Aquaporin-4 Functionality and Virchow-Robin Space Water Dynamics: Physiological Model for Neurovascular Coupling and Glymphatic Flow

UC Davis UC Davis Previously Published Works

Title Aquaporin-4 Functionality and Virchow-Robin Space Water Dynamics: Physiological Model for Neurovascular Coupling and Glymphatic Flow.

Permalink https://escholarship.org/uc/item/2rr5f0jj

Journal International journal of molecular sciences, 18(8)

ISSN 1422-0067

Authors Nakada, Tsutomu Kwee, Ingrid L Igarashi, Hironaka et al.

Publication Date 2017-08-18

DOI 10.3390/ijms18081798

Peer reviewed

eScholarship.org Powered by the California Digital Library University of California International Journal of Molecular Sciences

Review Aquaporin-4 Functionality and Virchow-Robin Space Water Dynamics: Physiological Model for Neurovascular Coupling and Glymphatic Flow

Tsutomu Nakada 1,2,* ID , Ingrid L. Kwee 1,2, Hironaka Igarashi 1 ID and Yuji Suzuki 1 ID

1 Center for Integrated Human Brain Science, Brain Research Institute, University of Niigata, Niigata 951-8585, Japan; [email protected] (I.L.K.); [email protected] (H.I.); [email protected] (Y.S.) 2 Department of Neurology, University of California, Davis, VANCHCS, Martinez, CA 94553, USA * Correspondence: [email protected]; Tel.: +81-25-227-0677; Fax: +81-25-227-0821

Received: 23 July 2017; Accepted: 16 August 2017; Published: 18 August 2017

Abstract: The unique properties of brain capillary endothelium, critical in maintaining the blood-brain barrier (BBB) and restricting water permeability across the BBB, have important consequences on fluid hydrodynamics inside the BBB hereto inadequately recognized. Recent studies indicate that the mechanisms underlying brain water dynamics are distinct from systemic tissue water dynamics. Hydrostatic pressure created by the systolic force of the heart, essential for interstitial circulation and lymphatic flow in systemic circulation, is effectively impeded from propagating into the interstitial fluid inside the BBB by the tightly sealed endothelium of brain capillaries. Instead, fluid dynamics inside the BBB is realized by aquaporin-4 (AQP-4), the water channel that connects astrocyte cytoplasm and extracellular (interstitial) fluid. Brain interstitial fluid dynamics, and therefore AQP-4, are now recognized as essential for two unique functions, namely, neurovascular coupling and glymphatic flow, the brain equivalent of systemic lymphatics.

Keywords: Hagen-Poiseuille equation; starling resister; rCBF; interstitial ﬂow; glia limitans externa

1. Introduction: Blood-Brain Barrier The blood-brain barrier (BBB) is the result of unique properties of brain capillary endothelium, namely the presence of tight junctions and active suppression of aquaporin-1 (AQP-1) expression, the water channel abundantly expressed in common capillaries and choroid plexus epithelium [1–4]. The BBB effectively isolates the brain from the systemic environment through its selectivity. Brain capillary endothelium has no fenestrations, and has highly restricted permeability to ions and water as evidenced by a significantly high electrical resistance [5,6]. There is virtually no non-specific transcellular and paracellular diffusion of hydrophilic compounds across the BBB. Instead, brain capillary endothelium has specifically localized receptors and transporters responsible for the active transport of nutrients to the brain interstitium. Tight junctions of the brain endothelial cells are structurally similar to epithelial tight junctions. The major transmembrane proteins in tight junctions include occludins and claudins [5]. While some accessory proteins specific to brain endothelium, such as cingulin, AF-6 and 7H6, have been identified, other epithelial tight junction proteins have thus far not been detected. Epithelial tight junctions, including choroid plexus epithelial cells, mostly contain claudin-1, -2 and -11. By contrast, brain endothelial tight junctions primarily express claudin-3 and -5. Accumulating evidence indicates that claudin-3 and -5, together with occludin, is responsible for controlling BBB paracellular permeability [5,6]. Thus far, only claudin-2, which is found on so-called leaky epithelium but not brain endothelium, has been shown to be a water channel [7]. Considering that expression of AQP-1, abundant in common

Int. J. Mol. Sci. 2017, 18, 1798; doi:10.3390/ijms18081798 www.mdpi.com/journal/ijms Int. J. Mol. Sci. 2017, 18, 1798 2 of 14 Int. J. Mol. Sci. 2017, 18, 1798 2 of 14 capillaries and choroid plexus epithelium, is actively suppressed in brain capillary endothelium [2,3], it is capillary endothelium [2,3], it is consonant that claudin‐2 is also absent in the tight junctions of brain consonant that claudin-2 is also absent in the tight junctions of brain capillary endothelium (Figure1). capillary endothelium (Figure 1). Studies to date have not identified any specific passage for water to the BBB. It appears that Studies to date have not identified any specific passage for water to the BBB. It appears that water entry into the inside of the BBB is non-specific, presumably slow movement through the lipid water entry into the inside of the BBB is non‐specific, presumably slow movement through the lipid membrane [8]. It follows that brain fluid dynamics would be distinct from systemic fluid dynamics. membrane [8]. It follows that brain fluid dynamics would be distinct from systemic fluid dynamics. We present here a brief review of the recent findings on the physiological and molecular mechanisms We present here a brief review of the recent findings on the physiological and molecular mechanisms of brain fluid dynamics and its impact on two essential processes, namely, neurovascular coupling and of brain fluid dynamics and its impact on two essential processes, namely, neurovascular coupling glymphatic flow. and glymphatic flow.

Figure 1.1. Capillaries.Capillaries. Common Common capillaries capillaries have have a aleaky leaky endothelium endothelium due due to to the presence ofof fenestrations.fenestrations. The The water channel aquaporin-1aquaporin‐1 (AQP (AQP-1)‐1) is also abundantlyabundantly expressed.expressed. Accordingly, waterwater dynamics between between intracapillary intracapillary and and interstitial interstitial fluid fluid space space is isdirectly directly connected connected (right). (right). In Incontrast, contrast, brain brain capillaries capillaries lack lackfenestrations fenestrations and have and havetight junctions. tight junctions. Expression Expression of AQP of‐1 is AQP-1 actively is activelysuppressed. suppressed. As a result, As water a result, dynamics water dynamics in the intr inacapillary the intracapillary space and space interstitial and interstitial fluid space fluid are spaceeffectively are effectively isolated from isolated each fromother each and othermust andbe analyzed must be independently. analyzed independently. Modified from Modified Reference from Reference[9]. [9].

2. Blood Flow Dynamics

2.1. Basics Classic blood flowflow dynamicsdynamics cancan bebe givengiven byby thethe Hagen-PoiseuilleHagen‐Poiseuille equationequation wherewhere volumetricvolumetric blood flow,flow, ΦΦ,, isis givengiven as:as: π ∆P 4 Φ = ∆ R Φ =8η L 8 where ∆P is pressure loss (differences in inflow and outflow pressure), L is the length of the vessel tube,whereη ΔisP blood is pressure viscosity, loss and (differencesR is the radius in inflow of the and vessel outflow [10]. pressure), L is the length of the vessel tube,Under η is blood physiological viscosity, conditions,and R is theL radiusand η ofcan the be vessel treated [10]. as constant, and regional blood flow, Φr, withinUnder the unit physiological area can be conditions, given as: L and η can be treated as constant, and regional blood flow, Φr, 4 within the unit area can be given as: Φr ∼ ∆P · R In the systemic circulation, the autonomicΦ nervous~∆ ∙ system is responsible for Φr by controlling both ∆P andIn theR through systemic neural circulation, and/or the chemical autono controlmic nervous of contractile system structuresis responsible of cardiac for Φr and by controlling circulatory systems.both ∆P Inand contrast, R through capillaries neural lack and/or contractile chemical structures control and, of contractile therefore, capillary structures blood of cardiac flow cannot and becirculatory directly controlled systems. In by contrast, the autonomic capillaries nervous lack system. contractile Although structures contractile and, therefore, function ofcapillary pericytes blood and itsflow neurotransmitter cannot be directly control controlled has been by proposed the autonomi [11], consideringc nervous system. its common, Although wide spreadcontractile functionality function relatedof pericytes to angiogenesis and its neurotransmitter and stem cell control like behavior has been [12 ],proposed it is difficult [11], toconsidering accept that its pericytes common, are wide the primaryspread functionality structural component related to forangiogenesis flow regulation. and stem It is, cell therefore, like behavior highly [12], plausible it is difficult to consider to accept that that pericytes are the primary structural component for flow regulation. It is, therefore, highly

Int. J. Mol. Sci. 2017, 18, 1798 3 of 14 capillary ﬂow dynamics are a passive phenomenon which are signiﬁcantly affected by the architectural properties of the capillaries per se.

