Numerous studies hint that exercise benefits the brain, but figuring out how has been a hard slog. Confounding factors riddle human results, complicating their interpretation. Enter the lean, mean, mouse machine. According to a report published October 29 in PLOS Biology, exercise prevented age-related breakdown of the brain vasculature in wild-type mice, and staved off the infiltration of inflammatory microglia. Animals lacking ApoE did not enjoy these benefits of “working out,” however. The findings reveal a complex interplay between aging, neurovascular function, and ApoE expression, and illuminate ways in which exercise might tip the balance away from decline. The researchers, led by Gareth Howell at the Jackson Laboratory in Bar Harbor, Maine, have yet to determine how human ApoE alleles, which differ from the rodent isoform, fit into the picture.

“These findings put ApoE and innate immunity at the crossroads of exercise and blood-brain barrier integrity,” said Terrence Town of the University of Southern California in Los Angeles. “It seems that the lion’s share of exercise benefits come from ApoE, at least in mice.” Town and other commentators pointed out that the story could unfold differently in humans because of their specific ApoE isoforms.

While their methods and findings vary greatly, many studies point to the cognitive benefits of exercise and even suggest that regular workouts may stave off Alzheimer’s disease (see July 2014 conference newsAug 2015 conference news; and AlzRisk analysis). It remains to be seen how working up a sweat benefits cognition, but some evidence suggests that preservation of cerebral blood flow and reduced hypoxia may be involved (see Ainslie et al., 2008; Paillard et al., 2015). 

To get a better handle on the subject, first author Ileana Soto and colleagues turned to mice. They first compared gene expression profiles of four-month-old and 21-month-old mice in several regions of the brain. A strong pattern emerged in the cortex: Genes involved in vascular function changed dramatically with age. The shifts were less pronounced in the hippocampus, however, the researchers noted that the vascular density there is lower than in the cortex, potentially masking vascular-related changes in gene expression in that region. In the cortex of aged mice, a threefold increase in deposits of the plasma protein fibrinogen, both in and outside of blood vessels, confirmed that the vasculature was compromised. 

Changing of the Guard.

In aged mice (right), pericytes (pink) decline while microglia (green) rush in. [Courtesy of Soto et al., PLOS Biology 2015.]

The researchers next zoomed in on the vasculature to further decipher age-related changes. Compared to young mice, aged mice lacked collagen IV and laminin, two key components of the basement membrane that supports the barrier, in the cortex and hippocampus. Immunostaining detected 20 percent fewer pericytes in the cortex, and they covered only half as many of the blood vessels there. Pericytes bolster the blood-brain barrier by preventing transcytosis across endothelial cells that line vessel walls, and they facilitate communication between endothelial cells and astrocytes. Astrocytes in aged mice expressed more glial fibrillary acidic protein (GFAP), a marker of reactivity, than those in younger mice, and expressed lower levels of aquaporin-4 (AQP4), a protein that regulates water transport in the brain and has been linked to clearance of solutes such as amyloid-β (see Aug 2012 news). 

Aged mice also had elevated numbers of microglia associated with blood vessels. Electron micrographs of cortical sections revealed that these microglia engulfed apoptotic cell debris, and tended to congregate in regions with fibrin deposits and degenerating pericytes. Aged mice had three times as many microglia expressing the complement protein C1qA, which has emerged as a major player in microglial-dependent pruning of synapses in AD and other diseases (see Nov 2015 conference news). Together, these findings indicate that aging alters key components of the neurovasculature and elevates neuroinflammation around blood vessels.

Freewheeling Mice Fight Age.

Mice that voluntarily exercised fended off neurovascular changes as they aged. [Courtesy of Soto et al., PLOS Biology 2015.]

Might exercise alter these age-related vascular changes? To find out, the researchers compared the neurovasculature in 18-month-old sedentary mice to animals of the same age that were given free access to a running wheel at 12 months. At 18 months, the runners performed like middle-aged mice in several measures of mouse daily living, including grip strength, nest construction, and burrowing behavior, while their sedentary counterparts had declined. In the elderly exercised mice, most components of the neurovasaculature—including basement membrane proteins, pericyte coverage, astrocyte reactivity, and microglial infiltration and complement expression—looked very similar to those in four-month-old mice. Running also prevented age-related reduction in the synaptic proteins synaptophysin and PSD-95. However, aged mice that never exercised displayed all the characteristic signs of neurovascular decline. Interestingly, exercise did not protect ApoE knockout mice. Instead, they looked the same as their sedentary brethren. However, because ApoE knockouts had already displayed neurovascular changes by 12 months of age, it is possible that exercise was unable to reverse this damage once it had begun, the researchers noted. Complicating interpretation even more, the initial gene profiling experiments revealed that ApoE expression in the cortex and hippocampus plummeted with age in normal mice. Could exercise reverse that? Further analysis indicated that astrocytes were the primary producers of ApoE in these brain regions, and that exercise prevented their age-related decline in ApoE production. The researchers concluded that loss of ApoE with age may lead to neurovascular changes, and that exercise may rescue this effect by somehow boosting ApoE expression. 

