Future Therapeutics in Alzheimer's Disease: Development Status of BACE Inhibitors

Alzheimer's disease (AD) is the primary cause of dementia in the elderly. It remains incurable and poses a huge socio-economic challenge for developed countries with an aging population. AD manifests by progressive decline in cognitive functions and alterations in behaviour, which are the result of the extensive degeneration of brain neurons. The AD pathogenic mechanism involves the accumulation of amyloid beta peptide (Aβ), an aggregating protein fragment that self-associates to form neurotoxic fibrils that trigger a cascade of cellular events leading to neuronal injury and death. Researchers from academia and the pharmaceutical industry have pursued a rational approach to AD drug discovery and targeted the amyloid cascade. Schemes have been devised to prevent the overproduction and accumulation of Aβ in the brain. The extensive efforts of the past 20 years have been translated into bringing new drugs to advanced clinical trials. The most progressed mechanism-based therapies to date consist of immunological interventions to clear Aβ oligomers, and pharmacological drugs to inhibit the secretase enzymes that produce Aβ, namely β-site amyloid precursor-cleaving enzyme (BACE) and γ-secretase. After giving an update on the development and current status of new AD therapeutics, this review will focus on BACE inhibitors and, in particular, will discuss the prospects of verubecestat (MK-8931), which has reached phase III clinical trials.

Products of the Alzheimer's disease amyloid precursor protein generated by,β-secretase are present in human platelets, and secreted upon degranulation

Proteolytic processing of Alzheimer's disease amyloid precursor protein (APP) by /3-secretase generates the Nterminus ofA/3 (which deposits in the brain) and releases a secreted ectodomain of APP (sAPP/3). We identified in human platelets a band at 125 kDa corresponding to APP ectodomain ending with C-terminal methionine residue (APP671) as characterized by an antibody specificfor the C-terminal methionine residue of sAPP/3. The same antibody also detected bands at -105 and U125 kDa in human brain homogenates. Platelet sAPP/3 is an isoform containing the Kunitzprotease inhibitor domain (sAPP/3-KPI+) and is released into the medium when platelets are induced to aggregate using agonists such as thrombin, collagen, phorbol 12-myristate 13-acetate, or calcium ionophore A2318 7. The release of sAPPB /from aggregatedplatelets is consistent with a role in regulation of the coagulation cascade and/or in platelet aggregation. These data together with previous reports suggest that human platelets contain the a-, /3-and y-secretase activities, and are a suitable system to study APP processing and Ap production, a pathway which is considered to be a prime targetfor therapeutic intervention in AD.

Critical analysis of the use of β-site amyloid precursor protein-cleaving enzyme 1 inhibitors in the treatment of Alzheimer's disease

Alzheimer’s disease (AD) is the major cause of dementia in the elderly and an unmet clinical challenge. A variety of therapies that are currently under development are directed to the amyloid cascade. Indeed, the accumulation and toxicity of amyloid-β (Aβ) is believed to play a central role in the etiology of the disease, and thus rational interventions are aimed at reducing the levels of Aβ in the brain. Targeting β-site amyloid precursor protein-cleaving enzyme (BACE)-1 represents an attractive strategy, as this enzyme catalyzes the initial and rate-limiting step in Aβ production. Observation of increased levels of BACE1 and enzymatic activity in the brain, cerebrospinal fluid, and platelets of patients with AD and mild cognitive impairment supports the potential benefits of BACE1 inhibition. Numerous potent inhibitors have been generated, and many of these have been proved to lower Aβ levels in the brain of animal models. Over 10 years of intensive research on BACE1 inhibitors has now culminated in advancing half a dozen of these drugs into human trials, yet translating the in vitro and cellular efficacy of BACE1 inhibitors into preclinical and clinical trials represents a challenge. This review addresses the promises and also the potential problems associated with BACE1 inhibitors for AD therapy, as the complex biological function of BACE1 in the brain is becoming unraveled.

Mild oxidative stress induces redistribution of BACE1 in non-apoptotic conditions and promotes the amyloidogenic processing of Alzheimer's disease amyloid precursor protein

BACE1 is responsible for β-secretase cleavage of the amyloid precursor protein (APP), which represents the first step in the production of amyloid β (Aβ) peptides. Previous reports, by us and others, have indicated that the levels of BACE1 protein and activity are increased in the brain cortex of patients with Alzheimer's disease (AD). The association between oxidative stress (OS) and AD has prompted investigations that support the potentiation of BACE1 expression and enzymatic activity by OS. Here, we have established conditions to analyse the effects of mild, non-lethal OS on BACE1 in primary neuronal cultures, independently from apoptotic mechanisms that were shown to impair BACE1 turnover. Six-hour treatment of mouse primary cortical cells with 10-40 µM hydrogen peroxide did not significantly compromise cell viability but it did produce mild oxidative stress (mOS), as shown by the increased levels of reactive radical species and activation of p38 stress kinase. The endogenous levels of BACE1 mRNA and protein were not significantly altered in these conditions, whereas a toxic H2O2 concentration (100 µM) caused an increase in BACE1 protein levels. Notably, mOS conditions resulted in increased levels of the BACE1 C-terminal cleavage product of APP, β-CTF. Subcellular fractionation techniques showed that mOS caused a major rearrangement of BACE1 localization from light to denser fractions, resulting in an increased distribution of BACE1 in fractions containing APP and markers for trans-Golgi network and early endosomes. Collectively, these data demonstrate that mOS does not modify BACE1 expression but alters BACE1 subcellular compartmentalization to favour the amyloidogenic processing of APP, and thus offer new insight in the early molecular events of AD pathogenesis.

