Abstract
Alzheimer's disease (AD) is a complex neurodegenerative disorder causing progressive cognitive decline, memory loss, and death of neural tissue. Current research suggests a connection between bulk flow of interstitial fluid and cerebrospinal fluid across the blood–brain barrier and the recently confirmed meningeal lymphatic channels of the brain. The main symptom of interest in AD is the spontaneous aggregation of amyloid beta (Aβ) proteins resulting from increased production or lack of clearance from brain tissues. These protein aggregates manifest as plaques in the capillary and artery lumina and the neuronal and dural tissues of the brain, and are known to contribute to cerebral amyloid angiopathy and a host of other neuroinflammatory conditions. The meningeal lymphatics contain a substantial population of immune cells and also serve as a drain into the deep cervical lymph nodes. In this study we discuss the molecular mechanisms by which Aβ could gain access to meningeal lymphatic channels through the blood–brain interface, including ways in which it can be cleared to preclude aggregation and plaque deposition.
Introduction
The discovery of meningeal lymphatics has prompted a re-examination of many neurological processes to determine their involvement in neurodegenerative disorders, the most common being Alzheimer's disease (AD). 1 The trademark neuropathological observations of AD are neuronal death, cerebral amyloid angiopathy (CAA), extracellular amyloid beta (Aβ) plaque deposition, and neurofibrillary tangles resulting from hyperphosphorylated tau.2,3 Widespread neural decay from AD results in progressive cognitive impairment, memory loss, spatial deficit, and general disorientation.3,4 There is no cure for AD, so current research efforts are aimed toward reversing the accumulation and migration of Aβ aggregates within the blood and lymph vasculature. 3 The molecular origins, intracellular/extracellular accumulation, and protein folding dynamics of Aβ fibrils are of great interest in the search for novel therapeutic targets and treatments.
The recently proposed “glymphatic system” is a major lymph drainage pathway by which water and solutes are dynamically exchanged between cerebrospinal fluid (CSF) and interstitial fluid (ISF). It is in immediate contact with Aβ secreting cells in the interstitium.5,6 Louveau et al. 7 noted that connections between the glymphatic system and the meningeal lymphatics are likely, as there are dural lymphatics flanking communicating veins that serve as receptacles for glymphatic fluid flow. These communicating veins with flanking meningeal vessels are evidence of CSF–ISF exchange and perhaps provide a route of exchange of Aβ with the adjacent meningeal lymphatics. 7 This review will discuss recent research connecting the clearance of Aβ through the meningeal lymphatics as a basis for further research to alleviate symptoms of AD. Furthermore, we provide a compendium of the clinical anatomy, molecular biology, and vascular biology relevant to the immediate connection between Aβ accumulation and clearance through the meningeal vessels.
Materials and Methods
Two electronic databases (MEDLINE and ScienceDirect) and one public search domain were searched by one reviewer. Studies pertaining to meningeal lymphatics, the glymphatic system, AD, Aβ accumulation, and clearance were included, and were selected on the basis of relevance to the intended topics for discussion in the review. Articles between 1900 and 2018 were surveyed, with emphasis on recent publications regarding the meningeal lymphatic vessels. The keywords used for the literature search were as follows: meningeal lymphatics; amyloid-beta; CNS lymphatics and Alzheimer's disease; amyloid-beta clearance; amyloid-beta therapies; glymphatic system; and brain lymphatics.
Results
Studies involving systematic reviews of the relevant literature were obtained. The selective criteria were as follows: studies involving CSF/ISF fluid dynamics, central nervous system (CNS) lymphatics, meningeal lymphatics, glymphatic system, and clearance of Aβ and/or therapeutic treatment of Aβ accumulation. Studies involving biochemistry, protein folding dynamics, and molecular biology were selectively excluded on the basis of immediate relevance. A total of 31 studies and 1 book chapter were reviewed. Eighteen of the included studies concerned new methods for observing Aβ clearance through fluorescent tracing; two involved direct discussion of the glymphatic system and its immediate connections to CSF/ISF fluid flow and the interstitium; 11 were function studies of the meningeal lymphatics; and the book chapter included discussion of aquaporin-4 (AQP-4) protein channels. Currently, meningeal lymphatics are being explored as a route for Aβ clearance, as studies have shown that Aβ can accumulate within these vessels and further drain into the deep cervical lymph nodes.
