The strongest genetic determinant of a person's risk for late-onset Alzheimer's is the APOE gene variant. Just like with other genes, every person inherits two copies of the APOE gene. There are three common variants of APOE, each carrying different levels of risk for developing Alzheimer's disease.

The most common variant, called APOE3, neither increases nor decreases one's chance for Alzheimer's. APOE2 is protective, and APOE4 carries an elevated risk for disease development. About 25% of people with European ancestry have one copy of APOE4, which more than doubles their chances of developing late-onset Alzheimer's. Another 2% to 3% of people have two copies of the variant, which renders them 8 to 10 times more likely to get the disease. 

That’s what makes the discovery of this mutation so exciting. Although the R251G mutation is rare, it’s co-inherited with the high-risk APOE4 variant, meaning that every person with a copy of the R251G mutation also has a copy of the APOE4 variant. But unlike most people with APOE4, these individuals have no increased risk of developing Alzheimer’s: The single amino acid change caused by R251G neutralizes the risk normally caused by APOE4.

“Identifying genetic variants counterbalancing the risk of APOE4 may shed new light on its role in Alzheimer’s disease development,” said Stanford neurology researcher Yann Le Guen, PhD, who shares lead authorship of the paper with Stanford neurology researcher Michael Belloy, PhD. “This finding could help develop new drugs mimicking the effect of the protective genetic variant to reduce the risk of disease.”

How APOE4 raises risk: a persistent mystery

If scientists could figure out the molecular mechanism by which APOE gene variants make people more or less susceptible to Alzheimer’s, they could develop therapies to prevent the disease, Greicius said. “We’re coming up with approaches now that are pretty good at knocking down a gene that we’re sure is bad.”

But in the case of Alzheimer’s, scientists don’t know whether the APOE4 variant increases risk by blocking normal activity of the APOE protein (a “loss of function” mutation) or by making APOE do things it doesn’t normally do (a “gain of function” mutation). So it’s not as simple as knocking out a “bad gene.”

“We’re essentially a full 30 years after those first studies linked APOE4 to Alzheimer’s disease,” Greicius said, “yet this basic question remains open.”

Determining exactly how APOE4 raises risk is tricky because APOE has so many different functions. In addition to raising the chances of Alzheimer’s, the APOE4 variant increases a person’s risk of heart disease and stroke, which can also contribute to cognitive decline.

“But because the effect of the R251G mutation is so potent,” Greicius said, “and because it’s co-inherited with the risk-increasing APOE4 variant, we can really start to decipher the critical features that convey this protection.” For instance, he said, you could activate the protective R251G mutation in cells growing in a lab dish and look at how the mutation changes the way APOE binds to receptors and transports fats.

Mining big datasets

In addition to R251G, the researchers reported a second protective mutation, called V236E, co-inherited with the APOE3 variant, that had been shown to reduce Alzheimer’s risk in a smaller study. In the current study, the V236E mutation decreased Alzheimer’s risk by about 60%, offering a similar level of protection as the protective APOE2 variant.

Because both the R251G and V236E mutations are quite rare, scientists were able to identify them using only very large datasets, obtained through collaboration with researchers around the world. Researchers analyzed more than 67,000 Alzheimer’s disease cases, 28,000 proxy cases (meaning individuals who have a first-degree relative with Alzheimer’s) and 340,000 healthy controls in this study.

Until the past few years, Le Guen said, the available libraries of genetic data from patients with Alzheimer’s did not include rare variants like R251G or V236E, because the existing technology relied on sampling relatively few variants, then inferring the rest. More recently, Le Guen said, new, affordable tools that look at a person’s entire genetic code have emerged, enabling scientists to scour genetic data of large and diverse cohorts of individuals with and without Alzheimer’s.