Investigation of matrix metalloproteinases, MMP-2 and MMP-9, in plasma reveals a decrease of MMP-2 in Alzheimer's disease

Pathological changes in the Alzheimer's disease (AD) brain include amyoid-β (Aβ) plaques, and neurofibrillary tangles, as well as neuronal death and synaptic loss. Matrix metalloproteinases MMP-2 and MMP-9 are known to degrade Aβ, and their expressions are increased in the AD brain, in particular in the astrocytes surrounding amyloid plaque. To investigate a possible association between plasma metalloproteinases and AD, we quantified MMP-2 and MMP-9 activities in the plasma of healthy controls (HC, n = 56), cases with mild cognitive impairment (MCI, n = 45), and AD (n = 50). All cases had previously been imaged with Pittsburgh compound B (PiB) and had a Mini-Mental Status Examination (MMSE) assessment. MMP-2 and MMP-9 activity was determined using gelatine-zymography. There was a significant 1.5-fold decrease in MMP-2 activity in the AD group compared to HC (p < 0.001) and a 1.4-fold decrease compared to MCI (p < 0.01). There was no difference in MMP-9 levels between the three groups. A positive correlation was identified between MMP-2 plasma activity and MMSE score (r = 0.16, p < 0.05), but there was no association with PiB. This is the first report of a change in MMP-2 activity in AD plasma and these findings may provide some insight into AD pathogenesis.

BACE inhibitors as potential drugs for the treatment of Alzheimer's disease: focus on bioactivity

Current drug development for the treatment of Alzheimer's Disease is principally based on the amyloid cascade theory, and aims to reduce the levels of Aβ amyloid peptide in the brain. This can be achieved, either by decreasing peptide production through inhibition of β-secretase (also known as BACE-1) or γ-secretase, or by interfering with Aβ aggregation, or by promoting Aβ clearance. Targeting BACE-1, the proteolytic enzyme that initiates Aβ formation, has generated a lot of research interest recently and is currently thought to be one of the most promising therapeutic approaches. In this review, we summarize and discuss the latest patents and publications describing BACE-1 inhibitors, principally focussing on their drug properties and performance in preclinical trials.

Presenilins promote the cellular uptake of copper and zinc and maintain copper chaperone of SOD1-dependent copper/zinc superoxide dismutase activity

Dyshomeostasis of extracellular zinc and copper has been implicated in β-amyloid aggregation, the major pathology associated with Alzheimer disease. Presenilin mediates the proteolytic cleavage of the β-amyloid precursor protein to release β-amyloid, and mutations in presenilin can cause familial Alzheimer disease. We tested whether presenilin expression affects copper and zinc transport. Studying murine embryonic fibroblasts (MEFs) from presenilin knock-out mice or RNA interference of presenilin expression in HEK293T cells, we observed a marked decrease in saturable uptake of radiolabeled copper and zinc. Measurement of basal metal levels in 6-month-old presenilin 1 heterozygous knock-out (PS1(+/-)) mice revealed significant deficiencies of copper and zinc in several tissues, including brain. Copper/zinc superoxide dismutase (SOD1) activity was significantly decreased in both presenilin knock-out MEFs and brain tissue of presenilin 1 heterozygous knock-out mice. In the MEFs and PS1(+/-) brains, copper chaperone of SOD1 (CCS) levels were decreased. Zinc-dependent alkaline phosphatase activity was not decreased in the PS null MEFs. These data indicate that presenilins are important for cellular copper and zinc turnover, influencing SOD1 activity, and having the potential to indirectly impact β-amyloid aggregation through metal ion clearance.

p75NTR regulates Abeta deposition by increasing Abeta production but inhibiting Abeta aggregation with its extracellular domain

Accumulation of toxic amyloid-β (Aβ) in the cerebral cortex and hippocampus is a major pathological feature of Alzheimer's disease (AD). The neurotrophin receptor p75NTR has been proposed to mediate Aβ-induced neurotoxicity; however, its role in the development of AD remains to be clarified. The p75NTR/ExonIII-/- mice and APPSwe/PS1dE9 mice were crossed to generate transgenic AD mice with deletion of p75NTR gene. In APPSwe/PS1dE9 transgenic mice, p75NTR expression was localized in the basal forebrain neurons and degenerative neurites in neocortex, increased with aging, and further activated by Aβ accumulation. Deletion of the p75NTR gene in APPSwe/PS1dE9 mice reduced soluble Aβ levels in the brain and serum, but increased the accumulation of insoluble Aβ and Aβ plaque formation. There was no change in the levels of amyloid precursor protein (APP) and its proteolytic derivatives, or α-, β-, and γ-secretase activities, or in levels of BACE1, neprilysin (NEP), and insulin-degrading enzyme (IDE) proteins. Aβ production by cortical neurons of APPSwe/PS1dE9 mice was reduced by deletion of p75NTR gene in vitro. Recombinant extracellular domain of p75NTR attenuated the oligomerization and fibrillation of synthetic Aβ(42) peptide in vitro, and reduced local Aβ plaques after hippocampus injection in vivo. In addition, deletion of p75NTR attenuated microgliosis but increased the microhemorrhage profiles in the brain. The deletion of p75NTR did not significantly change the cognitive function of the mice up to the age of 9 months. Our data suggest that p75NTR plays a critical role in regulating Aβ levels by both increasing Aβ production and attenuating its aggregation, and they caution that a therapeutic intervention simply reducing p75NTR may exacerbate AD pathology.

A domain level interaction network of amyloid precursor protein and Abeta of Alzheimer's disease

The primary constituent of the amyloid plaque, beta-amyloid (Abeta), is thought to be the causal "toxic moiety" of Alzheimer's disease. However, despite much work focused on both Abeta and its parent protein, amyloid precursor protein (APP), the functional roles of APP and its cleavage products remain to be fully elucidated. Protein-protein interaction networks can provide insight into protein function, however, high-throughput data often report false positives and are in frequent disagreement with low-throughput experiments. Moreover, the complexity of the CNS is likely to be under represented in such databases. Therefore, we curated the published work characterizing both APP and Abeta to create a protein interaction network of APP and its proteolytic cleavage products, with annotation, where possible, to the level of APP binding domain and isoform. This is the first time that an interactome has been refined to domain level, essential for the interpretation of APP due to the presence of multiple isoforms and processed fragments. Gene ontology and network analysis were used to identify potentially novel functional relationships among interacting proteins.

