William P. Esler

Activity-dependent isolation of the presenilin– γ-secretase complex reveals nicastrin and a γ substrate

Presenilin heterodimers apparently contain the active site of γ-secretase, a polytopic aspartyl protease involved in the transmembrane processing of both the Notch receptor and the amyloid-β precursor protein. Although critical to embryonic development and the pathogenesis of Alzheimers disease, this protease is difficult to characterize, primarily because it is a multicomponent complex of integral membrane proteins. Here the functional γ-secretase complex was isolated by using an immobilized active site-directed inhibitor of the protease. Presenilin heterodimers and nicastrin bound specifically to this inhibitor under conditions tightly correlating with protease activity, whereas several other presenilin-interacting proteins (β-catenin, calsenilin, and presenilin-associated protein) did not bind. Moreover, anti-nicastrin antibodies immunoprecipitated γ-secretase activity from detergent-solubilized microsomes. Unexpectedly, C83, the major endogenous amyloid-β precursor protein substrate of γ-secretase, was also quantitatively associated with the complex. These results provide direct biochemical evidence that nicastrin is a member of the active γ-secretase complex, indicate that β-catenin, calsenilin, and presenilin-associated protein are not required for γ activity, and suggest an unprecedented mechanism of substrate–protease interaction.

Transition-state analogue inhibitors of γ-secretase bind directlyto presenilin-1

The β-amyloid precursor protein (β-APP), which is involved in the pathogenesis of Alzheimer’s disease, and the Notch receptor, which is responsible for critical signalling events during development, both undergo unusual proteolysis within their transmembrane domains by unknown γ-secretases. Here we show that an affinity reagent designed to interact with the active site of γ-secretase binds directly and specifically to heterodimeric forms of presenilins, polytopic proteins that are mutated in hereditary Alzheimer’s and are known mediators of γ-secretase cleavage of both β-APP and Notch. These results provide evidence that heterodimeric presenilins contain the active site of γ-secretase, and validate presenilins as principal targets for the design of drugs to treat and prevent Alzheimer’s disease.

Point Substitution in the Central Hydrophobic Cluster of a Human β-Amyloid Congener Disrupts Peptide Folding and Abolishes Plaque Competence†

Alzheimers disease (AD) is pathologically characterized by the presence of numerous insoluble amyloid plaques in the brain composed primarily of a 40-43 amino acid peptide, the human beta-amyloid peptide (A beta). The process of A beta deposition can be modeled in vitro by deposition of physiological concentrations of radiolabeled A beta onto preexisting amyloid in preparations of unfixed AD cerebral cortex. Using this model system, it has been shown that A beta deposition is biochemically distinct from A beta aggregation and occurs readily at physiological A beta concentrations, but which regions and conformations of A beta are essential to A beta deposition is poorly understood. We report here that an active congener, A beta (10-35)-NH2, displays time dependence, pH-activity profile, and kinetic order of deposition similar to A beta (1-40), and is sufficiently soluble for NMR spectroscopy in water under conditions where it actively deposits. To examine the importance of the central hydrophobic cluster of A beta (LVFFA, residues 17-21) for in vitro A beta deposition, an A beta (10-35)-NH2 analog with a single point substitution (F19T) in this region was synthesized and examined. Unlike A beta (10-35)-NH2, the F19T analog was plaque growth incompetent, and NMR analysis indicated that the mutant peptide was significantly less folded than wild-type A beta. These results support previous studies suggesting that the plaque competence of A beta correlates with peptide folding. Since compounds that alter A beta folding may reduce amyloid deposition, the central hydrophobic cluster of A beta will be a tempting target for structure-based drug design when high-resolution structural information becomes available.

γ-Secretase/presenilin inhibitors for Alzheimer’s disease phenocopy Notch mutations in Drosophila

Signaling from the Notch (N) receptor is essential for proper cell‐fate determinations and tissue patterning in all metazoans. N signaling requires a presenilin (PS)‐dependent transmembrane‐cleaving activity that is closely related or identical to the γ‐secretase proteolysis of the amyloid‐β precursor protein (APP) involved in Alzheimers disease pathogenesis. Here, we show that N‐[N‐(3,5‐difluorophenacetyl)‐L‐alanyl]‐(S)‐phenylglycine t‐butyl ester, a potent γ‐secretase inhibitor reported to reduce amyloid‐β levels in transgenic mice, prevents N processing, translocation, and signaling in cell culture. This compound also induces developmental defects in Drosophila remarkably similar to those caused by genetic reduction of N. The appearance of this phenocopy depends on the timing and dose of compound exposure, and effects on N‐dependent signaling molecules established its biochemical mechanism of action in vivo. Other γ‐secretase inhibitors caused similar effects. Thus, the three‐dimensional structure of the drug‐binding site(s) in Drosophila γ‐secretase is remarkably conserved vis‐à‐vis the same site(s) in the mammalian enzyme. These results show that genetics and developmental biology can help elucidate the in vivo site of action of pharmacological agents and suggest that organisms such as Drosophila may be used as simple models for in vivo prescreening of drug candidates.

