Alzheimer's disease has a range of risk factors, but one of the clearest connections is the gene apoE4. Now, researchers at the Gladstone Institutes have peered closer at the protein encoded by this gene and uncovered how it affects the brain, how it increases the risks of Alzheimer's and most importantly, how the damage can be reversed.

The apoE gene comes in three variations, apoE2, E3 and E4, and everybody carries two copies in various combinations. The most common form is apoE3, and it doesn't seem to have any influence over a person's likelihood of developing Alzheimer's. But apoE4, present in up to 15 percent of people, is the real troublemaker: Having one copy increases the Alzheimer's risk by two to three times, while those unlucky enough to have two copies are 12 times more likely to develop the disease.

But why is that the case? The proteins created by these genes are extremely similar, with apoE4 differing from apoE3 at only one tiny point. So, the new study set out to examine what problems the former is causing in the brain, and whether that single change can be canceled out.

Rather than using mouse models, the results of which don't usually translate well to human biology, the Gladstone researchers experimented with human cells instead. The team gathered skin cells from Alzheimer's patients with two apoE4 genes, as well as some from people with two apoE3 genes without Alzheimer's. These were converted into induced pluripotent stem cells, and then turned into human neurons.

The team compared the neurons from the apoE3 and apoE4 donors, and found that the latter didn't function as well as they should. This means the protein breaks down into fragments in the cells, over time leading to the build-up of proteins in the brain that forms the calling card of Alzheimer's.

Interestingly, apoE4's devastating effects are clear in humans but not in mice. That illustrates the flaws in using animal models of human diseases, and may go a long way towards explaining why treatments that previously seemed so promising in mice haven't panned out in human trials.

"There's an important species difference in the effect of apoE4 on amyloid beta," says Chengzhong Wang, first author of the study. "Increased amyloid beta production is not seen in mouse neurons and could potentially explain some of the discrepancies between mice and humans regarding drug efficacy. This will be very important information for future drug development."

Having determined that apoE4 damages human brain cells, the team wanted to examine the root of the problem – namely, whether the problems were caused by the presence of apoE4 or perhaps just the absence of apoE3.

"It's fundamentally important to address this question because it changes how you treat the problem," says Yadong Huang, lead author of the study. "If the damage is caused due to the loss of a protein's function, you would want to increase protein levels to supplement those functions. But if the accumulation of a protein leads to a toxic function, you want to lower production of the protein to block its detrimental effect."

To figure that out, the team grew brain cells with no forms of apoE, and found they functioned much the same as those with the common apoE3 protein. As soon as apoE4 was added though, the neurons degraded in a familiar Alzheimer's fashion, indicating that this protein is actively the problem.

Best of all, the researchers were able to fix the damage after the fact, using a class of compounds that turn apoE4 into something closer to E3. Treating the brain cells with these structure-correcting molecules restored function to the neurons, and effectively reversed the signs of Alzheimer's. The scientists are now looking to the pharmaceutical industry to help with improving the compounds for future testing in human patients.

The research was published in the journal Nature Medicine.