The paper’s findings linking cerebrovascular and neuroinflammatory changes with the aging process confirm what many other studies have reported, commented Costantino Iadecola of Weill Cornell Medical College in New York. Cynthia Lemere of Brigham and Women’s Hospital in Boston agreed. She, too, noted a preponderance of findings from her own lab and others pointing to neurovascular decline and the damaging role of the complement cascade with age or during neurodegenerative disease (see Janota et al., 2015Shi et al., 2015Stephan et al., 2013). In a joint comment to Alzforum, Axel Montagne and Berislav Zlokovic of the University of Southern California in Los Angeles pointed out that the mouse study meshes with their recent findings linking BBB damage, as well as pericyte degeneration, with age and cognitive decline in people (see Feb 2015 Webinar; and Sagare et al., 2015). 

Nga Bien-Ly of Genentech in South San Francisco, who recently reported that the blood-brain barrier remains intact in mouse models of neurodegenerative disease, commented that more quantitative measures of BBB leakage could strengthen the mostly histological observations of the study (see Oct 2015 news).

All commentators agreed that the novel role of ApoE in manifesting the neurovascular benefits of exercise was the most important finding of the paper.

“What we now need to understand is the relationship between astrocytes, ApoE, and complement signaling through microglia,” Lemere said. “They must be involved in some feedback mechanism that ultimately leads to neurovascular dysfunction and cognitive decline, but we don’t know how it all ties together.”

Howell also sees astrocytes, which interact with myriad cells in the brain, as central players. He intends to use an army of transgenic mice to tease out the role of these and other cells in greater detail. He plans to use mice that express different isoforms of human ApoE to determine whether they modulate exercise differently. He also intends to use mice in which ApoE expression is switched off just as the animals embark on their midlife exercise routines to distinguish between pre- and post-exercise effects of the protein. Further studies in mice lacking complement proteins, or in animals in which resident microglia can be distinguished from peripheral monocytes, could further elucidate the role of microglia in this process, Howell told Alzforum.

How does any of this relate to AD? Howell is interested in that, but said that a deeper understanding of the changes that accompany aging—the primary risk factor for AD—is what is most needed in the field. Interestingly, a previous study reported that exercise reduced amyloid deposition people who carry an ApoE4 allele, but had no effect in non-carriers (although non-carriers had lower amyloid burden to begin with) (see Head et al., 2012). 

In their simplest interpretation, Lemere said the findings should at least motivate people to exercise. She added, “I plan to go on a brisk walk immediately following this call.”—Jessica Shugart


  1. Growing evidence suggests that maintaining a physically active lifestyle may be particularly beneficial in reducing the risk for dementia and Alzheimer's disease. In this paper, Soto et al. convincingly show that exercise in aging mice preserves the integrity and function of the neurovascular unit (NVU) leading to a healthier and “younger-like” brain.

    Using RNA-seq, in combination with histological analyses, Soto and colleagues showed that aging leads to significant deterioration of neurovascular structures, including basement membrane reduction, pericyte loss, and astrocyte dysfunction. Moreover, they show that age-dependent neurovascular decline leads to vascular leakages and blood-brain barrier (BBB) breakdown. Next, they provide important evidence that astrocytic APOE levels decrease in mice with aging that correlates with BBB disruption characterized by perivascular fibrin deposition, pericyte loss, gene expression changes, activation of microglia, and loss of aquaporin-4 from astrocytic endfeet. Interestingly, the authors found that exercise restores brain APOE levels in older mice, which rescues the vascular phenotype. The authors next studied the effects of exercise in APOE-null mice that develop pronounced BBB breakdown as shown by multiple independent laboratories, and show that exercise partially rescues vascular abnormalities in these mice by increasing the number of pericytes, but some BBB leakage, increased numbers of activated microglia, and behavioral deficits still persist.

    Overall, this paper also supports recent findings of age-dependent BBB breakdown in humans and worsening of BBB disruption in early stages of dementia, as we demonstrated using an advanced dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) method (Montagne et al., 2015). Interestingly, increased BBB permeability in elderly humans with mild dementia also correlates with pericyte injury, as shown by increased CSF levels of soluble platelet-derived growth factor receptor β (sPDGFRβ) – a marker of pericyte injury (Sagare et al., 2015). Again, this corroborates the present findings in the aging murine brain.

    Given the critical role of APOE in maintaining normal NVU integrity (Bell et al., 2012), Soto et al. data further extend previous findings by suggesting a link between astrocytic APOE, age-related neurovascular dysfunction, BBB breakdown, and microglia/monocyte activation. Future studies should address whether these findings in aging mice can translate to aging humans at genetic risk for AD carrying APOE4 allele and in APOE4 non-carriers with a lower risk for AD, and whether exercise will help ameliorate neurovascular changes and vascular phenotype during aging in humans.


    . Blood-brain barrier breakdown in the aging human hippocampus. Neuron. 2015 Jan 21;85(2):296-302. PubMed.

