The Cerebral Microvasculature: Basic and Clinical Perspectives on Stroke and Glioma

Abstract

Microvascular networks are vital components of the cardiovascular system, performing many key roles in maintaining the health and homeostasis of the tissues and organs in which they develop. As discussed in this review, the molecular and cellular components within the microcirculation orchestrate critical processes to establish functional capillary beds, including organization of endothelial cell (EC) polarity, guiding investment of vascular pericytes (PCs), and building the specialized extracellular matrix (ECM) that comprises the vascular basement membrane (vBM). Herein, we further discuss the unique features of the microvasculature in the central nervous system (CNS), focusing on the cells contributing to the neurovascular unit (NVU) that form and maintain the blood-brain barrier (BBB). With a focus on vascular PCs, we offer basic and clinical perspectives on neurovascular-related pathologies that involve defects within the cerebral microvasculature. Specifically, we present microvascular anomalies associated with glioblastoma multiforme (GBM) including defects in vascular-immune cell interactions, and associated clinical therapies targeting microvessels (i.e. vascular-disrupting/anti-angiogenic agents and focused ultrasound). We also discuss the involvement of the microcirculation in stroke responses and potential therapeutic approaches. Our goal was to compare the cellular and molecular changes that occur in the microvasculature and NVU, and to provide a commentary on factors driving disease progression in GBM and stroke. We conclude with a forward-looking perspective on the importance of microcirculation research in developing clinical treatments for these devastating conditions.

Introduction

The microcirculation plays an essential role in the overall health and functionality of the cardiovascular system as well as for many other organ systems. The microvasculature facilitates oxygen delivery, solute exchange, and the transport of hormones and nutrients to various tissues in the body, among other key roles¹. Microvascular networks are an intricate hierarchy of interconnected terminal arterioles, capillaries, and post-capillary venules, sustained by the complex interplay of their cellular and molecular components^2,3. Arterioles and venules are comprised of vascular smooth muscle cells (vSMCs), endothelial cells (ECs), and extracellular matrix (ECM) proteins within the vascular basement membrane (vBM), with pericytes (PCs) being the predominant mural cell type along capillaries. Coordination among these components maintains normal vascular function, sustaining the tissues and organ systems in which they reside^2,3. Therefore, when microvascular-related pathologies occur, they can have deleterious effects not only on the vasculature directly, but also on the host tissue as well, which, for organs such as the brain, can be debilitating, life threatening, and even fatal.

In this review, we will discuss key attributes of the microvascular endothelium with respect to its functionality, interactions with other cell types within the microcirculation, and relationship to the specialized ECM constituting the vBM. Acknowledging the challenges in capturing the breadth of literature on these topics, we selectively focus on foundational and recent insights into cerebral microvessels and their involvement in neurovascular-related illnesses, namely glioblastoma multiforme (GBM) and stroke. These pathologies exemplify how blood-brain barrier (BBB) maintenance is deeply intertwined with disease progression, and how defects in the neurovascular unit (NVU) exacerbate clinical outcomes. With respect to both GBM and stroke, we focus on the interplay between ECs, PCs, and junctional proteins contributing to each pathological state. By integrating perspectives from both basic science and clinical studies, we aim to illustrate the importance of both approaches in addressing these devastating conditions by considering the role of the microcirculation.

Fundamental Characteristics of Microvasculature

Endothelial cells are the most basic and essential “building block” of blood vessels. Lining the lumen of all vessels, they undergo arterial-venous specification early in development in response to various hemodynamic and genetic factors⁴. Vasculogenesis, or the de novo formation of vessels by angioblast-derived ECs, is the first process to occur during embryonic development. Vasculogenesis involves the differentiation, migration, and the coalescence of ECs to form a primitive network of vessels⁵. Beyond embryonic development, nascent vessels that arise from existing microvessels (i.e. during angiogenesis) also receive signals from various sources and inputs to adopt diverse roles and specialization⁶. During vasculogenesis and subsequent remodeling, ECs secrete factors such as platelet-derived growth factor-BB (PDGF-BB)^7,8 as well as transforming growth factor-β (TGFβ)⁹ to recruit mural cells to the vasculature, engaging additional mechanisms to stabilize the vessel wall and promote maturation (Figure 1). In many tissues, the quiescent endothelium then establishes a critical barrier separating the circulating blood and surrounding tissue parenchyma. Beyond early development, endothelial cells within mature vessels are constantly exposed to both chemical and mechanical inputs. The integration of these signals underlies both physiological and pathological aspects of the cardiovascular system^5,10. When these forces, or the mechanisms by which these forces are sensed, become dysfunctional, they have the capacity to initiate and contribute to diseases associated with blood vasculature¹¹. Therefore, ECs have received significant attention within the field of vascular biology to better understand their roles in normal vascular development and remodeling, as well as in cardiovascular physiology and pathogenesis.

