Aedvgsnkgai Iglmvggvv

This is the human amino acid sequence. In keeping with our focus on rodent models, I also have included in parentheses the rat sequence, which differs only at the three indicated amino acids. Current work focuses on the 42-amino acid variant as a principal culprit in AD, as we will return to later.

Ap42, being a fairly small peptide, can assume a number of different conformations in solution and even in protein crystals. One three-dimensional structure of an Ap peptide that has been determined

FIGURE 2 A crystal structure of a fragment of amyloid beta peptide. These two panels show a peptide backbone (A) and spacefill (B) rendering of amino acids 1-39 of amyloid beta peptide. Figures rendered using Rasmol and file 1BA6 from the Brookhaven National Protein Database. Note that amyloid beta peptide can assume many different conformations, and this particular structure is only representative of one known conformation of the peptide. In panel A, the shaded region indicates a central alpha-helical domain. Both panels are the same perspective of the molecule. The amino terminus of the molecule is at the lower right. Data for image is published in Watson et al. (111).

FIGURE 2 A crystal structure of a fragment of amyloid beta peptide. These two panels show a peptide backbone (A) and spacefill (B) rendering of amino acids 1-39 of amyloid beta peptide. Figures rendered using Rasmol and file 1BA6 from the Brookhaven National Protein Database. Note that amyloid beta peptide can assume many different conformations, and this particular structure is only representative of one known conformation of the peptide. In panel A, the shaded region indicates a central alpha-helical domain. Both panels are the same perspective of the molecule. The amino terminus of the molecule is at the lower right. Data for image is published in Watson et al. (111).

is shown in Figure 2 for your reference. However, Ap42 is highly "fibrillogenic," that is, it is subject to not remaining as a monomer in solution but rather to forming oligomers, multimers, and aggregates. These polymerized forms of Ap42, in complex with Ap40, make up the amyloid plaques that are characteristic of AD. Exactly which state of Ap (monomer versus multimer versus aggregate) is involved in AD pathogenesis is an area of much debate at present.

Work subsequent to the discovery of the sequence of Ap revealed that Ap is derived from p-amyloid precursor proteins (APPs; 20), yet another finding that has catalyzed progress in understanding the cellular and molecular underpinnings of AD. APPs are type I integral membrane proteins that are expressed at the cell surface (see Figure 3). The APP gene is located on chromosome 21 and contains 19 exons—over ~400 kb of DNA (21). Several alternatively spliced mRNAs encode APP in neurons and, to a lesser extent, glia. The 695 amino acid splice variant (APP695) is expressed exclusively in neurons—the APP751 and APP770 variants are more widely expressed.

In axons of the peripheral and central nervous system, APP is transported by the fast anterograde system to nerve terminals, where Ap peptides are generated and released into the extracellular space by mechanisms that are not clear at present (22, 23). Ap peptide is a normal constituent of the extracellular milieu and is present in your brain and CSF as you are reading this sentence. Only when it is aberrantly overproduced (or underdegraded) do amyloid plaques form. The normal physiologic role of Ap peptide and its precursor APP are completely unknown at present. Not surprisingly, almost all the work in this area has focused on the role of these proteins in AD pathogenesis.

Given that the overproduction of Ap appears to be such a critical factor in the development of AD, much important effort has been invested in understanding the production of this peptide. APPs are subject to alternative proteolytic processing by a family of "secretase" activities that are still being defined at the molecular level (24-26). The three relevant secretases for processing APP and other similar proteins (the developmental regulator Notch is another example) are termed alpha, beta, and gamma secretase (see Figure 3). The a-secretase cleaves APP within the Ap sequence to release the N-terminal ectodomain of APP (APPsa); thus, a-secretase cleavage within the Ap domain precludes production of Ap pep-tides. The alpha-secretase activity, therefore, dictates a route of APP processing distinct from the amyloidogenic, AD-related process.

Sequenz App Secretase

FIGURE 3 Amyloid precursor protein. The basic structure of APP is shown (upper section), along with several known mutations and the sites of alpha, beta, and gamma secretase cleavage (marked a, P, and y, lower section). See text for additional discussion. Adapted from Price, Sisodia, and Borchelt (112).

FIGURE 3 Amyloid precursor protein. The basic structure of APP is shown (upper section), along with several known mutations and the sites of alpha, beta, and gamma secretase cleavage (marked a, P, and y, lower section). See text for additional discussion. Adapted from Price, Sisodia, and Borchelt (112).

