The image of the “blood-brain barrier” as a static wall that shields the brain from the rest of the body is crumbling. A group of scientists argue, in a review published in Fluids and Barriers of the CNS earlier this year, that it’s time to adopt a more apt metaphor. Jan Pieter Konsman and Jerome Badaut of the University of Bordeaux, France; Jean-François Ghersi-Egea of Lyon Neurosciences Research Center, France; and Robert Thorne of Denali Therapeutics in South San Francisco, likened the blood-brain interface to a border crossing between countries, whereby people, goods, and taxes are exchanged to serve the needs of the countries on either side. Just as such borders adapt to changing times, the “blood-brain border” adapts in both time and space, adjusting its list of “approved” molecules depending on age and a healthy or diseased local environment. This is a better metaphor than a purely obstructive barrier standing in the way of CNS-targeted therapeutics.

  • Scientists suggest adopting the term “blood-brain border” to reflect the complexity and dynamism of the interface.
  • Rather than a purely obstructive barrier, the BBB is more like a highly selective geopolitical border crossing.
  • The molecules allowed across change with age and brain region, and shift depending on need.

“The term border does more justice to the flexibility observed in the (patho)physiology of the blood–brain interface, while avoiding the negative connotation of disruption or rupture,” the authors wrote. They see the new term as a way to “stimulate new drug approaches to modulate the properties of CNS endothelial and epithelial layers, or take advantage of endogenous systems, rather than finding ways to ‘break the barrier’ to facilitate drug delivery to the CNS.”

Bustling Border. In this schematic of the border between Switzerland and European Union nations, the exchange of people, taxes, and goods at crossing points adapts as relationships between countries evolve. Similarly, the blood-brain border shifts in time and space. [Courtesy of Badaut et al., Fluids and Barriers of the CNS, 2024.]

Scientists who study the intricacies of the brain’s extensive vasculature have long known that selective molecules traverse it via different mechanisms. Still, many emphasize the restrictiveness of the BBB, focusing on the barrier properties of the tight junctions that physically seal spaces between endothelial cells lining vessels. Yet squeezing across these crevices is not the way most molecules—particularly larger ones—manage to cross. Instead, a variety of selective paths, such as solute channels, receptor-mediated transcytosis systems, and vesicular trafficking are at work across the vasculature, the scientists emphasized.

Jürgen Götz of the University of Queensland in Brisbane, Australia, welcomed the change in terminology. With his work using ultrasound to temporarily open the BBB, Götz acknowledged that he initially viewed it as a barrier to be breached (Mar 2015 news). Yet further studies have continued to unveil the nuanced mechanisms underlying changes in so-called BBB “permeability,” including those that change with age and disease. Götz noted a study led by Tony Wyss-Coray of Stanford University. It showed that, at least in mice, while plasma proteins readily permeate the healthy brain parenchyma, their uptake flags in the aged brain due to a shift from ligand-specific, receptor-mediated modes of transport, to nonspecific caveolar transcytosis (Jul 2020 news).

Operative transport mechanisms vary substantially from segment to segment across the cerebral circulation, noted Costantino Iadecola of Weill Cornell Medical College in New York, who was not involved in the review. For example, at the capillary level, the BBB is dominated by solute transporter carriers, while larger vessels favor vesicular transport mechanisms, he said. “There is this zonal diversity to the interface with the blood, and that’s only one of the interfaces in the brain.”

Intricate Interfaces. In addition to the blood-brain border (top right), the brain’s other interfaces include the inner blood-CSF border within the choroid plexus (bottom left) and the outer blood-CSF border within the leptomeninges (bottom right). Each border has distinct molecular and structural attributes. [Courtesy of Badaut et al., Fluids and Barriers of the CNS, 2024.]

Other interfaces include the blood-CSF borders, such as the one between the blood and the choroid plexus, as well as between the blood and leptomeningeal spaces on the surface of the brain (image above). Denali’s Thorne takes the view that it is at these CSF borders, rather than capillary blood-brain interfaces, where minuscule amounts of therapeutic monoclonal antibodies tend to cross. After traversing the fenestrated blood vessels that predominate within the choroid plexus, these large macromolecules get snagged in the surrounding stroma before some of them cross into the CSF. There, they travel across the brain surface and into perivascular spaces around large caliber vessels that traverse the leptomeninges. These spaces are also where cerebral amyloid angiopathy occurs. Thorne and other scientists think that when anti-Aβ antibodies latch onto this vascular amyloid, the inflammatory consequences that manifest as ARIA ensue.

To avoid this risk and to improve delivery into the brain, some companies are taking advantage of the transferrin receptor to whisk monoclonal antibodies across. Notably, TfR expression is highest within the microvasculature, rather than at larger vessels where CAA accumulates. An appreciation for the diversity of transport mechanisms at work across the cerebrovasculature may thus allow scientists to side-step damaging responses and deliver their drugs more broadly throughout the brain parenchyma, as described in the previous story (Oct 2024 news).