2.2. Common Capillaries and Tissue Perfusion Common capillaries in the systemic circulation have a leaky endothelium. Water flow between intra-capillary and interstitial fluid is relatively free and follows the forces defined by the Starling equation: Jv = LpS [Pc − Pi] − σ πp − πi where Jv is the trans endothelial fluid filtration volume per second, Pc is the capillary hydrostatic pressure, Pi is the interstitial hydrostatic pressure, πp is the plasma protein oncotic pressure, πi is the interstitial oncotic pressure, Lp is the hydraulic conductivity of the membrane, S is the surface area for filtration, and σ is the Staverman’s reflection coefficient, respectively [13]. The term [Pc − Pi] − σ πp − πi in the Starling equation represents the net driving force for the trans endothelial fluid filtration volume per second, Jv. Under physiological conditions, σ πp − πi and Pi will be virtually constant and, therefore, regional flow per second, Φr per second, and hence averaged ∆P per second, will directly translate into Jv, defining tissue perfusion. Simply stated, for systemic, leaky capillaries, R remains constant, Kr.

4 Φr ∼ ∆P · R

R = Kr

Accordingly, Φr is a function of ∆P.

Φr ∼ ∆P Therefore, as far as tissues with common capillaries are concerned, the hydrostatic pressure field generated by the heart is by far the most important, if not sole, factor governing tissue perfusion. The systolic force of the heart effectively extrudes water out of the capillaries without much resistance into the interstitial fluid system as well as helps create significant interstitial fluid motion necessary for interstitial fluid circulation (see below). In clear contrast, brain capillaries have significantly different water permeability characteristics due to the presence of the BBB. Therefore, the principles of regional blood flow and interstitial fluid circulation described for common capillaries are not applicable to regional cerebral blood flow (rCBF).

2.3. Cerebral Autoregulation Cerebral autoregulation signifies an intrinsic ability of cerebral vasculature to maintain cerebral blood flow at a relatively constant rate of ca. 50 mL per 100 g brain tissue per minute in the face of blood pressure changes [14–19]. Autoregulation generally functions between mean blood pressures of 60 and 150 mm Hg. It is preserved in animals that have undergone parasympathetic and/or sympathetic denervation [14], and the system is independent from extrinsic neural control. Instead, intrinsic neural nitric oxide (NO) control [19] and release of vasoactive substrates by the brain are believed to play essential roles in maintaining constant cerebral perfusion [16,17]. Perfusion is held constant by means of cerebral vasculature smooth muscle constriction and dilation in response to elevated and decreased systemic pressure, respectively [14–19]. In the context of the principles of the blood flow dynamics described above, cerebral autoregulation translates to rigorous maintenance of a constant ∆P (differences in inflow and outflow pressure), Kp.

4 Φr ∼ ∆P · R Int. J. Mol. Sci. 2017, 18, 1798 4 of 14

∆P = Kp If brain capillaries have structural properties similar to common capillaries, rCBF would be essentially constant. Φr ∼ ∆P ∼ Kp This is indeed the primary functionality of cerebral autoregulation, which rigorously maintains a rather constant perfusion of the brain in the presence of signiﬁcant ﬂuctuation in systemic circulation associated with various physiological activities [14–19].

2.4. Neurovascular Coupling Increased rCBF associated with brain activation is a well-recognized phenomenon that is known as neurovascular coupling. It is relatively small increase in rCBF compared to steady rCBF, which is rigorously regulated by autoregulation [20]. Nevertheless, the rCBF increase associated with neurovascular coupling seemingly contradicts the purpose of cerebral autoregulation and, therefore, neurovascular coupling should have a role essential to maintain brain functionality and closely related molecular phenomena associated with neural activities. It was once thought that neurovascular coupling serves to ensure adequate neuronal nutrient supply. This intuitively appealing notion failed to account for the large quantitative discrepancy between supply and demand. The amount of essential nutrients delivered by the increased rCBF, such as oxygen and glucose, exceeds actual consumption by more than six times. Such a large discrepancy is virtually unknown in any other biological system, indicating that a factor other than nutrient supply underlies the observed disproportionate increase in rCBF [20]. Subsequent studies indicate that increased rCBF associated with brain activation serves as a heat removal mechanism. Information processing by brain generates considerable heat and water flow is the primary means of heat removal. It is therefore evident that the apparent surfeit in rCBF increase serves as a quick removal system of the additional heat generated by neural activities [21,22]. Neurovascular coupling is a micro, rather than macro environmental event occurring within an area limited to 250 µm around the site of neural activity [23]. Therefore, the regulatory mechanism for neurovascular coupling should be within the capillaries, independent from cerebral autoregulation per se. Brain capillaries have a very tight endothelium which severely restricts water permeability. Therefore, hydrostatic pressure differences between intra-capillary fluid and interstitial fluid affects the capillary’s structure as in the case of a Starling resistor [24]. This can provide the environment where small rCBF increase associated with neural activities, Φnvc, under rigorous control of rCBF by autoregulation is a function of small changes in capillary diameter ∂R.

4 Φnvc ∼ ∂R

2.5. Flow and Pressure in a Starling Resistor The experimental device consisted of an elastic tube clamped between two rigid pipes, surrounded by an outer pressure chamber, known as a Starling resistor, and was used in an isolated-heart preparation by Ernest Henry Starling. Studies using the device eventually led to the Frank-Starling law of the heart [13,24] (Figure2). The static pressure of the outer pressure chamber controls the degree of collapse of the tube, providing a variable resistor to simulate total peripheral resistance. As a rule, the device initially shows linear behavior, namely, linear changes in tube diameter that corresponds to the relationship between the pressure of the outer chamber and hydrodynamic pressure in the tube. It eventually shows non-linear behavior, leading to two non-linear phenomena known as the waterfall effect and self-excited oscillations, the biological equivalents of which represent expiratory ﬂow limitation in chronic obstructive pulmonary disease (COPD) and snoring, respectively [25]. Int.Int. J. J. Mol. Mol. Sci. Sci. 20172017, ,1818, ,1798 1798 55 of of 14 14