Discussion
The connection between the brain-wide glial-lymphatic (glymphatic) pathway proposed by Iliff et al. 6 and the meningeal lymphatic vessels, discovered recently by the efforts of Louveau et al. 1 and Aspelund et al., 8 is one route by which the soluble oligomers and insoluble aggregates of Aβ proteins could induce widespread inflammation of the brain parenchyma. 9 Recent research has demonstrated an association between the dynamic exchange of ISF and CSF, the glymphatic system, and the meningeal lymphatics serving as an effective route of Aβ clearance.1,6,7,9,10 Many questions about how Aβ can effectively be cleared through this route point toward the cellular processes that connect these channels.
Cerebrospinal fluid-interstitial fluid flow and solute clearance
Various reports have suggested that Aβ accumulates in ISF drainage pathways of the glymphatic system.2,4,11 Iliff et al. 6 were among the first to provide compelling evidence for this phenomenon, demonstrating that subarachnoid CSF enters the brain parenchyma rapidly through the perivascular lymphatic spaces and then effluxes along with ISF into the parenchyma through directional convective flow. Mathiisen et al. 12 and Iliff et al. 6 suggested that flux of water, CSF–ISF, and ions into the interstitium occurs through astroglial AQP-4 channels that are highly polarized to the astrocytic endfeet. It has been confirmed that astrocytic endfeet participate in a number of transport processes, and also filtration processes at the blood–brain barrier (BBB), by acting as a molecular sieve; however, the exact fluid dynamics remain ambiguous, with much disagreement during the past decade regarding the exact role of APQ-4 channels in fluid exchange. The abluminal surface of astrocytic endfeet exclusively and abundantly expresses AQP-4 channels. 13
Arterial pulsations propelling CSF/ISF along the perivascular spaces (PVSs) were also thought to contribute to the migration of Aβ across the BBB. 6 Using intracisternal injections of fluorescent tracers and in vivo two-photon imaging, Iliff et al. 6 determined that the flow of solutes within subarachnoid CSF and ISF was (i) unidirectional, (ii) bulk convective, and (iii) independent of solute size provided the ratio of solute to pore size is <0.5. They strongly suggested that APQ-4 channels are critical for conveying interstitial solutes along perivenous pathways, and they observed that the clearance of Aβ to followed the same APQ-4-mediated bulk convective efflux as interstitial solutes, in addition to Aβ-receptor-specific efflux, on the basis of the markedly lower rate of clearance of radiolabeled Aβ (∼55%) in AQP-null mice. These findings implied that Aβ clearance as well as fluid accumulation is connected to APQ-4 channels.
Smith et al.14,15 explored the mechanism by which solutes of various molecular weights cross the BBB, seeking to determine if APQ-4 channels have any connection to Aβ clearance. The authors suggested that the observations by Iliff et al. 6 were largely misinterpreted and lacked fundamental physiological support and, therefore, repeated the experiments of Iliff. Instead of the proposed vectorial convective flow from peri-arterial spaces to the interstitium, 6 Smith et al. 15 suggested that the direct visualization of solute transfer did not suggest directionality. They repeated similar experiments by Iliff et al., 6 examining solute flow within the cranial vault using two-photon fluorescence recovery after photobleaching to determine whether convection or diffusion was the primary mechanism driving solute transport into the brain parenchyma. They used fluorescent recovery of fluorescein isothiocyanate-dextrans within a 10 μm diameter disk, divided into four quadrants, proximal to a penetrating arteriole. Their experimental data revealed spatially homogenous kinetics of fluorescence recovery within the quadrated bleached area, in addition to sparse movement (0.4–1.0 μm) of the bleached molecules toward the arteriole ∼10 μm away. 15 These observations favored a diffusive model instead of convective flow. The same authors also tested the size dependence of solute flow from the PVS and visualized the recovery of fluorescent dextrans of various molecular weights injected into the parenchyma. The results strongly supported size-dependent migration of dextrans into the parenchyma and size-independent distribution within the PVS. These results opposed the “glymphatic” view of bulk convective flow into the interstitium and supported a size-dependent mechanism of solute entry into the parenchyma. 15 Furthermore, APQ-4-null mice demonstrated an increased rate of solute transfer across the blood–brain interface, nullifying another supposition of the “glymphatic” hypothesis that APQ-4 channels are intimately involved in solute transport across that interface.14,15