The concept of a blood-brain border matches the military-inspired words used to describe immune cells that “patrol” these borders, and “infiltrate” the brain when recruited by cells on the other side. For example, the role of T cells in exacerbating neurodegenerative disease has become increasingly appreciated in recent years (Mar 2023 news). Moreover, Iadecola and other scientists have found that macrophages residing in the perivascular space—aptly named “border-associated macrophages” (BAMs)—play an important part in surveilling border crossings, but are capable of unleashing a deadly inflammatory cascade of free radicals that exacerbates CAA and may even cause ARIA (Apr 2023 conference news; Aug 2024 conference news). 

“The roles that these brain macrophages play, in addition to their strategic positioning in what can be considered to be defensive buffer zones along blood–brain interfaces, fully justifies the name border-associated macrophages, and further encourages us to consider these interfaces as biological borders,” the authors wrote.

Iadecola applauded the authors for starting this discussion, and agreed that “border” is a more appropriate term than “barrier.” However, he added that “border” barely scratches the surface of the complexities of the brain’s diverse vasculature. “With so many different interfaces and different types of borders within them, maybe we should be splitting instead of lumping,” he said. Still, he said that the brain’s myriad interfaces do serve a common purpose, that is, to maintain homeostasis, whether by delivering nutrients, clearing junk, healing a wound, or quashing an infection.—Jessica Shugart

Comments

  1. Jerome Badaut and colleagues suggest using the term “blood-brain border” rather than “blood-brain barrier,” being kind in that the acronym BBB need not be abandoned. I must agree that I welcome this change in terminology.

    With our own work in using low-intensity ultrasound to overcome the blood-brain barrier (old terminology) for drug delivery, I find myself guilty in initially only thinking of this interface as being formed by tight junctions that form cis and trans interactions and need to be separated in order to allow for a more efficient drug uptake by the brain. In tissue culture systems, readings of the transepithelial/endothelial electrical resistance (TEER) are used to assess the leakiness of the barrier (separation of tight junctions); however, this reading does not factor in that low-intensity ultrasound also facilitates caveolin-mediated trans-cytoplasmic transport, and furthermore leads to the formation of pores and tunnels through which the exchange of cargoes can occur.  

    Another point in the context of Alzheimer’s disease is the ongoing discussion whether in AD the barrier is more leaky, or in fact becomes tighter as disease progresses and individuals age. Here, again the term "blood-brain border" would more accurately reflect the actual situation. Ryan Watts and colleagues had argued several years back that there is a lack of widespread blood-brain barrier disruption in AD, whereas the team of Berislav Zlokovic argued in multiple studies for the opposite. In my view, an illuminating paper was then published by Tony Wyss-Coray’s team who showed in mice that, while plasma proteins readily permeate the healthy brain parenchyma, their uptake is diminished in the aged brain, “driven by an age-related shift in transport from ligand-specific receptor-mediated to non-specific caveolar transcytosis.”

    A third aspect touched upon in Dr. Badaut’s article are differences at the BBB between brain areas. In fact, a widespread misconception of the brain is the idea that in a healthy person the brain is completely sealed off from the periphery whereas in a diseased situation it becomes leaky. This black-and-white thinking is incorrect, as even in a healthy brain, 0.1 percent of a peripherally administered antibody can enter the brain, with obvious relevance for immunotherapies. With its seven circumventricular organs that differ in their "border" function, the brain is a complex beast that cannot be perceived as a uniform barrier.

    Together, I welcome the proposed change in terminology from barrier to border and aim to adopt this wording in our future manuscripts.

    References:

    . Internalization of targeted microbubbles by endothelial cells and drug delivery by pores and tunnels. J Control Release. 2022 Jul;347:460-475. Epub 2022 May 19 PubMed.

    . Lack of Widespread BBB Disruption in Alzheimer's Disease Models: Focus on Therapeutic Antibodies. Neuron. 2015 Oct 21;88(2):289-97. PubMed.

    . Physiological blood-brain transport is impaired with age by a shift in transcytosis. Nature. 2020 Jul 1; PubMed.

    . Blood-Brain Barrier: From Physiology to Disease and Back. Physiol Rev. 2019 Jan 1;99(1):21-78. PubMed.

    . Characteristics of blood-brain barrier heterogeneity between brain regions revealed by profiling vascular and perivascular cells. Nat Neurosci. 2024 Oct;27(10):1892-1903. Epub 2024 Aug 29 PubMed.

    . Role for caveolin-mediated transcytosis in facilitating transport of large cargoes into the brain via ultrasound. J Control Release. 2020 Nov 10;327:667-675. Epub 2020 Sep 10 PubMed.

Make a Comment

To make a comment you must login or register.

References

News Citations

  1. Stop, Hey, What’s That Sound? ... Amyloid Is Going Down?
  2. Blood-Brain Barrier Surprise: Proteins Flood into Young Brain
  3. Tweaked, Aβ-Antibodies Cross Blood-Brain ‘Border’ (Bye-Bye, Barrier?)
  4. Neurodegeneration—It’s Not the Tangles, It’s the T Cells
  5. Macrophages Blamed for Vascular Trouble in ApoE4 Carriers
  6. Implicated in ARIA: Perivascular Macrophages and Microglia

Further Reading

No Available Further Reading

Primary Papers

  1. . Blood-brain borders: a proposal to address limitations of historical blood-brain barrier terminology. Fluids Barriers CNS. 2024 Jan 5;21(1):3. PubMed.