Alzheimer's Disease

The molecular origins of the disease are coming to light, suggesting several novel therapies

Vernon Ingram

Reducing the Overproduction of Aβ

Since the overproduction of the noxious Aβ1-42 fragment is now considered the root cause of Alzheimer's disease, the two enzymes that produce the peptide have become prime targets for therapy. The β-secretase enzyme produces the amino terminus, and the γ-secretase enzyme cleaves the carboxyl terminus of the peptide from the amyloid precursor protein.

The potential value of therapies based on the pattern of protein cleavage (or proteolysis) of APP is illustrated by the fact that no fewer than five Alzheimer's research groups in 1999 and 2000 independently identified the β-secretase enzyme—now called BACE1. Importantly, the β-secretase protein seems to be confined to specific regions within the central nervous system, making it a good therapeutic target. The hippocampus and cortex show the highest expression levels, and peripheral tissues have only trace amounts. As a consequence of its tight localization, systemic inhibition of the enzyme might be possible with relatively few undesirable side effects.

Additionally, mice without the BACE1 gene are healthy, reinforcing the enzyme's status as an exceptional pharmacological target. Indeed, powerful inhibitors of β-secretase have already been reported, but they are large molecules, making them unsuitable for therapeutic use. As a cautionary note, there has been some early evidence that β-secretase inhibitors can also interfere with other enzymes that cut proteins, leading to unwanted complications. Thus, more work is needed (and is certainly being done) to develop a smaller and more specific inhibitor.

The second critical enzyme, γ-secretase, has its own unusual properties that may aid in tailoring specific therapies. As shown in the figure, the γ cleavage site lies within the membrane-spanning domain of APP, meaning that the γ-secretase catalytic domain must work in the hydrophobic lipid layer of the cell membrane. Yet the chemistry of protein digestion requires the addition of water to cleave the peptide bond. How is this done? The mechanism is incompletely understood, but it involves several other proteins besides APP and the γ-secretase.

Two other molecules in the proteolytic complex are central: the so-called presenilins, PS1 and PS2. The presenilins have attracted attention because of the frequency with which mutations in the genes coding for PS1 or PS2 cause heritable, early-onset Alzheimer's disease. More than a hundred presenilin mutations are known, but they all seem to cause an increase in the Aβ1-42 product of amyloid precursor digestion. Fortunately, these sorts of autosomal dominant mutations occur only rarely.

Gamma secretase is also a less desirable candidate from a genetic standpoint. In mutant animals without the gene, the absence of the enzyme prevented normal development, and the embryos died. This finding hasn't deterred the concerted effort to identify γ-secretase inhibitors that might reduce levels of Aβ1-42 in the adult brain. Yet it remains to be seen whether a decrease in beta amyloid will lead to restoration of lost function in animal models and, of course, in people.