The apoε4 gene variant, carried by about 25 percent of the population, is found in 65-80 percent of people with Alzheimer's disease, which suggests that the protein encoded by the gene (called apoE4 ) somehow promotes the build-up in the brain of amyloid beta, the apparent trigger of the disease.
In addition, the apoE4 protein produces worse outcomes in all sorts of insults to the brain including multiple sclerosis, Parkinson's disease, traumatic brain injury, stroke, Lewy body dementia, epilepsy, and diabetes. Apoε4 carriers, even in their youth, display lower glucose metabolism in brain areas known to be particularly vulnerable to Alzheimer's pathology, as well as ominous declines in connectivity in the default mode network, which also is affected by the disease.
Such differences in apoε4 carriers have even been detected in children. A paper in the January JAMA Neurology describes brain changes that may provide a foothold for Alzheimer's in apoε4 carriers are present in children between 2 and 25 months old.
To try to mitigate such negative effects, researchers at the Gladstone Institutes have been developing a "structure corrector" that would transform the apoE4 protein into the shape of the protein encoded by the apoε3 gene, which does not increase the risk of Alzheimer's. (apoE2, the third human variation of the protein, appears to confer modest protection against Alzheimer's.)
The structure corrector, tentatively named PY-101, appears to work in animal models, and could begin human clinical trials within two years, says Robert Mahley, president emeritus of the Gladstone Institutes, and a professor of pathology and medicine at the University of California, San Francisco.
"The concept has been around for maybe 10 years," said Mahley, who has spent decades studying apoE's role in regulating cholesterol. "A very minor change in the apoE4 protein-a single amino acid out of 299-results in a significant structural defect that leads to functional problems."
Correcting that defect appears to eliminate the toxic effects of apoE4-at least in mice.
To explain how this structure corrector works, Mahley uses the human hand to represent the two "domains" of the apoE protein. The palm and four fingers held tightly together represent the largest domain, while the extended thumb, connected by a thin bridge, represents the other. In apoE2 and E3, the thumb remains comfortably apart from the larger protein domain, but in apoE4, a difference of one amino acid creates a bonding site that encourages the thumb to attach to the hand. This causes the protein to misfold. In response, the cell breaks down the protein for recycling through a process known as proteolysis, but then has difficulty ejecting the resulting fragments, which float in the cell's cytoplasm and damage the mitochondria, the structures that produce the cell's energy. Eventually, the mitochondria die.
The fragments also contribute to the disruption of the "rails" within the axons that carry vital molecules from the cell body to the synapse, which releases the neurotransmitters that propagate signals from one neuron to another. This process causes the rails to break apart and form the fibrillary tangles that constitute a hallmark of Alzheimer's disease.
This may be what makes apoε4 carriers so much more vulnerable to insults to the brain. The apoE protein contributes to the repair of damaged neurons by transporting fat molecules such as cholesterol to the site of damage. When people who carry one or two alleles of the apoε4 gene experience brain injury, the resulting increase in apoE4 production within neurons also increases the accumulation of debris, which causes its own type of damage to the neurons.
As Mahley and his longtime colleague, Yadong Huang, wrote in a 2012 paper in the Journal of Medicinal Chemistry, the structure corrector PY-101 would counteract this process by disrupting the chemical bond that causes the "thumb" to stick to the rest of the "hand."
"We can change structure of E4 to make it E3-like, and in so doing we correct the detrimental effects apoE4 has both in vitro and in vivo in mouse models," Mahley said.
PY-101 would almost certainly be delivered in a pill. "That's the only really efficient way to treat large numbers of patients with a complex disease," Mahley said. "We've already proven that PY-101 gets into the brain, so we're trying to develop a formulation that's orally available. We're working that out in our mice right now."
If PY-101 succeeds in mitigating Alzheimer's in people who carry one or two apoε4 alleles it would add a new dimension to the amyloid cascade hypothesis, which contends that the production of amyloid beta fragments in the brain somehow triggers a chain reaction of dysfunction that leads to memory problems and the death of neurons. While not disputing that amyloid beta, also known as Abeta, contributes to Alzheimer's, Mahley thinks it is not the entire explanation. The hypothesis "tries to force everything into one pathway where Abeta is the cause, and I do not believe that is the case," he said. "It could be part of the pathology, but I do not believe it's the only pathway. Otherwise, why aren't we getting some encouragement from all these clinical trials for drugs that lower Abeta?"
The amyloid cascade hypothesis, while widely embraced by researchers, tells a story that leaves several loose ends. For example, apoε4 carriers with mild cognitive impairment-a precursor to Alzheimer's-tend to have a smaller hippocampus and lower amounts of gray matter in certain brain regions even when their levels of amyloid beta are comparable to others who show less tissue loss.
Also, the hypothesis implies that amyloid beta somehow produces the other hallmark of Alzheimer's-the fibrillary tangles that result when the transport "rails" within axons deteriorate. However, the tangles bear a far stronger correlation to the actual symptoms of the disease. Autopsies show that many cognitively normal people die with brains filled with amyloid beta, while others with florid symptoms of Alzheimer's carry a relatively light burden of amyloid beta, but numerous fibrillary tangles within their brain cells.
So Mahley suspects that apoE4 strains the brains of people who carry the apoε4 gene. Then, a "second hit" such as aging, ischemia, trauma, inflammation, or the presence of amyloid beta, produces a cascade of neurodegeneration.
"ApoE sets the stage," he said during a presentation at the Society for Neuroscience annual meeting in San Diego in 2013. "A second hit determines the pathology."
Other researchers, while standing by the amyloid cascade hypothesis, agree with Mahley's contention that apoE4 might cause problems of its own in the brain.
William Jagust at the University of California, Berkeley, for example-a veteran researcher of aging and dementia-remains persuaded by the evidence linking amyloid beta to Alzheimer's pathology, but he also recognizes that apoE4 has effects that are unrelated to Alzheimer's pathology. So he finds the idea of a structure corrector that could mitigate the toxic effects of apoE4 to be very interesting.
"Does this therapy have the potential to affect Alzheimer's disease? Sure it does," Jagust said. "Those invested in the amyloid hypothesis might be a little skeptical, but I don't think (Mahley's) ideas necessarily exclude the amyloid hypothesis. I think you can accept both approaches."
Mahley recently received a $2.5 million "Seeding Drug Discovery Award" from Wellcome Trust so he and his colleagues can continue to study apoE4. They have formed a partnership with Numerate, Inc., a computational drug design firm near San Francisco, to develop a structure corrector for apoE4, and he believes this will expand the search for treatments for Alzheimer's disease.
"For 20 years the focus has been on Abeta, which ignores the fact that apoE4 alone, unrelated to Abeta, can have detrimental effects that cause neurodegeneration," he said. "I'm not saying the amyloid cascade hypothesis is totally wrong. This is a complex disease, and every complex disease has multiple pathways leading to pathology."