Paradoxical condensation of copper with elevated beta-amyloid in lipid rafts under cellular copper deficiency conditions: implications for Alzheimer disease

Redox-active copper is implicated in the pathogenesis of Alzheimer disease (AD), beta-amyloid peptide (Abeta) aggregation, and amyloid formation. Abeta.copper complexes have been identified in AD and catalytically oxidize cholesterol and lipid to generate H2O2 and lipid peroxides. The site and mechanism of this abnormality is not known. Growing evidence suggests that amyloidogenic processing of the beta-amyloid precursor protein (APP) occurs in lipid rafts, membrane microdomains enriched in cholesterol. beta- and gamma-secretases, and Abeta have been identified in lipid rafts in cultured cells, human and rodent brains, but the role of copper in lipid raft amyloidogenic processing is presently unknown. In this study, we found that copper modulates flotillin-2 association with cholesterol-rich lipid raft domains, and consequently Abeta synthesis is attenuated via copper-mediated inhibition of APP endocytosis. We also found that total cellular copper is associated inversely with lipid raft copper levels, so that under intracellular copper deficiency conditions, Abeta.copper complexes are more likely to form. This explains the paradoxical hypermetallation of Abeta with copper under tissue copper deficiency conditions in AD.

APP intracellular domain is increased and soluble Aβ is reduced with diet-induced hypercholesterolemia in a transgenic mouse model of Alzheimer disease

Cholesterol is one of multiple factors, other than familial genetic mutations, that can influence amyloid-beta peptide (Abeta) metabolism and accumulation in Alzheimer disease (AD). The effect of a high-cholesterol diet on amyloid precursor protein (APP) processing in brain has not been thoroughly studied. This study was designed to further investigate the role of cholesterol in the production of Abeta and APP intracellular domain (AICD) in 12-month-old Tg2576 transgenic mice. The mice were maintained on a high-cholesterol diet for 6 weeks. We found that diet-induced hypercholesterolemia increased the APP cytosolic fragment AICD and reduced sAPPalpha in the Tg2576 mice compared to the mice on a control basal diet. In addition, the levels of detergent-extracted Abeta40 were reduced, although no change in guanidine-extracted Abeta levels was observed. Full-length APP, alpha/betaC-terminal fragment (alpha/betaCTF), and beta-secretase (BACE) were not different in the cholesterol-fed mice compared to the control diet-fed mice. This study suggests that a high dietary cholesterol in aged mice may not only influence Abeta metabolism, but also regulate the AICD levels. AICD has a proposed role in signal transduction and apoptosis, hence modulation of AICD production could be an alternative mechanism by which cholesterol contributes to AD pathogenesis.

The proteolytic processing of the amyloid precursor protein gene family members APLP-1 and APLP-2 involves alpha-, beta-, gamma-, and epsilon-like cleavages: modulation of APLP-1 processing by n-glycosylation

Amyloid precursor protein (APP) processing is of major interest in Alzheimer's disease research, since sequential cleavages by beta- and gamma-secretase lead to the formation of the 4-kDa amyloid Abeta protein peptide that accumulates in Alzheimer's disease brain. The processing of APP involves proteolytic conversion by different secretases leading to alpha-, beta-, gamma-, delta-, and epsilon-cleavages. Since modulation of these cleavages represents a rational therapeutic approach to control amyloid formation, its interference with the processing of the members of the APP gene family is of considerable importance. By using C-terminally tagged constructs of APLP-1 and APLP-2 and the untagged proteins, we have characterized their proteolytic C-terminal fragments produced in stably transfected SH-SY5Y cells. Pharmacological manipulation with specific protease inhibitors revealed that both homologues are processed by alpha- and gamma-secretase-like cleavages, and that their intracellular domains can be released by cleavage at epsilon-sites. APLP-2 processing appears to be the most elaborate and to involve alternative cleavage sites. We show that APLP-1 is the only member of the APP gene family for which processing can be influenced by N-glycosylation. Additionally, we were able to detect p3-like fragments of APLP-1 and p3-like and Abeta-like fragments of APLP-2 in the media of stably transfected SH-SY5Y cells.

Altered expression of apolipoprotein E, amyloid precursor protein and presenilin-1 is associated with chronic reactive gliosis in rat cortical tissue

A major characteristic feature of Alzheimer's disease is the formation of compact, extracellular deposits of beta-amyloid (senile plaques). These deposits are surrounded by reactive astrocytes, microglia and dystrophic neurites. Mutations in three genes have been implicated in early-onset familial Alzheimer's disease. However, inflammatory changes and astrogliosis are also believed to play a role in Alzheimer's pathology. What is unclear is the extent to which these factors initiate or contribute to the disease progression. Previous rat studies demonstrated that heterotopic transplantation of foetal cortical tissue onto the midbrain of neonatal hosts resulted in sustained glial reactivity for many months. Similar changes were not seen in cortex-to-cortex grafts. Using this model of chronic cortical gliosis, we have now measured reactive changes in the levels of the key Alzheimer's disease proteins, namely the amyloid precursor protein, apolipoprotein E and presenilin-1. These changes were visualised immunohistochemically and were quantified by western blot analysis. We report here that chronic cortical gliosis in the rat results in a sustained increase in the levels of apolipoprotein E and total amyloid precursor protein. Reactive astrocytes in heterotopic cortical grafts were immunopositive for both of these proteins. Using a panel of amyloid precursor protein antibodies we demonstrate that chronic reactive gliosis is associated with alternative cleavage of the peptide. No significant changes in apolipoprotein E or amyloid precursor protein expression were seen in non-gliotic cortex-to-cortex transplants. Compared to host cortex, the levels of both N-terminal and C-terminal fragments of presenilin-1 were significantly lower in gliotic heterotopic grafts.The changes described here largely mirror those seen in the cerebral cortex of humans with Alzheimer's disease and are consistent with the proposal that astrogliosis may be an important factor in the pathogenesis of this disease.