Figure 1. Simplified schematic of pericyte interactions with endothelial cells in healthy, glioblastoma, and stroke conditions.

In healthy vasculature (top), endothelial cells (green) form junctions (yellow cylinders) with neighboring cells and recruit pericytes (red) to the blood vessel wall. The extracellular matrix (violet and blue lines) surrounds blood vessels, with the specialized vascular basement membrane (vBM) integrated in the vessel wall. In vasculature impacted by stroke (bottom left), dysregulated endothelial cell junctions and reduced pericyte coverage appear to be primary drivers of disrupted vascular integrity. The arrow and question mark are intended to suggest that PCs may be disassociating from the vessel wall during stroke, an important question in developing therapeutic targets for stroke treatment. Tumor vasculature (bottom right) contains abnormal branching patterns, uneven vBM, reduced expression of cell junctions, elevated levels of inflammatory cytokines (orange triangles), and a reduction of pericyte coverage.

An important step in vascular development and maturation is the establishment of endothelial cell polarity. Induced in part by the application of blood flow, cell polarity is a fundamental process that allows ECs to perform specialized functions such as transcytosis and junctional arrangement¹². Endothelial cell polarity facilitates recognition of extracellular cues and propagating these signals intracellularly as well as to neighboring cells. Hallmarks of EC polarity include the organization of luminal (apical) and abluminal (basal/basolateral) membrane domains separated by adherens and tight junctions, which are essential for establishing and maintaining their polarized morphology^13,14. Microvascular EC polarity is established by various external stimuli such as mechanical forces and cell-cell contacts. Shear stress for example causes ECs polarization along the plane of blood flow¹². Several studies further suggest that EC polarity is strongly influenced by the mechanical translocation of cell nuclei downstream under streamlined drag¹⁵. This displacement is proposed as one mechanism by which ECs integrate the direction and magnitude of blood flow¹⁶. Cell-cell junctions are also implicated as mechanosensing (i.e. flow-sensing) complexes that are contribute to the dynamic rearrangement of the cytoskeleton¹⁷. Given the importance of blood flow in shaping EC and microvessel biology, it is not surprising that many pathologies arise from, or are worsened by, disrupted blood flow and the mechanisms that mediate the transduction of physical forces into intracellular signals.

In many tissues, another essential cellular constituent of the microcirculation is the pericyte. PCs are vascular mural cells specifically associated with the microvasculature (i.e. terminal arterioles, capillaries, and post-capillary venules). They extend along and wrap around ECs to promote of blood vessel integrity^18,19, perhaps responding to transmural forces exerted by blood. Moreover, the potential involvement of microvascular PCs in EC polarization also remains to be fully elucidated, as PCs are likely to influence this process on the capillary level²⁰ as well as undergo their own polarization, though more work will be necessary to fully identify the crosstalk between EC and PC polarity mechanisms. PCs directly contact ECs at peg-and-socket junctions²¹ but are otherwise separated by a thin layer of ECM composing the vBM. While the ECM is the primary non-cellular component throughout all tissues and organs, the vBM consists of a select subset of ECM components such as Type IV Collagen (Col-IV) and laminins²². ECM proteins present not only a physical scaffold for cellular attachment and organization, they also orchestrate, and even generate, signals that are required for tissue morphogenesis including chemical and mechanical cues²³. Thus, the ECM, and specifically the vBM produced by microvascular ECs and PCs, functions as a “reservoir” for growth factors, pro-enzymes, and chemokines that guide vessel development. While this blend of ECs, PCs, and vBM within the microcirculation varies from tissue to tissue, one of the most specialized and tightly regulated configurations of the microvessel wall is found in the central nervous system (CNS) vasculature, namely the blood-brain barrier (BBB) (Figure 1).

