Rethinking Alzheimer's: How these tiny balls of fat factor in

Little fat-filled sacs

When, in 1906, Dr. Alois Alzheimer conducted his landmark microscopic inspections of brain tissue from a patient whose dementia would become known as Alzheimer’s disease, he observed not two but three defining aberrations. One was amyloid plaques, gummy blobs situated between brain cells and composed largely of the sticky substance A-beta.

Long thought to play a critical and even causal role in Alzheimer’s disease, amyloid plaques have encountered headwinds as drugs designed to remove them from the brain have failed to relieve or dramatically slow the progression of the condition’s symptoms. It’s virtually certain that Alzheimer’s disease involves A-beta, the amyloid-plaque starter material. But new thinking has it that this involvement may begin earlier in an A-beta molecule’s life, even as it starts assembling into small bundles and well before these stringy clusters coalesce into amyloid plaques.

The second aberration Alzheimer saw was neurofibrillary tangles, filamentous aggregates of a protein called tau that show up inside Alzheimer’s patients’ neurons. Neurofibrillary tangles’ presence is strongly predictive of the disease’s onset and progression. The last decade’s research has highlighted tau’s critical role in maintaining neurons’ shape and function and has shown how certain chemical modifications can speed tau molecules’ aggregation into neurofibrillary tangles.

But Alzheimer noticed a distinctive third feature in his autopsied patient’s brain tissue: tiny oily spheres, not in neurons but within other brain cells called microglia. Unlike amyloid plaques and neurofibrillary tangles, both of which have become objects of intense focus on the part of neuroscientists, not much has been said about these microscopic fatballs.

In part, that’s because when pathologists prepare brain-tissue samples from autopsied or biopsied patients for microscopic examination, the method they use to prepare the samples tends to wash away those so-called lipid droplets. (“Lipid” is chemists’ all-encompassing term for oils and fats.)

“By the time we examine an autopsied brain-tissue sample, a pathologist will have rinsed it with alcohol, removing lipids,” said Tony Wyss-Coray, PhD, the D. H. Chen Professor II, a professor of neurology and director of the Knight Initiative for Brain Resilience at the Wu Tsai Neuroscience Institute. “So, we can miss them.”

Besides, it’s not easy to see individual lipid droplets under a microscope, Wyss-Coray continued. “They’re fragile, hard to isolate.”

Dr. Alzheimer’s microscopic mission missed another culprit that remained unidentified until only a few decades ago. As discussed earlier in this series, about one out of every five people on Earth carry, in their genomes, one or two copies of the most powerful common genetic feature predisposing us to Alzheimer’s disease: a particular version, or variant, of a gene called APOE.

This variant, designated APOE4, is now well known — and known to be potent. Someone whose genome holds two copies of APOE4 (one inherited from the person’s mother and the other from the father) — referred to as an APOE4/4 genotype — is about 10 times as likely to get an Alzheimer’s diagnosis as if that person had instead inherited two copies of APOE3 (APOE3/3). Carrying even a single copy of APOE4 doubles or triples a woman’s Alzheimer’s risk. (For unknown reasons, men with a single APOE4 copy are relatively unaffected.)

A 2024 Nature study by a team Wyss-Coray led provides glimmerings of a unified-field theory of Alzheimer’s disease, revealing clear-cut connections between A-beta, lipid droplets in microglia, the APOE4 gene variant and nerve-cell death.

The brain’s moonlighting immune cells

Microglia, accounting for somewhere around 15% of all the cells in the brain, are its resident immune cells. They’re to the brain roughly what the immune cells called macrophages are to the body’s periphery. In fact, you could say microglia are the brain’s macrophages. They originate in bone marrow just as macrophages do and are genetically quite similar. But instead of hanging around below the neck, they migrate into the brain during early fetal development, before that organ becomes mostly off limits to intruders, including most of our own cells from outside the brain.

Microglia are responsible for keeping the brain free of pathogens, and they’re good at that. Not all that many pathogens get into the healthy brain to begin with, because the brain is encased in a kind of a rind made mostly of tightly knit cells and known as the blood-brain barrier. But woe unto those that do.

“If microglia detect a microbial pathogen, they get in there and gobble it up,” Wyss-Coray said.

In quiet times, microglia moonlight as the brain’s garbage crew, cleaning up waste, scavenging dead cells and debris. They also secrete growth factors that perk up neurons.

Plus, they eat A-beta molecules. You don’t want A-beta molecules to build up — they might bunch up into toxic stringy fibrils. Microglia scoop up A-beta molecules produced in minor amounts by neurons when they’re firing and churned out in bulk after an inflammatory event such as an infection or an injury to brain tissue (a concussion, say, or a stroke).

“But we’re finding that with age, these cells can go overboard, change their metabolism, get stuck in overdrive and become really toxic to the brain,” Wyss-Coray said.

Nano-fatballs arrive on the scene

Wyss-Coray and his colleagues had already discovered lipid droplets in mouse microglia and noticed that old mice’s microglia contain far more of them than young mice’s do. Studies by others had shown that inflammatory molecules like lipopolysaccharide — a material from the outer coats of some kinds of bacteria — can kick mouse microglia’s lipid-droplet production into high gear. A-beta can, too.

“We asked: What about humans?” Wyss-Coray said. “Do our microglia also make lipid droplets?” The answer: They do.

Not only that, but the new study demonstrated a link between people’s APOE genotype and the intensity with which their microglia produce lipid droplets, and showed that the little fat-filled sacs can kill neurons.

ApoE, the protein for which APOE is the recipe, is a bus whose passengers are molecules of fat in transit within and between cells. The researchers wondered, reasonably, whether different versions of ApoE might differentially influence lipid-droplet formation.