Figure 2. The experimental set-up known as a Starling resistor. An elastic tube clamped between Figurerigid tubes 2. The is experimental surrounded by set an‐up outer known pressure as a Starling chamber. resistor. Flow An is driven elastic throughtube clamped the tube between from leftrigid to tubesright byis surrounded an applied pressure by an outer differences, pressureP chamber.up-Pdn. The Flow pressure is driven of the through chamber, theP tubeext, can from be left adjusted to right to bycontrol an applied the average pressure diameter differences, of the Pup elastic‐Pdn. The tube. pressurePup: upstream of the chamber, pressure; PextP, dncan: downstream be adjusted to pressure; control thePext :average external diameter chamber pressure.of the elastic tube. Pup: upstream pressure; Pdn: downstream pressure; Pext: external chamber pressure. Brain capillaries, owing to tight endothelium, have virtually identical physiological conditions as aBrain Starling capillaries, resistor. owing The brain to tight capillary endothelium, is a biological have virtually pliable identical tube clamped physiological between conditions two rigid astubes, a Starling namely, resistor. arteriole The (artery)brain capillary and venule is a biological (vein) as depictedpliable tube in Figureclamped2. Cerebralbetween two autoregulation rigid tubes, namely, arteriole (artery) and venule (vein) as depicted in Figure 2. Cerebral autoregulation rigorously maintains hydrodynamic pressure corresponding to Pup-Pdn. Similar to the elastic tube in a rigorouslySterling resistor, maintains the diameterhydrodynam of whichic pressure is controlled corresponding by the pressure to Pup‐Pdn of. Similar the outer to pressurethe elastic chamber, tube in a Sterling resistor, the diameter of which is controlled by the pressure of the outer pressure chamber, Pext, in Figure2. The diameter of a given brain capillary, Rcap, is a function of the static pressure of the Pext, in Figure 2. The diameter of a given brain capillary, Rcap, is a function of the static pressure of the surrounding interstitial ﬂuid in the peri-capillary space (Virchow-Robin space), PVRS. surrounding interstitial fluid in the peri‐capillary space (Virchow‐Robin space), PVRS. 1 1 Rcap ~∼ PVRS

3. AQP-4 and Neurovascular Coupling 3. AQP‐4 and Neurovascular Coupling 3.1. Virchow-Robin Space and Interstitial Flow 3.1. Virchow‐Robin Space and Interstitial Flow Fluid-filled canals surrounding perforating arteries, capillaries, and veins in the brain parenchyma wereFluid recognized‐filled canals early in surrounding modern medicine perforating and are arteries, referred capillaries, to as the Virchow-Robin and veins in space the (VRS),brain parenchymabased on the were first two recognized scientists early who in described modern the medicine structures and in are detail, referred Rudolph to as Virchow the Virchow and Charles‐Robin spacePhilippe (VRS), Robin based [26 on,27 ].the It first was two soon scientists recognized who thatdescribed interstitial the structures fluid in VRS in detail, was dynamic,Rudolph Virchow and this andfluid Charles movement Philippe was termedRobin [26,27]. interstitial It was flow. soon Interstitial recognized fluid that in interstitial the peri-capillary fluid in VRS space was moves dynamic, along andthe perforatingthis fluid movement artery and was vein termed and eventually interstitial drains flow. into Interstitial the cerebrospinal fluid in fluidthe peri (CSF)‐capillary system. space movesUnlike along commonthe perforating capillaries, artery water and vein flow and from eventually brain capillary drains into to VRS the cerebrospinal is significantly fluid restricted (CSF) system.and not sufficient for generating interstitial flow. Recent studies disclosed that water dynamics in peri-capillaryUnlike common VRS and, capillaries, hence, interstitial water flow flow, from is maintained brain capillary by aquaporin-4 to VRS is significantly (AQP-4), a restricted water channel and notabundantly sufficient expressed for generating at the interstitial endfeet of flow. peri-capillary Recent studies astrocytes. disclosed AQP-4 that connects water dynamics the intracellular in peri‐ capillaryspace of astrocytesVRS and, hence, and extracellular interstitial flow, (interstitial) is maintained space at by capillary aquaporin VRS,‐4 (AQP thereby‐4), regulatinga water channel water abundantlyefflux essential expressed for peri-capillary at the endfeet VRS waterof peri dynamics‐capillary [28 astrocytes.–30]. As discussed AQP‐4 connects above, peri-capillary the intracellular VRS space of astrocytes and extracellular (interstitial) space at capillary VRS, thereby regulating water hydrostatic pressure, PVRS, is the determinant of brain capillary diameter, Rcap. It follows that AQP-4 effluxactivity essential is also the for main peri‐ factorcapillary governing VRS water brain dynamics capillary [28–30]. diameter As (Figure discussed3). above, peri‐capillary VRS hydrostatic pressure, PVRS, is the determinant of brain capillary diameter, Rcap. It follows that AQP‐4 activity is also the main factor1 governing1 brain capillary diameter (Figure 3). Rcap ∼ ∼ ∼ AQP4 Supression 1PVRS 1AQP4 Activities ~ ~ ~4 4 Since rCBF increases by a factor of four of the capillary diameter, Φ ∼ R4, rCBF increases Since rCBF increases by a factor of four of the capillary diameter, rΦ ~ , rCBF increases significantly when there is a minor reduction in AQP-4 activities, which consequently results in significantly when there is a minor reduction in AQP‐4 activities, which consequently results in expansion of capillary diameter. Indeed, the AQP-4 inhibitor, TGN-020, was found to increase rCBF expansion of capillary diameter. Indeed, the AQP‐4 inhibitor, TGN‐020, was found to increase rCBF significantly [31]. significantly [31].

Int.Int. J. J. Mol. Mol. Sci. Sci. 20172017, 18, 18, 1798, 1798 66 of of 14 14

Figure 3. Peri-capillary VRS water dynamics. Brain capillary endothelial tight junctions and active Figuresuppression 3. Peri‐ ofcapillary AQP-1 VRS expression water dynamics. highly restricts Brain capillary water movement endothelial across tight thejunctions BBB. Byand contrast, active suppressionsignificant water of AQP flow‐1 (blue expression arrow) ishighly present restricts inside thewater BBB movement within VRS across (interstitial the BBB. flow), By mediated contrast, by significantactive water water inflow flow through (blue arrow) AQP-4. is Modifiedpresent inside from Referencethe BBB within [9]. VRS (interstitial flow), mediated by active water inflow through AQP‐4. Modified from Reference [9]. 3.2. Neural Activities and AQP-4 Suppression 3.2. Neural Activities and AQP‐4 Suppression The role of AQP4 in neural signal transduction was first proposed following the observation that The role of AQP4 in neural signal transduction was first proposed following the observation that deletion of α-syntrophin in mice is associated with prolonged potassium (K+) clearance [32]. In the deletion of α‐syntrophin in mice is associated with prolonged potassium (K+) clearance [32]. In the hippocampus, the rapid water fluxes driven by AQP-4 are found to be required for maintaining K+ hippocampus, the rapid water fluxes driven by AQP‐4 are found to be required for maintaining K+ homeostasis during electrical activity [24]. Inwardly rectifying potassium channels, Kir4.1, responsible homeostasis during electrical activity [24]. Inwardly rectifying potassium channels, Kir4.1, for the rapid removal of K+ from extracellular (interstitial) fluid essential for maintaining neural responsible for the rapid removal of K+ from extracellular (interstitial) fluid essential for maintaining excitability, are co-localized with AQP-4 on astrocyte endfeet facing capillary endothelium at the neural excitability, are co‐localized with AQP‐4 on astrocyte endfeet facing capillary endothelium at VRS [33]. This K+ spatial buffer mechanism is responsible for the rapid redistribution of K+ by a the VRS [33]. This K+ spatial buffer mechanism is responsible for the rapid redistribution of K+ by a current flow termed potassium siphoning. current flow termed potassium siphoning. K+ entry generates a local depolarization, which propagates electrotonically through individual K+ entry generates a local depolarization, which propagates electrotonically through individual glial cells and through the glial cell syncytium [34,35]. The overall efficiency of the spatial buffer process glial cells and through the glial cell syncytium [34,35]. The overall efficiency of the spatial buffer is thought be dependent, in part, on the electrical space constant of the glial cell syncytium. However, process is thought be dependent, in part, on the electrical space constant of the glial cell syncytium. it is known that potassium siphoning within a single glial cell is sufficient. K+ influx occurring in one However, it is known that potassium siphoning within a single glial cell is sufficient. K+ influx region of the glial cell is balanced by efflux at another region of the astrocyte, typically, at the endfeet. occurring in one region of the glial cell is balanced by efflux at another region of the astrocyte, While K+ influx occurs primarily through Kir, K+ efflux occurs through the gap junctions [34–37]. typically, at the endfeet. While K+ influx occurs primarily through Kir, K+ efflux occurs through the Kir distribution corresponds to the highly polarized localization of AQP-4 [32–34]. gap junctions [34–37]. Kir distribution corresponds to the highly polarized localization of AQP‐4 [32– Kir4.1 and AQP-4 are considered to belong to the dystrophin-glycoprotein complex (DGC) 34]. family [34]. The HCO − independent proton pump, vacuolar ATPase (V-ATP), expressed abundantly Kir4.1 and AQP‐34 are considered to belong to the dystrophin‐glycoprotein complex (DGC) in astrocytes, is also considered part of the DGC family [38]. The functional coupling of Kir, family [34]. The HCO3− independent proton pump, vacuolar ATPase (V‐ATP), expressed abundantly AQP-4 and V-ATP with respect to neural activation has become evident [9]. As suggested by Hibino in astrocytes, is also considered part of the DGC family [38]. The functional coupling of Kir, AQP‐4 and Kurachi [39], this co-localization may be the result of a clustering of proteins in specialized lipid and V‐ATP with respect to neural activation has become evident [9]. As suggested by Hibino and raft domains which bring together functionally coupled molecules to a common site on the membrane. Kurachi [39], this co‐localization may be the result of a clustering of proteins in specialized lipid raft The implications of disruption in this coupled functionality, including AQP-4, with respect to neural domains which bring together functionally coupled molecules to a common site on the membrane. functionality are only beginning to be understood. The implications of disruption in this coupled functionality, including AQP‐4, with respect to neural AQP-4 suppression associated with neural activities can be seen as part of a cascade initiated by functionality are only beginning to be understood. rapid K+ flux into the extracellular space by production of action potentials. Rapid spatial buffering by AQP‐4 suppression associated with neural activities can be seen as part of a cascade initiated by Kir promotes release of its counterpart cation, H+, by V-ATP, resulting in extracellular acidification. rapid K+ flux into the extracellular space by production of action potentials. Rapid spatial buffering The increase in extracellular H+ inhibits AQP-4 activity at VRS [40] (Figure4). by Kir promotes release of its counterpart cation, H+, by V‐ATP, resulting in extracellular acidification. The increase in extracellular H+ inhibits AQP‐4 activity at VRS [40] (Figure 4).