Microcirculation of the Central Nervous System

While vascular barrier function within the CNS is not uniform (e.g. cortex vessels vs. choroid plexus, area postrema, etc.)²⁴, brain regions requiring a high level of barrier regulation are uniquely vulnerable to microvascular disruption. These selectively permeable microvascular networks depend on many cell types and signaling cascades²⁵, which together maintain the tightly-regulated neuronal microenvironment. The formation of the neurovascular unit (NVU) is completed by the interplay between the cell types discussed above (ECs, PCs, and vSMCs), along with CNS astrocytes (ASCs)^19,26 and perhaps even microglia. Within the NVU, tight junctions associated with these cell types greatly contribute to the regulation of vessel permeability. Capillary ECs comprising the BBB restrict the translocation of ions and macromolecules from the circulation into the brain, while simultaneously coordinating the transport of blood-derived hydrophilic molecules into the brain parenchyma²⁷. Although the BBB maintains homeostasis of the brain microenvironment under physiological conditions, this barrier forms early in CNS development and must be continually reinforced as neural tissue expands and becomes more specialized.

The BBB begins to emerge during earliest stages of embryogenesis with capillary networks developing around and within primordial and maturing CNS tissue. Prior to the differentiation and recruitment of brain ASCs, PCs are recruited to these nascent vessels, contributing to BBB formation^28,29. As the structural components of the BBB are established, ECs form critical tight junctions with one another concomitant with (i) the recruitment of PCs and ASC end-feet, and (ii) the deposition of ECM proteins within the vBM³⁰. PCs have been described as essential for the development and maintenance of the BBB. Detailed analysis of mice with defects in PDGF Receptor-β (PDGFRβ) signaling revealed BBB dysfunction with diminished PC coverage impacting vascular permeability²⁸. These studies have been corroborated by a thorough characterization of the CNS vasculature of developing rats. In this model, angiogenesis was found to begin at embryonic day 12 (E12), when ECs from the surrounding vascular plexus advanced into neural tissue. During construction of primitive vessels, PCs initiated and solidified their physical association with the microvascular endothelium. As these interactions developed, ECs further engaged mechanisms to secure the BBB including synthesis of tight junctions such as ZO-1, slowed rates of transcytosis across the vessel wall, and production of specialized molecular transporters¹³. These studies, as well as many others³¹, underscore how critical BBB formation is for proper neural development and functionality, as disruption of the mechanisms underlying cerebral microvessel formation can severely impact the sensitive microenvironment of the developing brain.

As described above, the breakdown in normal microvascular function and structure can fuel the progression of various disease phenotypes. Here, we highlight microvascular disruption and subsequent BBB dysfunction as it relates to the progression of neurological diseases such as glioblastoma multiforme (GBM) and stroke. We include a specific focus on how pericytes may play key roles in the modulating neurovascular permeability and may ultimately exacerbate these pathological scenarios. For each condition, we also highlight the current status of a subset of treatments, their relationship to the brain microvasculature, and we discuss strategies that are being used to design novel therapeutics.

The Microvasculature in Glioblastoma Onset and Progression

Microvascular disruption has been studied extensively in the formation and progression of several types of cancer³². GBM is one such cancer type that elicits remodeling of the surrounding blood vessels. An aggressive and invasive tumor type, GBM perturbs cerebral vasculature during its gradual expansion into surrounding tissues and invasion of perivascular locations^18,19. Tumor vasculature in general lacks the typical hierarchical organization found in normal vascular networks, often containing tortuous loops, blind-ended vessels, and an array of other defects²¹. With regard to the microvessels associated with solid tumors including GBM, development of these lesions includes several main features: (i) rapid vascularization that matches tumor growth, (ii) abnormal branching patterns leading to disorganized and non-productive vascular networks³³, (iii) an uneven and poorly distributed vBM, (iv) reduced expression of junctional proteins, and (v) a paucity of PC coverage on tumor vessels, ultimately yielding aberrant angiogenesis and accelerated metastasis^22,34. Angiogenesis is therefore a crucial mechanism for GBM growth and expansion, due to the elevated demand for nutrients and oxygen and probably correlates with prognosis.