Pharmacological knock-down of the presenilin 1 heterodimer by a novel gamma -secretase inhibitor: implications for presenilin biology

Intramembranous cleavage of the beta-amyloid precursor protein by gamma-secretase is the final processing event generating amyloid-beta peptides, which are thought to be causative agents for Alzheimer's disease. Missense mutations in the presenilin genes co-segregate with early-onset Alzheimer's disease, and, recently, a close biochemical linkage between presenilins and the identity of gamma-secretase has been established. Here we describe for the first time that certain potent gamma-secretase inhibitors are able to interfere with the endoproteolytic processing of presenilin 1 (PS1). In addition, we identified a novel gamma-secretase inhibitor, [1S-benzyl-4R-[1-(5-cyclohexyl-2-oxo-2,3-dihydro-1H-benzo[e][1,4]diazepin-3(R,S)-ylcarbamoyl)-S-ethylcarbamoyl]-2R-hydroxy-5-phenyl-pentyl]-carbamic acid tert-butyl ester (CBAP), which not only physically interacts with PS1, but upon chronic treatment produces a "pharmacological knock-down" of PS1 fragments. This indicates that the observed accumulation of full-length PS1 is caused by a direct inhibition of its endoproteolysis. The subsequent use of CBAP as a biological tool to increase full-length PS1 levels in the absence of exogenous PS1 expression has provided evidence that wild-type PS1 endoproteolysis is not required either for PS1/gamma-secretase complex assembly or trafficking. Furthermore, in cell-based systems CBAP does not completely recapitulate PS1 loss-of-function phenotypes. Even though the beta-amyloid precursor protein cleavage and the S3 cleavage of the Notch receptor are inhibited by CBAP, an impairment of Trk receptor maturation was not observed.

The C-terminal fragment of the Alzheimer's disease amyloid protein precursor is degraded by a proteasome-dependent mechanism distinct from gamma-secretase

The beta-amyloid protein (Abeta) is derived by proteolytic processing of the amyloid protein precursor (APP). Cleavage of APP by beta-secretase generates a C-terminal fragment (APP-CTFbeta), which is subsequently cleaved by gamma-secretase to produce Abeta. The aim of this study was to examine the cleavage of APP-CTFbeta by gamma-secretase in primary cortical neurons from transgenic mice engineered to express the human APP-CTFbeta sequence. Neurons were prepared from transgenic mouse cortex and proteins labelled by incubation with [35S]methionine and [35S]cysteine. Labelled APP-CTFbeta and Abeta were then immunoprecipitated with a monoclonal antibody (WO2) specific for the transgene sequences. Approximately 30% of the human APP-CTFbeta (hAPP-CTFbeta) was converted to human Abeta (hAbeta), which was rapidly secreted. The remaining 70% of the hAPP-CTFbeta was degraded by an alternative pathway. The cleavage of hAPP-CTFbeta to produce hAbeta was inhibited by specific gamma-secretase inhibitors. However, treatment with proteasome inhibitors caused an increase in both hAPP-CTFbeta and hAbeta levels, suggesting that the alternative pathway was proteasome-dependent. A preparation of recombinant 20S proteasome was found to cleave a recombinant cytoplasmic domain fragment of APP (APPcyt) directly. The study suggests that in primary cortical neurons, APP-CTFbeta is degraded by two distinct pathways, one involving gamma-secretase, which produces Abeta, and a second major pathway involving direct cleavage of APP-CTFbeta within the cytoplasmic domain by the proteasome. These results raise the possibility that defective proteasome function could lead to an increase in Abeta production in the AD brain.

Aspartyl protease inhibitor pepstatin binds to the presenilins of Alzheimer's disease

Mutations in the presenilin genes PS1 and PS2 cause early-onset Alzheimer's disease by altering gamma-secretase cleavage of the amyloid precursor protein, the last step in the generation of Abeta peptide. Ablation of presenilin (PS) genes, or mutation of two critical aspartates, abolishes gamma-secretase cleavage, suggesting that PS may be the gamma-secretases. Independently, inhibition experiments indicate that gamma-secretase is an aspartyl protease. To characterize the putative gamma-secretase activity associated with presenilins, lysates from human neuroblastoma SH-SY5Y and human brain homogenates were incubated with biotin derivatives of pepstatin, followed by immunoprecipitation of PS and associated proteins, and biotin detection by Western blotting. Precipitation with PS1 antibodies, directed to either N-terminal or loop regions, yielded the same 43 kDa band, of apparent molecular mass consistent with that of full-length PS1, although it may represent an aspartyl protease complexed with PS1. Incubation of cell lysates with pepstatin-biotin, followed by streptavidin precipitation and PS1 Western blotting, revealed PS1 fragments and full-length protein, indicating that pepstatin-biotin bound to both cleaved and uncleaved PS1. Binding could be competed by gamma-secretase inhibitor L-685,458 and could not be achieved with a PS1 mutant lacking the two transmembrane aspartates. Pepstatin-biotin was also shown to bind to PS2. PS1 was specifically absorbed to pepstatin-agarose, with an optimal pH of 6. Binding of pepstatin-biotin to PS1 from lymphocytes of a heterozygous carrier of pathologic exon 9 deletion was markedly decreased as compared to control lymphocytes, suggesting that this PS1 mutation altered the pepstatin binding site.

Presenilin 2 expression in neuronal cells: induction during differentiation of embryonic carcinoma cells

Mutations in the presenilin 1 and 2 (PS1 and PS2) genes cause most cases of early onset Alzheimer's disease. The genes encode two homologous multipass membrane proteins. Since the endogenous expression of PS2 has been poorly analyzed to date, we studied PS2 expression and localization in cultured human neuroblastoma cells and mouse neuronal cells. PS2 was mainly detected as a full-length protein of about 52 kDa in these cells and in brain, in contrast to PS1 that is mainly detected as endoproteolytic N-terminal and C-terminal fragments. Using immunofluorescence we found that like PS1, PS2 colocalized with markers of the endoplasmic reticulum-Golgi intermediate compartment, ERGIC-53 and beta-COP. Double labeling for PS1 and PS2 indicated that both proteins are colocalized in neuroblastoma SH-SY5Y cells. To study PS2 expression during differentiation, mouse embryonic carcinoma P19 cells were treated with retinoic acid. We found minimal PS2 expression in undifferentiated cells, an increase from day 2, and a maximum at day 8 after treatment. PS1 expression remained constant during this period. The differential expression of PS1 and PS2 within the P19 cells following retinoic acid treatment indicates different utilization or temporal requirements for these proteins during neuronal differentiation.

