BACE1: the beta-secretase enzyme in Alzheimer's disease

Data that have accumulated for well over a decade have implicated the beta-amyloid (Abeta) peptide as a central player in the pathogenesis of Alzheimer's disease (AD). Amyloid plaques, composed primarily of Abeta progressively form in the brains of AD patients, and mutations in three genes (amyloid precursor protein [APP] and presenilin 1 and 2 [PS1 and PS2]) cause early-onset familial AD (FAD) by directly increasing production of the toxic, plaque-promoting Abeta42 peptide. Given the strong association between Abeta and AD, it is likely that therapeutic strategies to lower the levels of Abeta in the brain should prove beneficial for the treatment of AD. One such strategy could involve inhibiting the enzymes that generate Abeta. Abeta is a product of catabolism of the large type-I membrane protein APP. Two proteases, called beta- and gamma-secretase, endoproteolyze APP to liberate the Abeta peptide. Recently, the molecules responsible for these proteolytic activities have been identified. Several lines of evidence suggest that the PS1 and PS2 proteins are gamma-secretase, and the identity of beta-secretase has been shown to be the novel transmembrane aspartic protease, beta-site APP-cleaving enzyme 1 (BACE1; also called Asp2 and memapsin 2). BACE2, a protease homologous to BACE1, was also identified, and together the two enzymes define a new family of transmembrane aspartic proteases. BACE1 exhibits all the functional properties of beta-secretase, and as the key enzyme that initiates the formation of Abeta, BACE1 is an attractive drug target for AD. This review discusses the identification and initial characterization of BACE1 and BACE2, and summarizes recent studies of BACE1 knockout mice that have validated BACE1 as the authentic beta-secretase in vivo.     

Proteolytic processing of the Alzheimer's disease amyloid precursor protein in brain and platelets

Proteolytic processing of the amyloid precursor protein by beta -and gamma-secretases results in the production of Alzheimer's disease (AD) Abeta amyloid peptides. Modulation of secretase activity is being investigated as a potential therapeutic approach. Recent studies with human brain have revealed that the beta-secretase protein, BACE, is increased in cortex of AD patients. Analysis of betaCTF (or C99), the amyloid precursor protein (APP) product of BACE cleavage that is the direct precursor to Abeta, shows it is also elevated in AD, underlying the importance of beta-secretase cleavage in AD pathogenesis. The C-terminal product of gamma-secretase cleavage of APP, epsilonCTF (or AICD), is enriched in human brain cortical nuclear fractions, a subcellular distribution appropriate for a putative involvement of APP cytosolic domain in signal transduction. Analysis of AD cortex samples, particularly that of a carrier of a familial APP mutation, suggests that processing of APP transmembrane domain generates an alternative CTF product. All these particularities observed in the AD brain demonstrate that APP processing is altered in AD. The transgenic mouse model Tg2576 seems to be a promising laboratory tool to test potential modulators of Abeta formation. Indeed, C-terminal products of alpha-, beta-, and gamma-secretase cleavage are readily detectable in the brain of these transgenic mice. Finally, the finding of the same secretase products in platelets and neurons make platelets a potentially useful and easily accessible clinical tool to monitor effects of novel therapies based on inhibition of beta- or gamma-secretase.     

Temporal memory deficits in Alzheimer's mouse models: rescue by genetic deletion of BACE1