Vascularization of the tumor mass requires EC and PC proliferation and migration, driven largely by mis-regulated signaling downstream of worsening and unresolved hypoxia²³. Once a GBM lesion exceeds a few cubic millimeters, a virtual “domino effect” unfolds with tumor cells manipulating their surrounding microenvironment to support tumor expansion. Growth factors, cytokines, and chemokines are released in substantial quantities to activate previously uninvolved, normal cells, leading to further dysregulation and accelerated GBM growth³⁵ (Figure 1). Additional angiogenic remodeling occurs within and around the tumor, drawing in more oxygen and nutrients into the GBM microenvironment, perhaps to support GBM stem cell expansion³⁶. The ultimate outcome of tumor angiogenesis is poor with respect to function. The tumor vessels become structurally defective and leaky, yielding irregular blood flow patterns and diffusion across the microvessel wall²⁴. Although systemic metastasis via dysfunctional vasculature is rarely observed in GBM compared to other aggressively angiogenic cancer types³⁷, the remodeling of the tumor microenvironment and associated microvasculature can allow GBM cells to initiate further dissemination throughout the brain.

Disrupted vasculature within and around GBM lesions can lead to additional deleterious consequences for GBM progression. For instance, tumor cells experiencing high shear stress from altered cerebral interstitial flow undergo changes that likely exacerbate tumor growth and migration^21,38. In addition, tumor microvasculature displays wide-ranging heterogeneity in diffusion across the vessel wall, suggesting oxygen delivery, solute exchange, hormone transport, etc. are likely to be reduced in GBM phenotypes³⁹. Various tracers can be used to determine vessel functionality by tracking tracer distribution within the lumen of vascular networks as well as dispersion within a tumor mass; though observations from these studies are often limited to tumor models wherein the vessel walls are largely intact and tumors are positioned ectopically or located peripherally²¹. Studies involving mouse tumor models have suggested that the disruption of larger vessels (>30 microns in diameter) may be more involved in the spread of tumor cells than previously appreciated. Additionally, tumor cells and tumor-associated macrophages⁴⁰ expressing growth factors such as vascular endothelial growth factor-A (VEGF-A) and heparin-binding epidermal growth factor-like growth factor (HB-EGF)⁴¹ have the ability to suppress endothelial barrier properties⁴² and further contribute to vascular-related dysfunction in the GBM microenvironment. GBM blood vessel defects are worsened by mis-regulation of the receptor tyrosine kinases (RTKs) that transduce angiogenic growth factor signals. In fact, three predominant signaling pathways dysregulated in GBM include the RTK / Ras / phospho-inositide 3-kinase (PI3K) axis and defects in p53 and retinoblastoma protein (Rb) activity⁴³. Thus, GBM progression can be attributed in part to consequences of compromised BBB function, suggesting other components of the NVU such as microvascular PCs may also play a role in the chaotic tumor microenvironment and specifically in GBM pathogenesis.

Perturbations in the brain microvasculature can catalyze immune cell contributions to GBM progression. Recent data suggest that microvascular PCs may play a more complex role in immune regulation than previously appreciated^44–46. In a study conducted by Valdor et al., complementary in vivo and in vitro experimental models revealed brain PCs interacting with GBM cells were found to secrete elevated levels of anti-inflammatory cytokines and suppress the anti-tumor response of the immune system^18,47. Thus, GBM cells were implicated in stimulating immune-modulatory changes in PCs through direct cell interactions, similar to observations from other tumor models such as melanoma⁴⁸. These observations supported the notion that GBM-associated PCs acquire an immunosuppressive function that facilitates evasion of anti-tumor responses, thereby promoting tumor growth¹⁸. Valdor and colleagues proposed that PCs might represent a viable target to modulate vascular-tumor cell interactions and further enhance anti-tumor cellular behaviors. This working model implies that GBM cells presumably induce immuno-tolerant properties of PCs, offering a potential explanation for why GBM detection by the immune system is compromised during tumor growth, permitting expansion into perivascular spaces. Studies such as the one by Valdor et al. further highlight that understanding PC dysregulation, and particularly disrupted crosstalk with the immune system, will likely provide greater insight into novel therapeutic approaches for GBM clinical management.

