If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Centre for Regenerative Medicine, Institute for Regeneration and Repair, Edinburgh BioQuarter, University of Edinburgh, Edinburgh, United KingdomUK Dementia Research Institute, Edinburgh Medical School, University of Edinburgh, Edinburgh, United Kingdom
Address correspondence to Axel Montagne, Ph.D., UK Dementia Research Institute, Centre for Clinical Brain Sciences, The University of Edinburgh, Chancellor's Bldg., 49 Little France Crescent, Edinburgh EH16 4SB, United Kingdom.
UK Dementia Research Institute, Edinburgh Medical School, University of Edinburgh, Edinburgh, United KingdomCentre for Clinical Brain Sciences, University of Edinburgh, Edinburgh, United Kingdom
Prevalence of dementia continues to increase because of the aging population and limited treatment options. Cerebral small vessel disease and Alzheimer disease are the two most common causes of dementia with vascular dysfunction being a large component of both their pathophysiologies. The neurogliovascular unit, in particular the blood-brain barrier (BBB), is required for maintaining brain homeostasis. A complex interaction exists among the endothelial cells, which line the blood vessels and pericytes, which surround them in the neurogliovascular unit. Disruption of the BBB in dementia precipitates cognitive decline. This review highlights how dysfunction of the endothelial-pericyte crosstalk contributes to dementia, and focuses on cerebral small vessel disease and Alzheimer disease. It also examines loss of pericyte coverage and subsequent downstream changes. Furthermore, it examines how disruption of the intimate crosstalk between endothelial cells and pericytes leads to alterations in cerebral blood flow, transcription, neuroinflammation, and transcytosis, contributing to breakdown of the BBB. Finally, this review illustrates how cumulation of loss of endothelial-pericyte crosstalk is a major driving force in dementia pathology.
Dementia is characterized by a gradual cognitive impairment with additional symptoms, such as depression and changes to balance and gait.
cSVD encompasses a group of pathologic conditions that affect the perforating cerebral venules, capillaries, and small arterioles, leading to white and gray matter damage in the central nervous system (CNS).
suggesting that vascular dysfunction is a major contributing factor in both. Vascular dysfunction can take many forms that involve many cells comprising the neurogliovascular unit (NVU) and include disruption of the blood-brain barrier (BBB), which plays a vital role in maintaining cerebral homeostasis. However, this review focuses on how the complex interactions between endothelial cells lining the blood vessels and pericytes closely apposed to the endothelial cells on the abluminal side contribute to BBB disruption, a central pathologic mechanism in these two important dementias.
The BBB in Health
Blood vessels throughout the body are normally partly permeable to allow the exchange of nutrients, solutes, and chemical signals between tissue and the blood, which are vital for keeping cells alive. However, the brain microenvironment is different from the rest of the body and requires tighter control provided by the BBB. Substance movement can occur paracellularly (between endothelial cells) or transcellularly (across endothelial cells), and the BBB tightly regulates the two types of substance movement via junctional complexes between endothelial cells and through expression of membrane receptors and pumps, with limited passive diffusion.
Pathogens, plasma proteins, and immune cells from the blood can have severe detrimental consequences if present within the brain, and disruption to the BBB is a major component of many neurologic diseases, including AD and cSVD.
The BBB comprises various components, commonly referred to as the NVU (Figure 1), including endothelial cells, pericytes, basement membrane, astrocyte end-feet, and surrounding oligodendrocytes and microglia, and links to neurons in a process called neurovascular coupling in which increased neuronal activity leads to increased blood flow to that area.
Endothelial cells line the blood vessels, and pericytes are mesenchymal-derived cells on the brain side of endothelial cells. Pericytes are found encircling capillaries as well as precapillary arterioles and postcapillary venules, whereas vascular smooth muscle cells (VSMCs) are found in larger vessels.
Collectively, pericytes and VSMCs are known as vascular mural cells (VMCs).
