Illustration of myelinated neuron axons in light blue, with spindly blue and purple cells interspersed among them.
Illustration of myelinated neuron axons in light blue, with spindly blue and purple cells interspersed among them.

A Gene Variant Linked to Alzheimer’s May Disrupt Myelin Production

The APOE4 variant causes cholesterol buildup in the cells that make protective fatty sheaths for neurons, possibly helping explain its role in neurodegeneration.

A black and white headshot of Andrew Carstens
Andy Carstens

Andy Carstens is a current contributor and past intern at The Scientist. He has a bachelor’s degree in chemical engineering from the Georgia Institute of Technology and a master’s in science writing from Johns Hopkins University. Andy’s work has also appeared in Audubon, Slate, Them, and Aidsmap.

View full profile.

Learn about our editorial policies.

ABOVE: Artist’s rendition of myelinated neurons and other brain cells © ISTOCK.COM, SELVANEGRA

Among people who fall prey to Alzheimer’s disease, between 40 and 65 percent have a gene variant called APOE4. Scientists have long known the variant significantly increases the risk of developing the neurodegenerative disease, especially when person inherits copies from both parents, but they’re still learning about molecular mechanisms that explain APOE4’s role.

A study published on November 16 in Nature suggests that one way APOE4 may contribute to Alzheimer’s disease is by causing cholesterol to accumulate inside oligodendrocytes, reducing the cells’ capacity to perform their main function: making fatty myelin sheaths that protect neurons and help them convey signals. The research further finds that, in mice with APOE4, clearing the clogged cholesterol—myelin’s main ingredient—and allowing the substance to cross the cells’ membranes partially restored myelin production and improved cognition, suggesting a therapeutic target to combat Alzheimer’s disease.

“They're showing that one of the things that APOE does, which hasn't been characterized so much before, is that it appears to influence lipid metabolism in oligodendrocytes,” says David Holtzman, a neurologist at Washington University in St. Louis, who was not involved in the work. Holtzman says that APOE4 has been shown by others to affect lipids and cholesterol in a number of different cell types, especially microglia and astrocytes, “but no one to my knowledge has really looked so specifically at the oligodendrocytes.”

See “Gut Molecule Linked to Decreased Myelination in Mouse Brains

scientific figures showing oligodendrocyte precursor cells producing myelin for neurons in vitro under various experimental conditions
Oligodendrocyte precursors cocultured with neurons in vitro, with myelin basic protein in green. APOE3 cells (left) shows signs of the most myelin, and APOE4 cells (center) show signs of reduced myelin. Treating APOE4 cells with cyclodextrin appears to promote myelin formation (right).

“Many biological pathways are perturbed by APOE4,” study coauthor Li-Huei Tsai, a neuroscientist at MIT’s Picower Institute for Learning and Memory, tells The Scientist. Because the APOE gene encodes a protein responsible for transporting lipids between cells, Tsai says that the researchers decided to focus on how the variant disrupts the metabolism of cholesterol and other lipids.

Using transcriptomic profiling, the researchers compared postmortem human brain tissue from people with and without Alzheimer’s disease and measured the expression level of various genes among those who had zero, one, or two copies of the APOE4 variant to reveal that“ APOE4 has a huge impact on gene expression,” says Tsai, especially on many genes that influence lipids. Conducting similar profiling of oligodendrocytes, she says, showed that genes associated with making cholesterol were significantly upregulated in the cells, while genes associated with myelin production were downregulated.

“What we saw from the gene expression data was counterintuitive,” says study coauthor Manolis Kellis, a computational biologist at MIT’s Computer Science and Artificial Intelligence Laboratory, adding that because the groups of genes weren’t regulated in the same direction, there appeared to be a mismatch of signals.

See “Inadequate Myelination of Neurons Tied to Autism: Study

Microscopy images showing the degree of myelination around neural axons under different experimental conditions
Transmission electron microscopy images of APOE4 mouse brain tissue, without (above) and with (below) the cyclodextrin treatment. The darker borders indicate thicker myelin sheaths around the axons.