Int. J. Mol. Sci. 2017, 18, 1798 7 of 14 Int. J. Mol. Sci. 2017, 18, 1798 7 of 14

Figure 4. MolecularMolecular processes processes within within the the Virchow–Robin space associated with neural activation. Neural activation induces induces potassium potassium efflux efflux into into th thee interstitial interstitial space space which which is quickly is quickly scavenged scavenged by byinwardly inwardly rectifying rectifying potassium potassium channel channel (Kir). (Kir). This This potassium potassium spatial spatial buffering buffering promotes promotes release release of a ofcounterpart a counterpart cation, cation, H+, Hby+ ,vacuolar by vacuolar ATPase, ATPase, resulting resulting in inprot protonon excess excess in in the the interstitial interstitial fluidfluid (extracellular acidosis).acidosis). AQP-4AQP‐4 is inhibited by higher proton dens densityity (red arrow). Excess protons move into thethe intracapillary intracapillary space space without without significant significant resistance resistance at the at tight the junctions tight junctions because ofbecause the Grotthuss of the protonGrotthuss tunneling proton mechanism.tunneling mechanism. Carbonic anhydrase Carbonic anhydrase type IV (CA-IV) type IV anchored (CA‐IV) toanchored the luminal to the surface luminal of capillariessurface of capillaries and NBC1 and sodium NBC1 bicarbonate sodium bicarbonate co-transporter co‐transporter that interacts that directly interacts with directly CA-IV with effectively CA‐IV scavengeeffectively these scavenge excess these protons. excess Modified protons. from Modified Reference from [9Reference]. [9].

3.3. AQP AQP-4‐4 Suppressionand Neurovascular Coupling AQP-4AQP‐4 suppressionsuppression initiated initiated by neuralby neural activities activities which triggerswhich triggers rapid K+ rapidflux into K+ the flux extracellular into the spaceextracellular described space above described is the above likely is underlying the likely underlying molecular mechanismmolecular mechanism of neurovascular of neurovascular coupling. Indeed,coupling. in Indeed, humans, in thehumans, area of the extracellular area of extracellular acidification acidification associated associated with regional with neural regional activity neural is confirmedactivity is confirmed to be identical to be to identical the area to of the rCBF area increase of rCBF [41 increase]. A decline [41]. in A interstitial decline in fluid interstitial hydrostatic fluid pressurehydrostatic due pressure to suppression due to ofsuppression AQP-4 results of AQP in caliber‐4 results expansion in caliber of the expansion tightly sealed of the brain tightly capillaries sealed and,brain hence,capillaries increased and, rCBF.hence, This increased phenomenon rCBF. alsoThis explainsphenomenon a curious also co-phenomenon,explains a curious namely, co‐ astrocytephenomenon, swelling namely, associated astrocyte with swelling neural activities. associated AQP-4 with suppressionneural activities. by extracellular AQP‐4 suppression proton excess by leadsextracellular to relative proton intracellular excess leads water to excess,relative in intr turnacellular producing water astrocyte excess, swelling.in turn producing This phenomenon astrocyte hasswelling. been shownThis phenomenon to be absent has in AQP-4 been shown knock to out be mice absent [42 in] (Figure AQP‐45 ).knock out mice [42] (Figure 5).