Figure 1Components of the neurogliovascular unit (NVU) and changes that occur in cerebral small vessel disease (cSVD) and Alzheimer disease (AD). The components of the NVU play an important role in maintaining the blood-brain barrier (BBB) in health. In dementia, changes, in particular to the endothelial cells and pericytes, lead to loss of function and BBB breakdown. Subsequently, the surrounding astrocytes, neurons, and oligodendrocytes get damaged, contributing to pathologic findings. Figure generated with Biorender.com (Biorender, Toronto, Ontario, Canada).
In human tissue, pericytes form two subtypes with enriched gene expression of either transmembrane transporters or extracellular matrix (ECM) regulation genes.
In addition, pericyte morphology plays an important role in maintaining BBB integrity because pericytes in the median eminence (ME), where BBB leakage naturally occurs, have a more irregular shape and less prominent nucleus compared with cortical tissue with a functional BBB.
subdivided pericytes into ensheathing pericytes located on larger diameter precapillary arterioles with shorter cell lengths and α-smooth muscle actin (α-SMA), and mesh pericytes and thin-strand pericytes located on smaller diameter capillaries with longer cell lengths, and no detectable α-SMA. They argue that ensheathing pericytes are transitional mural cells with characteristics of both VSMCs and pericytes, whereas mesh and thin-strand pericytes comprise the capillary pericytes.
Much of the literature makes no distinctions between pericyte subtypes because of not assessing morphologic differences. Furthermore, platelet-derived growth factor receptor β (PDGFR-β), which is expressed in all pericyte subtypes,
is frequently used as a pericyte marker. However, other CNS cell types have also been found to express PDGFR-β, such as perivascular fibroblast–like cells
which adds further complexity to distinguishing different pericytes subtypes and from surrounding perivascular cells.
Endothelial cells and pericytes have direct contact through gap junctions, which allow ionic currents to pass between their cytoplasms, and through adhesion plaques, which are involved in tethering pericytes to endothelial cells.
In addition, pericyte-endothelial direct connections are made through peg-and-socket junctions where evaginations interdigitate and provide a form of pericyte-endothelial anchorage.
It is their junctional proteins in particular that confer the barrier's characteristic properties of limiting paracellular permeability. Brain endothelial cells have a higher expression of certain junctional proteins localized at the cell-cell junctions compared with endothelial cells from nonneural tissues.
Endothelial cells express three types of junctional proteins that dimerize between two cells to form junctional complexes: adherens junctions (eg, vascular endothelial cadherin), tight junctions [eg, claudin-5 (CLDN-5) and occludin], and gap junctions (eg, connexins).
Already at this stage, endothelial cells and pericytes are seen in association, and both cell types invade the intraneural tissue to begin BBB formation. Endothelial cells initially possess fenestrations, which are lost following migration.
The close contact and synchronous migration of endothelial cells and pericytes during BBB formation are suggestive of the essential role of pericyte-endothelial cell interaction in BBB formation. Furthermore, mouse models deficient in pericytes do not survive to birth and have marked cerebral vasculature abnormalities that lack features of a mature functional BBB.
After the initial endothelial cell and pericyte migration, the BBB continues to mature with recruitment of other cell types to form the NVU (Figure 1).
In addition, endothelial-pericyte communication in adulthood is essential to maintain a functional BBB as acute ablation of pericytes results in BBB breakdown.
Clinical Evidence of BBB Disruption in AD and cSVD
The critical role of the BBB and its integrity in health is indicted by the fact that BBB breakdown is a component of many neurologic disorders, ranging from inflammatory to neoplastic to neurodegenerative conditions.
Loss of or reduced barrier function allows for ingress of toxic blood components, inflammatory cells, and pathogens, which disrupt the delicate brain environment and lead to pathologic events (Figure 1). In recent years, there has been mounting evidence of the involvement of vascular pathologic mechanisms in addition to the amyloid cascade in the development of AD
Extravascular fibrinogen in the white matter of Alzheimer's disease and normal aged brains: implications for fibrinogen as a biomarker for Alzheimer's disease.
Differential gene expression in multiple neurological, inflammatory and connective tissue pathways in a spontaneous model of human small vessel stroke.