Transgenic mouse models of Alzheimer's disease (AD) exhibit amyloid-beta (Abeta) accumulation and related cognitive impairments. Although deficits in hippocampus-dependent place learning have been well characterized in Alzheimer's transgenic mice, little is known about temporal memory function in these AD models. Here, we applied trace fear conditioning to two different Alzheimer's mouse models and investigated the relationship between pathogenic Abeta and temporal memory deficits. This behavioral test requires hippocampus-dependent temporal memory processing as the conditioned and unconditioned stimuli are separated by a trace interval of 30 s. We found that both amyloid precursor protein (APP) transgenic (Tg2576) and APP/presenilin (PS)1 transgenic (Tg6799) mice were impaired in memorizing this association across the time gap. Both transgenic groups performed as well as wild-type control mice in delay fear conditioning when the trace interval was removed, indicating that the trace conditioning deficits are hippocampus-specific. Importantly, Tg6799 mice engineered to lack the major Alzheimer's beta-secretase (beta-site APP-cleaving enzyme 1: BACE1) showed behavioral rescue from temporal memory deficits. Elevated levels of soluble Abeta oligomers found in Tg6799+ mouse brains returned to wild-type control levels without changes in APP/PS1 transgene expression in BACE1-/- * Tg6799+ bigenic mouse brains, suggesting Abeta oligomers as potential mediators of memory loss. Thus, trace fear conditioning is a useful assay to test the mechanisms and therapeutic interventions for Abeta-dependent deficits in temporal associative memory. Our gene-based approach suggests that lowering soluble Abeta oligomers by inhibiting BACE1 may be beneficial for alleviating cognitive disorders in AD.     

BACE (β-secretase) modulates the processing of APLP2 in vivo

BACE is an aspartyl protease that cleaves the amyloid precursor protein (APP) at the beta-secretase cleavage site and is involved in Alzheimer's disease. The aim of our study was to determine whether BACE affects the processing of the APP homolog APLP2. To this end, we developed BACE knockout mice with a targeted insertion of the gene for beta-galactosidase. BACE appeared to be exclusively expressed in neurons as determined by differential staining. BACE was expressed in specific areas in the cortex, hippocampus, cerebellum, pons, and spinal cord. APP processing was altered in the BACE knockouts with Abeta levels decreasing. The levels of APLP2 proteolytic products were decreased in BACE KO mice, but increased in BACE transgenic mice. Overexpression of BACE in cultured cells led to increased APLP2 processing. Our results strongly suggest that BACE is a neuronal protein that modulates the processing of both APP and APLP2.     

A novel immunotherapy for Alzheimer's disease: antibodies against the beta-secretase cleavage site of APP

One of the main neuropathological lesions observed at brain autopsy of Alzheimer's disease (AD) patients are the extracellular senile plaques mainly composed of amyloid-beta (Abeta) peptides. Abeta is generated by proteolytic processing of amyloid precursor protein (APP) via beta and gamma-secretases. The beta-secretase APP cleaving enzyme 1 (BACE1) has become a target of intense research aimed at blocking the enzyme activity. Recent studies showed that BACE1 is involved in processing other non-APP substrates, and that other proteases are involved in APP processing. We have recently established a novel approach to inhibit Abeta production via antibodies against the beta-secretase cleavage site of APP. These antibodies bind wild type and Swedish mutated APP expressed in transgenic mice brain tissues. The isolated antibodies do not bind any form of Abeta peptides. Antibody up-take experiments, using Chinese hamster ovary cells expressing wild-type APP, suggest that antibody internalization and trafficking are mediated via the endocytic pathway. Administration of antibodies to the cells growing media resulted in a considerable decrease in intracellular Abeta levels, as well as in the levels of the corresponding C-terminal fragment (C99). The relevance of intra-neuronal accumulation of mainly Abeta42 as an early event in AD pathogenesis suggests that this approach may be applicable as a novel therapeutic strategy in AD treatment.     

BACE (beta-secretase) modulates the processing of APLP2 in vivo

BACE is an aspartyl protease that cleaves the amyloid precursor protein (APP) at the beta-secretase cleavage site and is involved in Alzheimer's disease. The aim of our study was to determine whether BACE affects the processing of the APP homolog APLP2. To this end, we developed BACE knockout mice with a targeted insertion of the gene for beta-galactosidase. BACE appeared to be exclusively expressed in neurons as determined by differential staining. BACE was expressed in specific areas in the cortex, hippocampus, cerebellum, pons, and spinal cord. APP processing was altered in the BACE knockouts with Abeta levels decreasing. The levels of APLP2 proteolytic products were decreased in BACE KO mice, but increased in BACE transgenic mice. Overexpression of BACE in cultured cells led to increased APLP2 processing. Our results strongly suggest that BACE is a neuronal protein that modulates the processing of both APP and APLP2.     

