Two papers published this month claim to identify those at greatest risk for Alzheimer’s disease and illuminate the mechanisms behind it. In one, French researchers identify a gene that may be poised to join the ranks of the few that are blamed for early onset AD. Mutations in SORL1, which has previously been implicated in late-onset forms of the disease (see ARF related news story) showed up in seven out of 29 families who had early onset AD with no known genetic mutation, according to a Molecular Psychiatry paper published April 3. A paper published April 6 in the American Journal of Human Genetics by U.S. researchers aims to connect some of the most prominent AD risk alleles according to their past natural selection. It suggests that a protective module incorporating three of those alleles may have been beneficial in human ancestry.

Dominique Campion of INSERM in Rouen and colleagues have long sought genes responsible for familial, early onset AD. Many cases of this form of AD are not caused by mutations in the amyloid precursor protein (APP), presenilin-1 (PSEN1) and PSEN2 genes, and Campion is among those researchers who have gone looking for more culprits. His lab group previously added candidate genetic alterations to the list when they described a duplication in the APP gene (see ARF related news story on Rovelet-Lecrux et al., 2006) and rare copy number variants in genes involved in Aβ processing in some early onset families (see Rovelet-Lecrux et al., 2011). Of the 130 autosomal-dominant families they screened, however, 14 still lacked a genetic explanation for their disease.

In the current study, first author Cyril Pottier and colleagues sequenced the exomes of individuals from those 14 families. None carried any of the previously known mutations or copy number variants associated with early onset AD. But in five participants, the researchers found missense or nonsense mutations in SORL1. This gene encodes a transmembrane neuronal sorting receptor called LR11/SorLA that binds APP and controls its movement through the retromer/retrograde pathway. None of these variants surfaced in a control sample of 1,500 healthy individuals. To replicate their findings, the researchers looked at another early onset patient group of 15 cases with no known mutations. There, SORL1 was mutated in two patients—one missense and one nonsense. "This is the first evidence that SORL1 is involved in early onset forms and autosomal-dominant forms of Alzheimer's disease," said Anne Rovelet-Lecrux, also of INSERM, a coauthor on the article.

But were these mutations harmful? Scientists need to show that a mutation not only exists, but also is pathogenic. Previous research has shown that hampered SORL1 expression leads to more Aβ production (see Rogaeva et al., 2007). In one of the individuals with mutated SORL1 in the French study, RNA blood cell analysis revealed stunted SORL1 expression. Further, computational analyses predicted that the other mutations were likely damaging or pathogenic. But to really figure out whether the mutations are deleterious, the research team plans to develop cell models to analyze the pathogenic effects of each of these SORL1 mutations on Aβ release, Rovelet-Lecrux said. In addition, since segregation data were unavailable in most of the families with these mutations, the team cannot be sure yet whether SORL1 mutations result in fully penetrant early onset disease.

"These are promising candidate mutations," said Rudy Tanzi, Massachusetts General Hospital, Boston, who was not involved with the research. He agreed that crucial next steps will be to validate these mutations in cell models or animal models and show coinheritance in families. "We still need to see functional validation before we call these bona fide early onset mutations," he said.

Since there are many more families with early onset AD and unknown genetic mutations, more genes for early onset AD are likely yet to be found—some perhaps involved with Aβ in some way, said Rovelet-Lecrux. "The screen picked a novel gene in [the amyloid] pathway, thus reinforcing the amyloid cascade hypothesis in early onset forms at least," Campion told Alzforum in an e-mail. While researchers have likely found all the deterministic mutations for the earliest forms of the disease that become symptomatic in people's forties or even earlier, there probably are other genes still to be found for early onset AD that occurs between the ages of 50 and 65, Campion said. Compared to the known APP and presenilin mutations, the SORL1 gene seems to fit into this latter category.

Campion's lab group will continue to probe the exomes of early onset family members to seek out more mutations. Not only will the results help identify the people most at risk for the disease, but they may lead to insights about the cause and treatments for Alzheimer's, Tanzi told Alzforum. "The genes tell you what's broken and thus what you need to fix," he said. "Every new gene provides another opportunity to do that."