They compared autopsied brain tissue from three groups of deceased donors: Alzheimer’s patients with APOE4/4 genotypes, Alzheimer’s patients with APOE3/3 genotypes, and age-matched, Alzheimer’s-free APOE3/3 carriers.

Using fat-attracted red dye, the scientists were able to visualize myriad lipid droplets present in some, although not all, autopsied microglia. Lipid-droplet-rich microglia, which looked just as Alois Alzheimer had described them, were most profuse in APOE4/4 Alzheimer’s patients’ brains and scarcest in those of healthy APOE3/3 carriers.

They also found that the gene encoding an enzyme central to the production of fatty acids — the main building blocks from which most fats and oils are made — was especially activated in microglia from autopsied APOE4/4 Alzheimer’s donors’ brain tissue samples, somewhat less so in APOE3/3 microglia from Alzheimer’s donors, and least of all in Alzheimer’s-free donors’ microglia.

The investigators had access to cognitive test scores of their deceased brain-tissue donors. Those whose brain-tissue samples had the highest concentrations of lipid-droplet-rich microglia turned out to have had the worst performance on cognitive tests they’d taken while alive — and, on post-mortem inspection, the highest levels of amyloid plaques and neurofibrillary tangles.

Springtime for fatballs

Did any of this matter? Did it, for example, affect neurons?

To find out, Wyss-Coray and his associates used sophisticated but now-common methods to generate, in lab dishware, two batches of brand-new human microglia that were identical except for one thing: They carried different APOE gene variants in their genomes. One batch of microglia was APOE3/3, and the other was APOE4/4.

Knowing that A-beta can induce formation of lipid droplets in microglia, the scientists added A-beta fibrils to separate culture dishes, one holding APOE3/3 microglia and the other containing APOE4/4 microglia. There ensued a ho-hum uptick in APOE3/3 microglia’s overall lipid-droplet production, but a big jump in that process by their APOE4/4 counterparts.

Next, the scientists sorted their APOE4/4 microglia into lipid-droplet-rich and lipid-droplet-poor fractions and steeped the two collections in separate dishes full of nutrient broth. After 12 hours, they removed broth from both dishes and poured some from each dish, respectively, into one or the other of two other dishes containing lab-generated human neurons. To a third dish containing equivalent neurons, they added pristine nutrient broth unadulterated by microglial exposure.

In neurons given broth from lipid-droplet-rich microglia, tau molecules became extremely prone to chemical modifications that are known to speed tau’s clumping into neurofibrillary tangles. But this didn’t occur to any appreciable extent in neurons bathed in broth from lipid-droplet-poor microglia — and not at all in neurons basking in straight-from-the-bottle, microglia-unsullied broth.

What’s more, neurons bathed in the broth in which lipid-droplet-rich APO4/4 microglia had sat died in substantial numbers. This fate befell far fewer neurons steeped in lipid-droplet-poor microglia’s bathwater. And there was no effect at all when, instead, the scientists bathed the lab-grown neurons in the broth of A-beta-exposed microglia that had, however, been bioengineered to knock their APOE gene entirely out of working order.

Evidently, APOE is essential to something A-beta-annoyed microglia are spewing that harms neurons. What that something is, no one really knows yet. But scientists including Wyss-Coray want to find out, because this knowledge could open new avenues to treating or preventing the nerve damage that epitomizes Alzheimer’s disease.

Other promising therapeutic candidates might be compounds that put the brakes on microglial lipid-droplet formation without compromising all-important lipid production in all the body’s hundreds of other cell types.

Unified field theory confronts the elephant

Chances are good that tying together A-beta, microglia-manufactured lipid droplets, APOE4, tau warping and neuronal death is not the whole story. For example, there’s the brain-body connection. Macrophages seldom penetrate the blood-brain barrier. But they can send out inflammatory signals that reach the brain.

Macrophages are known to secrete lipid droplets in response to bacterial infections. Those lipid droplets are coated, as are the made-in-microglia variety Wyss-Coray’s study explored, with bacteria-killing substances called cathelicidins.

In Wyss-Coray’s examination of autopsied samples, cathelicidins were more abundant on lipid droplets in microglia from APOE4/4 Alzheimer’s brains than in those of their APOE3/3 counterparts, and least plentiful in APOE3/3 Alzheimer’s-free brains’ lipid-droplet-rich microglia. 

That observation, and microglia’s potent lipid-droplet-forming responsiveness to lipopolysaccharide — a purely bacterial product — might suggest that microglial lipid-droplet production could be an immune-like antibacterial response, with APOE4 providing an extra-potent punch.

Wyss-Coray speculates that lipid-droplets’ presence and actions in microglia may be an anachronistic hangover from those cells’ evolutionary origin as outside-the-brain immune cells. Unfortunately, in an adult brain those lipid droplets are far more likely to wipe out a nearby neuron than an invading bacterium.

“Bacteria seldom get inside the brain of a healthy person,” he said.

“We see, with age, increases in lipopolysaccharide levels in circulation,” Wyss-Coray added, maybe because our guts become leaky after many decades of high living and low fiber intake, allowing some of our commensal bacteria to escape through the intestine’s mucosal lining into the blood — potentially putting macrophages in an inflammatory funk.

Plus, the blood-brain barrier itself gets more porous with age. “Circulating lipopolysaccharide can affect brain vasculature, too, making it leakier,” he said. “We find more of it in Alzheimer’s brains.”

The upshot: It probably wouldn’t hurt to avoid life’s inflammation-inducing insults — faulty (suboptimal) diet, microbial infections and so many more — where possible. And carefully targeted new drugs, based on detailed new knowledge, may help a lot.

Alzheimer’ disease is one multifaceted malady. But a fuzzy picture of the elephant is emerging.