Cellular accumulation of extracellular amyloid beta
Soluble oligomeric Aβ (protofibrils) accumulates within the pyramidal cells of the hippocampus and is considered the most neurotoxic form of Aβ owing to its tendency to interfere with various organelle processes.3,16,17 The soluble protofibrils induce neuronal death through endoplasmic reticulum stress (the “Osaka” mutation) and mitochondrial dysfunction. 17 Intraneuronal aggregation of protofibrils within endosomes/lysosomes induces membrane leakage, also resulting in cytosolic acidification and apoptosis. 16 Aβ1–42 can be cleared within the cell through the ubiquitin-proteasome system (UPS) and by the autophagy-lysosome system.3,4,17 Intracellular Aβ aggregation disrupts and inhibits the UPS by interfering with associated UPS-chaperonins and competitively inhibiting the human 20S proteasome, which is the central core of the multisubunit 26S proteasome complex responsible for degrading the accumulating Aβ.18–20 UPS activity is especially decreased in cells of the hippocampus, peri-hippocampus, and aspects of the temporal and parietal gyri. 20 Microglial phagocytosis is also a route by which soluble Aβ is cleared; however, recent research 21 points to microglia being more active in interacting with fibrillar Aβ at the cell surface. This interaction is followed by an increase in secretory activity and chemotaxis by additional microglia to the location of fibrillar Aβ. Furthermore, pretreating fibrillar Aβ microglial cells with oligomeric Aβ inhibited the phagocytosis significantly, suggesting that soluble Aβ may interfere with the fibrillar Aβ microglial complement system. 21
ISF/CSF clearance of Aβ
The fluid dynamics of solutes in crossing the BBB, a crucial pathological barrier, is especially important as soluble Aβ is commonly observed in the human body at a range of molecular weights. 4 Extracellular clearance of Aβ is addressed in a similar manner as within the cell: secreted proteases as well as phagocytosing microglia and astrocytes can degrade Aβ, but chronic neuroinflammation due to increased chemical secretions by microglia results in inevitable neuronal death. 4 Aβ carried by CSF and ISF can gain access to the BBB by being conveyed along the PVSs with other solutes. Aβ traveling within the bloodstream can deposit around the lumen, resulting in plaque formation, increasing vascular resistance and interfering with solute exchange to the perivascular lymphatics, a marker of CAA. 9 Aβ within the interstitium can flow, presumably through the amended glymphatic mechanism, 15 into the perivenous space and further accumulate in the deep cervical lymph nodes upstream or recycle back into the CSF. Within the blood, Aβ is cleared by red blood cells and monocytes, and when bound to soluble low-density lipoprotein receptor-related protein-1 it is cleared in the periphery through the kidneys and liver. 4 Brain to blood transport of Aβ through CSF exchange and absorption through the arachnoid granulations are primary routes for clearance in the CNS. 4 Aβ in CSF interferes with various aspects of choroid plexus cellular machinery. It induces CSF stasis by suppressing the synthesis of AQP-1 channels, thus attenuating Aβ clearance within CSF and promoting oligomerization and formation of aggregate. 22
Aβ clearance by meningeal lymphatics
The discovery of the glymphatics and their connection to the meningeal lymphatic channels has prompted studies to confirm another route for solute flow and immune traffic, facilitate targeted drug therapies, and open prospects for clearing pathological molecules such as Aβ. Aspelund et al. 8 visualized the meningeal vessels using the classical lymphatic markers Prox1, Lyve1, and VEGFR-3, all expressed by meningeal lymphatic endothelium.1,10 These findings catalyzed further investigations into the molecular and embryological origins of the meningeal lymphatics, owing much to the studies of Louveau et al.1,8,10; however, the precise mechanisms of pathological traffic remained unexplored.