Presenilin I expression in yeast lowers secretion of the amyloid precursor protein

Presenilin (PS) mutations are associated with early-onset Alzheimer's disease and PS proteins are involved with gamma-secretase cleavage of the amyloid precursor protein, APP. We have shown previously that alpha-, beta- and gamma-secretase cleavages of APP are conserved in Pichia pastoris. Here, we report co-expression of APP and PSI in P. pastoris and show by immunoelectron microscopy colocalization of these two proteins in expanded endoplasmic reticulum. Western blot analysis indicates a drastic reduction of both alpha- and beta-secretase products. A relative increase in beta-secretase product derived from immature APP is also observed, pointing to a beta-secretase activity of P. pastoris associated with the early secretory pathway.

Interaction of the Presenilins with the Amyloid Precursor Protein (APP)

The genes encoding presenilin-1 (PS1) and presenilin-2 (PS2) were identified as the genes that harbour mutations that cause more than 60% of early onset familial Alzheimer's disease cases (FAD) (1-3). So far, more than 40 missense mutations have been described for presenilin-1 and two have been found in the gene coding for presenilin-2 (reviewed in refs. 4 and 5). Carriers of mutated presenilin genes develop in their brain neuropathological changes characteristic of Alzheimer's disease including the deposition of amyloid Aβ peptide. The latter is released from its cognate amyloid precursor protein (APP) by a two-step proteolytic conversion: first, proteolysis of APP by β-secretase, which releases the N-terminus of Aβ, and second, conversion of the remaining fragment by γ-secretase, which cleaves within the predicted transmembrane region of APP. This releases the C-terminus of Aβ, which may end either at position 40 or, to a lesser extent, at position 42 (reviewed in ref. 6). The latter species, Aβ(1-42), is more prone to aggregation and deposition than Aβ(1-40) and is produced at higher levels in the brains and primary fibroblasts of FAD patients carrying PS missense mutations (7). The same result was obtained when cultured cells transfected with mutated PS1 orPS2, or transgenic mice harboring missense PS1 were analyzed for the production of Aβ(1-42): in every case increased amounts of the longer Aβ(1-42) species were observed (8-10). The mechanisms by which mutations in the PS genes affect the proteolytic processing of APP by γ-secretase have not been resolved in detail. There are two possibilities by which the normal processing of APP may be disturbed: either mutations in the presenilins affect APP metabolism in an indirect way by modulation of proteases or interaction with proteins involved in APP intracellular routing, or presenilins may modulate APP processing directly through physical interactions with APP. Such a direct interaction between presenilins and APP was first demonstrated by us for PS2 (11). Later on, formation of stable complexes with APP was reported not only for PS2 but also for PS1 (12,13,13a).

Proteolytic processing of the Alzheimer's disease amyloid precursor protein within its cytoplasmic domain by caspase-like proteases

Alzheimer's disease is characterized by neurodegeneration and deposition of betaA4, a peptide that is proteolytically released from the amyloid precursor protein (APP). Missense mutations in the genes coding for APP and for the polytopic membrane proteins presenilin (PS) 1 and PS2 have been linked to familial forms of early-onset Alzheimer's disease. Overexpression of presenilins, especially that of PS2, induces increased susceptibility for apoptosis that is even more pronounced in cells expressing presenilin mutants. Additionally, presenilins themselves are targets for activated caspases in apoptotic cells. When we analyzed APP in COS-7 cells overexpressing PS2, we observed proteolytic processing close to the APP carboxyl terminus. Proteolytic conversion was increased in the presence of PS2-I, which encodes one of the known PS2 pathogenic mutations. The same proteolytic processing occurred in cells treated with chemical inducers of apoptosis, suggesting a participation of activated caspases in the carboxyl-terminal truncation of APP. This was confirmed by showing that specific caspase inhibitors blocked the apoptotic conversion of APP. Sequence analysis of the APP cytosolic domain revealed a consensus motif for group III caspases ((IVL)ExD). Mutation of the corresponding Asp664 residue abolished cleavage, thereby identifying APP as a target molecule for caspase-like proteases in the pathways of programmed cellular death.

Processing of the Alzheimer's disease amyloid precursor protein in Pichia pastoris: immunodetection of alpha-, beta-, and gamma-secretase products

betaA4 (Abeta) amyloid peptide, a major component of Alzheimer's disease (AD) plaques, is a proteolytic product of the amyloid precursor protein (APP). Endoproteases, termed beta- and gamma-secretase, release respectively the N- and C-termini of the peptide. APP default secretion involves cleavage within the betaA4 domain by alpha-secretase. To study the conservation of APP processing in lower eukaryotes, the yeast Pichia pastoris was transfected with human APP695 cDNA. In addition to the full-length integral transmembrane protein found in the cell lysate, soluble/secreted APP (sAPP) was detected in the culture medium. Most sAPP comprised the N-terminal moiety of betaA4 and corresponds to sAPPalpha, the product of alpha-secretase. The culture medium also contained minor secreted forms detected by a monoclonal antibody specific for sAPPbeta (the ectodomain released by beta-secretase cleavage). Analysis of the cell lysates with specific antibodies also detected membrane-associated C-terminal fragments corresponding to the products of alpha and beta cleavages. Moreover, immunoprecipitation of the culture medium with three antibodies directed at distinct epitopes of the betaA4 domain yielded a 4 kDa product with the same electrophoretic mobility as betaA4 synthetic peptide. These results suggest that the alpha-, beta-, and gamma-secretase cleavages are conserved in yeast and that P. pastoris may offer an alternative to mammalian cells to identify the proteases involved in the generation of AD betaA4 amyloid.

