ARTICLE IN BRIEF
Researchers used high-resolution imaging of the living human brain to show for the first time that the brain's protective barrier becomes leaky with age, starting at the hippocampus. Leakage was about 25 percent higher in older cognitively impaired subjects, compared with older subjects with normal cognition.
Age-related breakdown of the blood-brain barrier (BBB) begins in the hippocampus, according to a new study in the Jan. 21 issue of Neuron that reveals breaks in the integrity of the barrier in unprecedented detail. The study also shows a correlation between loss of hippocampal BBB integrity and mild cognitive impairment.
“Recent studies have shown that breakdown of the blood-brain barrier can allow entry of toxic substances and pathogens that can injure the brain,” said lead study author Berislav Zlokovic, MD, PhD, director of the Zilkha Neurogenetic Institute and professor of physiology and biophysics at the Keck School of Medicine at the University of Southern California in Los Angeles.
Loss of integrity of the BBB has been seen at autopsy in patients with Alzheimer's disease, “but at what stage breakdown occurs in the living human brain, and whether it contributes to cognitive impairment, remains controversial,” Dr. Zlokovic said.
The resolution of previous imaging studies of the BBB have been limited. To overcome this limitation, Dr. Zlokovic worked with his colleagues to develop an MRI protocol and image processing software to greatly increase the resolution. Among other advances, their analysis corrected for age-related differences in blood flow, which allowed them to compare subjects of different ages. Gadolinium, which binds to serum albumin and under normal circumstances does not cross the BBB, was used as a contrast agent. Taken together, these advances allowed the researchers to image extremely small increases in BBB permeability. “This was not possible even several years ago,” Dr. Zlokovic noted.
In the new study, Dr. Zlokovic measured BBB permeability in 12 regions of the brain in 64 individuals: six younger and 18 older subjects without cognitive impairment, 21 older subjects with cognitive impairment, and 19 younger subjects with multiple sclerosis (MS).
Unexpectedly, Dr. Zlokovic said, he found that individuals with no cognitive impairment had an age-related loss of BBB integrity in the hippocampus as a whole, as well as in the CA1 and dentate gyrus subregions. The measure of leakage, known as Ktrans, was about 40 percent greater in cognitively normal subjects in their late eighties, compared with those in their mid-twenties. There was a trend toward loss in CA3 as well, but it did not reach significance. The data were especially surprising because the barrier in the hippocampus is normally tighter than in other brain regions.
“Interestingly, there was not as much change in other regions of the brain,” he said, including cortical and subcortical regions, as well as white matter tracts.
Leakage was about 25 percent higher in older cognitively impaired subjects, compared with older subjects with normal cognition. There was no significant difference in hippocampal volume among any of the groups.
As a validation step, Dr. Zlokovic compared the BBB integrity in subjects with relapsing-remitting MS without cognitive impairment to age-matched cognitively normal subjects without MS. While there were strong differences in white matter regions, as expected, there was no difference between them in hippocampal BBB leakage, showing that BBB breakdown in one region does not necessitate breakdown in another.
Perivascular cells called pericytes play a key role in BBB formation and maintenance. Pericytes play a role in clearance of amyloid, Dr. Zlokovic noted, and have been shown to degenerate in Alzheimer's disease. In this study, he found that the level of a marker for pericytes, called soluble platelet-derived growth factor receptor beta, was increased in the cerebrospinal fluid (CSF) of subjects with mild cognitive impairment versus age-matched subjects with normal cognition, and its level correlated with the level of BBB breakdown.
Together, these results suggested that the CSF level of the receptor may serve as a surrogate marker for loss of BBB integrity. That conclusion was strengthened by additional experiments, which showed a strong increase in the CSF level of the marker in a transgenic mouse model of Alzheimer's disease with BBB breakdown.
“This all tells us that pericytes could be injured very early during the disease process, which may contribute eventually to cognitive impairment and dementia,” Dr. Zlokovic said.
What remains unanswered, he said, is what is driving the BBB breakdown and why the hippocampus appears to be differentially vulnerable. “The blood-brain barrier is everywhere, but it is not everywhere the same,” he noted.
Dr. Zlokovic has previously shown that breakdown of the blood-spinal cord barrier may be an early event in a mouse model of amyotrophic lateral sclerosis (ALS). “I think this may be a more general phenomenon, not just in dementia,” he said, with the symptomatology driven by the site of breakdown. If he is right, it suggests that treatments to protect the BBB may be widely applicable in neurodegenerative diseases.
“This interesting study raises a number of questions,” said Clifford Jack, MD, a professor of radiology at the Mayo Clinic in Rochester, MN. “The first is, what is the substrate of loss of blood-brain barrier impermeability? We know that cerebral amyloidosis is often accompanied by vascular amyloidosis, so is this a reflection of vascular deposition of amyloid?”
Second, he said, there is the question of diagnostic specificity. “Many conditions can lead to mild cognitive impairment, so the question is whether the loss of blood-brain barrier integrity is associated with cognitive impairment only in people who have Alzheimer's disease, or other forms of dementia as well.”
Third, he said, if it is specific for Alzheimer's, “does the loss of the blood-brain barrier anticipate other changes, or lead to them?”
That same question was foremost for Michael Weiner, MD, a professor of radiology and biomedical imaging, medicine, psychiatry, and neurology at the University of California, San Francisco, and a principal investigator for the Alzheimer's Disease Neuroimaging Initiative. Over the years, many investigators have suggested that vascular changes drive the disease, he said. But it is also possible that amyloid is driving the vascular changes. “We just don't know, mechanistically, what is the cause of dementia.”
While cautioning that the study will need to be replicated before gauging the likely impact of the finding, Dr. Weiner said, “it does raise the possibility of another approach” to examining the aging brain at risk for dementia.
Dr. Clifford concluded, “It is hard to say this is going to have clinical utility without understanding more about it. But to me it is a really interesting study, and it raises a number of interesting questions even before it becomes clinically useful.”
VIDEOS ON DEMAND: Contrast-enhanced MRI shows blood-brain barrier breakdown in the aging hippocampus: What does it mean for future research? In a video interview, Berislav Zlokovic, MD, PhD, director of the Zilkha Neurogenetic Institute and professor of physiology and biophysics at the Keck School of Medicine at the University of Southern California in Los Angeles, discusses the study published in the Jan. 21 issue of Neuron, which reports that people with mild cognitive impairment had much leakier blood vessels in the hippocampus and that blood vessels in the hippocampus tend to become leakier in all people as they age. Tap here for the video:.