Int. J. Mol. Sci. Sci. 20172017,, 1818,, 1798 1798 8 of 14

Figure 5. At the VRS, neural activation produces extracellular acidification accompanied by increase inFigure rCBF 5. and At astrocytethe VRS, neural swelling. activation Proton inhibitionproduces extr of AQP-4acellular results acidification in a reduction accompanied of water by flow increase from astrocytesin rCBF and into astrocyte the peri-capillary swelling. Proton Virchow-Robin inhibition of space, AQP‐ reduction4 results in of a peri-capillaryreduction of water fluid flow pressure, from capillaryastrocytes lumen into the expansion, peri‐capillary increase Virchow rCBF and‐Robin astrocyte space, swelling. reduction Modified of peri‐ fromcapillary Reference fluid [pressure,9]. capillary lumen expansion, increase rCBF and astrocyte swelling. Modified from Reference [9]. 4. AQP-4 and Interstitial Fluid Circulation 4. AQP‐4 and Interstitial Fluid Circulation 4.1. Interstitial Flow as Brain Lymphatic Equivalent–Glymphatic Flow 4.1. Interstitial Flow as Brain Lymphatic Equivalent–Glymphatic Flow Interstitial flow has been postulated to play a role similar to systemic lymphatics for the brain, whichInterstitial lacks a conventional flow has been lymphatic postulated system to [play43–46 a ].role The similar concept to has systemic recently lymphatics been revived for inthe relation brain, towhich CSF lacks dynamics a conventional and β-amyloid lymphatic clearance system [43,44 [43–46,47,48].]. The It is nowconcept clear has that recently the classic been view revived of CSF in circulationrelation to whereCSF dynamics the choroid and plexus β‐amyloid is solely clearance responsible [43,44,47,48]. for CSF productionIt is now clear is invalid that the [49 classic]. Although view choroidof CSF circulation plexus also where produces the choroid CSF, interstitial plexus is solely fluid entersresponsible the CSF for systemCSF production and contributes is invalid to [49]. CSF volumeAlthough significantly choroid plexus [29,49 also]. Thisproduces brain CSF, lymphatic interstitial equivalent fluid enters system the is CSF now system widely and referred contributes to as glymphaticsto CSF volume [48 significantly,50], denoting [29,49]. glial lymphatics. This brain lymphatic equivalent system is now widely referred to as Asglymphatics the lymphatic [48,50], equivalent denoting of theglial brain, lymphatics. the glymphatic system is anticipated to have a functional configurationAs the lymphatic similar to thatequivalent of the systemicof the brain, lymphatic the glymphatic system [51]. system In the systemic is anticipated circulation to have where a common,functional leakyconfiguration capillaries similar exist, hydrostaticto that of the pressure systemic generated lymphatic by thesystem systolic [51]. force In the of thesystemic heart effectivelycirculation extrudeswhere common, water out leaky ofthe capillaries capillaries exist, without hydrostatic much resistancepressure generated into the interstitial by the systolic fluid space,force of and the helps heart create effectively interstitial extrudes fluid water motion out necessary of the capillaries for interstitial without circulation much [resistance13,15]. Lymphatic into the capillariesinterstitial fluid have space, primary and valves helps whichcreate preventinterstitial backflow fluid motion effecting necessary one way for flow.interstitial The hydrostaticcirculation pressure[13,15]. Lymphatic created by capillaries the systolic have force primary of the heart valves is usuallywhich prevent sufficient backflow to return effecting interstitial one fluid way toflow. the systemicThe hydrostatic circulation pressure although created an occasionalby the systolic build-up force of of interstitial the heart fluid is usually can occur sufficient causing to edema,return especiallyinterstitial influid areas to vulnerablethe systemic to gravitationalcirculation although force [13 an,51 occasional]. build‐up of interstitial fluid can occurIn causing contrast, edema, the endothelium especially in of areas brain capillariesvulnerable is to effectively gravitational sealed force by tight[13,51]. junctions, the primary structureIn contrast, responsible the endothelium for the BBB [ 1of–4 ].brain Active capillaries suppression is effectively of AQP-1 sealed expression by tight in brain junctions, capillaries the furtherprimary ensures structure restriction responsible of waterfor the movement BBB [1–4]. across Active the suppression BBB [5]. Fromof AQP a‐ hydrostatic1 expression dynamics in brain standpoint,capillaries further the BBB ensures also impedesrestriction propagation of water movement of the systolic across force the ofBBB the [5]. heart From across a hydrostatic capillary endothelium.dynamics standpoint, Accordingly, the BBB the combinedalso impedes restricted propagat waterion mobility of the systolic from capillaries force of tothe the heart interstitial across spacecapillary of the endothelium. brain and blocked Accordingly, hydrostatic the combined force are insufficientrestricted water to produce mobility interstitial from capillaries fluid movement to the withininterstitial the BBB.space Therefore, of the brain to and achieve blocked interstitial hydrostatic fluid dynamicsforce are insufficient in the brain to similar produce to interstitial that of systemic fluid lymphatics,movement within additional the BBB. mechanisms Therefore, for to increasingachieve inters watertitial volume/contents fluid dynamics in and the hydrostatic brain similar pressure to that withinof systemic the peri-capillarylymphatics, additional interstitium mechanisms are essential for increasing to ensure wate glymphaticr volume/contents flow. Water and flux hydrostatic through AQP-4pressure into within peri-capillary the peri‐capillary VRS provides interstitium the requisite are essential water flow to ensure within theglymphatic peri-capillary flow. interstitiumWater flux (Figurethrough6 ).AQP‐4 into peri‐capillary VRS provides the requisite water flow within the peri‐capillary interstitium (Figure 6).

Int. J. Mol. Sci. 2017, 18, 1798 9 of 14 Int. J. Mol. Sci. 2017, 18, 1798 9 of 14

FigureFigure 6. 6. SchematicSchematic presentation presentation of of wave wave propagation propagation simulation simulation for for conceptualizing conceptualizing water water dynamics dynamics insideinside the the BBB. BBB. From From a a hydrodynam hydrodynamicic standpoint, standpoint, the the tight tight endothelium endothelium of of brain brain capillaries capillaries forming forming thethe BBB BBB is seriouslyseriously ﬂawed.flawed. While While hydrostatic hydrostatic pressure pressure generated generated by by the the systolic systolic force force of the of heartthe heart (blue (bluearrow) arrow) provides provides the necessary the necessary kinetics kinetics in the casein the of fenestratedcase of fenestrated capillaries capillaries (left), sealed (left capillaries), sealed capillariesimpede propagation impede propagation of such a force of such (middle a force). The (middle addition). ofThe AQP-4 addition driven of waterAQP‐4 ﬂow driven reconstitutes water flow the reconstitutesnecessary environment the necessary (right). Diagramsenvironment are created (right using). Diagrams wave propagation/interference are created using simulation wave propagation/interferencefor two windows. simulation for two windows.

4.2.4.2. Glymphatic Glymphatic Flow Flow and and β‐βAmyloid-Amyloid Clearance Clearance TheThe concept thatthat glymphatic glymphatic flow flow is theis systemicthe systemic lymphatic lymphatic equivalent, equivalent, in relation in relation to CSF dynamics to CSF dynamicsand β-amyloid and clearance,β‐amyloid has clearance, recently becomehas recently a major become focus of a medical major investigationfocus of medical [30,43 investigation,47,50,52]. It is [30,43,47,50,52].now clear that glymphatic It is now flowclear withinthat glymphatic VRS is responsible flow within for clearance VRS is ofresponsibleβ-amyloid, for and clearance disturbance of β‐ in amyloid,glymphatic and flow disturbance likely plays in an glymphatic important roleflow in likely the pathogenesis plays an important of Alzheimer’s role in disease the pathogenesis (AD) [30,50, 52of]. Alzheimer’sSubsequently, disease it has been(AD) confirmed [30,50,52]. that Subsequently, patients with it AD has indeed beenhave confirmed impaired that glymphatic patients flowwith [53AD]. indeedGlymphatic have impaired flow glymphatic is dependent flow on[53]. AQP-4 function for β-amyloid clearance. In animal studies,Glymphatic conditions flow which is dependent can produce on AQP potential‐4 function decline for in β‐ AQP-4amyloid function clearance. all result In animal in β-amyloid studies, conditionsaccumulation which [44– 52can]. Inproduce humans, potential anti-AQP-4 decline antibody in AQP positive‐4 function autoimmune all result neuromyelitis in β‐amyloid optica accumulation(NMO) does [44–52 not affect]. In AQP-4humans, functionality anti‐AQP‐4 perantibody se. Consonant positive autoimmune with this, there neuromyelitis is no β-amyloid optica (NMO)accumulation does not observed affect inAQP anti-AQP-4‐4 functionality antibody per positive se. Consonant autoimmune with NMO this, [54 there]. In contrast, is no β‐ Duchenneamyloid accumulationmuscular dystrophy observed (DMD) in anti is accompanied‐AQP‐4 antibody by AQP-4 positive dysfunction autoimmune and also NMO associated [54]. In with contrast, excess Duchenneβ-amyloid muscular accumulation dystrophy [55]. DMD (DMD) patients is accompanied studied by single by AQP photon‐4 dysfunction emission computed and also tomography associated with(SPECT) excess demonstrated β‐amyloid GABA-Aaccumulation abnormalities [55]. DMD [56], apatients finding studied consonant by with single the knownphoton βemission-amyloid computedimpairment tomography of synaptic (SPECT) inhibition demonstrated via GABA-A GABA receptor‐A abnormalities endocytosis [57 [56],]. a finding consonant with the known β‐amyloid impairment of synaptic inhibition via GABA‐A receptor endocytosis [57]. 4.3. AQP-4 as Interstitial Fluid Circulator 4.3. AQPAQP-4‐4 as inInterstitial the brain Fluid is highly Circulator localized to astrocyte endfeet at the glia limitans externa (GLE) andAQP peri-capillary‐4 in the brain VRS is (Figure highly7)[ localized58–60]. to This astrocyte polarized endfeet location at the of glia AQP-4 limitans expression externa is (GLE) frequently and perisubject‐capillary of erroneous VRS (Figure speculation 7) [58–60]. that AQP-4 This facilitatespolarized thelocation flow of of water AQP into‐4 expression and out of is the frequently brain [60]. subjectAs discussed of erroneous above, speculation the existence that of AQP the BBB‐4 facilitates precludes the this flow possibility of water into [1–4 and]. Indeed, out of the brain absence [60]. of AsAQP-4 discussed does notabove, alter the water existence entry intoof the the BBB BBB preclude from thes this systemic possibility circulation, [1–4]. althoughIndeed, the it does absence reduce of AQPwater‐4 entrydoes not into alter the CSFwater system entry tointo a level the BBB similar from to the brain systemic parenchyma circulat [29ion,]. Polarizedalthough it localization does reduce of waterAQP-4, entry therefore, into the has CSF another system functionality to a level similar for water to brain dynamics parenchyma within [29]. the BBB. Polarized localization of AQP‐4, therefore, has another functionality for water dynamics within the BBB.