Matrix metalloproteinase-2-mediated occludin degradation and caveolin-1-mediated claudin-5 redistribution contribute to blood-brain barrier damage in early ischemic stroke stage.
Cerebral small vessel endothelial structural changes predate hypertension in stroke-prone spontaneously hypertensive rats: a blinded, controlled immunohistochemical study of 5- to 21-week-old rats.
Extravascular fibrinogen in the white matter of Alzheimer's disease and normal aged brains: implications for fibrinogen as a biomarker for Alzheimer's disease.
Extravascular fibrinogen in the white matter of Alzheimer's disease and normal aged brains: implications for fibrinogen as a biomarker for Alzheimer's disease.
Differential gene expression in multiple neurological, inflammatory and connective tissue pathways in a spontaneous model of human small vessel stroke.
Differential gene expression in multiple neurological, inflammatory and connective tissue pathways in a spontaneous model of human small vessel stroke.
Differential gene expression in multiple neurological, inflammatory and connective tissue pathways in a spontaneous model of human small vessel stroke.
Differential gene expression in multiple neurological, inflammatory and connective tissue pathways in a spontaneous model of human small vessel stroke.
Matrix metalloproteinase-2-mediated occludin degradation and caveolin-1-mediated claudin-5 redistribution contribute to blood-brain barrier damage in early ischemic stroke stage.
Cerebral small vessel endothelial structural changes predate hypertension in stroke-prone spontaneously hypertensive rats: a blinded, controlled immunohistochemical study of 5- to 21-week-old rats.
Examples from selected references of neuropathology are seen in tissue and biofluids, such as serum and cerebrospinal fluid, as well as on neuroimaging, such as magnetic resonance imaging.
Clinical diagnosis of AD or cSVD is supported through visualization of changes on magnetic resonance imaging (MRI) (Table 1), with BBB leakage detectable with dynamic contrast-enhanced MRI, and intravenous injection of gadolinium-based tracers. Gadolinium tracer leakage is seen in AD,
In addition, many of the MRI signs, such as white matter hyperintensities (WMHs), lacunes, microbleeds, and infarcts, are shared between the diseases, emphasizing their overlap.
In patients with cSVD, occurrence of BBB breakdown in WMHs is indicated by increased gadolinium leakage compared with that in the surrounding normal-appearing white matter,
Furthermore, disruption to the BBB and leakage has also been found in the normal-appearing white matter of patients with cSVD, indicating more generalized vascular dysfunction
and demonstrating the importance of BBB breakdown as a core pathologic mechanism in dementia. Of interest, BBB leakage in the hippocampus of patients with mild cognitive impairment correlates with cerebrospinal fluid (CSF) levels of soluble PDGFR-β,
There is much debate about the extent of cerebral pericyte-endothelial cell coverage, primarily because of the wide variety of ways of measuring it. Some studies use electron microscopy to define coverage as a percentage of endothelial cell encircled by pericytes,
although more recent studies use PDGFR-β immunofluorescence to define coverage as a percentage of the vessel area that is PDGFR-β-positive and assess the number of pericytes by counting PDGFR-β–positive cell bodies.
A study of retinal and brain cortical tissue found that pericytes covered 85% of the capillary circumference in human and monkey retinas, but coverage was significantly less in monkey cortex.
This study used electron microscopy to determine the extent of pericyte coverage, thereby examining very small areas and perhaps not detecting variation between blood vessels in different regions. Pericytes cover up to 80% of brain capillaries in the human cortex and hippocampus, as indicated by immunofluorescence.
In three regions of mouse brain, coverage was approximately 80%, whereas in the spinal cord, pericyte coverage varied from 48% to 68%, depending on location.
Pericyte coverage is heterogeneous and dependent on subtype, which relates to vessel diameter, leading to variations in vessel coverage. In mice, ensheathing pericytes cover up to 95% of the vessel, whereas mesh and thin-strand pericytes cover 71% and 51%, respectively.