Digging deeper into the mystery, the researchers conducted lipid analyses in brain tissue from people with at least one copy of APOE4 or APOE3, the most common APOE variant, which is not associated with increased risk of Alzheimer’s. The results indicated that APOE4 brains had higher levels of cholesteryl esters, which Tsai says are insoluble forms of cholesterol known to accumulate within cells. When the researchers stained brain tissues from humans and mice homozygous for APOE4, they found that cholesterol droplets accumulated inside the oligodendrocytes of both. Meanwhile, cholesterol was found outside the cells more prevalently in samples with APOE3.

The researchers began to suspect that the APOE4 variant hindered cholesterol transport across oligodendrocyte membranes. Imagining that oligodendrocytes function like myelin printers, the emerging hypothesis was that the cells had plenty of ink (cholesterol), but “the ink is getting stuck inside the cartridge,” says Kellis.

To verify APOE4’s association with reduced myelin production, the researchers conducted a variety of experiments. First, they induced pluripotent stem cells to form oligodendroglia, the precursors of oligodendrocytes, then they cocultured them with lab-induced neurons. Typically, these cocultured oligodendroglia begin to form myelin within several weeks, says Tsai. The researchers noted this effect in the APOE3 coculture, “but in the APOE4 coculture, we saw much-reduced . . . myelin basic protein signature,” she says, indicating less myelin production. Transmission electron microscopy of brain tissue also showed that mice with APOE4 had fewer neurons with myelinated axons, and that the neurons that were myelinated had thinner sheaths as compared to APOE3 mice.

The team then attempted to reduce cholesterol accumulation and restore myelin production in vitro and in the APOE4 mice using cyclodextrin, a drug known to reduce intracellular cholesterol accumulation. APOE4 oligodendroglia-neuron cocultures treated with cyclodextrin over two weeks saw myelin increase to levels approaching APOE3 cocultures. The researchers gave mice twice-weekly subcutaneous injections of the drug over 8 weeks and then compared their brain tissue to untreated APOE4 mice. In the treated mice, less cholesterol accumulated within oligodendrocytes, and the researchers observed higher levels of myelin basic protein outside of the cells, suggesting the treatment improved both cholesterol flow and myelination. Lastly, the researchers performed several cognitive tests on the mice, and “what we saw is that, indeed, it actually improves cognition,” says Kellis.

Holtzman says that he’d like to see others confirm the results and that he’s not yet convinced the findings are clinically relevant to humans. “APOE4 has many mechanisms as to how it likely alters Alzheimer’s risk,” he says, “I’m not sure this is one of the important ones or not. It’s really hard to tell.” It’s possible that changes in cholesterol and myelin contribute to the decline or degeneration associated with Alzheimer’s disease, he says, “but there’s so many other things that APOE does that we know has a massive effect on Alzheimer’s pathology that’s independent of any of these findings.”

Holtzman writes in a follow-up email to The Scientist that while APOE4 is linked to an increased risk of developing Alzheimer’s disease and to steeper cognitive declines after symptoms appear, “these effects of APOE4 on Alzheimer’s disease are seen only in the presence of Alzheimer’s pathology,” specifically the characteristic accumulation of amyloid plaques in the brain or of tau protein tangles inside of neurons. He adds that having APOE4 isn’t known to affect myelin levels in adults without Alzheimer’s pathology, and to his knowledge, nobody has investigated that potential. Therefore, the findings of reduced myelin and cyclodextrin-mediated restoration of myelin and cognition in mice with APOE4 but without Alzheimer’s pathology may or may not be clinically relevant to humans, he says.

See “Alzheimer’s Risk Gene Paradoxically Protects Against Memory Loss

Tsai explains via email that past fMRI studies have linked the gene variant to reduced white matter, which is mainly composed of myelinated axons. She adds that the work of her and her colleagues highlights the need to study Alzheimer’s risk mechanisms that may begin before disease and cognitive decline set in. “Our results suggest earlier opportunities for therapeutic intervention,” she says. “I think this is exactly what is interesting about the study.”

Tsai says that her lab won’t be involved with any potential clinical trials exploring whether cyclodextrin helps reduce symptoms in people with Alzheimer’s disease. Instead, she and her colleagues will continue to investigate mechanisms behind why the genetic variant causes this clogging effect in oligodendrocytes. “The more we know about how [cholesterol] gets there, the more potential targets for drug discovery will emerge.”