Present and Potential Vascular-based GBM Treatments

Structural heterogeneity in tumor vessels poses numerous challenges in effectively delivering chemotherapeutic agents to GBM lesions. Highly porous vessels compromise diffusion such that drug penetrance is limited and non-uniform, while microvasculature retaining BBB properties restrict pharmacological agents from crossing the vessel wall altogether⁴⁹. Two frequently utilized classes of GBM drug treatments are vascular disrupting agents (VDAs) and anti-angiogenic therapies. These approaches have their respective strengths and weaknesses when targeting tumors and associated vasculature. Several studies have shown that cancer-associated blood vessels, and specifically those within GBM tumors, can retain a high level of PC coverage, which limits their responsiveness to VDAs⁵⁰. The primary mechanism of action of VDAs is to disrupt the intracellular tubulin cytoskeleton, aiming to destroy tumor vasculature⁵¹. Therefore, because PCs promote capillary stability, their sustained presence within the GBM microcirculation⁵² can reduce the efficacy of these drugs in binding their target and eliciting a beneficial effect.

Anti-angiogenic therapies have also been explored to disrupt GBM vessel formation and slow tumor progression⁵³. One of the pathogenic hallmarks of GBM is outgrowing its blood supply to the point of causing necrosis, as well as extensive neovascularization. The intent of an anti-angiogenic therapy is to cut off the blood supply to the tumor, starving the tumor of the nutrients needed for survival and propagation. Avastin^® (Bevacizumab^®) is used for treatment of recurrent glioma and can alter the tumor vasculature to the point it no longer generates contrast-enhancement by MRI, perhaps by reducing overall vascular density and/or vessel leakage. Avastin^® also reduces vasogenic edema (as do steroids), attributed in large part to the inhibition of VEGF-A-induced effects on vascular permeability⁵⁴. Refractoriness to these strategies however has suggested an incomplete understanding of tumor vascularization mechanisms and the potential need to consider combinatorial approaches such as alongside traditional chemotherapeutics (e.g. temozolomide) or novel immunotherapies⁵⁵. Vasogenic edema around gliomas is often substantial, presumably due to infiltrating “tendrils” of the tumor as it expands. GBM often gives rise to small extensions into adjacent tissue, which is thought to underlie its rapid recurrence. Disrupting these tendrils surgically is unfortunately not feasible as these extensions are not visible by eye or by fluorescent labeling with 5-Aminolevulinic Acid (5-ALA). Radiation or temozolamide also fail to completely remove these extensions. Thus, successfully targeting tumor blood vessels to treat GBM clinically will require a more nuanced understanding of the mechanisms underlying their formation and the crosstalk between the tumor and vascular compartments.

Essential criteria must be met for the optimal design of a GBM drug – 1) the compound must reach all aspects of the tumor, 2) these compounds must be capable of efficiently crossing the BBB, and 3) P-glycoprotein (P-gp) efflux pumps must remove foreign material that bypasses the BBB⁵⁶. Due to the high threshold for realizing a successful GBM-drug design, CNS drugs are less than ~45% likely to enter Phase III clinical trials when compared to drugs that do not target the CNS. The current delivery methods for GBM drugs often result in poor delivery via the cerebral microcirculation, off-target effects, and inability to cross the BBB. In recent years, there have been two major drug delivery advances that show promise for treatment of GBM, nanoparticle-based delivery and pro-drugs. As molecules that are either organic or polymeric, nanoparticles can be engineered to facilitate better transport of the drug across the BBB. Nanoparticles are intended to cross the BBB via receptor-mediated endocytosis⁵⁶. In contrast, a pro-drug is a compound formed between a constituent of molecules and a drug that increases the solubility of the compound. Pro-drugs are intended to cross the BBB by either diffusion or carrier-mediated transport. Current therapies have primarily targeted one of the pathways that have been dysregulated by GBM. Preliminary studies show that next-generation therapies hold substantial promise, and that combinatorial approaches may be more effective forms of GBM treatment. Nonetheless, more attention needs to be given to drug design and pharmacokinetics for critical advancements in the armamentarium for treating GBM.

