Immunology: Priming Cellular Pathway May Lead to New AIDS and Cancer Vaccines Structural Biology: Image of Epidemic Dengue Virus Reveals Drug Target Public Health: Common Industrial Ingredient Appears Able to Stifle Sperm Neurology: Alzheimer's Culprit Fingered as Gang of Four Immunology: New Vaccine Role Discovered for Anthrax Health Disparities: Minority Health Policy Program Points to Need for Dollars and Data AIDS Research: Advanced AIDS Research Facility to Open in South Africa in Collaboration with HMS New Books: The Summer Bookshelf   Rare Disease Offers Explanation of Ovarian Cancer's Drug Resistance High Colorectal Surgery Volume Tied to Better Outcomes Study Suggests Timely Control Efforts Can Halt Spread of SARS   Neuro Center Awards Innovation Grants Medical School Presents Faculty Teaching Awards New Chair in Women's Health Established at Brigham and Women's Hospital Delbanco Receives Glaser Award from the Society of General Internal Medicine HMS Faculty Named to American Academy of Arts and Sciences Alfred Sommers Wins Warren Alpert Prize HMS Faculty Council Reynolds Foundation Awards $24 Million for Heart Research at HMS and Brigham and Women's Hands-on Training Offered for GenBank and NCBI Molecular Resources Countway Redesigns Website CDC Warns of Squeeze on Disease Prevention Funds Honors and Advances Save the Date   Operation Mouthguard Protects Oral Health   Troubleshooting Allocation of Transplant Organs   Front Page

NEUROLOGY Alzheimer's Culprit Fingered as Gang of Four The enzyme known as gamma-secretase, notorious for its purported role in Alzheimer's disease, has eluded identification for years. Now, it seems that the protein players collectively responsible for gamma-secretase's activity have been caught. A study led by Dennis Selkoe and Michael Wolfe and published in the May 27 Proceedings of the National Academy of Sciences is one of a trio of recent papers all showing that gamma-secretase is really a close-knit gang of four proteins: presenilin, nicastrin, aph-1, and pen-2. Finally identifying this foursome will allow further studies on the role of the enzyme in several cell functions and may point toward better therapies for Alzheimer's. Gamma-secretase, an enzyme implicated in Alzheimer's disease, is really a joint venture among four proteins, according to a study by (l to r) Michael Wolfe, W. Taylor Kimberly (seated), Matthew LaVoie, Dennis Selkoe, and Wenjuan Ye. (Photo by Steve Gilbert) Gamma-secretase was initially discovered for its role in turning the amyloid precursor protein (APP) into amyloid-beta (A-beta), the plaque-forming protein that many believe is responsible for the neurodegeneration in Alzheimer's. APP is a membrane protein that undergoes two proteolytic cuts before ending up as A-beta; gamma-secretase is responsible for the final cut that casts A-beta off into the space outside the cell. "We think that most of the causative action has not yet been discovered, that it's in the A-beta clearance mechanisms." --Dennis Selkoe Selkoe, who is the Vincent and Stella Coates professor of neurologic diseases at HMS and Brigham and Women's Hospital; Wolfe, HMS associate professor of neurology at BWH, and their colleagues speculated in 1999 that gamma-secretase may actually be presenilin, a protein that is mutated in many cases of inherited, early-onset Alzheimer's disease (see Focus, April 16, 1999). But there was a complication to this theory: boosting the levels of presenilin expressed in cells did not also increase A-beta production, which would be expected if presenilin alone were performing gamma-secretase's action. It was assumed that there were "limiting factors," one or more proteins associated with gamma-secretase that were required to release the brakes on the enzyme's function. The protein nicastrin seemed to be an accomplice, but the two alone still did not result in increased gamma-secretase activity. A breakthrough came last year when a team from Exelixis, Inc. found that the C. elegans genes aph-1 and pen-2 are somehow necessary for gamma-secretase function. Several groups then took on the task of seeing what role these two new proteins play in the enzyme. Critical Players In Selkoe's lab, MD-PhD student W. Taylor Kimberly and coworkers used a line of hamster cells that express the human form of presenilin to evaluate the role of the other proteins in cleaving APP. Kimberly overexpressed the proteins alone and in different combinations to see which would result in more gamma-secretase activity. Only the entire foursome resulted in more A-beta production. "We found that when all four are at an excess level in the