Platelet amyloid precursor protein processing: a bio-marker for Alzheimer's disease

The amyloid precursor protein (APP) in brain is processed either by an amyloidogenic pathway by beta-secretase and gamma-secretase to yield Abeta (beta-amyloid 4 kDa) peptide or by alpha-secretase within the beta-amyloid domain to yield non-amyloidogenic products. We have studied blood platelet levels of a 22-kDa fragment containing the Abeta (beta-amyloid 4 kDa) peptide, beta-secretase (BACE1), alpha-secretase (ADAM10), and APP isoform ratios of the 120-130 kDa to 110 kDa peptides from 31 Alzheimer's disease (AD) patients and 10 age-matched healthy control subjects. We found increased levels of Abeta4, increased activation of beta-secretase (BACE1), decreased activation of alpha-secretase (ADAM10) and decreased APP ratios in AD patients compared to normal control subjects. These observations indicate that the blood platelet APP is processed by the same amyloidogenic and non-amyloidogenic pathways as utilized in brain and that APP processing in AD patients is altered compared to control subjects and may be a useful bio-marker for the diagnosis of AD, the progression of disease and for monitoring drug responses in clinical trials.     

Polymorphism in the BACE gene influences the risk for Alzheimer's disease

Pathological characteristics of Alzheimer's disease (AD) are neurofibrillary tangles and amyloid-beta (Abeta) plaques. Abeta is generated by cleavage of the amyloid precursor protein by beta- and gamma-secretases. BACE (beta-site APP cleaving enzyme) was identified as the beta-secretase. Variations of the BACE gene might influence activity and function of the protein and, thus, might influence the pathogenesis of AD. Consequently, we investigated the association of different BACE polymorphisms with AD. BACE exon 5 polymorphism influenced the risk of AD. This effect was most pronounced in apolipoprotein E4 allele carriers. Furthermore, Abeta(42) CSF levels were influenced by BACE genotype. It appears that BACE polymorphism plays a more important role in the development of AD than previously assumed, possibly by influencing Abeta(42) levels.     

BACE1 structure and function in health and Alzheimer's disease

Amyloid plaques, hallmark neuropathological lesions in Alzheimer's disease (AD) brain, are composed of the beta-amyloid peptide (Abeta). Much evidence suggests that Abeta is central to the pathophysiology of AD and is likely to play an early role in this intractable neurodegenerative disorder. Given the strong correlation between Abeta and AD, therapeutic strategies to lower cerebral Abeta levels should prove beneficial for AD treatment. Abeta is derived from amyloid precursor protein (APP) via cleavage by two proteases, beta- and gamma-secretase. The beta-secretase has been identified as a novel aspartic protease named BACE1 (beta-site APP Cleaving Enzyme 1) that initiates Abeta formation. Importantly, BACE1 appears to be dysregulated in AD. As the rate-limiting enzyme in Abeta generation, BACE1, in principle, is an excellent therapeutic target for strategies to reduce the production of Abeta in AD. While BACE1 knockout (BACE1-/-) mice have been instrumental in validating BACE1 as the authentic beta-secretase in vivo, data indicates that complete abolishment of BACE1 may be associated with specific behavioral and physiological alterations. Recently a number of non-APP BACE1 substrates have been identified. It is plausible that failure to process certain BACE1 substrates may underlie some of the reported abnormalities in the BACE1-/- mice. Here we review the basic biology of BACE1, focusing on the regulation, structure and function of this enzyme. We pay special attention to the putative function of BACE1 during normal conditions and discuss in detail the relationship that exists between key risk factors for AD and the pathogenic alterations in BACE1 that are observed in the diseased state.     