    . Shedding of soluble platelet-derived growth factor receptor-β from human brain pericytes. Neurosci Lett. 2015 Oct 21;607:97-101. Epub 2015 Sep 25 PubMed.

    . Apolipoprotein E controls cerebrovascular integrity via cyclophilin A. Nature. 2012 May 24;485(7399):512-6. PubMed. Correction.

    View all comments by Berislav Zlokovic
  2. The authors’ findings complement extant literature suggesting that exercise attenuates age-related changes associated with Alzheimer’s disease. The study’s results support a mechanism by which exercise provides resilience to neurovascular decline. The proposed involvement of neuroinflammatory pathways highlights the need for further investigation, in humans, of aberrant immune response and inflammation in AD. Elucidating the biological mechanisms of the protective effects of engagement in physical activity, in humans, on these specific pathways would provide valuable insight for future non-pharmacological interventions for AD.

    Interestingly, the diminishing age-related effects were only observed in APOE-producing animals. These findings extend recent human studies that show an interaction between physical activity and APOE4-carrier status on AD biomarkers. Future studies examining the effects of APOE, exercise, and neurovascular decline on the risk for AD are needed, and may help identify at-risk individuals who would benefit from lifestyle intervention.

    View all comments by Stephanie Schultz
  3. This is an incredibly interesting article demonstrating that exercise can limit age-related decline in neurovascular integrity and cognitive function. While it is clear that neurovascular integrity and function decline with age, and that this can lead to declining cognitive function, it is not known what molecular mechanism mediates this pathway or whether lifestyle changes can attenuate this issue. This article shows an important role of exercise in limiting neurovascular decline, demonstrating that indeed environmental and lifestyle changes can influence this process. Furthermore, the authors demonstrate that ApoE, a known susceptibility locus for Alzheimer’s disease, is a key mediator that helps maintain neurovascular integrity through life, linking normal age-related decline with pathophysiology of AD. Therefore, this new finding has important implications for understanding the molecular mechanism of age-related cognitive decline as well as Alzheimer’s disease, and for finding methods to potentially attenuate these processes.

    This study also raises several really interesting questions, most notably, how does ApoE mediate the ability of exercise to regulate neurovascular aging, and how can we harness this to limit age-related cognitive decline as well as the onset and progression of AD? It will be very interesting to identify the sequence of events that leads to disrupted neurovascular integrity and age-related cognitive decline to determine which cell(s) ApoE acts on, and how these signals are then transduced to other neural, vascular, and immune cells. Does ApoE signal to all cells within the neurovascular unit to maintain integrity or does it target one cell type, such as pericytes, which then maintain the integrity of the unit? Which are the critical signaling interactions for the decrease in cognitive function? Which interactions can be targeted by pharmacological intervention? Furthermore, it will be of particular interest to understand how different ApoE isoforms can influence these processes, and how they may go awry in AD.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Healthy Lives, Healthy Minds: Is it Really True?
  2. Exercise Boosts Cognition In Symptomatic Disease
  3. Brain Drain—“Glymphatic” Pathway Clears Aβ, Requires Water Channel
  4. Microglia Control Synapse Number in Multiple Disease States
  5. Does the Blood-Brain Barrier Stand Up to Alzheimer’s? Study Finds No Breach

Webinar Citations

  1. Leaky Blood-Brain Barrier a Harbinger of Alzheimer's?

Paper Citations

  1. . Elevation in cerebral blood flow velocity with aerobic fitness throughout healthy human ageing. J Physiol. 2008 Aug 15;586(16):4005-10. Epub 2008 Jul 17 PubMed.
  2. . Protective Effects of Physical Exercise in Alzheimer's Disease and Parkinson's Disease: A Narrative Review. J Clin Neurol. 2015 Jul;11(3):212-9. PubMed.
  3. . Glio-vascular changes during ageing in wild-type and Alzheimer's disease-like APP/PS1 mice. Brain Res. 2015 Sep 16;1620:153-68. Epub 2015 May 9 PubMed.
  4. . Complement C3-Deficient Mice Fail to Display Age-Related Hippocampal Decline. J Neurosci. 2015 Sep 23;35(38):13029-42. PubMed.
  5. . A Dramatic Increase of C1q Protein in the CNS during Normal Aging. J Neurosci. 2013 Aug 14;33(33):13460-74. PubMed.
  6. . Shedding of soluble platelet-derived growth factor receptor-β from human brain pericytes. Neurosci Lett. 2015 Oct 21;607:97-101. Epub 2015 Sep 25 PubMed.
  7. . Exercise Engagement as a Moderator of the Effects of APOE Genotype on Amyloid Deposition. Arch Neurol. 2012 Jan 9; PubMed.

External Citations

  1. AlzRisk analysis

Further Reading


  1. . Dissecting the Contribution of Vascular Alterations and Aging to Alzheimer's Disease. Mol Neurobiol. 2015 Jul 5; PubMed.

Primary Papers

  1. . APOE Stabilization by Exercise Prevents Aging Neurovascular Dysfunction and Complement Induction. PLoS Biol. 2015 Oct;13(10):e1002279. Epub 2015 Oct 29 PubMed.