cell, you release the limit on gamma-secretase," Selkoe said. The study established in mammalian cells what two other groups were finding simultaneously in invertebrates. A group from the University of Tokyo found that blocking each protein in fruit fly cells using RNA interference disrupted gamma-secretase activity. Another group from Ludwig Maximilians University in Munich employed yeast, which lack APP and all the components for making A-beta. When all four ingredients were added, even yeast turned APP into A-beta. The Selkoe-Wolfe team also used a purification technique that isolates the active form of gamma-secretase to show that each of these proteins remains associated with the active complex, supporting the notion that all of them serve a direct role in proteolysis. While other players may still be required to traffic or stabilize the gamma-secretase complex in cells, these three studies together suggest that this quartet of proteins are the essential components. The end result is a beast of a complex, crossing the membrane 18 times. "We envision that the 18 transmembrane domains form some kind of porelike structure" surrounding the substrate to be snipped, Selkoe said. He believes presenilin is still the lead player, the part of the complex that actually severs the substrate. The Final Cut Gamma-secretase has already helped change some assumptions about cell signaling. APP sits astride the plasma membrane of cells; gamma-secretase must perform the final cut inside the membrane through hydrolysis, a feat that requires getting a water molecule into a membrane made of lipids. The discovery that gamma-secretase also clips the developmental protein Notch showed that this unusual form of proteolysis serves an important signaling function in cells, a phenomenon only seen before in an aspect of cholesterol homeostasis. Now, several proteins have been unearthed that rely on gamma-secretase, and the form of protein-cutting it employs has its own name: regulated intramembrane proteolysis (RIP). Classical signaling pathways can seem like a game of telephone, in which the binding of a membrane receptor triggers a signal that is passed down in a cascade of proteins inside the cell and eventually affects gene transcription. With RIP, the liberated fragment in the cell simply travels to the nucleus to turn on genes. Bruce Yankner, HMS professor of neurology at Children's Hospital, said, "This kind of mechanism results in quick and direct activation of gene transcription, bypassing adapter proteins and kinase cascades. It is less modifiable but possibly more definitive in its end result than classic kinase signaling cascades." Selkoe believes that gamma-secretase may explain why Alzheimer's disease arose in the human population. The enzyme has a vital role in processing Notch, which may be among the most important proteins to guide development. The processing of APP may have come later--and some researchers believe that the protein's tiny tail, which is left inside the cell once A-beta is disengaged, itself serves a signaling function. A-beta, Selkoe argues, is the leftover, and it has the unfortunate ability to aggregate if it accumulates outside neurons. Sound like a poorly designed system? In evolutionary terms, the slow buildup of A-beta was of little consequence before people lived long enough to experience the resulting neurodegeneration. Most cases of late-onset Alzheimer's disease probably have nothing to do with a defect in gamma-secretase. "We think that most of the causative action has not yet been discovered, that it's in the A-beta clearance mechanisms," Selkoe said. But targeting A-beta production, through gamma-secretase and beta-secretase, is still a high priority for drug development. Rudolph Tanzi, HMS professor of neurology at Massachusetts General Hospital, said that a drug could not wipe out gamma-secretase without causing harm, but it might make a slight modification to inhibit the cleavage of APP selectively. For instance, gamma-secretase can produce two versions of A-beta, 40 and 42, the longer of which has a greater propensity to aggregate. "One idea is to try to find gamma-secretase inhibitors that target the production of 42 over 40," Tanzi said. The discovery of four key players in gamma-secretase, while complicating the picture of how the enzyme works, now offers more drug targets. Selkoe, who cochairs the Harvard Center for Neurodegeneration and Repair, hopes to take advantage of the Center's Laboratory for Drug Discovery in Neurodegeneration to screen for compounds that interfere with gamma- secretase in different ways. --Courtney Humphries