Int.Int. J. Mol. Sci. 2017, 18, 1798 1010 ofof 1414 Int. J. Mol. Sci. 2017, 18, 1798 10 of 14

Figure 7.7. Astrocyte and polarized localization of AQP-4.AQP‐4. ( A):): Expression of AQP-4AQP‐4 in the brain isis highlyFigure polarized 7. Astrocyte to endfeetendfeet and polarized of astrocytesastrocytes localization atat twotwo specificofsp ecificAQP ‐ locations,locations,4. (A): Expression thethe gliaglia limitanslimitans of AQP externa‐externa4 in the (GLE)(GLE) brain at atis thethehighly corticalcortical polarized surfacesurface to andand endfeet peri-capillaryperi‐ capillaryof astrocytes Virchow-RobinVirchow at two‐Robin specific spacespace locations, (VRS). VRS the glia constitutes limitans fluid-filledfluid externa‐filled (GLE) canalscanals at surroundingsurroundingthe cortical surface perforatingperforating and peri arteries,‐capillary capillaries, Virchow and‐Robin veins space in brainbrain (VRS). parenchyma.parenchyma. VRS constitutes While fluid pia‐ materfilledmater canals endsends nearsurrounding the brain perforating surface, VRS arteries, continues capillaries, into the brainandbrain veins parenchyma in brain accompanyingparenchyma. While a perforating pia mater artery; ends B ((nearB):): AstrocyteAstrocyte the brain endfeetendfeet surface, attachattach VRS continues toto manymany structures.structures. into the br However,However,ain parenchyma AQP-4AQP‐4 accompanying isis foundfound onlyonly atat a thetheperforating GLEGLE andand artery; VRS.VRS. (B): Astrocyte endfeet attach to many structures. However, AQP‐4 is found only at the GLE and VRS. AQP-4AQP‐4 at peri-capillaryperi‐capillary VRS connects astrocyte intracellular space and extracellular (interstitial) space.space.AQP Water Water‐4 at efflux efflux peri‐ capillaryfrom astrocytes VRS connects into the astrocyte interstitial intracellular space must space be and balancedbalanced extracellular byby anan (interstitial) equivalentequivalent amountspace. Water ofof water efflux influxinflux from in order astrocytes to maintain into the water interstitial equilibrium. space Since must astrocyte be balanced AQP-4AQP ‐4by is an also equivalent localized totoamount endfeetendfeet of atat water thethe GLE,GLE, influx itit in isis order crediblecredib tole maintain thatthat waterwater water movementmovement equilibrium. fromfrom thetheSince interstitialinterstitial astrocyte spacespace AQP ‐ intointo4 is also intracellularintracellular localized spacespaceto endfeet ofof astrocytesastrocytes at the GLE, occursoccurs it is atatcredib thethe GLE.leGLE. that InIn water thisthis schema,movementschema, thethe from AQP-4AQP the‐4 systeminterstitialsystem functionsfunctions space into asas circulatorcirculator intracellular ofof interstitialinterstitialspace of astrocytes fluidfluid withinwithin occurs thethe BBB,BBB, at th ande GLE. is responsible In this schema, for producing the AQP interstitial‐4 system fluidfluidfunctions movement as circulator essential of forforinterstitial glymphaticglymphatic fluid flowflow within (Figure(Figure the BBB,8 8).). and is responsible for producing interstitial fluid movement essential for glymphatic flow (Figure 8).

Figure 8. AQP-4 circulator model. The AQP-4 system provides additional water flow into peri-capillary VRSFigure (thick 8. AQP blue‐ arrow).4 circulator Water model. enters The astrocytes AQP‐4 through system AQP-4provides at theadditional GLE. This water internal flow circulation into peri‐ furthercapillaryFigure promotes8. VRS AQP (thick‐4 appropriate circulator blue arrow). interstitialmodel. Water The fluid entersAQP dynamics‐ 4astrocytes system including provides through flow AQPadditional through‐4 at the VRSwater GLE. (interstitial flow This into internal flow), peri‐ otherwisecirculationcapillary VRS known further (thick as promotes glymphatic blue arrow). appr flow. Wateropriate enters interstitial astrocytes fluid throughdynamics AQP including‐4 at the flow GLE. through This internal VRS (interstitialcirculation flow),further otherwise promotes known appr opriateas glymphatic interstitial flow. fluid dynamics including flow through VRS (interstitial flow), otherwise known as glymphatic flow. The AQP-4 system appears to serve to facilitate water dynamics within the BBB. In addition to providingThe AQP additional‐4 system waterappears into to theserve peri-capillary to facilitate water space, dynamics this system within also createsthe BBB. a waterIn addition density to providingThe AQP additional‐4 system water appears into tothe serve peri ‐tocapillary facilitate space, water this dynamics system within also creates the BBB. a waterIn addition density to gradientproviding within additional the interstitial water into space the further peri‐capillary promoting space, movement this system of interstitial also creates fluid a towater the surfacedensity ofgradient the brain. within Indeed, the interstitial 7T MRI microscopic space further analysis promot ofing the movement relaxation propertiesof interstitial of waterfluid to molecules the surface in ofgradient the brain. within Indeed, the interstitial7T MRI microscopic space further analysis promot of ingthe movementrelaxation propertiesof interstitial of waterfluid tomolecules the surface in of the brain. Indeed, 7T MRI microscopic analysis of the relaxation properties of water molecules in

Int. J. Mol. Sci. 2017, 18, 1798 11 of 14 humans demonstrated that the cortical area around GLE has a lower water density, effectively creating an electrostatic environment [61].

5. Conclusions The tight endothelium of brain capillaries plays a critical role in maintaining the BBB, a unique functional property of the brain. Water permeability through the BBB is highly restricted through suppression of expression of AQP-1 and the presence of tight junctions. As a result, brain water dynamics is effectively isolated from the systemic circulation, and AQP-4, a water channel abundantly expressed at the endfeet of astrocytes, plays a critical role in the water dynamics inside the BBB. Two unique functional features characterize AQP-4 facilitated brain water dynamics, namely neurovascular coupling and glymphatic flow. The former is initiated by neural activities that trigger rapid K+ flux into the extracellular space that eventually results in extracellular acidosis and AQP-4 suppression, which in turn reduces interstitial pressure, increases capillary diameter and leads to an increase in rCBF. The latter occurs in spite of the tight endothelium of brain capillaries which effectively impedes propagation of the hydrostatic pressure created by the systolic force of the heart to the interstitial fluid inside the BBB. Functionality of AQP-4 is necessary for water influx into peri-capillary VRS and the generation of interstitial fluid dynamics essential for glymphatic flow, the lymphatic equivalent in the brain. Considering that β-amyloid clearance is dependent on glymphatic flow, proper AQP-4 functionality is likely a key factor in the prevention and/or treatment of AD.

Acknowledgments: The study was supported by grants from the Ministry of Education, Culture, Sports, Science, and Technology (Japan). Conﬂicts of Interest: The authors declare no conﬂict of interest.