Zinc-Induced Aggregation of Human and Rat β-Amyloid Peptides In Vitro

Abstract: The major pathological feature of Alzheimers disease is the presence of a high density of amyloid plaques in the brain tissue of patients. The plaques are predominantly composed of human β‐amyloid peptide (Aβ), a 39–43‐mer peptide the neurotoxicity of which is related to its aggregation state. Previous work has demonstrated that certain metals that have been implicated as risk factors for Alzheimers disease (Al, Fe, and Zn) also cause substantial aggregation of Aβ. In particular, we reported that zinc cations at concentrations of >10−4M dramatically accelerate the rate of Aβ aggregation at physiological peptide concentrations at 37°C in vitro. In the present study, we investigate the effect of Zn2+ on aggregation of radiolabeled and unlabeled human and rat Aβ over a wide range of peptide concentrations in the presence and absence of salt and blocking protein. Aggregation was assayed by centrifugation and filtration using amino acid analysis, immunoassay, and γ‐counting for quantification over a wide range of concentrations of Zn2+ and Aβ above and below physiological values. The results of this study demonstrate the following: (a) Radio‐iodinated Aβ accurately tracked unlabeled Aβ, (b) zinc concentrations of at least 10−4M were required to induce significant aggregation of Aβ, and (c) rat and human Aβ species were cleared from aqueous solutions by similar concentrations of zinc. These results stand in significant quantitative disagreement (∼100‐fold in zinc concentration) with one previous study that reported significant aggregation of Aβ by <1 µM Zn2+. Differences between the present study and the latter study from another laboratory appear to result from inappropriate reliance on optical density to measure Aβ concentrations and nonspecific loss of Aβ to plastic in the absence of blocking protein.

Aβ deposition inhibitor screen using synthetic amyloid

The formation, growth, and maturation of brain amyloid “senile” plaques are essential pathological processes in Alzheimers disease (AD) and key targets for therapeutic intervention. The process of in vitro deposition of Aβ at physiological concentrations onto plaques in AD brain preparations has been well characterized, but is cumbersome for drug discovery. We describe here a high-throughput screen for inhibitors of Aβ deposition onto a synthetic template (synthaloid) of fibrillar Aβ immobilized in a polymer matrix. Synthaloid is indistinguishable from plaques in AD brain (the natural template) in deposition kinetics, pH profile, and structure-activity relationships for both Aβ analogs and inhibitors. Synthaloid, in contrast to current Aβ aggregation screens, accurately predicted inhibitor potency for Aβ deposition onto AD cortex preparations, validating its use in searching for agents that can slow the progression of AD and exposing a previously inaccessible target for drug discovery.

Stereochemical specificity of Alzheimer's disease β-peptide assembly

The formation and growth of insoluble amyloid deposits composed primarily of the human β-amyloid peptide (Aβ) in brain is an essentially invariant feature of Alzheimers disease (AD) and is widely believed to contribute to the progressive neurodegeneration of the disorder. To probe the specificity of amyloid formation and growth, we synthesized and examined the self-assembly of D- and L-stereoisomers of Aβ in vitro. While both enantiomers formed insoluble aggregates at similar rates with amyloid-like fibrillar morphology, deposition of soluble Aβ peptide onto preexisting Aβ aggregates was stereospecific. Although the L-peptide deposited readily onto immobilized L-Aβ aggregates with first-order kinetic dependence on soluble peptide concentration, essentially no association between the D-peptide and L-template was observed. Similarly, the D-peptide deposited with first-order kinetics onto a D-Aβ aggregate template but did not deposit onto a similar template composed of aggregates of the L-enantiomer. Furthermore, although the L-Aβ isomer deposited onto authentic AD amyloid in preparations of unfixed AD brain, no focal association between the D-peptide and brain amyloid was detected. These results establish that deposition of soluble Aβ onto preexisting amyloid template is stereospecific, likely involving direct docking interactions between peptide backbone and/or side chains rather than simple hydrophobic association.