Cerebrolysin decreases amyloid-beta production by regulating amyloid protein precursor maturation in a transgenic model of Alzheimer's disease

Cerebrolysin is a peptide mixture with neurotrophic effects that might reduce the neurodegenerative pathology in Alzheimer's disease (AD). We have previously shown in an amyloid protein precursor (APP) transgenic (tg) mouse model of AD-like neuropathology that Cerebrolysin ameliorates behavioral deficits, is neuroprotective, and decreases amyloid burden; however, the mechanisms involved are not completely clear. Cerebrolysin might reduce amyloid deposition by regulating amyloid-beta (Abeta) degradation or by modulating APP expression, maturation, or processing. To investigate these possibilities, APP tg mice were treated for 6 months with Cerebrolysin and analyzed in the water maze, followed by RNA, immunoblot, and confocal microscopy analysis of full-length (FL) APP and its fragments, beta-secretase (BACE1), and Abeta-degrading enzymes [neprilysin (Nep) and insulin-degrading enzyme (IDE)]. Consistent with previous studies, Cerebrolysin ameliorated the performance deficits in the spatial learning portion of the water maze and reduced the synaptic pathology and amyloid burden in the brains of APP tg mice. These effects were associated with reduced levels of FL APP and APP C-terminal fragments, but levels of BACE1, Notch1, Nep, and IDE were unchanged. In contrast, levels of active cyclin-dependent kinase-5 (CDK5) and glycogen synthase kinase-3beta [GSK-3beta; but not stress-activated protein kinase-1 (SAPK1)], kinases that phosphorylate APP, were reduced. Furthermore, Cerebrolysin reduced the levels of phosphorylated APP and the accumulation of APP in the neuritic processes. Taken together, these results suggest that Cerebrolysin might reduce AD-like pathology in the APP tg mice by regulating APP maturation and transport to sites where Abeta protein is generated. This study clarifies the mechanisms through which Cerebrolysin might reduce Abeta production and deposition in AD and further supports the importance of this compound in the potential treatment of early AD.     

Retinoic acid normalizes nuclear receptor mediated hypo-expression of proteins involved in beta-amyloid deposits in the cerebral cortex of vitamin A deprived rats

Recent data have revealed that disruption of vitamin A signaling observed in Alzheimer's disease (AD) leads to a deposition of beta-amyloid (Abeta). The aim of this study was to precise the role of vitamin A and its nuclear receptors (RAR) in the processes leading to the Abeta deposits. Thus, the effect of vitamin A depletion and subsequent administration of retinoic acid (RA, the active metabolite of vitamin A) on the expression of RARbeta, and of proteins involved in amyloidogenic pathway, e.g., amyloid precursor protein (APP), beta-secretase enzyme (BACE), and APP carboxy-terminal fragment (APP-CTF) was examined in the whole brain, hippocampus, striatum, and cerebral cortex of rats. Rats fed a vitamin A-deprived diet for 13 weeks exhibited decreased amount of RARbeta, APP695, BACE, and of APP-CTF in the whole brain and in the cerebral cortex. Administration of RA is able to restore all expression. The results suggest that fine regulation of vitamin A mediated gene expression seems fundamental for the regulation of APP processing.     

BACE1 gene deletion prevents neuron loss and memory deficits in 5XFAD APP/PS1 transgenic mice

Evidence suggests that beta-amyloid (Abeta) peptide triggers a pathogenic cascade leading to neuronal loss in Alzheimer's disease (AD). However, the causal link between Abeta and neuron death in vivo remains unclear since most animal models fail to recapitulate the dramatic cell loss observed in AD. We have recently developed transgenic mice that overexpress human APP and PS1 with five familial AD mutations (5XFAD mice) and exhibit robust neuron death. Here, we demonstrate that genetic deletion of the beta-secretase (BACE1) not only abrogates Abeta generation and blocks amyloid deposition but also prevents neuron loss found in the cerebral cortex and subiculum, brain regions manifesting the most severe amyloidosis in 5XFAD mice. Importantly, BACE1 gene deletion also rescues memory deficits in 5XFAD mice. Our findings provide strong evidence that Abeta ultimately is responsible for neuron death in AD and validate the therapeutic potential of BACE1-inhibiting approaches for the treatment of AD.     

