Brain changes associated with leaky capillaries suggest new, potential drug targets as well as a way to diagnose the disease sooner

USC Alzheimer's research: Illustration of leaky blood vessel
This image depicts a blood vessel in the brain that has become leaky, or permeable. (Illustration/Courtesy of Arthur Toga)

Leaky capillaries in the brain portend early onset of Alzheimer’s disease as they signal cognitive impairment before hallmark toxic proteins appear, new USC research shows.

The findings, which appear in the Jan. 14 issue of Nature Medicine, could help with earlier diagnosis and suggest new targets for drugs that could slow or prevent the onset of the disease.

The number of Americans with Alzheimer’s is expected to more than double to about 14 million in 40 years, according to the Centers for Disease Control and Prevention. Five Alzheimer’s drugs are approved by the U.S. Food and Drug Administration to temporarily help with memory and thinking problems, but none treats the underlying cause of the disease or slow its progression. Researchers believe that successful treatment will eventually involve a combination of drugs aimed at multiple targets.

USC Alzheimer’s research: five-year study

USC’s five-year study, which involved 161 older adults, showed that people with the worst memory problems also had the most leakage in their brain’s blood vessels — regardless of whether abnormal proteins amyloid and tau were present.

“The fact that we’re seeing the blood vessels leaking, independent of tau and independent of amyloid, when people have cognitive impairment on a mild level, suggests it could be a totally separate process or a very early process,” said senior author Berislav Zlokovic, director of the Zilkha Neurogenetic Institute at the Keck School of Medicine of USC. “That was surprising that this blood-brain barrier breakdown is occurring independently.”

In healthy brains, the cells that make up blood vessels fit together so tightly they form a barrier that keeps stray cells, pathogens, metals and other unhealthy substances from reaching brain tissue. Scientists call this the blood-brain barrier. In some aging brains, the seams between cells loosen, and the blood vessels become permeable.

If the blood-brain barrier is not working properly, then there is the potential for damage.

Arthur Toga

“If the blood-brain barrier is not working properly, then there is the potential for damage,” said co-author Arthur Toga, director of the USC Mark and Mary Stevens Neuroimaging and Informatics Institute at the Keck School of Medicine. “It suggests the vessels aren’t properly providing the nutrients and blood flow that the neurons need. And you have the possibility of toxic proteins getting in.”

Series of tests determines ‘clinical dementia rating score’ in USC Alzheimer’s research

Participants in the study had their memory and thinking ability assessed through a series of tasks and tests, resulting in measures of cognitive function and a “clinical dementia rating score.” Individuals diagnosed with disorders that might account for cognitive impairment were excluded. The researchers used neuroimaging and cerebral spinal fluid analysis to measure the permeability, or leakiness, of capillaries serving the brain’s hippocampus, and found a strong correlation between impairment and leakage.

“The results were really kind of eye-opening,” said first author Daniel Nation, an assistant professor of psychology at the USC Dornsife College of Letters, Arts and Sciences. “It didn’t matter whether people had amyloid or tau pathology; they still had cognitive impairment.”

The researchers cautioned that their findings represent a snapshot in time. In future studies, they hope to get a better sense of how soon cognitive problems occur after blood vessel damage appears. Zlokovic said it’s unlikely that scientists will soon abandon amyloid and tau as Alzheimer’s biomarkers, “but we should be adding some vascular biomarkers to our toolkit.”


In addition to Zlokovic, Toga and Nation, the study’s authors are Melanie Sweeney and Axel Montagne of USC, who are co-first authors with Nation; Abhay Sagare, Lina D’Orazio, Maricarmen Pachicano, Farshid Sepehrband, Amy Nelson, John Ringman, Lon Schneider, Helena Chui and Meng Law, all of USC; Tammie Benzinger, John Morris and Anne Fagan of the Washington University School of Medicine in St. Louis; and Michael Harrington and David Buennagel of Huntington Medical Research Institutes in Pasadena.

The study was supported with grants from the National Institutes of Health (R01AG023084, R01NS090904, R01NS034467, R01AG039452, 1R01NS100459, 5P01AG052350, 5P50AG005142, P01AG052350, RO1AG054434, R21AG055034, R01AG055770, P50AG05681, P01AG03991 and P01AG026276), an Alzheimer’s Association Strategic 509279 Grant, the Cure Alzheimer’s Fund, the Fondation Leducq and the L.K. Whittier Foundation.

More stories about: