To paraphrase Winston Churchill, Alzheimer's disease is a riddle wrapped in a mystery inside an enigma. Thousands of researchers in labs around the world are hard at work every day trying to unlock its secrets. But how does one begin to unravel the cause of a disease that arises from the interplay of dozens of genes plus a number of environmental factors? To date, 900 scientific papers have identified 350 candidate genes that may be involved in late-onset Alzheimer's, the form of the disease that accounts for roughly 95 percent of cases. Yet researchers have reached a consensus on only one of them—the APO E4 gene variant. That's why a paper appearing Sunday in the online version of the journal Nature Genetics is drawing attention. In it, an international team of 41 scientists has provided strong evidence for the involvement of another gene, called SORL1. The new gene appears to confer only a modest degree of susceptibility for Alzheimer's, but simply knowing that it is involved in the disease process could potentially help researchers find new drug targets for treatment.
SORL1 is a logical player in Alzheimer's. Scientists already knew that it makes a protein—similarly dubbed sorL1—that performs many tasks in a brain cell. Among these is the job of binding to a substance called amyloid precursor protein (APP) and making sure it goes to a part of the cell where it will be chopped into harmless segments. If there is too little sorL1, APP strays instead into a different part of the cell, where it is transformed into a gummy substance called beta amyloid, or A-beta. Strands of A-beta clump together into toxic fibers, which appear to contribute to the death of brain cells in patients with Alzheimer's. The analogy that some scientists use is that sorL1 acts as a sort of traffic cop. "It forces APP away from the parts of the cell that make A-beta," said Dr. Peter St. George-Hyslop of the University of Toronto, the paper's senior author, in a pre-publication press conference.
For proponents of the so-called amyloid hypothesis of Alzheimer's, the new finding provides one more piece of evidence that A-beta is a critical player in the development of the disease. It's a piece of evidence that's hard to dismiss. Unlike most studies, which involve just one population group and need to be replicated by other labs before scientists will accept them, this study was conducted in nine different groups comprising 6,800 individuals from several ethnic populations—namely, European and American whites, African-Americans, Dominican Hispanics and Israeli Arabs. "We provided built-in replication," says Lindsay Farrer, chief of the genetics program at Boston University School of Medicine and another of the paper's lead authors. "When you see the same result again and again in different populations, it makes the data more compelling. We deliberately didn't publish immediately but waited until we had confirmation."
Equally important, the discovery could provide a target for the development of new drugs. An earlier study in the Archives of Neurology had found that individuals with late-onset Alzheimer's have lower levels of sorL1 in brain tissue—but had left open the question of whether lower sorL1 was a cause or an effect of the disease. The new genetic evidence suggests that it may indeed have been causal. "I can imagine screening for drugs that elevate sorL1," says Dr. Samuel Gandy, director of the Farber Institute for Neurosciences at Philadelphia's Thomas Jefferson University and chair of the Alzheimer's Association's medical and scientific advisory council. "We wouldn't have given that a lot of attention last week. We knew it was a step, but one of dozens." Of course, it's a long way from an idea to an actual drug. The fact that sorL1 is involved in many other "housekeeping" functions in the brain may mean that elevating its levels throws something else out of whack. "Or maybe other functions will also benefit from elevating sorL1," says Farrer. Only testing in lab cultures, then animals and finally people will provide the answer.
One thing that won't come out of the discovery is a new genetic test for Alzheimer's. That's partly because the scientists haven't yet located the actual mutations on the SORL1 gene that contribute to the disease risk and partly because the increased risk conferred by this gene alone is probably relatively small. It is certainly not as powerful as the APO E gene, which is widely considered the most important of the susceptibility genes for late-onset Alzheimer's. If you have just one copy of APO E4, your chances of getting the disease are two to three times higher. If you carry two copies, your risk is increased 10-fold. Even so, scientists can point to people in their 80s and 90s who have APO E4 but have not contracted Alzheimer's. SORL1's impact is clearly smaller, placing it in the second tier of risk genes. "This is just one small piece of a very large puzzle," says Dr. Rudolph Tanzi, professor of neurology at Harvard Medical School and one of the leading researchers in the genetics of Alzheimer's. He should know. He and Lars Bertram at Harvard have set up a Web site at AlzGene.org with data on hundreds of candidate genes for Alzheimer's. "As soon as we have testing in multiple study populations, we look at the data collectively to see whether the candidate gene poses a significant risk or not," Tanzi says. "Of the 26 genes in this category, SORL1 will rank 13th in its influence on risk." Still, as he points out, even a gene with a small effect can sometimes lead to an important drug.
Given the complexity of late-onset Alzheimer's, a genetic test will probably have to await the completion of a so-called Alzheimer's genome project. One such endeavor is being headed by Tanzi with backing from the nonprofit Cure Alzheimer's Fund. Rather than studying one gene at a time, he is using powerful new DNA chips to screen the entire genomes of thousands of people with and without Alzheimer's disease in an attempt to spot the key differences in a very complex pattern. He hopes to have the profile completed by the summer of 2008 and a prototype gene test ready within five years. Farrer and others have similar projects underway. But in the end, Farrer expects that a full understanding of all the genetic players in Alzheimer's will require sustained collaboration among many groups of scientists. "In order to see further advances in the field, we need to move beyond the old way of doing science—where someone goes into a lab, makes a discovery and shouts 'eureka'," he says. "We need large-scale cooperation, as we had in this study."
However progress is achieved, it needs to come soon. There are already 4.5 million Americans with the disease—and a very large population of baby boomers heading into the age when they will be at risk. Many people are depending on a swift decoding of the enigma.