| STUDY |
Alzheimer Enzyme No Longer So Secretive Mairin Brennan Researchers are homing in on the elusive enzyme that makes the final cut that releases the Alzheimer peptide, amyloid -peptide (A), from its transmembrane amyloid precursor protein (APP). Two groups have independently shown that the active site of the elusive enzyme---secretase--is contained in a transmembrane protein dubbed presenilin. One group was led by chemist Michael S. Wolfe, an associate professor of neurology at Boston's Brigham & Women's Hospital and Harvard Medical School [Nat. Cell Biol., 2, 428 (2000)], and the other by Stephen J. Gardell, director of biological chemistry at Merck Research Laboratories , West Point, Pa., and his colleagues there and at Merck's neuroscience facility in Terlings Park in the U.K. [Nature, 405, 689 (2000)]. Presenilin threads through the membrane in a serpentine fashion. Shortly after being synthesized, it is cleaved by an unknown mechanism to form two subunits. Mutations in presenilin are associated with familial Alzheimer's disease. A is generated after -secretase, which was identified by several pharmaceutical companies last year, cleaves APP at a site outside the membrane and -secretase snips it within the membrane. The peptide is either 40 or 42 amino acids long depending on where -secretase makes its cut. Both forms are produced normally in small amounts, although their function is not known. In Alzheimer's disease, the ratio of A40 to A42 is skewed in favor of A42, and that triggers the peptides to form fibrillar deposits in amyloid plaque in the brain. Although -secretase has a name and a function, the enzyme itself has never been isolated. Evidence that presenilin might be -secretase began to appear last year when Wolfe and coworkers showed that mutations in two aspartyl residues in presenilin inhibited -secretase activity in cells growing in culture [Nature, 398, 513 (1999)]. Earlier this year, Gardell and colleagues precipitated presenilin with an antibody to it and showed that the precipitate displays -secretase activity [Proc. Natl. Acad. Sci. USA, 97, 6138 (2000)]. The Merck findings revealed that this activity is catalyzed by a presenilin-containing complex that is much larger in size than presenilin itself. Other researchers also have shown that presenilin and -secretase are intricately related: Both are involved in releasing a membrane-tethered segment of a protein called Notch that travels to the nucleus to activate genes associated with embryonic development [Nat. Cell Biol., 2, 463 (2000)]. Embryonic stem cells deficient in presenilin lack -secretase activity [Nat. Cell. Biol., 2, 461 (2000)]. And mice lacking presenilin genes produce no -secretase and die in utero similar to Notch-deficient mice [Genes Dev., 13, 2801 (1999)]. In parallel approaches, Gardell and Wolfe and their coworkers now have definitively coupled -secretase and presenilin. They are presenting their findings at the World Alzheimer Congress 2000, being held this week in Washington, D.C. Both groups have shown that inhibitors intended to bind to the active site of -secretase bind directly to presenilin. The Merck compound--L-685,458--was identified by scientists at the firm's Terlings Park facility during a high-throughput screen for compounds that would inhibit A production. The compound was originally developed as an HIV protease inhibitor, notes Jules A. Shafer, executive director of basic research at Merck. It was a poor inhibitor of HIV protease but a potent inhibitor of -secretase, he says. The Harvard compound is a substrate-based -secretase inhibitor that exploits a rationally designed difluoroalcohol group the team previously had shown to be an analog of -secretase's transition state. To find out whether the inhibitor would bind to presenilin, the team attached a bromoacetamide to one end of the compound, reasoning that bromoacetamide would undergo nucleophilic attack by residues near the active site to form a covalent bond with presenilin. At the other end, they tacked on biotin so the complex could be easily detected with streptavidin, which binds to biotin. In various experiments they confirmed that the derivatized inhibitor, dubbed BrA-1-Bt, bound to each of two presenilin subunits and inhibited -secretase activity. The Merck team prepared two photoreactive benzophenone derivatives of L-685,458 that would separately label each subunit of presenilin and tagged them with biotin. Both compounds proved to be potent inhibitors of -secretase in the dark, but when photoactivated, they bound to different subunits. The researchers concluded that the active site in presenilin is shared between the subunits and represents the active site of -secretase. "We feel that presenilin contains the catalytic machinery of -secretase," Shafer says. He suggests that the behavior of presenilin may be similar to that of an enzyme--Factor Xa--in the blood-clotting cascade that inhabits a large complex and has little activity when isolated from it. Both the Merck and Harvard inhibitors are transition-state analogs. They would therefore be expected to bind selectively to active forms of -secretase, the researchers point out. But the Merck group has shown that their compound binds to "full-length" wild-type presenilin--presenilin that's not been cleaved into subunits. Failure to do this suggests that full-length presenilin is an inactive precursor protein, they say. The group also has shown that the compound binds to a mutant full-length presenilin associated with early-onset Alzheimer's disease. The mutant lacks the loop that contains the subunit cleavage site, and the team suggests that the loss of this strategic loop may cause the mutant to mimic the conformational change in wild-type prenesilin that's triggered by cleavage into subunits. Compounds that would inhibit either - or -secretase could potentially block A production and be useful in treating Alzheimer's disease. Indeed, -secretase inhibitors are already under development. Not much is known about -secretase, though, other than its role in cleaving APP, Wolfe notes. -Secretase, on the other hand, is involved in processing the Notch protein that's crucial for embryonic development. That "raises a red flag" about whether -secretase could be blocked without affecting Notch activity, he says. But he believes "the answer is 'yes,' because you don't have to completely shut down -amyloid production to get a therapeutic effect." Cholesterol-lowering drugs don't completely shut down cholesterol synthesis, he notes. Likewise, -secretase inhibitors could reduce A production without depleting the enzyme. "The insights of the current studies go beyond research into Alzheimer's disease and are highly relevant to cell biology," writes cell biologist Bart De Strooper at Louvain & Flanders Interuniversity Institute for Biotechnology, Louvain, Belgium, in a commentary on the dual achievement [Nature, 405, 627 (2000)]. But much more work will be needed to understand how hydrolysis of peptide bonds in the hydrophobic membrane can occur, he states. Presenilin's putative catalytic site is especially intriguing, he observes, because it doesn't resemble that of classical aspartyl proteases. "The hunt is now on for other substrates of presenilin, alias -secretase, and for proteins that regulate its activity," he notes. -------------------------------------------------------------------------------- |
| UPDATE | 07.00 |
| COMPANY | Merck |
| LITERATURE REF. | This data is not available for free |
|
Want more information ? Interested in the hidden information ? Click here and do your request. |