In contrast to the drug strategies discussed above that result in overt disruption or destruction of the tumor microcirculation, alternative vascular-based therapies for GBM have been proposed. Among these strategies are tumor vessel normalization and ultrasound-based delivery methods. Tumor vessel normalization has emerged as a potentially powerful concept, with the goal of attenuating vascular dysfunction and improving chemotherapeutic and/or immunotherapy delivery and efficacy⁵⁷. The goal of tumor vessel normalization is to alleviate the hypoxia in the microenvironment and stabilize vascular networks within cancerous lesions. A drawback of this method however is that the time-frame for this therapy to be effective is extremely short⁵⁸. Another potential approach to improve drug interactions with the tumor microenvironment revolves around transiently modulating the BBB and GBM microcirculation with mechanical energy, as applied in focused ultrasound therapy. Ultrasound-based methods are being developed to enhance the delivery of therapeutic agents into the tumor parenchyma⁵⁹. Focused ultrasound can transmit sufficient energy to temporarily open the GBM microvasculature and facilitate more widespread drug penetration⁶⁰, while inducing the acoustic cavitation of circulating microbubbles offers another approach to localize mechanical disruption of the vessel wall⁶¹. It is currently unclear exactly how focused ultrasound affects the cells and vBM associated with the NVU. Ultrasound-enhanced drug delivery for GBM can also be utilized with magnetic resonance (MR) imaging⁶², allowing an even greater level of control in distributing agents with specific regions of the tumor mass and limiting off-target delivery. The GBM microvessels therefore represent important conduits for minimally invasive drug delivery as well as potential targets for enhancing GBM treatment.

Stroke and the Microcirculation

Representing a significant cause of disability, morbidity, and mortality worldwide, stroke is defined as neurological impairment arising from a localized injury to the CNS secondary to vascular injury or dysfunction. The specific modes of damage to the CNS vasculature can include an overt blockage (i.e. cerebral infarction) or a breach of the vessel wall, which encompasses intra-cerebral hemorrhages (ICHs) and subarachnoid hemorrhages (SAHs)²⁵. The clinical phenotype of stroke is highly variable and can depend on the neuroanatomical location of the vascular deficit, infarction, or hemorrhage. Thus, a comprehensive view of the vascular neuroanatomy, specifically concerning each unique patient, is critical for diagnosis and clinical management⁶³. Carefully considering the neuroanatomy of each stroke scenario can illuminate whether the cause is vascular-based as well as the specific vascular regions involved. Complex collateral circuits within the cerebral vascular system facilitate general brain perfusion⁶⁴. Improved understanding of pathological disruptions to this collateral circulation system and the downstream microcirculatory networks they supply, forms the basis for advancing stroke treatment and prognosis⁶⁵. Insight into the neurovascular architecture must also be coupled with an in-depth understanding of stroke-induced tissue damage on the cellular and molecular levels. For example, brain ischemia triggers the production of highly reactive and unstable free radicals as well as the release and activation of proteases (e.g. matrix metalloproteinases, MMPs)⁶⁶. A subset of these molecular species fuel tissue degeneration including unraveling the microvasculature and BBB during the acute phase of stroke; these same molecules however perform important functions in the recovery phase²⁶. It is critical therefore to establish the cellular and molecular events that occur immediately after cerebral blood flow ceases within an affected region, and how both the neural and vascular compartments contribute to both acute and recovery phases.

As described above, a primary clinical feature during the early stages of stroke is the loss of BBB integrity, allowing the unregulated movement of water and solutes into the brain parenchyma. Ischemic injuries such as stroke are known to rapidly induce edema¹⁴. The dysfunction of ion channels appears to be the main driver of edema after stroke⁶⁷. In various models, PC deficiency has been shown to compromise microvascular integrity, enhancing BBB permeability, and leading to increased fluid accumulation in perivascular spaces in vivo^28,29. Though a complete understanding of the role PCs play in BBB maintenance continues to emerge, PCs may disassociate from the microvascular ECs in certain disease states such as in stroke (Figure 1). Additionally, it has been proposed that PCs regulate astrocyte end-feet polarization and distribution via regulation of aquaporin-4 (AQP4), α-syntrophin and laminin α2²⁸. In PC-deficient mutants, these markers appeared to be expressed at much lower levels relative to control animals. Aquaporin-4 signal for example was localized to the opposing region of the astrocyte (i.e. within the cell soma and less at the end-feet), indicating aberrant polarization of astrocytes during PC loss²⁸. As astrocytes collaborate with PCs and the brain endothelium to maintain this critical barrier, it becomes critical to determine how astrocytic end-feet organize these unique channels and transporters controlling water and ion homeostasis at the neurovascular interface.

