NEP) in young transgenic mice expressing the AD-causing amyloid precursor protein mutations (SwAPP) led to reduction of brain Abeta levels and delayed Abeta plaque deposition. In contrast, a comparable increase of brain NEP levels in aged SwAPP mice with pre-existing plaque pathology did not result in a significant reduction of plaque pathology. Therefore, we suggest that the potential of NEP for AD therapy is age-dependent and most effective early in the course of AD pathophysiology.     

A(beta) generation in autophagic vacuoles

Alzheimer's disease (AD) is the most common form of dementia among older people. It is characterized by the extracellular accumulation of beta-amyloid (Abeta) deposits called senile or neuritic plaques. Abeta is generated by the proteolytic cleavage of Abeta precursor protein (APP) by beta and gamma-secretases localized in the secretory and endocytic compartments. In this issue, Yu et al. (on p. 87) report a novel mechanism for the generation of Abeta peptides, which takes place in autophagic vacuoles (AVs) that accumulate in AD brains.     

Dietary deficiency increases presenilin expression, gamma-secretase activity, and Abeta levels: potentiation by ApoE genotype and alleviation by S-adenosyl methionine

Apolipoprotein E4 (ApoE4) is a risk factor for Alzheimer's disease (AD). Whether this risk arises from a deficient function of E4 or the lack of protection provided by E2 ir E3 is unclear. Previous studies demonstrate that deprivation of folate and vitamin E, coupled with dietary iron as a pro-oxidant, for 1 month displayed increased presenilin 1 (PS-1) expression, gamma-secretase, and Abeta generation in mice lacking ApoE (ApoE−/− mice). While ApoE−/− mice are a model for ApoE deficiency, they may not reflect the entire range of consequences of E4 expression. We therefore compared herein the impact of the above deficient diet on mice expressing human E2, E3, or E4. As folate deficiency is accompanied by a decrease in the major methyl donor, S -adenosyl methionine (SAM), additional mice received the deficient diet plus SAM. E2 was more protective than murine ApoE or E3 and E4. Surprisingly, PS-1 and gamma-secretase were over-expressed in E3 to the same extent as in E4 even under a complete diet, and were not alleviated by SAM supplementation. Abeta increased only in E4 mice maintained under the complete diet, and was alleviated by SAM supplementation. These findings suggest dietary compromise can potentiate latent risk factors for AD.     

APP processing is regulated by cytoplasmic phosphorylation

Amyloid-beta peptide (Abeta) aggregate in senile plaque is a key characteristic of Alzheimer's disease (AD). Here, we show that phosphorylation of amyloid precursor protein (APP) on threonine 668 (P-APP) may play a role in APP metabolism. In AD brains, P-APP accumulates in large vesicular structures in afflicted hippocampal pyramidal neurons that costain with antibodies against endosome markers and the beta-secretase, BACE1. Western blot analysis reveals increased levels of T668-phosphorylated APP COOH-terminal fragments in hippocampal lysates from many AD but not control subjects. Importantly, P-APP cofractionates with endosome markers and BACE1 in an iodixanol gradient and displays extensive colocalization with BACE1 in rat primary cortical neurons. Furthermore, APP COOH-terminal fragments generated by BACE1 are preferentially phosphorylated on T668 verses those produced by alpha-secretase. The production of Abeta is significantly reduced when phosphorylation of T668 is either abolished by mutation or inhibited by T668 kinase inhibitors. Together, these results suggest that T668 phosphorylation may facilitate the BACE1 cleavage of APP to increase Abeta generation.