BACE1 interacts with lipid raft proteins

A neuropathological hallmark of Alzheimer's disease is the presence of amyloid plaques in the brain. Amyloid-beta peptide (Abeta) is the major constituent of the plaques and is generated by proteolytic cleavages of amyloid precursor protein (APP) by beta- and gamma-secretases. Growing evidence shows that lipid rafts are critically involved in regulating the Abeta generation. In support of this, APP, Abeta, and presenilins have been found in lipid rafts. Although cholesterol plays a crucial role in maintaining lipid rafts, functions of other components in the generation of Abeta are unknown. Caveolins (CAVs) and flotillins (FLOTs) are principal proteins related to lipid rafts and have been suggested to be involved in APP processing. Here, we report that FLOT-1 binds to BACE1 (beta-site APP cleaving enzyme 1) and that overexpression of CAV-1 or FLOT-1 results in recruiting BACE1 into lipid rafts and influence on beta-secretase activity in cultured cells. Our results show that both CAV-1 and FLOT-1 may modulate beta-secretase activity by interacting with BACE1.     

A genome wide analysis of ubiquitin ligases in APP processing identifies a novel regulator of BACE1 mRNA levels

Proteolysis of beta-amyloid precursor protein (APP) into amyloid beta peptide (Abeta) by beta- and gamma-secretases is a critical step in the pathogenesis of Alzheimer's Disease (AD), but the pathways regulating secretases are not fully characterized. Ubiquitinylation, which is dysregulated in AD, may affect APP processing. Here, we describe a screen for APP processing modulators using an siRNA library targeting 532 predicted ubiquitin ligases. Seven siRNA pools diminished Abeta production. Of these, siRNAs targeting PPIL2 (hCyp-60) suppressed beta-site cleavage. Knockdown of PPIL2 mRNA decreased BACE1 mRNA, while overexpression of PPIL2 cDNA enhanced BACE1 mRNA levels. Microarray analysis of PPIL2 or BACE1 knockdown indicated that genes affected by BACE1 knockdown are a subset of those dependent upon PPIL2; suggesting that BACE1 expression is downstream of PPIL2. The association of PPIL2 with BACE expression and its requirement for Abeta production suggests new approaches to discover disease modifying agents for AD.     

Beta-secretase protein and activity are increased in the neocortex in Alzheimer disease

CONTEXT: Amyloid plaques, a major pathological feature of Alzheimer disease (AD), are composed of an internal fragment of amyloid precursor protein (APP): the 4-kd amyloid-beta protein (Abeta). The metabolic processing of APP that results in Abeta formation requires 2 enzymatic cleavage events, a gamma-secretase cleavage dependent on presenilin, and a beta-secretase cleavage by the aspartyl protease beta-site APP-cleaving enzyme (BACE). OBJECTIVE: To test the hypothesis that BACE protein and activity are increased in regions of the brain that develop amyloid plaques in AD. METHODS: We developed an antibody capture system to measure BACE protein level and BACE-specific beta-secretase activity in frontal, temporal, and cerebellar brain homogenates from 61 brains with AD and 33 control brains. RESULTS: In the brains with AD, BACE activity and protein were significantly increased (P<.001). Enzymatic activity increased by 63% in the temporal neocortex (P =.007) and 13% in the frontal neocortex (P =.003) in brains with AD, but not in the cerebellar cortex. Activity in the temporal neocortex increased with the duration of AD (P =.008) but did not correlate with enzyme-linked immunosorbent assay measures of insoluble Abeta in brains with AD. Protein level was increased by 14% in the frontal cortex of brains with AD (P =.004), with a trend toward a 15% increase in BACE protein in the temporal cortex (P =.07) and no difference in the cerebellar cortex. Immunohistochemical analysis demonstrated that BACE immunoreactivity in the brain was predominantly neuronal and was found in tangle-bearing neurons in AD. CONCLUSIONS: The BACE protein and activity levels are increased in brain regions affected by amyloid deposition and remain increased despite significant neuronal and synaptic loss in AD.     

