With age, the brain’s ability to clear aggregating proteins such as Aβ wanes. In the November 9 Nature, researchers led by Jonathan Kipnis at Washington University in St. Louis suggest a reason why. They found that the macrophages that cozy up to arteries in the brain help thin out extracellular matrix around these vessels. This allows the arteries to fully dilate. In turn, the pulsatility of these vessels moves the surrounding cerebrospinal fluid, driving it through the brain and rinsing away waste. In mice, killing off these perivascular macrophages caused blood vessels to stiffen and CSF flow to dwindle. Conversely, stimulating them with a growth factor restored arterial suppleness. Because protein clearance slows with aging, boosting perivascular macrophage function could improve brain health and potentially delay the onset of diseases such as Alzheimer’s, Kipnis suggested.

  • Depleting perivascular macrophages weakens CSF flow and protein clearance.
  • The cells die off with age and are altered in Alzheimer’s brain.
  • Stimulating macrophages restores flow, suggesting these cells could be a therapeutic target.

Other researchers were enthusiastic. “This study is exciting, because it suggests that elements of the perivascular space may be therapeutically targetable by simple, minimally invasive techniques,” Rupal Mehta at Rush University in Chicago wrote to Alzforum. Per Kristian Eide at the University of Oslo, Norway, agreed this has potential, and added that studying these cells may help researchers pinpoint what goes wrong with clearance in late-onset AD. “These novel findings may have great impact,” Eide wrote.

Suppressed Flow. In a healthy mouse (left), tracer injected into the CSF at the base of the brain (red) travels beside blood vessels to penetrate throughout the parenchyma (top) and far into tissue around arteries (bottom), but after depleting perivascular macrophages (right), it barely infiltrates. [Courtesy of Drieu et al., Nature.]

Macrophages inhabit spaces that border the brain parenchyma, namely along blood vessels and in the meninges. In both locations, they make direct contact with CSF. Although these cells have been implicated in conditions such as hypertension, stroke, and AD, no one knew exactly what they did in healthy brain (May 2017 news; Thanopoulou et al., 2010; Faraco et al., 2016). 

To explore this, first author Antoine Drieu killed off border macrophages in wild-type mice. He did this by injecting liposomes containing a toxin into the CSF, where they were selectively taken up by macrophages. One week later, Drieu and colleagues injected a fluorescent tracer into CSF at the base of the brain, and tracked its diffusion along vessels. In control mice, arterial pulsing pushed the tracer into parenchyma. In liposome-treated mice, however, it penetrated only half as far as it did in control mice, indicating weaker CSF flow (see image above). To see why this was, the authors directly examined arterial movement through a cranial window while they stimulated brain activity by tickling the mice’s whiskers. In mice lacking perivascular macrophages, blood vessels dilated less than they did in controls.

Why might this be? Perivascular macrophages are known to pump out matrix metalloproteases (MMPs), which chew up extracellular matrix proteins. The authors found that MMP activity around brain blood vessels was suppressed after macrophage depletion, and the ECM was thicker. This overgrowth hindered dilation of blood vessels, in effect making them stiffer. In addition, nearby fibroblasts released more ECM proteins in the absence of macrophages. The authors concluded that macrophages keep vessels supple both by breaking down ECM and via crosstalk with fibroblasts that regulates their output.

Intriguingly, a previous study from Mehta’s group had also tied perivascular macrophages to the extracellular matrix, but in that case, thicker overgrowth correlated with an excess of macrophages (Jul 2022 news). The conflicting data may reflect a bimodal response, where both too many and too few macrophages are deleterious, Mehta suggested.

Drieu and colleagues found that not all perivascular macrophages contribute to CSF flow. An analysis of gene expression revealed two subsets. One expressed the immune marker MHCII, and clustered around veins. Based on their expression profile, the authors believe these cells may recruit circulating leukocytes to brain. The other group expressed LYVE1, a membrane glycoprotein. These cells resided mainly around arteries and arterioles, and were scavengers, engulfing nearby debris. LYVE1 cells seem to be the ones controlling CSF flow, as genetically ablating only this subset thickened ECM and stiffened vessels.

LYVE1 macrophages die off with age, perhaps explaining the impaired CSF flow in older mice. When the authors gave macrophage colony-stimulating factor to 2-year-old mice to activate their remaining macrophages, MMP activity rose, ECM thinned, and CSF flow strengthened. This suggests these cells could be manipulated therapeutically, Kipnis said.

Burgeoning Plaque. The brains of 3-month-old 5xFAD mice (left) are scattered with amyloid plaques (gold), but in the absence of perivascular macrophages (right), the plaques become much more abundant. [Courtesy of Drieu et al., Nature.]