Subcellular localization of the Alzheimer's disease amyloid precursor protein and derived polypeptides expressed in a recombinant yeast system

Different isoforms and derived polypeptides of the Alzheimer's disease amyloid protein precursor (A beta PP) have been expressed in the yeast Pichia pastoris. The expression characteristics of the different A beta PP polypeptides were studied by post-embedding immunogold electron microscopy with various A beta PP antibodies. The site of intracellular expression could be readily identified with specific antibodies. Full length A beta PP was expressed in association with the nuclear membrane and the endoplasmic reticulum. Secretory derivatives of A beta PP were localized in membrane-bound secretory vesicles. A construct encoding two copies of beta A4[1-42] linked head-to-tail (beta A4duplex) accumulated as irregular dense cytoplasmic and intranuclear inclusions which reacted with all beta A4 antibodies tested. A beta A4-C-terminal construct accumulated into membranous structures in the cytoplasm and nucleus and reacted with most antibodies to beta A4 and the cytoplasmic domain of A beta PP. The two shorter constructs containing the beta A4 sequence formed similar intranuclear aggregates to those reported for intranuclear inclusions of polyglutamine peptides from huntingtin (in Huntington's disease) and ataxin protein fragments (in spinocerebellar ataxia). This is of interest because intracellular aggregation of the polyglutamine and beta A4 peptides may affect cells by similar toxic mechanisms. These studies demonstrate clear differences in the expression properties of different A beta PP polypeptides.

Secretion of Alzheimer's disease Abeta amyloid peptide by activated human platelets

Alzheimer's disease (AD) is characterized by the deposition of Abeta (betaA4) peptides of 39 to 43 amino acid residues, which are normal cellular metabolic products derived by proteolysis of the amyloid precursor protein (APP). The physiologic function of Abeta/APP in vivo is poorly understood. We analyzed human platelets for Abeta production by immunoprecipitation coupled to immunoblotting. A 4-kd Abeta fragment that comigrates with an Abeta40 synthetic peptide and reacts with several antibodies specific for the N- and C-termini of Abeta is detected. The majority of platelet Abeta appears to end at residue 40, as determined by immunoreactivity with an Abeta40-specific antibody. Furthermore, Abeta is secreted upon platelet stimulation with the physiologic agonists thrombin and collagen, together with secretion of soluble APP (sAPP). A comparison between serum and plasma shows a 1.6-fold increase in Abeta levels and a 2.4-fold increase in sAPP levels in serum. This is consistent with the view that platelets are the primary source of circulating Abeta and APP. The release of platelet Abeta by physiologic stimuli suggests that it may play a role in platelet aggregation and coagulation or in the repair mechanisms associated with injury.

Alzheimer's disease-associated presenilin 1 in neuronal cells: evidence for localization to the endoplasmic reticulum-Golgi intermediate compartment

The recently identified Alzheimer's disease-associated presenilin 1 and 2 (PS1 and PS2) genes encode two homologous multi membrane-spanning proteins. Rabbit antibodies to the N-terminal domain of PS1 detected PS1 in human neuroblastoma SH-SY5Y wild type and PS1 transfectants (SY5Y-PS1) as well as in mouse P19, in CHO-K1 and CHO-APP770 transfected cells, in rat cerebellar granule and hippocampal neurons, and astrocytes. Immunoblotting detected full-length protein of 50 kDa, and a major presumptive cleavage product of 30 kDa. The immunofluorescence pattern resembled labeling of the endoplasmic reticulum-Golgi intermediate compartment (ERGIC) marker protein ERGIC-53. PS1 distribution showed slight condensation after brefeldin A and more marked condensation after incubation of cells at 16 degrees C, characteristic of the ERGIC compartment. Double labeling showed colocalization of ERGIC-53 with PS1 in the SY5Y-PS1 cells. PS1 labeling of SY5Y-PS1 and P19 cells showed overlap of the cis-Golgi marker p210 and colocalization with p210 after brefeldin A which causes redistribution of p210 to the ERGIC. Expression of PS1 did not change in level or cellular distribution during development of neurons in culture. Double labeling for the amyloid precursor protein (APP) and PS1 on SY5Y-PS1 cells and CHO-APP770 cells showed some overlap under control conditions. These results indicate that PS1 is a resident protein of the ERGIC and could be involved in trafficking of proteins, including APP, between the ER and Golgi compartments.

Formation of stable complexes between two Alzheimer's disease gene products: presenilin-2 and beta-amyloid precursor protein

Mutations in the presenilin genes are associated with early onset familial Alzheimer's disease and lead to increased accumulation of beta A4 peptide, the proteolytic product of the amyloid precursor protein (APP). To test whether presenilins interfere with APP metabolism, presenilin-2 (PS2) was coexpressed with APP in mammalian cells. Analysis of PS2 immunoprecipitates revealed that a fraction of APP was associated with the PS2 immunocomplexes. This non-covalent association was specific for the APP family of proteins and restricted to immature forms, occurring probably during transit through the endoplasmic reticulum. Additionally, coexpression with PS2 resulted in a decrease of APP secretion, suggesting a direct participation of presenilins in the intracellular sorting, trafficking and processing of APP molecules.

A novel metalloprotease in rat brain cleaves the amyloid precursor protein of Alzheimer's disease generating amyloidogenic fragments

The amyloid protein (A beta or beta A4) is the major constituent of amyloid plaques in the Alzheimer's disease brain. A beta is cleaved from the amyloid precursor protein (APP) by a mechanism which is poorly understood. Cell culture studies suggest that APP may be cleaved by secretases within the late Golgi compartment. Studies performed so far have mainly used exogenous APP and synthetic peptides as substrates. For this study, a Golgi and plasma membrane-enriched fraction was isolated from rat brain and incubated at 37 degrees C at pH 7.2 to study the degradation of endogenous APP. The breakdown of APP was accompanied by the concomitant generation of A beta-containing C-terminal fragments, in a time-dependent fashion. The metal ion chelators EDTA and 1,10-phenanthroline inhibited this degradation. The inhibition by EDTA was reversed by 50 microM Zn2+ but not by other metal ions. The protease activity was not inhibited by cysteine, serine or aspartic protease inhibitors nor was it inhibited by compounds which are inhibitors of known metalloendopeptidases and matrix metalloproteinases (cFP, phosphoramidon and TIMP-2). Our data suggest that a novel Zn(2+)-dependent metalloprotease activity associated with a Golgi and plasma membrane-enriched fraction can degrade endogenous APP to generate A beta containing C-terminal fragments. This protease may generate amyloidogenic fragments of APP which may serve as precursors for A beta.

