Rethinking Alzheimer's: Untangling the sticky truth about tau

In just the past couple of years, pharmaceutical companies have slogged through lengthy, costly and sometimes futile rounds of clinical trials, finally gaining approval by the Food and Drug Administration for a handful of “plaque attack” therapies — slow biweekly or monthly infusions that work, to the extent that they do work, by pulling amyloid plaques out of Alzheimer’s patients’ brains.

Despite these recent approvals, the new plaque-attack drugs have proved underwhelming and controversial. Yes, they remove amyloid plaques as promised. But any clinical improvement, if there is one, is barely observable to a trained neurologist.  Plus, these drugs carry a real risk of serious side effects.

That swings the door open for alternative ways of countering Alzheimer’s disease. One promising approach targets an entirely different type of brain clump: neurofibrillary tangles — elongated, twisted filaments that are every bit as essential to an Alzheimer’s diagnosis as amyloid plaques, perhaps even more so.

Stanford Medicine researcher Irina Skylar-Scott, MD, clinical assistant professor of memory disorders, is conducting a clinical trial to see if a drug that impedes the formation of neurofibrillary tangles in Alzheimer’s patients’ brains slows the disease’s progression. Another investigator, Emmanuel Mignot, MD, PhD, professor of psychiatry and behavioral sciences, is exploring the intriguing possibility that a sizeable chunk of the world’s population could someday benefit from a vaccine that achieves a similar result, providing a layer of protection from the ravages of Alzheimer’s.

The second hallmark

Early in the 20th century, Alois Alzheimer, amid his pathbreaking neuroanatomical explorations, described both amyloid plaques and neurofibrillary tangles as defining features of the disease that took his name.

Whereas amyloid plaques pile up between nerve cells, neurofibrillary tangles precipitate inside them — and they’re thought to be directly lethal to the nerve cells in which they build up.

The number of neurofibrillary tangles, not the number of amyloid plaques, in a person’s brain closely tracks with that person’s degree of cognitive impairment, according to Skylar-Scott.

“Some people’s brains have plenty of amyloid plaques but seem to work just fine,” she said. “Where you see amyloid plaques in the brain often aren’t the parts of the brain that start seriously misfiring and failing.”

Neurofibrillary tangles turn up in precisely those places, and their presence there always spells trouble. Where they form, nerve cells die.

Neurofibrillary tangles are composed largely of a single protein named tau (it rhymes with “cow”). Like A-beta, tau is clearly there for a reason. Unlike A-beta, we pretty much know what that reason is.

Tau’s day job

Every human cell is a gel.

Far from mere water-filled bags with a bunch of random stuff floating around inside, our cells are models of organization. Their contours are precisely sculpted by their cytoskeletons, dynamic three-dimensional grids along which molecular express trains speed in all directions, carrying cargo — from wherever in a cell it’s been made, modified or stored — to wherever it’s needed.

Each cell’s cytoskeleton is composed of virtually identical Lego-like building blocks that hook together to form hollow tubes called microtubules. A microtubule grows longer when new units are patched to either of its ends. When end units fall off, the microtubule shrinks.

Tau’s job is to regulate the construction, maintenance and remodeling of the cytoskeleton by stabilizing microtubules so they don’t fall apart, or by helping them grow or shrink, as the situation demands. It does this not by just sitting there, but by jumping onto a microtubule, hanging for a while, hopping off, then jumping on again. That way, it also avoids collisions with the cargo-carrying “trains” shooshing along the cytoskeleton’s microtubule tracks.

Nerve cells in particular depend on tau’s cytoskeleton-stabilizing action because of their characteristic long, winding, branching extensions along which impulses travel toward other, often-distant nerve cells.

Many proteins, including tau, undergo chemical modifications during their lifetimes, thanks to enzymes in the cell stapling small chemical caps — pieces of small molecules that can be attached to larger ones — onto them. Certain kinds of chemical caps can cause tau molecules to start bunching up into clusters; clumps; and, finally, neurofibrillary tangles, damaging the very nerve cells tau was meant to protect.

What to do with too much tau

With all the hoopla surrounding amyloid plaques as the Alzheimer’s drug target of choice, comparatively few clinical trials have focused on tau as an alternative target. Skylar-Scott is the principal investigator of Stanford Medicine’s site, one of 138 around the country and throughout the world. The phase 2 trial, sponsored by Biogen, is designed to see if preventing tau from accumulating in the brains of people with early-stage Alzheimer’s disease will slow, halt or reverse the buildup of neurofibrillary tangles and, it’s hoped, counter symptom progression.

The trial has enrolled more than 300 50- to 80-year-old patients diagnosed with early-stage Alzheimer’s disease or mild cognitive impairment, a subtler condition that precedes Alzheimer’s. Every 12 or 24 weeks for about 18 months, either a placebo or a biomolecule called an antisense nucleotide is injected directly into the fluid surrounding their spinal cords.

The antisense nucleotide is designed to impede the production of new tau molecules, but to have no effect on tau molecules already in existence.

“Reducing the new supplies of tau may make it a little scarcer and less likely to bunch up,” Skylar-Scott said.

