Why is a common gene variant bad for your brain?

Alzheimer’s disease’s suspected causes are diverse, and its cures are, today, nonexistent. What’s all but certain is that many who have the mental chops to wade through a detailed article about the disorder’s drivers and demographics will nevertheless succumb to it someday.

With no cure available, despite numerous attempts to find one, researchers are looking down new roads for treatments. A recent discovery by Stanford Medicine neurologist Mike Greicius, MD, may help clear one of those roads for faster passage.

Gummy clumps, plaque-attack drugs and luck
of the genetic draw

While its most visible outward symptoms include memory loss and confusion, a key defining feature of Alzheimer’s at the molecular level is the overabundance in patients’ brains of a substance called A-beta, which aggregates into gummy clumps, collectively called amyloid plaque, situated between their nerve cells. These plaques begin showing up in the brain years before mental decline becomes noticeable.

So, hopes were high for a class of new drugs based on the idea that amyloid plaque is the smoking-gun cause — or at least one cause — of the slow, but steady, crumbling of memory that’s one of Alzheimer’s behavioral hallmarks.

But simply removing amyloid deposits, or plaque, from the brains of people with Alzheimer’s disease hasn’t been the game changer some thought it would be, leaving more than 6 million people with this condition and their caretakers and physicians looking for alternative treatments.

The recent failure of a slew of “plaque attack” drugs to provide clinically significant improvements in Alzheimer’s patients’ condition puts the spotlight on scientists who’ve been thinking outside the amyloid-plaque box. One of them is Greicius, who recently spearheaded a genetics study described in a paper published in January in Neuron. The study focuses on variants of a gene called APOE and ventures into the realm of personalized medicine: A drug that works for someone carrying one variant may not necessarily be effective against people carrying other variants.

At least one-fifth of all people on the planet are carrying a gene variant that predisposes them to Alzheimer’s. Known as APOE4, it’s one of four versions of a gene called APOE. Which APOE version you’re carrying makes a big difference in your Alzheimer’s risk.

Most people whose genome includes a copy of APOE4 don’t wind up with an Alzheimer’s diagnosis. But people with a single copy are at double or triple the risk for Alzheimer’s compared with people who have two copies of the most common variant, APOE3. Those with two copies of APOE4 (one inherited maternally, the other paternally) develop Alzheimer’s at more like 10 times the frequency that people with two APOE3 copies do.

“About 25% of people of European ancestry are APOE4 carriers, but this variant is present in 50% to 60% of Alzheimer’s patients with European ancestry,” said Greicius, the Iqbal Farrukh and Asad Jamal Professor and a professor of neurology and neurological sciences.

(A third, less-common variant, APOE2, is actually protective in comparison with APOE3. The fourth, APOE1, is so rare that fewer than 10 people carrying it have ever been identified.)

Of the people who develop Alzheimer’s disease, the ones with an APOE4 copy tend to start showing symptoms soonest — about five to 10 years earlier, on average, than those with two APOE3 copies.

“APOE4 starts the ball rolling well before it would normally start,” Greicius said.

The unwanted connection

The APOE4 variant was first recognized in the 1970s as a risk factor in cardiovascular disease. In the early 1990s, studies directed by a Duke University neuroscientist, the late Allen Roses, PhD, showed that APOE4 also increased Alzheimer’s risk. At the time, researchers were mainly laser-focused on amyloid plaque and A-beta — the protein snippet that aggregates to form these brain deposits — and were skeptical about any APOE4 connection to Alzheimer’s. But now it’s written in stone.

Yet, three decades later, nobody really understands why APOE variants differentially affect Alzheimer’s risk. It’s not even clear what the gene’s protein product (designated “ApoE”) does in the first place. Be that as it may, genes similar to APOE have been identified in all animals from amoebas to mammals, so you know it must be doing something important — it may be doing different things in our brains than what it’s doing elsewhere in our bodies.

It is known that ApoE shuttles various fatty substances within and between cells, both inside and outside the brain, like passengers on a bus. And it’s suspected to be involved in our immune response to infections, as some of the fatty acids it shuttles have antimicrobial properties.

That antimicrobial capacity, if it’s for real, could help explain an intriguing ethnic distribution of APOE variants, whose prevalence and harmfulness seem to follow opposing geographic gradients.

Your likelihood of carrying APOE4 depends, in part, on where your ancestors came from. At least one copy of APOE4 in one’s genome shows up in roughly 1 in every 3 people of African descent, for example; about 1 in 4 people of European descent; and a scant 1 in 10 (or even only 1 in 20, according to some research) Japanese people.

But APOE4 risk runs in the opposite direction. Among those of African descent, carrying a single APOE4 copy is barely observable as an Alzheimer’s risk factor. For someone of European descent, having a single copy of APOE4 in one’s genome translates to two to three times the risk of having two APOE3 copies. And Japanese people with a single copy of APOE4 are at five times the risk for Alzheimer’s disease as their compatriots with two APOE3 copies. Having two APOE4 copies in your genome always boosts your risk, but much more so if you’re Japanese, less so if you’re of African ancestry.