A separate trend in AD genetics research is to try to find patterns between AD risk genes. "We have these susceptibility genes, but we don't understand how they work in the context of Alzheimer's disease," said Philip De Jager of Brigham and Women’s Hospital and Harvard Medical School in Boston, Massachusetts. He and his colleagues decided to look for evidence of recent natural selection in the top 11 gene variants associated with sporadic Alzheimer's disease (see AlzGene Top Results). The goal was to see if, like the ApoE locus (see Drenos and Kirkwood, 2010), any of these top hits had undergone recent selection. Three of the 11 genes they examined genes—PICALM, BIN1, and CD2AP—showed evidence of recent natural selection in people of East Asian descent. "The odds of that are pretty small," said De Jager. The genes may have been under selection pressure together because of some non-AD-related event in that particular population's history—perhaps a pathogen or metabolic or environmental challenge. "If true, the genes may be part of the same pathway," he said.

To test that hypothesis, the group used a computational tool that looks for connections between genes based on whether their proteins interact and whether they are expressed in the same tissue. Though the proteins did not connect directly to one another, they did fit into the same pathway. Several intermediate proteins also seem to have undergone natural selection in the same East Asian population. One such intermediate, GAB2, has been associated with AD already. "We offer robust statistical evidence that these genes are interacting, and we link in new genes that were not suspected to be part of the pathway," De Jager said. "These offer targets for further genetic work to see if they do contribute to susceptibility of Alzheimer's." The researchers don't know exactly what the proteins do together; however, several of them are generally involved in vesicular trafficking.

The team also found that some of the gene variants with evidence of natural selection seemed to influence gene expression in immune cells. These results suggest that the variants may have once been under selection because they altered immune cell function and protected individuals against a pathogen. Though speculative, "it is a plausible explanation [for how] this functional gene module, which is important for Alzheimer's disease today, probably has other uses in other contexts," De Jager said. Using mRNA data from immune and non-immune cells, he and colleagues plan to investigate this functional module to define its component genes and characterize the consequences of AD-associated genetic variation on its function.

"How the proteins coded by these genes interact could be an important part of the mechanism in Alzheimer's disease," said Caleb Finch, University of Southern California in Los Angeles. "The association of this cluster with modern Alzheimer's risk factors is intriguing and will lead to productive experiments to test the hypothesis."—Gwyneth Dickey Zakaib


No Available Comments

Make a Comment

To make a comment you must login or register.


News Citations

  1. SORLA Soars—Large Study Links Gene to Late-onset AD
  2. APP Double Dose Causes Early Onset AD

Paper Citations

  1. . APP locus duplication causes autosomal dominant early-onset Alzheimer disease with cerebral amyloid angiopathy. Nat Genet. 2006 Jan;38(1):24-6. Epub 2005 Dec 20 PubMed.
  2. . A genome-wide study reveals rare CNVs exclusive to extreme phenotypes of Alzheimer disease. Eur J Hum Genet. 2011 Dec 14; PubMed.
  3. . The neuronal sortilin-related receptor SORL1 is genetically associated with Alzheimer disease. Nat Genet. 2007 Feb;39(2):168-77. PubMed.
  4. . Selection on alleles affecting human longevity and late-life disease: the example of apolipoprotein E. PLoS One. 2010;5(4):e10022. PubMed.

Other Citations

  1. early onset AD

External Citations

  1. SORL1
  2. AlzGene Top Results
  4. BIN1
  5. CD2AP
  6. GAB2

Further Reading


  1. . Association of CR1, CLU and PICALM with Alzheimer's disease in a cohort of clinically characterized and neuropathologically verified individuals. Hum Mol Genet. 2010 Aug 15;19(16):3295-301. PubMed.

Primary Papers

  1. . High frequency of potentially pathogenic SORL1 mutations in autosomal dominant early-onset Alzheimer disease. Mol Psychiatry. 2012 Apr 3; PubMed.
  2. . Alzheimer disease susceptibility loci: evidence for a protein network under natural selection. Am J Hum Genet. 2012 Apr 6;90(4):720-6. PubMed.