Tau’s sticky patch

Meanwhile, Mignot has discovered that people whose genomes carry a genetic variant called Human Leukocyte Antigen DR4, or just DR4, are less likely to develop Alzheimer’s than those without this variant. Mignot theorizes that DR4 protects people from Alzheimer’s by interrupting tau’s aggregation into neurofibrillary tangles.

In a study published in August 2023 in the Proceedings of the National Academy of Sciences, Mignot and his associates examined genetic and health data collected from, among others, 100,000 people with Alzheimer’s, more than 40,000 with Parkinson’s disease and more than 400,000 with neither disorder. The study showed that those carrying DR4 — about 30% of people of European ancestry, and upward of 20% of Earth’s overall population — were 10% less likely, on average, to develop either condition than those who didn’t carry DR4.

“DR4 is one of hundreds of versions of a single gene residing in the No. 1 most-variable stretch of gene sequences in our genome, by far,” said Mignot, the Craig Reynolds Professor in Sleep Medicine and director of the Stanford Center for Narcolepsy. “Each version of each gene in this multigene region codes for a slightly different molecule.”

That adds up to an immense assortment of differently shaped molecules, with resultingly different biochemical affinities that qualify them for service as ultra-specialized jewel cases. Collectively, they can bind to miniscule snippets, or peptides, of virtually any protein a cell carries — from the ones it normally makes and should make (to do its job) to the oddball ones it makes but shouldn’t (due to cancer or viral infection) — and display those peptides on that cell’s outermost surface in a way that makes all those myriad peptides highly visible to roving inspector cells of the immune system.

This is absolutely critical to the immune system’s ability to recognize diseased or abnormal cells and zap them before the bug that’s crawled inside of them, or before their potential to turn into tumors, zaps us.

In one of a series of follow-on experiments, Mignot and his co-authors looked at autopsied Alzheimer’s patients’ brains in which the number of neurofibrillary tangles and number of plaques had been counted.

“DR4 carriers, by and large, had fewer neurofibrillary tangles in their brains than non-DR4 carriers, as well as an increased age of onset,” Mignot said.

Mignot and his colleagues wondered if DR4 might latch onto some piece of tau that’s crucial to clumping, blocking tau’s aggregation. They were especially curious about a short segment of the tau molecule, called PHF6, that’s known to be highly prone to pairing up with corresponding PHF6 segments on other tau molecules and to be critical to neurofibrillary-tangle formation.

Some chemical modifications make PHF6 particularly sticky, Mignot noted. For example, PHF6 becomes hundreds of times more aggregation-prone when it’s been chemically modified by the addition of a kind of chemical cap called an acetyl group (a close relative of acetic acid, the pungent ingredient in vinegar).

“Acetylated PHF6 is the most common post-production chemical modification of tau you find in Alzheimer’s patients’ neurofibrillary tangles,” Mignot said.

In a lab experiment, Mignot and his teammates chopped the entire tau protein into its constituent peptides, tossed each of them into separate dishes, and added still more dishes containing tau peptides wearing one or another of the various chemical caps that might normally accrue inside a living cell. Then they dropped DR4 into each of the 448 resulting dishes to see if it bound to any tau peptide (whether sporting a chemical cap or not) in a way that could prevent tau’s aggregation.

“Of all those peptides, the only one that bound strongly to DR4 was PHF6,” Mignot said, “and only when that peptide was acetylated.”

That makes acetylated PHF6 a prime suspect in tau’s aggregation into neurofibrillary tangles.

Autoimmunity with a minus sign

Mignot thinks possession of DR4 acts as a kind of “benign autoimmune” disease by optimizing the display of acetylated PHF6, priming the immune system to attack and destroy any such complex it comes across, and preventing the aggregation of individual tau molecules into clumps inclined to become neurofibrillary tangles.

“It’s like autoimmunity with a minus sign,” he said. “It’s like having an autoimmune disease that is good for you.”

Mignot seeks to amplify the strength of lucky DR4-carriers’ resistance to Alzheimer’s disease, and he thinks he may know how to do it: by further sensitizing the immune system to the DR4/acetylated-PHF6 junction so it attacks this interface faster and more furiously.

“Vaccination with acetylated PHF6 or a close look-alike might at least slow disease progression,” Mignot said. (This would work only for people carrying the DR4 variant, he observed. But 20% to 30% of all the people on our planet still amounts to a lot of people.)

“A simple blood test can identify which people have DR4 and could benefit from a vaccine — and probably are already benefiting modestly from their having DR4 in their genomes,” Mignot said.

It so happens that there’s an existing bioengineered mouse lineage that carries a human version of the DR4 variant. Mignot plans to conduct an experiment on such mice that are also inclined to tau hyper-aggregation, as occurs in Alzheimer’s disease. The idea is to vaccinate these mice with acetylated PHF6 and see if it elicits an immune response that can mitigate the neurofibrillary-tangle pileup.

“If it works, it could be a very natural way to delay the progression of the disease,” Mignot said. “It will probably have very low side-effect risk, because it’s super simple.”

Mignot thinks DR4’s interaction with the tau molecule may play a protective role not only in Alzheimer’s but also in Parkinson’s and maybe even some other neurodegenerative disorders. If that turns out to be true, there could be a multipurpose anti-neurodegenerative vaccine in our future.