However, it is generally accepted that in the CNS a high proportion of endothelial cells are encircled by pericytes, hinting at the vital role they play in the NVU and regulation of the BBB,
but the exact figures vary among species, brain regions, and measurement techniques. Of importance, pericyte coverage appears to be related to BBB leakage because naturally leaky capillaries of the ME have a reduced pericyte coverage in mice compared with cortical capillaries.
with patients with AD having up to a 60% reduction in pericyte number and 30% reduction in capillary coverage when compared with controls, as indicated by PDGFR-β immunostaining.
Furthermore, a recent study found an overall reduction in VMCs and endothelial cells in patients with AD, in particular a subtype of pericytes involved in regulating the ECM.
The loss of pericytes may occur through disruption of the endothelial sphingosine-1-phosphate (S1P) receptor, which is involved in regulating pericyte vessel coverage
by controlling N-cadherin trafficking to the membrane needed for endothelial-pericyte adhesion. A reduction in S1P receptors occurs in AD brain tissue,
Therefore, S1P and S1P receptor play a role in the pathophysiologic mechanisms of dementia by altering amyloid-induced apoptosis, neuroinflammation, and endothelial-pericyte adhesion.
Interestingly, recent evidence from a genetic form of vascular dementia, cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL), has found no reduced pericyte coverage in patients or a mouse model.
These patients harbor mutations in the NOTCH3 gene; therefore, the BBB disruption seen may be attributable to the requirement of NOTCH for adhesion of endothelial cells and VSMCs to the basement membrane, affecting vessel stability.
In addition, pluripotent stem cell (iPSC)–derived VMCs in patients with CADASIL have reduced PDGFR-β and can induce apoptosis in neighboring endothelial cells, leading to decreased blood vessel stability. This phenotype is specific to the vascular mural iPSCs because the endothelial iPSCs are able to form normal vessel networks.
Furthermore, differentially expressed genes in the VMCs of patients with AD are similar to those seen in CADASIL, and ECM maintenance in AD involves loss of a pericyte subtype.
Together, these data suggest that BBB disruption in dementia occurs not only through loss of pericyte coverage and endothelial-pericyte contact but also through ECM disruption and/or loss of PDGFR-β signaling (Figure 2).
Figure 2Loss of pericyte-endothelial crosstalk in cerebral small vessel disease and Alzheimer disease. Disruption to endothelial-pericyte signaling leads to a wide variety of changes: (1) pericyte loss; (2) endothelial cell activation and extravasation of immune cells; (3) loss of balanced amyloid-β (Aβ) transport caused by increased receptor for advanced glycation end products and loss of low-density lipoprotein receptor–related protein 1 (LRP1), leading to Aβ build-up in the brain; (4) loss of LRP1 also contributes to loss of inhibition of the cyclophilin A–matrix metalloproteinase (MMP)-9 pathway, leading to extracellular matrix/basement membrane breakdown by MMPs; (5) MMP activation contributes to loss of junctional proteins, changes in organization, and widening junctional gap between endothelial cells; (6) transcriptional changes that lead to a more venous phenotype; (7) reduced major facilitator superfamily domain containing 2a, which contributes to increased transcytosis; (8) BBB breakdown and ingress of neurotoxic substances such as fibrin(ogen) into the brain; (9) dysregulation of vasoactive substances (eg, nitric oxide) and subsequent reduced cerebral blood flow (CBF); and (10) morphologic changes to endothelial cells with cellular hyperplasia and aberrant angiogenesis. Figure generated with Biorender.com (Biorender, Toronto, Ontario, Canada). CAMs, cellular adhesion molecules; CLDN-5, claudin-5; ICAM-1, intercellular adhesion molecule 1; PDGFR-β, platelet-derived growth factor receptor β; PLVAP, plasmalemma vesicle–associated protein; VCAM-1, vascular cell adhesion molecule 1; ZO-1, zonula occludens protein 1.
