New gene in Alzheimer's disease

A new gene, sortilin-related receptor SORL1, is joining the ranks of other genes scientists have associated with late-onset Alzheimer's disease, thanks to a report in this week's Nature Genetics. Many of the two dozen genes already linked to the disease have disappointed researchers by demonstrating inconsistent correlation across data sets, sparking debate over the significance of the latest findings.

Still, some scientists are optimistic. "The main difference is the size of the study [linking SORL1 to Alzheimer's], the replication, and the diversity of the sample," Creighton Phelps, director of the Alzheimer's Disease Centers Program at the National Institute on Aging (NIA), told The Scientist. The NIA funded a portion of the study, but Phelps was not involved in the study.

The project, which included teams from Columbia University, the University of Toronto, and Boston University, sampled over 6,000 participants from seven data sets and four ethnic groups. These groups included Caribbean Hispanic and Israeli Arab -- which are known to have an elevated prevalence of Alzheimer's disease -- Northern European, Caucasian, and African American.

The teams searched for single nucleotide polymorphisms (SNPs) among several genes involved in processing amyloid precursor protein (APP), the molecule that generates the toxic amyloid beta peptide (Abeta) involved in Alzheimer's. One gene showed up positive: Six SNPs in two distinct regions of the gene for sorLA/LR11, a receptor that regulates the recycling of APP, were associated with late-onset Alzheimer's disease.

The authors speculate that variants in SORL1 reduce the expression of the gene, sending APP away from the recycling pathway and down the pathway toward Abeta. "It all makes biological sense, it's certainly very plausible," Allen Levey, director of the Center for Neurodegenerative Disease and Alzheimer's Disease Center at Emory University, told The Scientist. In 2004, Levey showed that brains of patients with Alzheimer's disease were about two-fold less abundant in sorLA/LR11. His group is looking into the role sorLA/LR11 might play during mild cognitive impairment, a precursor to Alzheimer's disease.

Lindsay Farrer, chief of the Genetics Program at Boston University and one of the authors on the study, points out that the group has not yet shown that the variants they found result in any reduction in sorLA/LR11 expression. "This is where we hit the limits of the work we've done," Farrer told The Scientist. "As enthusiastic as we are, we have not identified yet any variants that are causally or functionally related to the disease."

This group is not alone: Among other Alzheimer's-associated genes, only one -- APOE -- has yielded evidence of mutants functionally related to the disease, Rudolph Tanzi reported in a meta-analysis in the January issue of Nature Genetics. The analysis, led by Lars Bertram at Massachusetts General Hospital, compiled data from 900 publications on Alzheimer's genetics.

Possessing just one copy of the mutant APOE allele increases the risk of developing Alzheimer's two to three-fold, Tanzi, a Neurology professor at Massachusetts General Hospital, told The Scientist. Relative to APOE, all of the other AD-associated genes described in these publications carry a much smaller risk of developing Alzheimer's, and Tanzi said SORL1 appears no different. The AlzGene Web site developed from the meta-analysis ranks Alzheimer's-associated genes in terms of risk of developing the disease; SORL ranks at around 13th, Tanzi said, giving it a "middle-of-the-pack effect."

Mutations in APOE account for 30% of the genetic risk of Alzheimer's disease, Tanzi said, and the rest he expects results from a combination of risk from many of these other associated variants. "It's still important to discover all of these genes that are associated, because it's all a part of the puzzle. All of them are important, but is [SORL1] more important than any of those that have come before it? No it is not," Tanzi said.

The authors of the SORL1 study say they are eager to see the results of independent analyses on the gene's association with Alzheimer's disease. The next step, Farrer said, is to identify the SORL1 mutations responsible for the elevated risk. He's confident his colleagues will find them. "I believe it's a matter of time, [but] it won't necessarily be easy."

Kerry Grens
kgrens@the-scientist.com

Links within this article:

E Rogaeva et al., "The neuronal sortilin-related receptor SORL1 is genetically associated with Alzheimer disease," Nature Genetics, published online January 14, 2007.
http://www.nature.com/ng/index.html

National Institute on Aging
http://www.nia.nih.gov

MW Anderson, "Amending the amyloid hypothesis," The Scientist, October 25, 2004.
http://www.the-scientist.com/article/display/15006

N Johnston, " Gamma-Secretase makes a splash," The Scientist, July 4, 2005.
http://www.the-scientist.com/article/display/15596

Allen Levey
http://neurology.emory.edu/Faculty/Levey.htm

CR Scherzer et al., "Loss of apolipoprotein E receptor LR11 in Alzheimer disease," Archives of Neurology 61:1200-1205, 2004.
http://archneur.ama-assn.org/cgi/content/abstract/61/8/1200

Lindsay Farrer
http://www.bu.edu/alzresearch/team/faculty/farrer.html

R Mahley and Y Huang, "Going beneath the fold," The Scientist, August 1, 2006.
http://www.the-scientist.com/article/display/24105

Rudolph Tanzi
http://www.hms.harvard.edu/dms/neuroscience/fac/tanzi.html

L Bertram et al., "Systematic meta-analysis of Alzheimer disease genetic association studies: the AlzGene database," Nature Genetics 39:17-23.
http://www.nature.com/ng/journal/v39/n1/abs/ng1934.html

AlzGene
http://www.alzforum.org/res/com/gen/alzgene/default.asp