Login to Your Account

Alzheimer's and Amyloid: What Exactly is Going On?

By Anette Breindl
Science Editor

As far as the relationship between Alzheimer's disease and amyloid plaques is concerned, the mystery continues.

At a recent talk at the National Institutes of Health, Harvard University's Jie Shen summed up the evidence regarding the so-called amyloid hypothesis, that is, the notion that beta-amyloid is the cause of neurodegeneration and memory loss in Alzheimer's disease.

She also gave an overview of her own research, which has led her to conclude that loss-of-function mutations in presenilin-1 are the driving force behind the clinical symptoms of Alzheimer's disease.

"I think the strongest evidence [for the amyloid hypothesis] is actually the fact that mutations in [amyloid precursor protein or] APP are causal," she told the audience. That causality has recently been confirmed by a genomewide association study. (See BioWorld Today, July 23, 2012.)

"Definitely A-beta and APP have something to do with the pathogenesis of Alzheimer's disease," she concluded.

But what exactly?

There is also a fairly long list of experimental results that are at odds with the idea that amyloid plaques cause Alzheimer's. For one thing, the amount of amyloid plaques in patients' brains is not the best correlate of how severe their cognitive impairment is.

Many transgenic mice strains overproduce APP and A-beta, and "most of them, if not all, do not produce significant neurodegeneration."

And "most damning from a practical standpoint," Shen said, "is the failure of clinical trials" based on the amyloid hypothesis.

At last count, the numbers had added up to more than a dozen failed Phase III trials, and zero disease-modifying drugs approved for Alzheimer's disease.

Additionally, Shen said, Alzheimer's is known as a memory disease. "But actually, it goes far beyond that." In advanced stages, patients cannot take care of themselves in the most basic ways, necessitating constant care.

Altogether, Shen, who studies Parkinson's as well as Alzheimer's disease, concluded that "If I could choose one disease to avoid, I would definitely pick Alzheimer's disease."

Shen studies the protein presenilin-1, which is part of the gamma-secretase complex that cuts amyloid precursor protein into A-beta 42 and A-beta 40.

When she first began working on the protein, she told her audience, "what caught my attention . . . was the distribution of the mutations in presenilin-1."

Those mutations are very broadly distributed throughout the coding sequence, while APP mutations tend to cluster around the site where it is cleaved.

"As a biologist," Shen said, "when I look at this [distribution], my gut feeling is that this protein is so important in the central processes of the disease that a change in a single nucleotide that leads to a change in a single amino acid is enough for a near hundred percent chance of developing the disease."

Shen said that those mutations are likely to lead to a loss of function on presenilin-1, because they do not cluster around functional sites.

Presenilin can regulate long-term potentiation, which is the way in which the brain stores memories, through two different mechanisms. And presenilin knockout mice also show the neurodegeneration that accounts for the ravages of late-stage Alzheimer's disease.

Such animals "recapitulate the neuropathology of AD well, except for the plaques," Shen said – another piece of evidence that plaques and neurodegeneration may be symptoms of the same underlying problem, rather than the former causing the latter.

At 2 months of age, the number of synapses and neurons is not significantly reduced in presenilin knockouts compared to controls. But "neurodegeneration is already beginning in a small percentage of neurons."

Shen and her team found that there was an eightfold increase in apoptosis of brain cells, though this is still "only a tiny percentage of cortical neurons." Still, cumulatively, that increase in apoptosis led to large neuronal losses. By 6 months of age, such mice had lost 18 percent of their neurons and were unable to learn any task the experimenters set them.

The developmental protein Notch and APP are both physiological targets of presenilin-1. Shen and her team looked at the consequences of knocking out Notch, but came up empty.

"Basically, we did not find anything in these mice" even when they knocked out both forms of Notch, she noted. The team is now looking at APP knockouts, but has not yet published those results.