Columbia University Medical Center
In Vivo - The Newsletter of Columbia University Medical Center
Back Issues
Contact Us
In Vivo
 New Alzheimer’s Gene Found

Photo: Charles Manley
CUMC researchers
CUMC researchers involved in the discovery of a new Alzheimer’s gene gather around sheets showing genetic histories of Alzheimer’s patients from the Dominican Republic. From left to right, Scott Small, Joseph Lee, Rafael Lantigua, Vincent Santana and Richard Mayeux.
An international team of researchers from CUMC, University of Toronto, Boston University and the Mayo Clinic has discovered the first Alzheimer’s gene since 1993, the year apoE was linked to disease risk.
   The discovery of the gene, dubbed SORL1, may provide a new way to stop the disease because it identifies a previously unknown pathway within cells by which the toxic amyloid beta peptides that cause the disease can build up in the brain.
   “Identifying genes that raise the risk of Alzheimer’s helps us diversify our portfolio for ways to treat the disease,” says Richard Mayeux, M.D., Gertrude H. Sergievsky Professor of Neurology, Psychiatry and Epidemiology, director of the Gertrude H. Sergievsky Center, and co-director of the Taub Institute for Research on Alzheimer’s Disease and the Aging Brain.
   Dr. Mayeux led the study with Peter St. George-Hyslop of Toronto and Lindsay Farrer of Boston University. “We can’t predict if this one will lead to a treatment, but it is best to identify as many pathogenic pathways as possible to generate as many ideas for treatment as possible,” he says.
   The new gene was discovered by comparing single nucleotide polymorphisms (SNPs) in the DNA of Alzheimer’s patients and unaffected people from the Dominican Republic. Certain patterns of SNPs inside the SORL1 gene were more prevalent in Alzheimer’s patients than in unaffected people.
   The same patterns were found in four other separate groups of subjects: two groups of white Americans, one group of African Americans, and one group of Israeli Arabs. In total, more than 6,000 people participated in the study.
   “Hundreds of genes have been linked to Alzheimer’s in the past 15 years but those findings have not been replicated,” Dr. Mayeux says. “We think SORL1 is different because we built several replicates into the study, and those replicates include different ethnic groups. But even with the replication we have, we all want to see this finding repeated by other researchers before we accept it completely.”
   When today’s baby boomers were born, Alzheimer’s disease was considered a rare disease that primarily affected young people in their 30s and 40s. Not until the 1960s did researchers learn that most cases of senile dementia were actually caused by Alzheimer’s.
   Far from rare, Alzheimer’s affects one out of every 10 people over the age of 65 and one of every two over the age of 85.
   By the 1990s, the cause of Alzheimer’s disease was generally attributed to a build-up of amyloid beta peptides, which are released when a protein called APP is cut into pieces. But the genetic and environmental factors that determine how much amyloid beta accumulates in the brain have been difficult to determine and remain largely unknown.
   The new study shows that certain variants of the SORL1 gene increase the risk of developing Alzheimer’s because of the effect they have on APP.
   Inside cells, SORL1 transports APP to “safe” compartments where APP is hidden from enzymes that cut it into amyloid beta peptides.
   But the cells of Alzheimer’s patients carrying the risk-increasing variants produce fewer SORL1 proteins than normal. Cells with low levels of SORL1 leave more APP proteins stranded outside safe compartments, and the end result is more toxic amyloid beta peptides.
   “The job now is to find compounds that increase SORL1 expression and reduce amyloid beta production,” Dr. Mayeux says. “That will be difficult but not impossible. And we don’t feel we have to elevate SORL1 very much to achieve a benefit.”
   Scott Small, M.D., who provided an important clue that led to the gene’s discovery (see below), is now rapidly screening hundreds of thousands of molecules for such compounds.
   The discovery of SORL1 also may prove to be important in calculating each individual’s risk of developing the disease. Right now, the researchers do not know how much SORL1 increases the chances of getting Alzheimer’s, but they estimate the gene increases risk by 10 percent to 20 percent (for comparison, the e4 allele of the apoE gene raises risk by approximately 20 percent).
   “In the future, there may be prognostic value in knowing your SORL1 status, just as now there is prognostic value in knowing your cholesterol levels,” Dr. Mayeux says. “We hope to have preventative treatments we can give to people with the greatest risk of developing the disease, much the same way we now give statins to prevent cardiovascular disease.”
   The research was published in the February issue of Nature Genetics. It was supported, in part, by the Howard Hughes Medical Institute, the National Institute on Aging, and the Alzheimer’s Association.

—Susan Conova

New Technology Key to Alzheimer’s Gene Discovery

The search for a new Alzheimer’s gene succeeded because of the hard work of many geneticists. But the discovery of SORL1 also depended on a big clue from a new type of microarray analysis developed by a CUMC physician-scientist.
   The power of microarrays to measure the activity of tens of thousands of genes all at once has also been its biggest liability. “Everyone thought microarrays had great promise to uncover the pathogenic molecules underlying complex disorders like Alzheimer’s, but it generates so much data you’re left wondering what’s relevant,” says Scott Small, M.D., associate professor of neurology. “It has generated a lot of frustration. What it really underscores is the need to be more sophisticated when analyzing the data.”
   To filter out more irrelevant data, Dr. Small created a model of how an Alzheimer’s-causing protein should be distributed in the brain over time and then looked in microarray data for proteins that fit the model. The model was developed from fMRI images made from living patients. The microarray data came from post-mortem brains of Alzheimer’s patients and control subjects. Dr. Small worked closely with Dr. Mayeux and the Taub Institute for Research on Alzheimer’s Disease and the Aging Brain to collect the images and data.
   Several proteins fit the model, but VPS35, a protein that works with SORL1, stood out. This finding, together with other clues, led Dr. Mayeux to focus his search for Alzheimer’s genes on VPS35, SORL1 and five other related genes.
   “This shows that it’s much easier to find genes for complex genetic disorders when you have an idea what you’re looking for,” Dr. Small says.
   Dr. Small is now applying his technique, which he calls model-guided microarray, to Parkinson’s and other neurodegenerative diseases. “It may turn out that our success with Alzheimer’s was just luck,” Dr. Small says, “but I don’t think so.”
A feature story highlighting the search for SORL1 appeared in the Dec. 12, 2005, issue of the New Yorker magazine.