However, the actions of the beta and gamma secretases lead to production of Ap peptides (reviewed in reference 27). The y-secretase is unusual because it cleaves proteins, including APP, within the lipid bilayer. That is, this protease actually acts to cut the APP at its alpha-helical transmembrane domain while it is still in the membrane. The p-secretase is more pedestrian, cleaving the APP in a soluble domain. However, even the p-secretase has the unusual attribute that it is an ecto-protease, that is, it acts upon the extracellular domain of the APP molecule.

The p-secretase has been identified and termed BACE for Beta-site APP Cleaving Enzyme (28; reviewed in Vassar and Citron, 29).This important discovery was made by Mark Citron's group at Amgen. BACE is a single-transmembrane domain aspartyl protease that can cleave APP at several sites including the one responsible for generating Ap peptide. BACE also cleaves other proteins. Knockout mice deficient in BACE produce essentially no amyloid beta peptide (30). These knockout mice also have no discernable behavioral or developmental phenotype, a good sign in terms of the possibility of utilizing BACE inhibitors as a potential AD therapy.

What about the y-secretase? There is a clear and compelling candidate for the y-secretase; a family of proteins termed the presenilins (PSs). There are two homologous human PS genes, presenilin 1 (PS1) and presenilin 2 (PS2). As might be expected for an enzyme capable of proteolyzing a transmembrane alpha helix, PSs have multiple transmembrane domains (see reference 31). These proteins also undergo proteolysis themselves as part of their conversion to the active state. Presenilins are hypothesized to be the gamma secretase, although there is some discussion in the literature that PSs might instead act indirectly to promote gamma secretase activity. Certainly other proteins such as nicastrin are necessary in addition to presenilins in order to achieve full gamma-secretase activity.

Presenilins act not only on APP but also on other proteins such as Notch, a general role referred to as Regulated Intramembrane Proteolysis (RIP; reviewed in reference 32). PS-mediated RIP is involved in a variety of cellular signaling processes that are important in neurodevelopment and homeostasis. For our purposes here we will focus on the role of Presenilins in regulating the production of Ap peptides in AD.

The APP fragments produced by the combined activities of the beta- and gamma-secretases are generally 40 or 42-43 amino acids in length. There is some variability in the site of cleavage of the gamma secretase—it can cleave APP at any of three sites. Ap40 comprises ~90% of the Ap population while the rest is usually made up of Ap42(43) (33). The minor Ap species (42/43) is highly fibrillogenic, readily aggregates, and is neurotoxic (34-39).

C. AP42 as the Cause of AD

Briefly stated, decades of work indicate that Ap42 is the likely causative agent of AD. This idea is commonly referred to as the amyloid hypothesis of AD (see Hardy and Selkoe, 31, for a review). The essential findings supporting this hypothesis follow:

1. As we have already discussed, amyloid senile plaque number (i.e., Ap deposition) in the neocortex is the primary criterion for the post-mortem diagnosis of Alzheimer's disease (40-43). The initial deposits in senile plaques are the Ap42 and Ap43 peptides (39).

Moreover, Ap burden is an early indiator of cognitive decline in AD. Post-mortem studies of total Ap (nonaggregate and aggregates in diffuse or mature senile plaques combined) in the brains of recently deceased patients correlate with recent pre-morbid Clinical Dementia Rating scale values for those individuals. Quantitative histopathological studies have shown that the number of senile plaques correlates with dementia scores in AD patients (44). This is true even for patients not yet in advanced stages of the disease. These findings support the idea that extracellular Ap levels are elevated in at-risk individuals even prior to gross plaque deposition and severe cognitive impairment.

2. Recent studies from animal and in vitro models have made clear the capacity of aberrant Ap production to elicit pathological features of AD. For example, animals genetically engineered to overproduce Ap exhibit some of the pathological features of AD, such as amyloid plaque production (see references 45 and 46 and Figure 4). In vitro studies have shown that Ap can elicit cytotoxic effects as well.

Tg 2576 Tg 2576 + PS1 A246E

Was this article helpful?

0 0
Advanced Memory Techniques

Advanced Memory Techniques

A course in techniques and skills for mentalists, magicians and students. For students, improve your grades with less effort! But this book is also.... The ideal for any stage mentalist or magician by establishing credibility of amazing skills with an easy to follow instructional book on using the amazing power of your memory.

Get My Free Ebook


Post a comment