Mutations don’t always cause disease, quite the opposite

A mutation in APP protecting against Alzheimer’s disease. Alzheimer’s disease is one of the leading causes of dementia in the Western world. In rare familial forms of Alzheimer’s disease, variants in the APP gene are well-known to be disease-causing. This led Jonnson and colleagues to search for additional rare variants in the APP gene that might be associated with further cases of Alzheimer’s disease. When they analysed their datasets, they stumbled across an associated APP variant. However, this variant does not increase the risk for Alzheimer’s, it reduces it…

Plaques and APP. Histologically, Alzheimer’s disease is characterized by an accumulation of so-called amyloid plaques in the brain.  Amyloid plaques are protein deposits that can be identified through specific staining methods and that have a characteristic appearance. While the existence of these plaques is long known, the mechanism how these plaques form had largely been unresolved. However, in the last decades, rare familial forms of Alzheimer’s disease have been characterized and the underlying genetic changes have been identified. Most familial forms of Alzheimer’s disease are due to mutations in the APP gene, the gene coding for the amyloid precursor protein. Physiologically, the APP protein is cleaved at several sites. Cleaving the APP protein at the so-called beta-site creates amyloid-beta-proteins, which aggregate into plaques. The cleaving of APP is a finely balanced process, and the APP mutations known to cause familial Alzheimer’s disease result in excessive cleaving at the b-site, resulting in abundant amyloid-b-proteins and subsequently plaques.

In silico genotypes. The Icelandic population is unique in several respects. For one, this population is a founder population with a known and well-documented family history. Therefore, the connection of virtually every Icelander to the founder population and the relatedness between different individuals can be assessed. When combined with systematic medical data that can be assessed, this population provides a powerful tool for genetic studies. The company deCODE genetics is using this knowledge on family structure and available medical data to systematically screen for genes associated with disease. For example, the known family structures in Iceland provides the unique opportunity to infer genotypes in family members without actually genotyping them. These virtual genotypes appear to be highly reliable.

A screen for APP variants with a surprise. In a recent study in Nature, Jonsson and colleagues from deCODE genetics assessed the range of variants in APP in the Icelandic population in 1,800 individuals and followed these variants up in the larger population. They inferred virtual genotypes, demonstrated that these virtual genotypes are reliable and then tested for an association with the identified APP variants with Alzheimer’s disease. What they found for the A673T variant was perplexing – this rare variant was more frequent in controls that in cases. Usually, a rare disease associated variants is more frequent in patients with the disease, but A673T was more frequent in healthy individuals (0.45% vs. 0.13%).

Frequencies of the A673T variants in APP in various patients groups. This variants is least frequent in patients with Alzheimer’s disease and most common in the elderly older than age 85 without evidence with cognitive decline, i.e. this variant protects against Alzheimer’s disease and cognitive decline.

Not only protective, but leading to a longer life. The authors followed their finding up in other samples including elderly individuals without Alzheimer’s disease. Again, this variant was more frequent than in unscreened control individuals, suggesting that it (a) protects against Alzheimer’s and (b) leads to a higher chance of reaching the age of 85 (i.e. in increases the “risk” of reaching age 85 and older).

Not only a longer life, but a longer life without cognitive decline. In the next step, the authors wanted to investigate whether the A673T variant only protects against Alzheimer’s or whether it also is associated with common cognitive decline during again. Not every cognitive decline, i.e. loss of cognitive abilities, is due to Alzheimer’s disease. It is still debated, however, whether common mechanisms are involved. In order to test the role of the A673T variant, the authors then looked at carriers and non-carriers and their cognitive measurements over time. Iceland offers a routine cognitive test for individuals living in nursing homes. The authors could use this data to combine this with their genotypes. Again, the results were stunning. Carriers of the rare A673T variant had better scores than the more common non-carriers. This suggests that this variant does much more than just protect against Alzheimer’s: it protects against a general cognitive decline during aging.

The biology behind it. As the fundamental pathological mechanisms connecting APP and Alzheimer’s disease are known, the authors investigated the effect of the A673T mutation. They found that b-cleavage is drastically reduced. Less beta-cleavage leads to less amyloid beta-protein and fewer plaques. This is in stark contrast to the other APP mutations that enhance beta-cleavage. By connecting this variant with the rate of APP cleavage, the authors demonstrated that this mechanism does not only apply to Alzheimer’s disease, but is also involved in normal cognitive decline.

The lessons for EuroEPINOMICS. There are at least two important messages that this paper send beyond the field of dementia genetics. First, genetic variants can be protective. While often hypothesized -including some of our own speculations that the 15q13.3 microduplication might project against IGE-, the paper by Jonsson and colleagues is one of the few papers to show the existence of such variants convincingly. Secondly, this paper was only possible due to the unique infrastructure of deCODE. Without the possibility to trace back genotypes and connect these genotypes with clinical data, this work could not have been done. While epilepsy genetics lacks many of those resources, we have tried to generate facilities such as the BENCH database for central clinical data repository prior to genetic analysis within EuroEPINOMICS. While there will always be many possibilities for “hunting epilepsy genes” on a smaller scale, we should keep in mind that identifying meaningful genetic risk factors will require large projects with a broad collaborative network and common infrastructure.

8 thoughts on “Mutations don’t always cause disease, quite the opposite

  1. Pingback: Endophenotypes for Alzheimer’s disease – GWAS on CSF tau levels | Beyond the Ion Channel

  2. Pingback: Transmission of rare variants in parent-offspring trios – power or no power? | Beyond the Ion Channel

  3. Pingback: Guilt by association: SCN1A in Temporal Lobe Epilepsy | Beyond the Ion Channel

  4. Pingback: The genetics of treatment response in newly diagnosed epilepsy | Beyond the Ion Channel

  5. Pingback: 2B or not 2B – mutations in GRIN2B and Infantile Spasms | Beyond the Ion Channel

  6. Pingback: Modifier genes in Dravet Syndrome: where to look and how to find them | Beyond the Ion Channel

  7. Pingback: Mining GWAS mountains for missing heritability | Beyond the Ion Channel

  8. Pingback: The age of mega-genomics, type 2 diabetes, and protective variants in SLC30A8 | Beyond the Ion Channel

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s