Lactacystin decreases amyloid-beta peptide production by inhibiting beta-secretase activity

The human amyloid precursor protein (APP) is processed by the nonamyloidogenic and the amyloidogenic catabolic pathways. The sequential cleavage of APP by the beta- and gamma-secretase activities, known as the amyloidogenic processing of APP, leads to the formation of the amyloid-beta peptide (Abeta). Abeta is the main constituent of the amyloid core of senile plaques, a typical hallmark of Alzheimer's disease. In addition to secretases, other cellular proteolytic activities, like the proteasome, might participate in the metabolism of APP. We investigated the consequence of proteasome inhibition on the amyloidogenic processing of human APP. CHO cells and primary cultures of rat cortical neurons expressing human APP or a protein corresponding to its beta-cleaved C-terminal fragment (C99) were treated with lactacystin, an irreversible inhibitor of the chymotrypsin-like activity of the proteasome. Lactacystin significantly decreased the level of Abeta produced from APP in both cellular models, whereas the production of Abeta from C99 was not affected. Lactacystin did not inhibit gamma-secretase activity but was found to inhibit the beta-cleavage of APP, leading to a proportional decrease in Abeta production. Although lactacystin did not inhibit the catalytic activity of recombinant BACE1, a decrease in neuronal beta-secretase activity was measured after treatment with lactacystin.     

BACE/APPV717F double-transgenic mice develop cerebral amyloidosis and inflammation

Most of the transgenic mice generated to model Alzheimer's disease express human amyloid precursor protein (APP) mutants alone or in conjunction with presenilin mutants. We have generated a mouse model by overexpressing human BACE and human APP with the V717F mutation. The combination of a mutation at the gamma-secretase cleavage site of APP and of increased beta-secretase activity should favour the production of amyloid peptides. We analysed double BACE/APPIn and single APPIn transgenic mice at 16-18 months for amyloid load, brain histopathology and behavioural deficits. We show that overexpression of BACE induces an increase in APP CTFbeta and total brain Abeta peptides. Brain histopathology shows clearly enhanced amyloid deposits in the cortex, hippocampus and in brain vasculature when compared to single APPIn transgenic mice. Amyloid deposits are mostly diffuse and predominantly composed of Abeta(42). A strong inflammatory reaction is evidenced by the presence of microglial cells around the most mature amyloid deposits and astrocytosis over the entire cerebral cortex. At the same age, the APPIn single-transgenic mice show only very limited pathology. When assessed for their cognitive performance at 12 months, BACE/APPIn mice show impaired spatial acquisition in the Morris water maze test. However, these deficits are not greater than those observed in the APPIn single-transgenic animals.     

Novel mutations introduced at the beta-site of amyloid beta protein precursor enhance the production of amyloid beta peptide by BACE1 in vitro and in cells

Abnormal production and accumulation of amyloid-beta peptide (Abeta) plays a major role in the pathogenesis of Alzheimer's disease (AD). beta-secretase (BACE1) is responsible for the cleavage at thebeta-site in amyloid beta protein precursor (AbetaPP/APP) to generate the N-terminus of Abeta. Here we report the stepwise identification and characterization of a novel APP-beta-site mutant, "NFEV" (APP_NFEV) in vitro and in cells. In vitro, the APP_NFEV exhibits 100-fold enhanced cleavage rate relative to the "wild-type" substrate (APPwt) and 10-fold increase relative to the Swedish-type mutation variant (APPsw). In cells, it was preferably cleaved among 24 APP beta-site mutations tested. More importantly, the APP_NFEV mutant failed to generate any detectable Abeta peptides in BACE1-KO mouse fibroblast cells. The production of Abeta peptides was restored by co-transfecting human BACE1, demonstrating that BACE1 is the only enzyme responsible for the processing of APP_NFEV in these cells. Analysis of APP_NFEV cleavage products secreted in the media revealed that in cells BACE1 cleaves APP_NFEV at the position between NF and EV, identical to that observed in vitro. A BACE inhibitor blocked the processing of the APP_NFEV beta-site in vitro and in cells. Our data indicates that the "NFEV" mutant is not only an enhanced substrate for BACE1 in vitro, but also a specific substrate for BACE1 in cells.