Platelet-derived growth factor β (PDGF-β)/PDGFR-β signaling is one of the most studied and important pathways in pericyte-endothelial crosstalk. Brain endothelial cells are enriched for and secrete PDGF-β,
) or postnatally in hypomorphic Pdgfbret/ret- or Pdgfrb-deficient models. Of importance, these models have helped further the understanding of possible mechanisms that contribute to BBB breakdown seen in cSVD and AD.
supporting the involvement of PDGF-β/PDGFR-β signaling in disease pathology. Further evidence of the clinical importance of disruption to the PDGF-β/PDGFR-β signaling pathway can be seen after surgically-induced large vessel stroke in adult Pdgfrb knockout mice, which indicates reduced pericyte coverage, significantly larger stroke lesions, and more BBB leakage, as compared with controls. This phenotype is significantly rescued with the preservation of PDGFR-β expression in pericytes.
(Figure 2). The findings of BBB impairment in patients with reduced pericyte vessel coverage are supported by reduced pericyte coverage in the naturally leaky ME
and studies in rodents using pericyte-deficient mice seen by dextran tracer leakage around pericyte-deficient blood vessels, which worsens with increasing age,
a major risk factor for cSVD and AD dementias. In these mice, fibrin(ogen) and IgG leakage can occur with as little as 20% reduction in pericyte coverage.
Interestingly, fibrin(ogen) accumulation is much greater in the white matter regions of the brain compared with gray matter. Furthermore, fibrin(ogen) deposition in the white matter leads to loss of myelin, oligodendrocytes, and neuronal axons,
The loss of oligodendrocytes may be, in part, attributable to the reduced vascular density and corresponding reduction in blood flow, leading to increased levels of hypoxia in white matter regions compared with that in the gray matter.
This study highlights that regional differences in pericyte coverage within the CNS likely contribute to the differences in tissue susceptibility in dementia.
In addition to overexpression of the Aβ precursor protein in transgenic mice, pericyte deficiency leads to reduced brain Aβ clearance mediated by low-density lipoprotein receptor–related protein 1 (LRP1). The reduction of LRP1-mediated clearance is not through changes in pericyte LRP1 expression but through reduction in pericyte numbers,
thus demonstrating an important role for pericytes in AD pathology, and in particular, limiting the formation of Aβ plaques. Endothelial cell LRP1 is important in maintaining BBB integrity; loss of endothelial LRP1 leads to BBB leakage, neuronal loss, and altered cognition in mice. This phenotype is attributable to cyclophilin A (CypA) matrix metalloproteinase (MMP)-9 pathway activation (discussed subsequently) and can be rescued by CypA inhibition or LRP1 reexpression.
Loss of pericytes as well as changes to pericyte and endothelial protein expression likely affect not only BBB leakage and toxin clearance but also cerebral blood flow.
The integrity of the brain vasculature is important in maintaining perfusion to and oxygenation of the brain, which are vital because of a high metabolic demand and low energy storage capacity of the brain.
It is thought that pericytes lack contractile α-SMA and therefore have the ability to contract in response to stimuli. However, the presence of α-SMA in pericytes could be underreported because of filamentous-actin fixation or low cellular expression.
demonstrating endothelial-pericyte crosstalk in cerebral blood flow control. Disparities in the literature may be attributable to inconsistencies in terminology with first- to fourth-order branches from perforating arterioles termed capillaries instead of precapillary arterioles and their associated ensheathing pericytes being termed α-SMA–positive VSMCs. When different subtypes of pericytes are taken into account, studies appear to agree that precapillary ensheathing pericytes are able to regulate blood flow in vivo.
Furthermore, recent studies have found that capillary pericytes modulate blood vessel diameter more slowly after direct in vivo stimulation compared with surrounding ensheathing pericytes.
Furthermore, imaging techniques may contribute to the ability to detect capillary diameter changes, and regional differences may occur; cerebral in vivo studies are currently limited to the cortex. Despite the debate over the ability of capillary pericytes to modulate cerebral blood flow, there are strong data supporting the role of ensheathing pericytes, particularly at vessel junctions,
00340-0&id=gr1.jpg){kind=link}
00340-0&id=gr2.jpg){kind=link}