Meningeal lymphatics drain into the deep cervical lymph nodes, whereby cellular debris, solutes, and pathological molecules can be collected. 8 The accumulation and clearance of Aβ through the lymphatic system remained unexplored, as Aβ plaques were only shown to accumulate within the cell,3,4,16,17 in the extracellular space and parenchyma, 21 within CSF and choroid tissue, 22 and within in the bloodstream as symptoms of CAA. 2 There is evidence for Aβ clearance within the lymphatics as suggested by Pappolla et al., 11 demonstrating that in amyloidosis-induced transgenic mice after 15 months of age, plasma Aβ levels had decreased, whereas Aβ within the cervical lymph nodes continued to increase. These findings provide evidence for Aβ accumulation within the lymphatic system of the brain, but the superficial lymph nodes were analyzed, and these serve as a sink for the olfactory spinal nerves and not the meningeal lymphatic. 9
Wen et al. 23 injected recombinant human vascular epithelial growth factor-C (rhVEGF-C) into the cisterna magna of amyloid precursor protein/presenilin 1 (“Swedish” mutation amyloid precursor protein) transgenic mice. rhVGEF-C/VEGFR-3 promoted tube formation (lymphangiogenesis). The total superfical area of lymphatics was significantly higher in rhVEGF-C mice than the control group, suggesting that exogenous rhVEGF-C induces substantial lymphatic regrowth within the dura mater. There was also significant growth of lymphatic vessels within the deep cervical lymph nodes (in direct connection with the meningeal lymphatics), further suggesting total remodeling of the meningeal lymphatic drainage route to the deep cervical lymph nodes. The Aβ plaque burden and the associated cognitive deficits also decrease with administration of rhVGEF-C; 7 days postinjection there was a 23% reduction in parenchymal soluble Aβ, and a 31% reduction in CSF soluble Aβ. Cognitive deficits were also restored as indicated by the Morris water maze, which tests spatial learning and memory: escape latency was greater in the rhVGEF-C mouse group. 23 This pioneering study substantiates the prospect for exploring gene therapies in humans based on the effectiveness of rhVGEF-C in stimulating widespread meningeal lymphatic growth, significant Aβ clearance in overexpressed transgenic mice, and restoration of cognitive deficits from soluble Aβ oligomeric accumulation.
Of special interest is the mechanism by which Aβ produced by the brain reaches the meningeal lymphatics; the meningeal lymphatic vessels do not penetrate the brain parenchyma, so the direct, blind-ended connection of classical lymphatics is not a plausible option. The glymphatic system is a pathway by which Aβ within the CSF could gain access to the interstitial space6,24; soluble Aβ can reach the interstitial space through CSF that is ejected into the brain as it is driven along the PVSs through arterial pulsations and pressure gradient. 24 Aspelund et al. 8 demonstrated a connection between the glymphatic system and the meningeal vessels, revealing a new route by which Aβ could be traced and studied further. 7 Molecular targets along the glymphatic-meningeal lymphatic pathway should, therefore, be investigated as a means of reversing Aβ accumulation and promoting clearance along this route.
The meningeal lymphatics contain a significant population of antigen-presenting cells, providing evidence for their role in immunosurveillance and similarity to classical lymphatics in the periphery. 1 Kim et al. 25 provided further evidence for the connection of Aβ within the meningeal lymphatics, showing that human leukocyte immunoglobulin-like receptor B1, expressed on the surfaces of immune cells, strongly binds Aβ oligomers and induces a signaling cascade ending in synaptic loss and cognitive dysfunction. Neuroinflammation caused by chemical signaling and immune traffic is a hallmark of AD, causing more harm than good; often the inevitable death of neuronal tissue is due to apoptosis induced through the immune response, possible influence by VEGF-C stimulation, or other aberrations in growth factor pathways that involve CNS vasculature.26,27 Research into the development of molecular therapies that suppress immune cells directly interacting with soluble Aβ oligomers should, therefore, be considered.
Conclusion
The inevitable accumulation and plaque formation of Aβ in AD remains seriously misunderstood in many ways. The meningeal lymphatics, which are intimately connected to the glymphatic system, to immune traffic of the brain, and to CSF/ISF and solute reflux, constitute another route by which Aβ can be cleared from brain tissues. Understanding the precise mechanism of Aβ flow within the meningeal lymphatics could provide further insight into ways of preventing accumulation, widespread neuroinflammation, and further neuronal death. Clinical studies involving administration of therapies that induce widespread lymphangiogenesis, whereas also mitigating the chance of fluid accumulation should, therefore, be explored.
Footnotes
Acknowledgment
We thank our editor Katherine N. Brooks.
Author Disclosure Statement
The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this article. No competing financial interests exist.