27 June 2013. Scientists hunting for new Alzheimer’s genes usually focus on single nucleotide changes, but a large amount of human genetic variation takes the form of structural rearrangements of DNA, such as deletions and duplications. To look for disease-associated copy number variations, researchers led by Rudy Tanzi at Massachusetts General Hospital, Charlestown, examined the genomes of 261 families with early-onset AD for whom no genetic cause was yet known. In the June 11 Molecular Psychiatry, they report finding 10 novel variations that co-segregate with disease and implicate 18 new genes in its pathogenesis. Some of these genes have known roles in neuronal functions such as axon guidance and neurotransmitter production; others were previously linked to other neurological conditions, including frontotemporal dementia. One variation duplicates the tau gene. Tau tangles are a hallmark of Alzheimer’s pathology, but the gene had not previously been shown to affect AD risk. Further research will be needed to determine how these variations might increase risk for the disease.
Copy number variations (CNVs) occur frequently in the human genome (see ARF related news story). Researchers estimate they account for about 18 percent of the genetic variation in gene expression (see Stranger et al., 2007). Scientists predict that structural DNA variations may play a role in numerous neurological diseases (see ARF related news story; ARF related news story). For example, CNVs may increase risk for schizophrenia (see Schizophrenia Research Forum related news story; Vacic et al., 2011), autism (see Sebat et al., 2007; Sanders et al., 2011; Levy et al., 2011), and Parkinson’s disease (see ARF related news story). Duplications of the amyloid precursor protein (APP) gene cause early-onset AD (see ARF related news story).
First author Raj Hooli examined the genomes of more than 1,000 people from the families with early-onset Alzheimer’s. Half the genomes came from the National Institute of Mental Health Alzheimer’s Disease Genetics Initiative and half from the National Cell Repository for Alzheimer’s Disease. Roughly two-thirds of the people had AD; the rest were unaffected siblings. Coauthor Zsolt Kovacs Vajna developed a new algorithm for comparing genomic segments between affected and unaffected siblings, and for identifying rare CNVs that segregate exclusively with disease. This program, called the Genomic Segment Correlator, will be made freely available to other scientists, Tanzi said.
The researchers found some 15,000 CNVs. Of those, 10 were new ones that segregated with disease within one family. They focused on CNVs that occurred only in single early-onset AD families because they wanted to find rare variations that were strongly associated with disease rather than common polymorphisms, Tanzi explained. Family members with these mutations developed AD at an average of 60 years of age. Five of the CNVs represented deletions averaging about 224 kilobases in size, the other five, duplications around 500 kilobases long. Some of the 18 genes in the vicinity of these 10 variations were previously linked to neurological disorders such as mental retardation, autism, epilepsy, cerebral palsy, and cerebral ataxia. Others have known roles in synaptic adhesion, axon guidance, neurotransmitter production, and learning and memory.
Some of the genes have ties to AD. For example, the very low density lipoprotein receptor (VLDLR) binds ApoE, the primary genetic risk factor for sporadic AD. Previous genetic studies have found no strong association between polymorphisms in VLDLR and AD, but the fact that the loss of one copy of the gene can lead to early-onset AD implies that this receptor may play an important role in helping ApoE to clear Aβ from the brain, Tanzi suggested. Another example is ataxin 2-binding protein 1 (A2BP1), which also binds ataxin 1, a known risk factor for AD. Knockout of ataxin 1 dramatically increases the level of BACE1, the secretase that helps produce Aβ, Tanzi said. Since loss of A2BP1 leads to early-onset AD, this binding protein may help ataxin 1 to throttle BACE1 production, Tanzi speculated.
Tanzi believes the study helps identify molecular players involved in Alzheimer’s pathology, suggesting that further research into the affected genes may turn up therapeutic targets that will benefit people with both early- and late-onset AD. The researchers are using the same methodology to look for disease-associated CNVs in people with sporadic AD.
It remains unclear how much of a genetic contribution CNVs might make to overall AD risk. A recent study (see Chapman et al., 2013) suggests CNVs may play less of a role in AD than they do in autism or schizophrenia, noted John Hardy at University College London, U.K. “None of the results reported are statistically significant, and so it is difficult to evaluate whether anything useful has been found or not. Chapman et al. came to the conclusion that CNVs had little or no role to play beyond rare APP duplications,” Hardy wrote to Alzforum. That study used different methodology from the current paper, looking at whether CNVs were overrepresented in more than 3,200 people with sporadic AD compared to about 1,300 controls. The authors found no excess of CNVs in AD and concluded that these structural rearrangements do not make a significant contribution to the disease.––Madolyn Bowman Rogers.
Hooli BV, Kovacs-Vajna ZM, Mullin K, Blumenthal MA, Mattheisen M, Zhang C, Lange C, Mohapatra G, Bertram L, Tanzi RE. Rare autosomal copy number variations in early-onset familial Alzheimer’s disease. Mol Psychiatry. 2013 Jun 11. Abstract