Gelatinase A possesses a beta-secretase-like activity in cleaving the amyloid protein precursor of Alzheimer's disease

The ability of the 72 kDa gelatinase A to cleave the amyloid protein precursor (APP) was investigated. HeLa cells were transfected with an APP695 plasmid. The cells were incubated with gelatinase A, which cleaved the 110 kDa cell-surface APP, releasing a 100 kDa form of the protein. A peptide homologous to the beta-secretase site was cleaved by gelatinase A adjacent to a glutamate residue at position -3 (beta A4 numbering system). A peptide homologous to the alpha-secretase site was not cleaved. The results demonstrate that 72 kDa gelatinase A is not an alpha-secretase, but that it may have a beta-secretase activity.

Candidate gamma-secretases in the generation of the carboxyl terminus of the Alzheimer's disease beta A4 amyloid: possible involvement of cathepsin D

beta A4 amyloid peptide, the main constituent of amyloid plaques and cerebrovascular amyloid deposits associated with Alzheimer's disease, derives from a large precursor protein (APP) by the action of beta- and gamma-secretases, the unidentified endoproteases which release the amino and carboxyl termini of beta A4, respectively. Several gamma-secretase cleavage sites exist which yield the more soluble (1-39/40) forms of beta A4 and the longer forms (1-42/43) which have a greater tendency to aggregate into amyloid plaques. gamma-Secretase activity may therefore be critical in amyloid formation. In this study, a synthetic peptide which encompasses the various gamma-secretase cleavage sites was used as a substrate to probe proteases of various classes and specificities. Elastase, collagenase, and cathepsin D cleaved at the amyloidogenic sites (after Ala42 or after Thr43) to release the carboxyl termini of the aggregating forms. In addition, collagenase and pepsin released the carboxyl terminus of the more soluble forms. Human brain fractions enriched in lysosomes contained a proteolytic activity that cleaved the substrate at the amyloidogenic site(s). This activity was more active at acidic pH and was inhibited by pepstatin, two characteristics of the lysosomal aspartyl proteinase cathepsin D. The same lysosomal fractions were found to contain APP carboxyl-terminal fragments which are potentially amyloidogenic. These were degraded, only in acidic conditions, by an endogenous protease activity inhibited by pepstatin. Thus, a cathepsin D-like activity from human brain is a candidate for APP gamma-secretase(s).

Proteolytic processing of Alzheimer's disease beta A4 amyloid precursor protein in human platelets

The processing of amyloid precursor protein (APP) and production of beta A4 amyloid are events likely to influence the development and progression of Alzheimer's disease, since beta A4 is the major constituent of amyloid deposited in this disorder. Our previous studies showed that human platelets contain full-length APP (APPFL) and are a suitable substrate to study normal APP processing. In the present study, we show that a 22-kDa beta A4-containing carboxyl-terminal fragment (22-CTF) of APP is present in unstimulated platelets. Both APPFL and 22-CTF are proteolytically degraded when platelets are activated with thrombin, collagen, or calcium ionophore A23187. Complete cleavage of APPFL and 22-CTF require the presence of extracellular calcium. Following stimulation in the presence of calcium, a new CTF of 17 kDa is generated, and the NH2-terminal epitope of beta A4 amyloid is lost. Preincubation of platelets with the cell-permeable cysteine protease inhibitors calpeptin, (2S,3S)-trans-epoxysuccinyl-L-leucyl-amido-3-methylbutane ethyl ester (E64d), Na alpha-p-tosyl-L-lysine chloromethyl ketone, or calcium chelator EGTA before platelet stimulation inhibits the degradation of both APPFL and 22-CTF. Divalent metal ions including zinc, copper, and cobalt inhibit the degradation of APPFL and 22-CTF. This study suggests that a calcium-dependent neutral cysteine protease is involved in the proteolytic processing of an amyloidogenic species of APP in human platelets.

A synthetic substrate assay for the gamma-secretase of the beta-A4 amyloid of Alzheimer's disease

gamma-secretase, the endoprotease which releases the C-terminus of beta A4 amyloid peptide, cleaves within the hydrophobic transmembrane domain of the amyloid precursor protein. In order to obtain a substrate for gamma-secretase, a dodecapeptide which spans the cleavage site was synthesized, labelled with 125-iodine and conjugated to an agarose gel. A radiometric solid-phase assay was developed using this immobilized substrate. Peptide products were separated by reverse-phase HPLC and TLC to allow characterization of the cleavage site(s).

A crossreactive antipeptide monoclonal antibody with specificity for lysyl-lysine

Synthetic peptides meeting certain guidelines have been used as immunogens to generate antibodies with predefined specificity. We have raised and characterized using established methods a monoclonal antibody against a synthetic peptide corresponding to the 18-amino acid carboxyterminal sequence (A194-211) of the platelet-derived growth factor (PDGF) A chain expressed by the U343 human glioma cell line. This antibody was generated in order to carry out structure-function studies on this region of PDGF whose biological significance is not yet clear. Anti-PDGF-A194-211 was found to be a low titre, IgM kappa molecule, with a Kd of 2.8 x 10(-7) M. When antibody reactivity was tested with parent PDGF-AAL (A chain homodimer containing a carboxyterminal extension) significant binding was observed. Surprisingly, 125I-PDGF-AAS, consisting of truncated A chains but lacking the extension was also bound. Moreover, poly-L-lysine, beta-thromboglobulin, PDGF-A194-211, and myoglobin competed dose-dependently with 125I-PDGF-AAL for antibody. 125I-bovine serum albumin was also bound. Examination of the primary sequence of proteins and peptides bound by the antibody revealed only one shared structural motif: a lysyl-lysine moiety. Selected small synthetic peptides containing this and other sequences were used as potential competitors of 125I-PDGF-A194-211 in antibody binding. Lysyl-lysyl-glycyl-glutamic acid [corrected] and lysyl-lysine competed, whereas lysyl-leucine did not. These results suggest that as few as two amino acid residues constitute a functional antigenic determinant and contrast with most previous estimates of the minimum number of residues required. Furthermore, we show that guidelines governing the design of synthetic peptides for their use as antigens to produce monoclonal antibodies of predetermined specificity may be unreliable.

