Subscribe to eTOC

Deficits in the Blood-Brain Barrier Play a Role in Alzheimer's Disease Pathology
A New 3D Tissue Model Provides a Potential Target for Developing Therapy

Article In Brief

Researchers have developed a three-dimensional human neural cell model of the blood-brain barrier that enabled them to see the changes that occur with amyloid and other factors involved in Alzheimer's pathology.


No caption available.

A new Alzheimer's disease (AD) model suggests that defects in the blood-brain barrier (BBB) play a key role in worsening amyloid plaques and neurofibrillary tangles, and that therapies to reduce gaps in the BBB hold promise for treatment of the disease.

The new model uses “microfluidics” to examine how a layer of endothelial cells—an essential component of the BBB—are altered by and contribute to the presence of disease pathology.

“Traditionally, vascular abnormalities were considered to be consequences of the neuropathology, even though it was very well known that vascular dysfunction occurred early and correlated with worsening neurodegeneration,” said Katerina Akassoglou, PhD, of the Gladstone Institutes at University of California San Francisco, who was not involved in the work. “So there was always the chicken-and-egg question, of which came first, whether the vascular dysfunction was contributing to the neuropathology or the other way around. But we didn't have the tools to answer this question.”

Systems like the one in the new study may provide those tools, she said.

The new system uses a microfluidic chamber, in which cells are allowed to grow within multiple narrow porous tubes in close juxtaposition to one another. The flow of fluids through the tubes, and the degree of contact between adjacent tubes, can be controlled, allowing the researcher to investigate whether and how perturbations in one set of cells affect the other.

The new study, led by Rudolph E. Tanzi, PhD, professor of neurology at Harvard Medical School in Boston, and Roger Kamm, PhD, professor of biological and mechanical engineering at Massachusetts Institute of Technology, was an outgrowth of previous work from Dr. Tanzi's lab, which used hydrogels to examine how neurons carrying AD-causing mutations behave in the three-dimensional hydrogel environment.

“If you use a two-dimensional liquid system, such as cell culture in a Petri dish, you don't emulate Alzheimer's disease,” Dr. Tanzi said. “Cells in the brain interact, but if you don't study them in a facsimile of their natural habitat, you are not going to see those interactions.”

In 2014, in Nature, Dr. Tanzi's group showed that neurons grown in three dimensions make both plaques and tangles. In 2018 in Nature Neuroscience, he showed that when neurons were grown in a microfluidic chamber in the presence of migrating glial cells, they could also see evidence of neuroinflammation, the third pathological hallmark of AD.

Work by Berislav Zlokovic, MD, PhD, of the Keck School of Medicine at University of Southern California, and others, has established that the BBB is compromised in AD, Dr. Tanzi said. “The question is how does it get compromised, and how does it contribute to the disease?”

Study Design, Findings

Dr. Tanzi and his colleagues addressed that question in the latest study, published in the August 12 online edition of the journal Advanced Science. They mimicked the BBB by lining one tube of a microfluidic chamber with brain endothelial cells, and another with neurons expressing a gene for familial AD. The two tubes were separated by a third tube, which was filled with a hydrogel, allowing materials to flow across the gel between the two cell types.

The team immediately saw that, when the endothelial cells were in contact with the AD gene-expressing neurons, their permeability increased significantly, both for small (3 kDa) and large (40 kDa) proteins, and that increase was accompanied by reduced expression of multiple components of the endothelial junctions that are responsible for the barrier's tight regulation of cross-membrane flow, including cadherin and claudins.

Additionally, they found that the level of matrix metalloproteinase-2 (MMP-2) rose when endothelial cells were exposed to AD neurons. MMP-2 degrades the extracellular matrix and has previously been shown to increase BBB permeability and disrupt tight junctions, the cell-cell linkages that help maintain membrane impermeability. Markers of neuroinflammation also rose.

“We saw that the blood-brain barrier began to break down in our model, and this was partially due to the accumulation of beta-amyloid from the neurons,” Dr. Tanzi said. “The bottom line is that we learned something from this system that we couldn't learn from mice, or from any other system, that the pathology of amyloid, tangles, and neuroinflammation triggers the release of matrix metalloproteinases, which contribute to breakdown of the barrier.”

The investigators also saw that amyloid from the neurons was deposited directly on the cells of the endothelium but did not pass through.

What happens next in the pathological cascade was not covered in the new study, but Dr. Tanzi said a key step is likely that the breach of the BBB “allows the entry of blood-borne pathogens, including bacteria and viruses,” and these help seed deposition of yet more amyloid, leading to yet more tangles and neuroinflammation. “It's a vicious cycle.” Before, it was unclear whether the compromise of the barrier was “the chicken or the egg”—whether the pathology causes the breach or vice versa, he said. “It turns out it is both.”