Regarding stroke-induced increases in vessel permeability to solutes, breach of endothelial cell-cell junctions and endothelial transcytosis defects appear to be primary drivers of BBB dysfunction. Like the GBM scenario, data obtained from Krueger et al. using murine and rat models suggest that EC degradation and damage leads to that ischemia-related BBB breakdown²⁷. This group characterized BBB degeneration in a 4 stage classification system by grading vascular damage via tracer extravasation (FITC-albumin): s1 (minimal leakage) to s4 (increased border zones and greater leakage); s0 served as a control (no leakage). This study provides a helpful framework for correlating the severity of vascular injury with the extent of vessel leakage, providing insight into the relationship between cellular damage and disease outcomes. Along with ECs, PCs and astrocytes are responsible for maintaining BBB integrity with respect to solute exchange in the brain^14,26,27. In the context of stroke, PCs are likely involved in pathological alterations to the microvessel wall, as PCs are known to influence EC gene expression patterns including BBB-related mediators (e.g. ZO-1). There is no direct evidence that PC deficiency causes stroke; however, it is suggested that PCs may contribute to stroke pathologies by their effect on vascular integrity. PC modulation of BBB integrity may also stem from their expression of VEGF-A²⁹. Sodium Cyanide (NaCN) treatments substantially increase VEGF-A expression from brain PCs, which can elicit BBB disruption. VEGF-A produced from a variety of cellular sources has been implicated as an inducer of BBB leakage during ischemia and has also been shown to enhance BBB integrity post-ischemia²⁹. Thus, elucidating the exact mechanisms to which PCs contribute in (i) the polarization of astrocytes, (ii) expression of VEGF-A, and (iii) overall BBB maintenance will be of clinical significance in designing next-generation stroke therapies.

One challenge that has been observed in a subset of stroke cases is the inability to completely restore perfusion to downstream brain tissue following clot removal, similar to the coronary no-reflow problem in treating myocardial infarction patients⁶⁸. Recent studies have suggested that microvascular PCs in the brain may be a contributing factor underlying this impairment, specifically by undergoing substantial and persistent vasoconstriction, perhaps even dying and remaining in this constricted state⁶⁹. Recent studies suggest that PC constriction of microvessels causes dramatic changes in oxygen and nitrogen distribution, yielding free radicals that ultimately exacerbate tissue ischemia⁷⁰. This ischemia feeds-forward in the disruption of tight junctions, leading to greater permeability of the BBB⁷¹. These data suggest that PCs within the NVU may be an important target in the development of next-generation stroke therapies. A recent observation from our lab offers an additional consideration in developing this working model. Specifically, live imaging of microvascular PCs within the retina, an extension of the CNS, demonstrated in real-time a PC reducing capillary diameter just before it died and subsequently fragmented (Figure 2). Our findings suggest that, although PC constriction of cerebral capillaries may occur following the loss of blood flow (i.e. in stroke or myocardial infarction), PCs undergoing cell death in these contexts are more likely to experience cytoplasmic fragmentation, as described here and in other scenarios⁷². Thus, while sustained vasoconstriction by PCs in rigor following ischemic stroke may not fully explain residual perfusion deficits following blockage removal, PCs are likely to play critical roles in acute and long-term stroke recovery and therefore represent potential targets for therapeutic stroke intervention on the level of the cerebral microcirculation.

Figure 2. Representative time-lapse imaging of an explanted retina from a “double-reporter” mouse.