References

1. Reese, T.S.; Karnovsky, M.J. Fine structural localization of a blood-brain barrier to exogenous peroxidase. J. Cell Biol. 1967, 34, 207–217. [CrossRef][PubMed] 2. Mack, A.F.; Wolburg, H. A novel look at astrocytes: Aquaporins, ionic homeostasis, and the role of the microenvironment for regeneration in the CNS. Neuroscientist 2013, 19, 195–207. [CrossRef][PubMed] 3. Nielsen, S.; Smith, B.L.; Christensen, E.I.; Agre, P. Distribution of the aquaporin CHIP in secretory and resorptive epithelia and capillary endothelia. Proc. Natl. Acad. Sci. USA 1993, 90, 7275–7279. [CrossRef] [PubMed] 4. Dolman, D.; Drndarski, S.; Abbott, N.J.; Rattray, M. Induction of aquaporin 1 but not aquaporin 4 messenger RNA in rat primary brain microvessel endothelial cells in culture. J. Neurochem. 2005, 93, 825–833. [CrossRef] [PubMed] 5. Weiss, N.; Miller, F.; Cazaubon, S.; Couraud, P.O. The blood-brain barrier in brain homeostasis and neurological diseases. Biochim. et Biophys. Acta 2009, 1788, 842–857. [CrossRef][PubMed] 6. Günzel, D.; Yu, A.S.L. Claudins and the modulation of tight junction permeability. Physiol. Rev. 2013, 93, 525–569. [CrossRef][PubMed] 7. Rosenthal, R.; Milatz, S.; Krug, S.M.; Oelrich, B.; Schulzke, J.D.; Amasheh, S.; Günzel, D.; Fromm, M. Claudin-2, a component of the tight junction, forms a paracellular water channel. J. Cell Sci. 2010, 123, 1913–1921. [CrossRef][PubMed] 8. Mathai, J.C.; Tristram-Nagle, S.; Nagle, J.F.; Zeidel, M.L. Structural determinants of water permeability through the lipid membrane. J. Gen. Physiol. 2008, 131, 69–76. [CrossRef][PubMed] 9. Nakada, T. The molecular mechanisms of neural ﬂow coupling: A new concept. J. Neuroimage 2015, 25, 861–865. [CrossRef][PubMed] 10. Faber, T.E. Fluid Dynamics for Physicists; Cambridge University Press: Cambridge, UK, 1995. 11. Hall, C.N.; Reynell, C.; Gesslein, B.; Hamilton, N.B.; Mishra, A.; Sutherland, B.A.; O’Farrell, F.M.; Buchan, A.M.; Lauritzen, M.; Attwell, D. Capillary pericytes regulate cerebral blood ﬂow in health and disease. Nature 2014, 508, 55–60. [CrossRef][PubMed] Int. J. Mol. Sci. 2017, 18, 1798 12 of 14

12. Munde, P.B.; Khandekar, S.P.; Dive, A.M.; Upadhyaya, N.R. Pericytes in Health and Disease. Int. J. Oral Maxillofac. Pathol. 2014, 5, 2–7. 13. Levick, J. Introduction to Cardiovascular Physiology, 5th ed.; Hodder Arnold: London, UK, 2010. 14. Paulson, O.B.; Strandgaard, S.; Edvinsson, L. Cerebral autoregulation. Cerebrovasc. Brain Metab. Rev. 1990, 2, 161–192. [PubMed] 15. Heistad, D.D.; Kontos, H.A. The Cardiovascular System III. In Handbook of Physiology; Berne, R.M., Sperelakis, N., Eds.; American Physiological Society: Bethesda, MD, USA, 1979; pp. 137–182. 16. Mellander, S. Functional aspects of myogenic vascular control. J. Hypertens. 1989, 7, S21–S30. 17. Osol, G.; Brekke, J.F.; McElroy-Yaggy, K.; Gokina, N.I. Myogenic tone, reactivity, and forced dilatation: A three-phase model of in vitro arterial myogenic behavior. Am. J. Physiol. Heart Circ. Physiol. 2002, 283, H2260–H2267. [CrossRef][PubMed] 18. Busija, E.W.; Heistad, D.D. Factors involved in the physiological regulation of the cerebral circulation. Rev. Physiol. Biochem. Pharmcol. 1984, 101, 161–211. 19. Talman, W.T.; Dragon, D.N. Neuronal nitric oxide mediates cerebral vasodilatation during acute hypertension. Brain Res. 2007, 1139, 126–132. [CrossRef][PubMed] 20. Roland, P.E. Brain Activation; Wiley-Liss: New York, NY, USA, 1993. 21. Nakada, T. Virchow-Robin space and aquaporin-4: New insights on an old friend. Croat. Med. J. 2014, 55, 328–336. [CrossRef][PubMed] 22. Nakada, T.; Suzuki, K.; Kwee, I.L. Excess heat removal is likely to be the primary role of increase in regional cerebral blood flow associated with brain activation. In Proceedings of the Society of Neuroscience Annual Meeting, San Diego, CA, USA, 13–17 November 2010. 23. Silver, I.A. Cellular microenvironment in relation to local blood flow. In Ciba Foundation Symposium 56–Cerebral Vascular Smooth Muscle and Its Control; John Wiley & Sons: Chichester, UK, 2008. 24. Whittaker, R.J.; Heil, M.; Jensen, O.E.; Waters, S. Predicting the onset of high-frequency elf-excited oscillations in elastic-walled tubes. Proc. R. Soc. A 2010, 466, 3635–3657. [CrossRef] 25. Armitstead, J.P.; Bertram, C.D.; Jensen, O.E. A study of the bifurcation behaviour of a model of flow through a collapsible tube. Bull. Math. Biol. 1996, 58, 611–641. [CrossRef][PubMed] 26. Virchow, R. Ueber die Erweiterung kleinerer Gefaesse. Arch. Pathol. Anat. Physiol. Klin. Med. 1851, 3, 427–462. [CrossRef] 27. Robin, C. Recherches sur quelques particularites de la structure des capillaires de l’encephale. J. Physiol. Homme. Animaux. 1859, 2, 537–548. 28. Haj-Yasein, N.N.; Jensen, V.; Ostby, I.; Omholt, S.W.; Voipio, J.; Kaila, K.; Ottersen, O.P.; Hvalby, O.; Nagelhus, E.A. Aquaporin-4 regulates extracellular space volume dynamics during high-frequency synaptic stimulation: A gene deletion study in mouse hippocampus. Glia 2012, 60, 867–874. [CrossRef][PubMed] 29. Igarashi, H.; Tsujita, M.; Kwee, I.L.; Nakada, T. Water influx into cerebrospinal fluid (CSF) is primarily controlled by aquaporin-4, not by aquaporin-1: O-17 JJVCPE MRI Study in Knockout Mice. Neuroreport 2014, 25, 39–43. [PubMed] 30. Xie, L.; Kang, H.; Xu, Q.; Chen, M.J.; Liao, Y.; Thiyagarajan, M.; O’Donnell, J.; Christensen, D.J.; Nicholson, C.; Iliff, J.J.; et al. Sleep drives metabolite clearance from the adult brain. Science 2013, 342, 373–377. [CrossRef] [PubMed] 31. Igarashi, H.; Tsujita, M.; Huber, V.J.; Kwee, I.L.; Nakada, T. Inhibition of Aquaporin-4 significantly increases regional cerebral blood flow. NeuroReport 2013, 24, 324–328. [CrossRef][PubMed] 32. Amiry-Moghaddam, M.; Williamson, A.; Palomba, M.; Eid, T.; de Lanerolle, N.C.; Nagelhus, E.A.; Adams, M.E.; Froehner, S.C.; Agre, P.; Ottersen, O.P. Delayed K+ clearance associated with aquaporin-4 mislocalization: Phenotypic defects in brains of α-syntrophin-null mice. Proc. Natl. Acad. Sci. USA 2003, 100, 13615–13620. [CrossRef][PubMed] 33. Lee, D.J.; Hsu, M.S.; Seldin, M.M.; Arellano, J.L.; Binder, D.K. Decreased expression of the glial water channel aquaporin-4 in the intrahippocampal kainic acid model of epileptogenesis. Exp. Neurol. 2012, 235, 246–255. [CrossRef][PubMed] 34. Olsen, M.L.; Sontheimer, H. Functional implications for Kir4.1 channels in glial biology: From K+ buffering to cell differentiation. Neurochemmistry 2008, 107, 589–601. [CrossRef][PubMed] Int. J. Mol. Sci. 2017, 18, 1798 13 of 14