Amyloid-lowering isocoumarins are not direct inhibitors of γ-secretase

To the Editor — Recent reports have described the interaction of specific phosphoinositides (PIs) with the phox homology (PX) domain of several proteins, including the phagocyte NADPH oxidase components p47-phox and p40-phox, after which the module was named. These papers, and a subsequent Nature Cell Biology News and Views article referred to a study in which we identified rare mutations in the p47phox gene of a small group of patients with chronic granulomatous disease (CGD). In CGD, superoxide generation by NADPH oxidase is absent or severely diminished. In two compound heterozygous patients, a point mutation on one allele (G125A) predicted the amino acid change R42Q. Arg 42 is the first residue of a motif that is highly conserved among a subset of PX domains and is believed to be directly involved in ligating one of the phosphate groups of PIs. Indeed, R42Q abolished PI binding to the PX domain of p47-phox in vitro. This led one group of authors to speculate that the inability of PIs to bind to the mutated protein may account for the absence of NADPH oxidase activity in these CGD patients. Other authors cited our data as providing evidence for the critical importance of the PX domain and PIs in phagocyte antimicrobial action. However, we clearly stated in our Blood paper that neutrophils from these patients had no detectable p47-phox protein and suggested that the mutation must lead to instability in either the message or, more likely, the resulting R42Q polypeptide. We have subsequently detected mutant mRNA in a third patient who is heterozygous for G125A, indicating it is the polypeptide that is unstable. This is consistent with the observation that the majority of CGD missense mutations in phox genes result in a loss of protein. In the absence of p47-phox from these CGD phagocytes, G125A cannot be considered a simple loss-of-function mutation and no inferences can be drawn from the predicted amino acid change regarding the functional significance of the domain. Therefore, our findings do not provide any evidence that the PI-PX domain interaction is vital for phagocyte superoxide production and that its disruption can result in CGD. CGD studies have shed much light on the biology of the normal NADPH oxidase system. In some cases, the identification of specific mutations has provided insights at a molecular level. R42Q is not such a case, and care should be taken to avoid misinterpreting genetic data to support functional hypotheses. Paul G. Heyworth* and Andrew R. Cross

Deposition of soluble amyloid-beta onto amyloid templates: with application for the identification of amyloid fibril extension inhibitors.

Publisher Summary Several human diseases are now recognized to be caused by the formation of insoluble amyloid from naturally occurring peptides and proteins. Amyloid-forming monomers are not intrinsically pathological; rather it is their conformation and assembly state that are most important to their cytotoxicity. Because the chemical information for forming these amyloids apparently lies in the primary structure of the monomers, there has been significant interest in using in vitro systems to study amyloid formation and growth. A β deposition, the process of amyloid growth by the association of individual soluble A β molecules with a preexisting amyloid template, can be modeled readily in vitro under physiologically relevant conditions and importantly at physiological (nM) A β concentrations. Such assays may elucidate key molecular interactions in amyloidosis and may be exploitable in the design of antiamyloid therapeutics. High-throughput A β deposition assays may provide a means for the identification and optimization of such compounds. The methods described in this chapter have been developed for the examination of A β deposition, but are adaptable for the study of other amyloids.

Apolipoprotein E affects amyloid formation but not amyloid growth in vitro: mechanistic implications for apoE4 enhanced amyloid burden and risk for Alzheimer's disease.

The transition from the partially folded soluble Aβ monomer to insoluble Aβ amyloid fibrils is seminal to the formation and growth of amyloid plaques in Alzheimers disease (AD). A detailed understanding of the role of AD risk factors in these processes is essential to understanding the physiochemical nature of this conformational rearrangement- The apolipoprotein E ϵ4 allele, a risk factor for AD, affects AD pathology by increasing amyloid burden relative to the much more common ϵ3 allele. In the present study, in vitro models were employed to probe the effect of these proteins on kinetically distinct steps in Aβ fibrillogenesis. Formation of Aβ amyloid was faster in the presence of apo E4 than apoE3, while growth of existing plaques was unaffected by either isoform. Further, experiments with Aβ stereoiso-mers establish that this effect ofapoE3 is mediated through interaction with oligomeric fibrillogenic intermediates rather than through specific contacts with monomeric Aβ. Consistent with the altered pathology and enhanced risk for AD associated with inheritance of the ϵ4 allele, we conclude that APOE ϵ4 is a risk factor for AD not due to a pathological gain of function of apo E4 but to a loss of protective function of apo E3.