Science Editor

It's a grim truism that the only 100 percent reliable diagnosis of Alzheimer's disease (AD) is by post-mortem. Only autopsy of the brain can confirm the presence of AD's hallmarks - senile plaques and fibrillary tangles.

"In the experience of our clinical unit," observed neurologist Michael Irizarry at Harvard-affiliated Massachusetts General Hospital, "the accuracy rate for AD diagnosis is about 90 percent - by examining patients and with frequent follow-up." Irizarry, whose research week includes seeing patients 20 percent of the time, is senior author of a paper in the September issue of the Archives of Neurology (a publication of the Journal of the American Medical Association). His paper is titled: "b-secretase protein and activity are increased in the neocortex in Alzheimer disease."

"The main finding of our paper," Irizarry told BioWorld Today, "was that in areas of the brain most affected by the changes of AD we found the function of this beta-secretase enzyme was increased. And this increase persisted even in Alzheimer's cases of long duration, measured in years from the onset of dementia symptoms.

"The novelty," he continued, "is it suggests that one of the main reasons for accumulation of amyloid plaques in AD is increased production of the amyloid-beta protein - possibly in addition to the body's inability to clear it. The implications are that targeting that enzyme for therapy may be a promising approach to block amyloid-beta production, and ultimately prevent the formation of amyloid plaques. Our purpose was mainly to see why particular areas of the brain are vulnerable to AD changes. It seemed to our group that one of the changes is that beta-secretase protein, in those particular areas of the brain, is more active.

"We measured this activity by capturing the enzyme from the brain onto a plate. Then we had a substrate that modeled where amyloid precursor protein is cut by this enzyme. And based on fluorescence we could tell how active that enzyme is. That's a little different than just measuring the level of the protein. The beta-secretase itself cuts APP, so that apparently is its major function."

Assaying 61 AD Brains Plus 33 Controls

To investigate why amyloid plaques appear only in certain regions of the brain, Irizarry and his co-authors designed new assays to measure the quantity and functional activity of the beta-secretase in brain tissue from 61 patients who had died with Alzheimer's disease. "These 61 came out of the Alzheimer Research Center Brain Bank at MGH," Irizarry recounted. "The tissues were from patients who were seen in clinic and followed for a long time. The 33 controls," he went on, "were age-matched patients who did not have dementia and no changes of Alzheimer's disease in their brains and did not have AD when they died.

"The controls' tissues let us distinguish changes that are more specific to AD," Irizarry pointed out. "If we just had the AD brains we could compare levels between different brain areas but those changes would just be normal. And we mainly used the controls to distinguish changes that are specific to AD and to certain brain areas.

"We found that the activity of the secretase enzyme was increased 63 percent in the temporal cortex of AD patients compared with controls and increased in patients with longer-term disease. Increased amounts and activity of the enzyme went up 13 percent in the frontal cortex. However, we saw no response in the cerebellar cortex, a part of the brain where amyloid plaques do not develop."

The co-authors also noted that increased production of amyloid-beta protein by gene mutations in the enzyme gamma-secretase, and its parent molecule, APP - amyloid precursor protein - is a mechanism of plaque formation in rare inherited variants of Alzheimer's disease. "And the current findings about enzyme changes in the brains of AD patients," Irizarry said, "provide compelling evidence that increased amyloid-beta production is also important in sporadic AD, by far the commonest form of the disease, as well as familial."

In AD, amyloid-beta is released when the large APP molecule is cleaved in one location by beta-secretase and in another site by gamma-secretase. The amyloid-beta fragments collect in plaques, which are thought toxic to brain cells. Normal APP processing by another enzyme, alpha-secretase, produces an alternative, nontoxic protein.

One key tool of the research reported in the journal is an antibody-capture strategy. "We used a commercial antibody and coated plates with it. Then we added the brain sample to those plates, and the antibody bound any of the beta-secretase in the brain. After washing off the rest of the plates, we had the antibody binding the enzyme that was only in that brain sample.

"When we added the substrate, which is a protein similar to APP, it fluoresces upon being cut. That told us how efficiently the enzyme is working. It's an interesting mechanism, a peptide of about 10 amino acids, with a fluorescent tag on each end. When the peptide is all connected, it turns out that a technique called FRET - fluorescence resonance energy transfer - quenches the two light-emitting tags. The antibody's epitope, or antigenic target," Irizarry went on, "is the carboxy-terminal of the beta-secretase molecule. That allows the other terminal, which contains the enzymatic activity, to remain functional. The antibody capture showed us in a sensitive and quantitative manner how to measure both the amount and function of this enzyme in a large number of brain samples."

Inhibiting Beneficial' Enzymes For Therapy

"Our next step," he observed, "will be to investigate factors underlying that increased enzyme activity in AD brains. One therapeutic application arising from this work is that the assays can be adapted to look at ways of modifying beta-secretase activity. So the assays themselves can be modified to screen for medications that might be effective on these enzymes. Our results suggest that these would be beneficial enzymes to inhibit for drug therapy because they're abnormal in the AD brain, and stay abnormal even late in the disease. Two more questions would be: Is there any way to measure this increase in activity without requiring brain surgery? Is there some sort of AD marker," Irizarry concluded, "such as a blood sample?"