ARTICLE IN BRIEF
From the largest genome-wide scans to date for Alzheimer disease, two independent investigative teams identified three new genes as risk factors for sporadic disease.
Two new whole-genome scans for Alzheimer disease (AD) genes, the largest such studies ever conducted in the disease, have identified three genes as new risk factors for sporadic disease. They join apolipoprotein E (APOE) in offering the best clues to date for understanding the origin of AD in most patients.
The genome-wide association scans were conducted by two international groups, one led by Julie Williams, PhD, professor of neuropsychological genetics in the department of psychological medicine at Cardiff University, Wales, and the other by Philippe Amouyel, MD, PhD, of the Institut Pasteur de Lille, in Lille, France. The findings were reported Sept. 6 online in advance of the print edition of Nature Genetics
Each study enrolled approximately 4,000 patients, and determined their genotype for over a half million single nucleotide polymorphisms (SNPs) - the common sequence variants found in all individuals - using DNA microarrays (gene chips). A SNP found significantly more often in people with AD than in control populations indicates that the gene containing the SNP is associated with the disease.
This approach can be successful when you have a powerful enough sample, Dr. Williams said. The need for such large samples has been clear for several years, because no other gene has been discovered with as large an effect on risk as APOE, the only common risk factor for AD that has been identified before these studies. Because other genes were likely to have smaller effects, more patients were needed for them to emerge from the background noise. In the past, many other genes have been suggested as contributing to the disease, but have been found only in much smaller studies, and have not been confirmed.
NEW GENES, FAMILIAR PATHWAYS
Both of the new studies found that a gene called CLU, for the protein clusterin, is associated with AD. When we discovered that one of the two genes we found was in common with those found by Dr. Williams's group, we were very excited, Dr. Amouyel said. It was an immediate replication.
Clusterin is also known as apolipoprotein J, and is a major brain apolipoprotein, suggesting that susceptibility genes are not randomly distributed through functional pathways, Dr. Williams said. These genes form patterns. They are implicating particular biological mechanisms.
Other studies have shown that clusterin is elevated when there is brain injury or inflammation. In AD, clusterin is elevated in affected cortical areas, and is present in amyloid plaques. It can bind amyloid beta (Abeta), and appears to contribute to its clearance across the blood-brain barrier, as APOE does.
Previous studies of rare genes in AD have made it clear that amyloid overproduction is important in disease pathogenesis, at least in people with these mutations. Problems with clearance seem to be at work in late-onset disease, Dr. Williams said. This seems to be another route into the disease.
Dr. Amouyel also found a gene called CR1, encoding a complement receptor in the immune system. One of its functions appears to include clearance of Abeta as well. The specific complement molecule that CR1 binds to, called C3b, binds to aggregated Abeta, and overexpression of C3b ameliorates disease in animal models. Altogether, these data support a protective role for CR1, Dr. Amouyel said, through C3b-mediated amyloid clearance.
But CR1 also forms part of the innate immune response, and inflammation has long been implicated in worsening neurodegenerative diseases, including AD. Dr. Williams pointed out that clusterin also has an immune system role, in dampening inflammation. We know that subtle changes in the immune system can have quite serious implications, and could well result in damage to the brain, she said. It is possible, she added, that rather than being a secondary phenomenon, inflammation may be a primary contributor to disease pathogenesis. The genes tell us that these are important.
Finally, Dr. Williams found a gene called PICALM, which encodes a protein that mediates endocytosis at the synapse and elsewhere. It is possible it is linked to the amyloid beta story as well, Dr. Williams said. We know from other research that in distress, endocytosis patterns change, and trafficking tends to put more amyloid precursor protein into compartments that process it abnormally, producing more beta amyloid.
So far, that specific mechanism is speculative, she emphasized, as is the suggestion that the CR1 role in AD is through inflammation. The next step for each of the new genes is to determine what specific sequence changes - which alleles - are associated with the disease, just as was done for APOE, in determining it was the epsilon-4 allele that conferred the greatest risk. That information can be used to develop new animal models and to ask how changes in the encoded proteins cause changes in cell function.
How important are these new genes, compared to APOE? Dr. Amouyel calculated the attributable risk fraction to be 25.5 percent for APOE, 8.9 percent for CLU, and 3.8 percent for CR1. Dr. Williams estimated the contribution of PICALM to about 9 percent. Given that genes are thought to contribute between 60- and 80-percent of the risk for AD, more genes are waiting to be found, Dr. Amouyel said. That will require even larger samples, and such studies are already being planned.
NEW GENES: 'TIP OF THE ICEBERG'
These new genes are not the end of line, they are the tip of the iceberg, according to Margaret Pericak-Vance, PhD, director of the Hussman Institute for Human Genomics at the University of Miami Miller School of Medicine in Florida.
She noted that APOE was discovered in 1993, and that it has been frustrating since then, because we haven't been able to really add to the confirmed gene pile, at least not with genes that everybody could agree on. These new studies show you the power of a genome-wide association scan to pull out information.
Gratifyingly, she said, the two studies confirm one another in their identification of CLU. That pathway looks interesting. I think we will see a coherent story coming out of it. We are starting to see it all come together.
It will be important to look now for the same genes in populations of non-European descent, to see how widely implicated the genes are. If we can do that, then we are ready to look at function, she said.
While the newer genes don't have the same effect size as APOE, they are still very important. Anything you learn about the disease is important, because it can take you into new pathways, Dr. Pericak-Vance said. The more we understand the genetics, the further we are towards figuring out a target for treatment or prevention.
REFERENCES