“The bottom line is that we learned something from this system that we couldnt learn from mice, or from any other system, that the pathology of amyloid, tangles, and neuroinflammation triggers the release of matrix metalloproteinases, which contribute to breakdown of the barrier.”—DR. RUDOLPH E. TANZI

Work by Dr. Akassoglou of the Gladstone Institutes has shown that leakage of fibrogen into the brain can trigger neurodegeneration. Here, Dr. Tanzi showed that thrombin, with a molecular weight of 36 kDa, introduced into the endothelial lumen, increased neuronal cell death, suggesting it passed through the faulty BBB.

Finally, the team tested whether two compounds that promote BBB integrity could reduce neuronal injury from AD gene expression. Treatment of the endothelium with the inhaled steroid beclomethasone had no effect, but etodolac, a non-steroidal anti-inflammatory drug, decreased thrombin-induced neuronal death, suggesting it or related compounds may have some potential in treatment of AD.

“We think that if you can find compounds that are able to prevent the breakdown of the barrier in the presence of Alzheimer's pathology, you can nip that vicious cycle in the bud,” Dr. Tanzi said. “This is important because in the patient, the plaques and tangles form anywhere from five to fifteen years before symptoms develop,” meaning very early presymptomatic therapy is the most likely route to preventing the damage the pathology causes.

Expert Commentary

“This is a very exciting study that has explored an innovative new platform for the study of blood-brain barrier dysfunction in Alzheimer's disease,” said Dr. Akassoglou, who is senior investigator at the Gladstone Institute of Neurological Disease and professor of neurology at the University of California, San Francisco.

“This model could be the basis for high throughput screens for drug discovery and for developing a more complete model of the disease,” one that is supplemented with the other cells that comprise the barrier, including pericytes and astrocytes. “This model adds to the studies that are changing the way we think of blood-brain barrier disruption and vascular dysfunction, from merely a marker of the pathology, to being a real contributor to the neurodegenerative process.”


“This model adds to the studies that are changing the way we think of blood-brain barrier disruption and vascular dysfunction, from merely a marker of the pathology, to being a real contributor to the neurodegenerative process.”—DR. KATERINA AKASSOGLOU

Michael Workman, a PhD graduate student in the lab of Clive Svendsen, PhD, at Cedars-Sinai Medical Center in Los Angeles, has developed microfluidic systems and was co-first author on a recent publication describing a system with a blood-brain barrier and neurons derived from individual patient cells, that the lab is now using to study neurodegeneration.

“One advantage of microfluidic systems is that you can study the effect of fluid flow and shear stress,” he said. “This is an important component to capture in understanding how the BBB may take part in neurodegenerative disease.”

“Breakdown of the barrier is only one contributor to the disease,” Dr. Workman noted, and pathology can progress, at least in model systems, even in the absence of a leaky BBB. “But there is certainly cross-talk between the barrier and neurons. Cause and effect are still up for debate.”

Microfluidic systems are still not standardized enough, or scalable enough, to use for high throughput screening, he added. “But they have come a long way. They are still a little tricky to use, but I think we are headed in the right direction.”


Dr. Tanzi has no conflicts of interest. Dr. Akassaglou is a co-founder and consultant of MedaRed, Inc. She is a named inventor on patents and patent applications related to fibrinogen. Her interests are managed by the Gladstone Institutes in accordance with its conflict of interest policy. Dr. Workman is named as an inventor on a patent application (PCT/US2017/016098) entitled “Systems And Methods For Growth Of Intestinal Cells In Microfluidic Devices.” Cedars-Sinai is an investor in Emulate Inc., but he has not received any stocks, royalty payments, or anything of monetary value from any company, physician, or commercial entity related to this work.

Link Up for More Information

• Choi SH, Kim YH, Hebisch M, et al. A three-dimensional human neural cell culture model of Alzheimer's disease Nature 2014;515(7526):274–278.
• Park J, Wetzel I, Marriott I, et al. A 3D human triculture system modeling neurodegeneration and neuroinflammation in Alzheimer's disease Nat Neurosci 2018;21(7):941–951.
• Vatine GD, Barrile R, Workman MJ, et al. Human iPSC-derived blood-brain barrier chips enable disease modeling and personalized medicine applications Cell Stem Cell 2019;24(6):995–1005.
• Shin Y, Choi SH, Kim E, et al. Blood–brain barrier dysfunction in a 3D in vitro model of Alzheimer's disease Adv Sci 2019; Epub 2019 Aug 12.