An alternative strategy for the radioimmunoassay of angiotensin peptides using amino-terminal-directed antisera: measurement of eight angiotensin peptides in human plasma

We describe here a method of measuring angiotensin peptides and their carboxy-truncated metabolites in human plasma using N-terminal-directed antisera. Antisera raised against N-acetylated angiotensin (Ang) II and N-acetylated Ang III analogues were used to develop two radioimmunoassays. Extracted plasma samples were acetylated prior to separation of cross-reacting angiotensin peptides by high-performance liquid chromatography (HPLC). Fractions were assayed with both antisera to obtain measurements for eight angiotensin peptides. Angiotensin levels measured in normal males were (fmol/ml plasma, mean +/- s.e.m., n = 14): Ang-(1-7) 1.0 +/- 0.2, Ang II 13.9 +/- 2.0, Ang-(1-9) less than 0.4, Ang I 19.5 +/- 2.4, Ang-(2-7) less than 1.1, Ang III 2.9 +/- 1.0, Ang-(2-9) less than 2.1, Ang-(2-10) 2.4 +/- 0.8. Hypertensive patients receiving angiotensin converting enzyme (ACE) inhibitor therapy (n = 8) had an increase in Ang I to 187.3 +/- 107.2 fmol/ml (P = 0.002), and a reduction in Ang II to 4.8 +/- 1.2 fmol/ml (P less than 0.001). Furthermore, these patients showed a ninefold increase in Ang-(1-7) to 9.7 +/- 4.3 fmol/ml (P less than 0.001), indicating a role for prolylendopeptidase in the metabolism of Ang I in vivo. These N-terminal assays have demonstrated that carboxy-truncated metabolites of Ang I and Ang II make little contribution to plasma angiotensin peptides, except during ACE inhibitor therapy. Furthermore, these antisera allow the measurement of Ang I and Ang II in the same radioimmunoassay of fractions from HPLC, providing a highly reliable estimate of the Ang II:Ang I ratio.

[Production and characterization of human antirenin antibodies prepared with synthetic immunogens]

Specific blockade of renin angiotensin system can be obtained both by enzymatic inhibitors and by passive and active immunization against renin. Recent studies have shown that synthetic peptides mimicking a protein segment can be used as immunogens to elicit antibodies which react with the parent molecule. In order to develop synthetic antirenin antigens we have selected peptidic sequences from human active renin and synthesized the corresponding peptides to produce antibodies able to recognize the entire human molecule and to inhibit its enzymatic activity. Three dimensional models of human renin were used to predict 17 putative epitopes. Then 18 peptides related to 13 potential antigenic determinants were synthesized by solid or liquid phase technic and 11 were shown to be antigenic when tested by their binding to several polyclonal and monoclonal human renin antibodies. The peptides were injected into rabbits and antisera tested by radio-immunoassay (RIA) and enzyme linked immunosorbent assay (ELISA) showed that nine peptides were immunogenic. Mainly, antiserum raised against the peptide mimicking the beta-hairpin 81-50 of active human renin which lies across the catalytic cleft, produced a 25 p. 100 inhibition of plasma renin activity at a 1: 50 dilution. Immunoglobulins, purified from antibodies raised against this epitopes, bound labelled renin and inhibited enzymatic activity of pure human renin on its synthetic tetradecapeptide substrate, in a dose dependent manner.

Characterization of five epitopes of human renin from a computer model

Antigenic determinants in proteins have been found to coincide with surface regions accessible to a large probe of 1-nm radius, which approximates the size of an antibody binding domain. To predict epitopes of human renin, a computer algorithm was used to calculate the accessibility of a model of the three-dimensional structure of human renin to a 1-nm probe. Of the 17 segments predicted to be epitopes or parts of epitopes, 6 were synthesized and tested for their recognition by 11 polyclonal antibodies prepared against pure human renin. Four peptides, Y-133-144, C-180-188, Y-211-224, and Y-300-310, were bound by some polyclonal anti-renin antibodies when tested in both a solution assay and an enzyme-linked immunosorbent assay (ELISA), and one peptide, C-290-296-G, was bound only in an ELISA. Of the five epitopes identified, peptide Y-211-224 seems dominant, since it was bound by all antisera in the solution assay and by 8 of the 11 in the ELISA. This peptide contains a disulfide loop, and the corresponding linear peptide obtained after reduction of the disulfide was not recognized by any anti-renin antisera in the solution assay. This suggests that the epitope is conformationally dependent and that a disulfide bridge linking cysteines-217 and -221 is present in the native structure of human renin. Purified immunoglobulins raised to peptides Y-133-144 and Y-211-224 (located near, but not at, the catalytic site) were found to inhibit the cleavage of human angiotensinogen by purified human renin but not the cleavage of a tetradecapeptide substrate representing the N-terminal region of angiotensinogen.(ABSTRACT TRUNCATED AT 250 WORDS)

Production and characterization of human renin antibodies prepared with synthetic immunogens

Renin is an aspartyl protease that plays a key role in regulating blood pressure. The aim of this study was to use synthetic peptides to produce antibodies able to recognize the entire human renin molecule and to inhibit its enzymatic activity. Two three-dimensional renin models were used to predict 17 putative epitopes. Then 18 peptides related to 13 potential antigenic determinants were synthesized by solid- or liquid-phase technique, and 11 were shown to be antigenic when tested by their binding to several polyclonal human renin antibodies. The peptides were injected into rabbits, and antisera tested by radio-immunoassay (RIA) and enzyme-linked immunosorbent assay (ELISA) showed that nine peptides were immunogenic. The peptide related to the flap region which lies across the renin catalytic cleft showed potentially useful characteristics. Immunoglobulin G, purified from antibodies raised against this epitope, bound labelled renin and inhibited its enzymatic activity in a dose-dependent manner. This approach constitutes the basis for the development of a synthetic antirenin vaccine able to inhibit the renin-angiotensin system specifically.