… Simply removing amyloid deposits, or plaque, from the brains of people with Alzheimer’s disease hasn’t been the game changer some thought it would be.

APOE4’s combined higher frequency but lower risk among people whose recent ancestors inhabited Africa — the continent where humans originated — suggests to biologists that APOE4 was the first APOE variant carried in humans. Some theorize that its initial importance was in combating infectious microbes, which abound in warmer climates. As humans migrated out of Africa to or through colder climes with less microbial exposure, the theory goes, other variants — first the now-dominant APOE3 and, later, the protective APOE2 — came along and, over time, became more common.

APOE4’s power to boost the likelihood of Alzheimer’s disease varies not only by ancestry but also by sex. Women of European descent between age 50 and 80 who carry one APOE4 copy and one APOE3 copy are at three or four times as much risk as those with two copies of APOE3, while same-age men with the same APOE status are at only marginally increased risk, according to a review Greicius co-authored in Neuron in 2019.

A no-brainer confronts a brain-teaser

Scientists agree that APOE4 is “bad” in the sense of hiking people’s risk for cognitive decline in advanced age. But whether that’s because ApoE4 — the protein for which APOE4 is a recipe — is an underachiever (not doing enough of some good thing it’s supposed to be doing in the brain) or because ApoE4 itself is a bad actor (doing some bad thing it’s not supposed to be doing there) is an open question.

Knowing the answer would tell researchers and drug developers whether their goal should be to punch it up or to tone it down — a key step toward finding a drug to deal with it.

That’s what Greicius and his colleagues, including University of Washington professor of medicine Chang-En Yu, PhD, who was Greicius’s co-senior author, set out to determine.

For their study, they gained access to a giant registry of people with and without Alzheimer’s whose genes had been carefully scrutinized for APOE status, then they zeroed in on people age 65 and older. Of the 56,684 people in this cohort, a fair number were APOE4 carriers — no surprises there — but precisely two carried an APOE4 copy that was so defective it couldn’t direct the production of its correspondingly malfunctioning ApoE protein.

“I was shocked to learn that the 90-year-old, on postmortem inspection, had no appreciable buildup of beta-amyloid plaque in his brain.”

Michael Greicius, MD, neurologist and a professor of neurology and neurological sciences

Those two people turned out to be carrying, along with a nonworking copy of APOE4, a perfectly normal APOE3 copy. Neither of them, despite their advanced years (one was 90 at the age of death, the other 79 and still alive at the time), had evidenced any signs of mental decline. To the contrary.

“They were in great shape,” Greicius said. “I was shocked to learn that the 90-year-old, on postmortem inspection, had no appreciable buildup of beta-amyloid plaque in his brain.”

The cerebrospinal fluid of the younger of the two was likewise devoid of any significant A-beta changes when last checked at age 76. (By age 75, two-thirds of even asymptomatic APOE3/APOE4 carriers — much less the ones diagnosed with cognitive symptoms of Alzheimer’s disease — typically have abnormal A-beta levels in their cerebrospinal fluid.)

Evidently APOE4 wasn’t simply too wimpy to get the job done; it was actually bad news. If you’re carrying APOE4, it seems, you’re better off if this gene variant isn’t making any ApoE4 than if it is.

“This is the first human study to make a strong case that ApoE4 is toxic and that its loss may be protective,” Greicius said.

He noted that a complete absence of ApoE activity could be damaging in peripheral organs such as the heart. “Rare cases have been found of people with zero functioning copies of any variant,” Greicius said. They had very high cholesterol levels, he said.

But neither of these two broken-APOE4-carrying individuals, each of whom carried a working copy of APOE3, had sky-high cholesterol. “Apparently, one copy of out-of-order APOE4 doesn’t hurt you,” Greicius said.

The road ahead

So far, no great small molecules that could be used as drugs have been shown to safely and selectively inhibit the production or activity of the problem protein, ApoE4. Finding such a finely discriminating drug could prove daunting.

But in the near term, it may not be necessary. A drug that knocks APOE down but not out, so ApoE production isn’t entirely stamped out, might be safe. The robust health of the people in the new study who had only a single working copy of an APOE gene implies that, Greicius said.

Nor would a drug’s inability to distinguish between different APOE variants pose a problem for treating those carrying two copies of APOE4 (2-3% of all people), he observed.

The new research has begun to resolve scientists’ uncertainty as to whether to put more muscle into ApoE4 or put it out of commission. That should give some direction to drug-development efforts, said Greicius, who is following up in a collaboration with other Stanford Medicine scientists to learn more about ApoE’s interactions with other key fat-shuttling proteins and tease out differences in how ApoE4 and its alternatively numbered counterparts select which fatty substances they take aboard.

“Now, we know which way to go,” he said. 

ApoE4 is not a wimp. It’s a cutthroat. Get rid of it.