35. Strohschein, S.; Hüttmann, K.; Gabriel, S.; Devin, K.; Binder, D.K.; Heinemann, U.; Steinhäuser, C. Impact of aquaporin-4 channels on K+ buffering and gap junction coupling in the hippocampus. Glia 2011, 59, 973–980. [CrossRef][PubMed] 36. Kofuji, P.; Newman, E.A. Potassium buffering in the central nervous system. Neuroscience 2004, 129, 1045–1056. [CrossRef][PubMed] 37. Giaume, C.; Koulakoff, A.; Roux, L.; Holcman, D.; Rouach, N. Astroglial networks: A step further in neuroglial and gliovascular interactions. Nat. Rev. Neurosci. 2010, 11, 87–99. [CrossRef][PubMed] 38. Chen, J.; Skinner, M.A.; Shi, W.; Yu, Q.C.; Wildeman, A.G.; Chan, Y.M. The 16 kDa subunit of vacuolar H+-ATPase is a novel sarcoglycan-interacting protein. Biochim. Biophys. Acta 2007, 772, 570–579. [CrossRef] [PubMed] 39. Hibino, H.; Kurachi, Y. Distinct detergent-resistant membrane microdomains (lipid rafts) respectively harvest K(+) and water transport systems in brain astroglia. Eur. J. Neurosci. 2007, 26, 2539–2555. [CrossRef] [PubMed] 40. Kaptan, S.; Assentoft, M.; Schneider, H.P.; Fenton, R.A.; Deitmer, J.W.; MacAulay, N.; de Groot, B.L. H95 is a pH-dependent gate in aquaporin 4. Structure 2015, 23, 2309–2318. [CrossRef][PubMed] 41. Magnotta, V.A.; Heo, H.Y.; Dlouhy, B.J.; Dahdaleh, N.S.; Follmer, R.L.; Thedens, D.R.; Welsh, M.J.; Wemmie, J.A. Detecting activity-evoked pH changes in human brain. Proc. Natl. Acad. Sci. USA 2012, 109, 8270–8273. [CrossRef][PubMed] 42. Kitaura, H.; Tsujita, M.; Huber, V.J.; Kakita, A.; Shibuki, K.; Sakimura, K.; Kwee, I.L.; Nakada, T. Activity-dependent glial swelling is impaired in aquaporin-4 knockout mice. Neurosci. Res. 2009, 64, 208–212. [CrossRef][PubMed] 43. Esiri, M.M.; Gay, D. Immunological and neuropathological significance of the Virchow-Robin space. J. Neurol. Sci. 1990, 100, 3–8. [CrossRef] 44. Weller, R.O. Pathology of cerebrospinal fluid and interstitial fluid of the CNS: Significance for Alzheimer’s disease, prion disorders and multiple sclerosis. J. Neuropathol. Exp. Neurol. 1998, 57, 885–894. [CrossRef] [PubMed] 45. Johnston, M.; Papaiconomou, C. Cerebrospinal fluid transport: A lymphatic perspective. News Physiol. Sci. 2002, 17, 227–230. [CrossRef][PubMed] 46. Abbott, N.J. Evidence for bulk flow of brain interstitial fluid: Significance for physiology and pathology. Neurochem. Int. 2004, 45, 545–552. [CrossRef][PubMed] 47. Weller, R.O.; Djuanda, E.; Yow, H.Y.; Carare, R.O. Lymphatic drainage of the brain and the pathophysiology of neurological disease. Acta Neuropathol. 2009, 117, 1–14. [CrossRef][PubMed] 48. Iliff, J.J.; Wang, M.; Liao, Y.; Plogg, B.A.; Peng, W.; Gundersen, G.A.; Benveniste, H.; Vates, G.E.; Deane, R.; Goldman, S.A.; et al. A paravascular pathway facilitates CSF flow through the brain parenchyma and the clearance of interstitial solutes, including amyloid β. Sci. Transl. Med. 2012, 4, 147ra111. [CrossRef] [PubMed] 49. Orešković,D.; Klarica, M. The formation of cerebrospinal fluid: Nearly a hundred years of interpretations and misinterpretations. Brain Res. Rev. 2010, 64, 241–262. [CrossRef][PubMed] 50. Tarasoff-Conway, J.M.; Carare, R.O.; Osorio, R.S.; Glodzik, L.; Butler, T.; Fieremans, E.; Axel, L.; Rusinek, H.; Nicholson, C.; Zlokovic, B.V.; et al. Clearance systems in the brain-implications for Alzheimer Disease. Nat. Rev. Neurol. 2015, 11, 457–470. [CrossRef][PubMed] 51. Margaris, K.N.; Black, R.A. Modelling the lymphatic system: Challenges and opportunities. J. R. Soc. Interface 2012, 9, 601–612. [CrossRef][PubMed] 52. Igarashi, H.; Suzuki, Y.; Kwee, I.L.; Nakada, T. Water influx into cerebrospinal fluid is significantly reduced in senile plaque bearing transgenic mice, supporting β-amyloid clearance hypothesis of Alzheimer disease. Neurological Res. 2014, 36, 1094–1098. [CrossRef][PubMed] 53. Suzuki, Y.; Nakamura, Y.; Yamada, K.; Igarashi, H.; Kasuga, K.; Yokoyama, Y.; Ikeuchi, T.; Nishizawa, M.; Kwee, I.L.; Nakada, T. Reduced CSF water influx in Alzheimer’s disease supporting the β-amyloid clearance hypothesis. PLoS ONE 2015, 10, e0123708. [CrossRef][PubMed] 54. Nicchia, G.P.; Mastrototaro, M.; Rossi, A.; Pisani, F.; Tortorella, C.; Ruggieri, M.; Lia, A.; Trojano, M.; Frigeri, A.; Svelto, M. Aquaporin-4 orthogonal arrays of particles are the target for neuromyelitis optica autoantibodies. Glia 2009, 57, 1363–1373. [CrossRef][PubMed] Int. J. Mol. Sci. 2017, 18, 1798 14 of 14

55. Salam, E.A.; Abdel-Meguid, I.E.; Shatla, R.; Korraa, S. Evaluation of neural damage in Duchenne muscular dystrophy patients. Acta Myol. 2014, 33, 13–18. [PubMed] 56. Suzuki, Y.; Higuchi, S.; Aida, I.; Nakajima, T.; Nakada, T. Abnormal distribution of GABAA receptors in brain of Duchenne muscular dystrophy patients. Muscle Nerve 2017, 55, 591–595. [CrossRef][PubMed] 57. Ulrich, D. Amyloid-β impairs synaptic inhibition via GABA (A) receptor endocytosis. J. Neurosci. 2015, 35, 9205–9210. [CrossRef][PubMed] 58. Rash, J.E.; Yasumura, T.; Hudson, C.S.; Agre, P.; Nielsen, S. Direct immunogold labeling of aquaporin-4 in square arrays of astrocyte and ependymocyte plasma membranes in rat brain and spinal cord. Proc. Natl. Acad. Sci. USA 1998, 95, 11981–11986. [CrossRef][PubMed] 59. Neely, J.D.; Christensen, B.M.; Nielsen, S.; Agre, P. Heterotetrameric composition of aquaporin-4 water channels. Biochemistry 1999, 38, 11156–11163. [CrossRef][PubMed] 60. Papadopoulos, M.C.; Verkman, A.S. Aquaporin water channels in the nervous system. Nat. Rev. Neurosci. 2013, 14, 265–277. [CrossRef][PubMed] 61. Suzuki, K.; Yamada, K.; Suzuki, Y.; Kwee, I.L.; Nakada, T. 7.0 Tesla MRI reveals electrostatic environment of the glia limitans. In Proceedings of the Society of Neuroscience Annual Meeting, Chicago, IL, USA, 17–21 October 2010.

© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).