BioWorld International Correspondent

LONDON - The search is on for molecules to stimulate a brain enzyme that can prevent the formation of the toxic deposits that characterize Alzheimer's disease, following encouraging results in an animal model.

Rather than focusing on trying to break down the toxic amyloid b-peptides that lead to the formation of senile plaques in the brain, the strategy aims to prevent amyloid b-peptides from forming in the first place. The approach focuses on boosting the activity of an enzyme called a-secretase, which breaks down amyloid precursor proteins in such a way that amyloid b-peptides cannot form.

Falk Fahrenholz, director of the Institute of Biochemistry at the University of Mainz in Germany, told BioWorld International: "Our study shows that enhancing the expression or activity of a-secretase is a worthwhile approach to preventing the pathology of Alzheimer's disease."

The paper, by Rolf Postina and colleagues in Belgium and at the University of Mainz, is published in the May 17, 2004, issue of Journal of Clinical Investigation. Its title is "A disintegrin-metalloproteinase prevents amyloid plaque formation and hippocampal defects in an Alzheimer disease mouse model."

In a "Commentary" in the same issue, Stefan Lichtenthaler and Christian Haass at the Ludwig-Maximilians-Universitat in Munich, Germany, said the study provides "compelling evidence" that the proposed strategy might work. In their article - titled "Amyloid at the cutting edge: activation of a-secretase prevents amyloidogenesis in an Alzheimer disease mouse model" - they wrote, "The work by Fahrenholz and colleagues is a major step forward in exploring the therapeutic potential of a-secretase targeting."

Amyloid precursor protein is the substrate for three proteases. Two of those, called b-secretase and g-secretase, generate amyloid b-peptides when they break down amyloid precursor protein, and many research teams have focused on ways of interfering with those reactions. The third, a-secretase, also acts on amyloid precursor protein, but snips it in the middle of the section that goes to form the amyloid b-peptide.

The old idea of activating a-secretase to ensure that amyloid precursor protein is broken down into harmless fragments, Lichtenthaler and Haass said, had been virtually forgotten until Fahrenholz and his colleagues decided to resurrect it.

The previous work of the team had identified a member of the ADAM (a disintegrin and metalloproteinase) family, known as ADAM10, as an a-secretase. They also knew that ADAM10 was expressed in the brains of both mice and humans.

They set out to establish if overexpressing ADAM10 could increase the breakdown of amyloid precursor protein into harmless products, while at the same time decreasing levels of amyloid b-peptides and slowing the formation of amyloid plaques. They also were keen to see the effect of such changes on the symptoms of animals that provide a mouse model of Alzheimer's disease.

They developed two different strains of transgenic mice. The first overexpressed ADAM10 selectively in neurons. The second overexpressed a mutant form of ADAM10, which was unable to break down amyloid precursor protein.

Each of those strains was crossed with a mouse model of Alzheimer's disease, in which human amyloid precursor protein was overexpressed in neurons, leading to higher levels of amyloid b-peptides, amyloid plaques and memory deficits.

The first of those experiments showed that ADAM10 works as an a-secretase in vivo. The offspring of that cross had lower than normal levels of amyloid b-peptides in their brains, and the amyloid plaques that would be expected to be seen in the Alzheimer's disease mouse model were almost completely absent.

By contrast, breakdown of amyloid precursor protein was lower than normal in the offspring of the second cross, levels of amyloid b-peptides increased and both the number and size of amyloid plaques in the brains of those animals also increased.

Lichtenthaler and Haass concluded that perhaps the most important finding of the study was that it provided evidence that "overexpression of ADAM10 [also] alleviates the deficits in spatial learning and synaptic plasticity observed in the control animals, suggesting that an activation of a-secretase cleavage may also improve cognitive status in humans."

Fahrenholz and colleagues are looking for novel activators of a-secretases by screening libraries of low-molecular-weight compounds. "We are looking for certain neuropeptides that stimulate the a-secretase pathway, and neuropeptides for which receptors exist in brain areas involved in Alzheimer's disease," he said. "We have also cloned the promoter for ADAM10 and we are looking for activators of gene expression."

Lichtenthaler and Haass noted that to reproduce the results of the experiments in humans, it would be necessary to use gene therapy to overexpress ADAM10 in neurons. Fahrenholz said he and his group currently are working on a nonviral approach to gene therapy.