(A) Live imaging of ECs (Flk1-eGFP, green) and PCs (NG2-DsRed, red) interacting in a retina explant demonstrates that PCs can constrict vessels as they undergo apoptosis; however, the vessel diameter returns to a previous level of tone as denoted by the small white lines. PCs do not remain intact following cell death as seen by cellular debris near the vessel (arrowheads). Time in the upper left corner of the left column indicates hours and minutes as h:mm. Scale bar, 10 microns. Non-consecutive images were taken from the time-lapse sequence. (B) Vessel diameter changes over time for the movie shown in (A). The arrowheads and gray region indicate the associated time-frame in (A). See Darden et al. Angiogenesis 2019 for full experimental details of tissue explant experiments and animal use certifications.

Present and Potential Ischemic Stroke Treatments

The primary treatment modalities post-ischemic stroke has an emphasis on thrombolysis, mechanical clot removal (mechanical thrombectomy), and clot prevention^73–75. Mechanical thrombectomy has become the primary treatment for acute ischemic stroke resulting from a large vessel occlusion with or without intravenous thrombolytics to re-establish antegrade blood flow^75–77. Other than permissive hypertension to increase cerebral perfusion pressure and augment collateral flow, there does not appear to be an emphasis on treatments that would further enhance blood flow in the collateral vessel system post-stroke, although there is some ongoing work in this arena. The main drugs currently used in ischemic stroke patients include (i) thrombolytics such as tissue plasminogen activators (TPA) including Activase^® (Alteplase) and TNKase^® (Tenecteplase) and (ii) anti-platelet agents such as aspirin and clopidogrel⁷⁸. Thrombolytics, also known as “clot busters,” work by breaking up the offending clots, albeit with some limitations. Clinical practice guidelines recommend Activase^® for acute ischemic stroke, with TNKase^® considered in select cases including those eligible for mechanical thrombectomy and cases with mild symptomology and no overt intracranial occlusions. For longer-term prevention of clot formation, patients suffering an ischemic stroke may be prescribed anti-platelet medications (e.g. aspirin or clopidogrel) or in some cases anti-coagulants (e.g. for atrial fibrillation)⁷⁸. Anti-platelet agents work by “thinning” the blood and preventing additional clot formation by platelet aggregation whereas anti-coagulants target clotting factors that facilitate thrombus formation⁷⁸. While these clinical approaches are the first-line therapies for ischemic stroke patients and have proven effective in improving patient outcomes, deeper insight into the effects of stroke on the microcirculation may yield additional targets to further enhance these strategies.

One hallmark of vascular dysfunction during stroke is the degradation of BBB-associated structures such as endothelial tight junctions^79,80. Disruption of these junctions exacerbates the permeability of solutes across the BBB and is associated with a greater expression of hypoxia-inducible factor-1α (HIF-1α), likely due to the hypoxic environment created by the stroke. Though these symptoms are known and are well established, there currently seems to be a lack of a targeted approach in resolving tight junction disruption or reducing HIF-1a expression post-stroke. The clinical management of stroke has included administering agents that can cross the BBB and providing neuro-protection in models of brain ischemia and injury. For example, 4-hydroxy-2,2,6,6-tetramethylpiperidine-N-oxyl, or TEMPOL, is an anti-oxidant that is stable, membrane-permeable, and elicits perseveration of neural tissue⁷⁹. TEMPOL has also been shown to preserve localization of occludin proteins, which are affiliated with tight junctions even during periods of stress⁷⁹. Additionally, during a stroke, PC coverage is reduced in the blood vessels affected. For an ideal stroke therapy, it would make sense to have a cocktail of drugs that have the capability of recruiting PC, re-establishing tight junction barriers, and reducing hypoxia.

Perspectives

As we learn more about the fundamental mechanisms that orchestrate the development and maintenance of the microcirculation in the CNS, we will certainly draw from those discoveries to advance clinical management of neurovascular-based pathologies. Insight into how blood flow itself organizes cell polarity for ECs and PCs, for instance, may shed light on how these cells respond during the sudden cessation of flow as occurs during ischemic stroke. These mechanisms may fuel BBB dysfunction that disrupts crosstalk with the immune system during onset and progression of GBM, resulting in downstream defects in cerebral microvascular networks. Thus, in GBM, stroke, and other NVU-related diseases, the brain microcirculation offers (i) unique access points to the affected neurological tissue, and (ii) a range of cellular and molecular targets to improve current therapeutic strategies.