Do the findings have implications for Alzheimer’s? Killing off border macrophages in 5XFAD mice worsened plaque load. This agrees with a previous study, which found worse cerebral amyloid angiopathy after perivascular macrophage depletion in TgCRND8 mice, a CAA model (Hawkes and McLaurin, 2009). 

To investigate whether something similar might happen in human brain, the authors turned to the Dominantly Inherited Alzheimer Network. In postmortem brain sections of people with a familial AD gene, Drieu and colleagues found gene expression changes in perivascular macrophages similar to those the authors saw in 5XFAD mice, hinting at relevance for human disease.

“We believe targeting these macrophages could be a new avenue to treat patients,” Drieu told Alzforum. Kipnis noted that because perivascular macrophages inhabit the outskirts of the brain, they can be readily reached with drugs. Because these cells bathe in CSF and are scavengers, they take up the bulk of solutes from this fluid. Kipnis thinks delivering drugs to them would be feasible. Costantino Iadecola at Weill Cornell Medical College, New York, agreed this approach has potential. He suggested targeting specific receptors on macrophages to fine-tune their function, rather than boosting cell activity in general.

The authors are also exploring whether the decrease in LYVE1 macrophages with age heightens risk for other neurodegenerative disorders. Tal Iram at Stanford University thinks the findings may be generally applicable. “This is a truly novel and exciting new frontier in the study of debris clearance in aging-related diseases … improving CSF flow by targeting macrophages would affect overall clearance of debris, and therefore could be a robust approach for [treating] neurodegenerative disease,” she wrote.—Madolyn Bowman Rogers


  1. One of the major roles of the CSF is waste clearance, which, as stated by the authors, becomes inefficient with aging.

    This paper unravels a novel regulator of CSF flow—border macrophages that phagocytose extracellular matrix around arteries to allow proper pulsation of CSF through the brain parenchyma. By using very creative tools, the authors deplete this population, or manipulate their function, to gain more insight into underlying mechanisms.

    I think that this is a truly novel and exciting new frontier in the study of debris clearance in aging-related diseases. As opposed to many of the current approaches that focus on enhancing clearance of specific aggregates (such as anti-amyloid antibodies), improving CSF flow by targeting macrophages could affect overall clearance of debris, and therefore could be a much more robust approach for neurodegenerative diseases.

  2. This is another hallmark study by the Kipnis group. Here, they demonstrate a role for parenchymal border macrophages in the regulation of CSF flow along blood vessels in the brain, commonly referred to as para- (or peri-)vascular transport of CSF. Indeed, they show that a particular set of macrophages samples CSF content on its way in and out of the brain. They report that reduced macrophage function was accompanied with reduced CSF, or tracer, transport. The function of the macrophages further affected arterial motion, and the authors presented evidence for their role in Aβ clearance.

    These results are intriguing for several reasons, not least linking immunological cells with CSF transport, brain-clearance processes, and mechanisms behind altered arterial motion in aging and in the abnormal accumulation of substances such as Aβ. Further, these cells can be therapeutic targets to modify CSF transport, and thereby, parenchymal clearance. The findings also have implications for our understanding of impaired CSF transport during aging, and of Aβ accumulation in Alzheimer’s disease.

    From my perspective, these novel findings may have great impact.

  3. This study, along with some other recent studies, indicates that perivascular space stromal components, which have been poorly investigated, are much more heterogeneous and dynamic than was once recognized. Exploring this heterogeneity and the physiology of these spaces is critical, because research in this area may inform on novel treatment targets and disease predispositions.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Do Perivascular Macrophages Mediate Aβ Pathology?
  2. Mesh, not Membrane: Pia Filters CSF, Gets Clogged by Aβ Fibrils

Research Models Citations

  1. 5xFAD (B6SJL)
  2. TgCRND8

Paper Citations

  1. . Scavenger receptor class B type I (SR-BI) regulates perivascular macrophages and modifies amyloid pathology in an Alzheimer mouse model. Proc Natl Acad Sci U S A. 2010 Nov 30;107(48):20816-21. PubMed.
  2. . Perivascular macrophages mediate the neurovascular and cognitive dysfunction associated with hypertension. J Clin Invest. 2016 Dec 1;126(12):4674-4689. Epub 2016 Nov 14 PubMed.
  3. . Selective targeting of perivascular macrophages for clearance of beta-amyloid in cerebral amyloid angiopathy. Proc Natl Acad Sci U S A. 2009 Jan 27;106(4):1261-6. PubMed.

Further Reading

Primary Papers

  1. . Parenchymal border macrophages regulate the flow dynamics of the cerebrospinal fluid. Nature. 2022 Nov;611(7936):585-593. Epub 2022 Nov 9 PubMed.