This chapter summarizes the current knowledge regarding the regulation of the tone of cerebral resistance arteries under conditions of normal health and with the development of chronic diseases (e.g., metabolic disease). The work integrates the myogenic (pressure-induced) regulation of vascular tone, the impact of elevated luminal flow or shear stresses, that of local tissue metabolic activity on vascular tone and the concept of neurovascular coupling (linking neuronal activity to the impacts on vascular diameter). In addition, this work summarizes some of the recent work on how diseases such as type 2 diabetes impact the mechanisms of cerebrovascular tone regulation. It is anticipated that the current review will provide the reader with an up-to-date understanding of how the cerebral resistance vessels respond to changes in their local environment and contribute to the regulation of blood flow within the brain.
- perfusion resistance
- metabolic disease
- vascular function
- vascular disease risk
The brain has a remarkably high metabolic rate and thus requires a highly disproportional amount of blood flow. Although its only 2% of body weight, the brain takes up 15–20% of cardiac output , making it one of the most highly perfused organs in the human body. This high metabolic rate coupled with its limited capacity for energy storage  necessitates heavy reliance on oxidative metabolism and thus requires constant blood flow to maintain nutrient and oxygen supply, remove waste products, and maintain a state of cerebral metabolic homeostasis. Severe underperfusion can quickly result in unconsciousness  and if prolonged, death ; while chronic mild under perfusion is associated with cognitive decline . In addition to its high perfusion and metabolic rate, the cerebral circulation faces a unique challenge of being enclosed in the skull. This rigid structure prevents the expansion of tissue and extracellular fluid. Swelling within the skull from vasogenic edema leads to an increase in intracranial pressure which in turn can lead to neurologic complications or in more extreme cases death . The unique challenges of the cerebral circulation, including intolerance to ischemia and edema, coupled with the paramount importance of maintaining constant nutrient and oxygen supply to cerebral tissue creates a need for precise regulation of cerebral blood flow and therefore the presence of redundant intrinsic mechanisms for its regulation. The anatomy of the brain vasculature ensures multiple routes for blood and oxygen delivery potentially allowing for perfusion even in cases of a blocked blood vessel ; however, acute regulation of flow is done primary by altering the diameter of blood vessels, and thus the resistance to flow. The major mechanisms of local regulation of vascular tone intrinsic to the cerebral vasculature include myogenic, shear, and metabolic based regulation. Although each mechanism has a discrete effect on vascular tone the integration of the different contributors to determine an appropriate level of tone is much more difficult to discern, especially in the cerebral circulation. These complex interactions allow for highly accurate control of cerebral blood flow in addition to protecting vulnerable downstream capillaries from high pressures and flow rates that could otherwise lead to edema; but, they also introduce several potential areas for failure. The intimate interactions of the various mechanisms of regulation of flow mean that the failure of one mechanism has the potential to initiate a cascade of events that results in inappropriate regulation of flow. As such abnormal execution of vascular tone regulation may form the basis of vascular pathologies .
One of these pathologies with a significant vascular component associated with impaired cerebral vascular tone regulation is metabolic syndrome (MetS). The incidence and prevalence of MetS is growing in Western society [8, 9, 10] and is contributing to decreased quality of life and increased economic burden. Thus an understanding of how it alters the cerebral circulation is crucial. MetS is categorized by a collection of metabolic risk factors including obesity, hypertension, atherogenic dyslipidemia and impaired glycemic control creating a pro-oxidant pro-inflammatory environment that raises the risk of developing impaired vascular structures and function [11, 12, 13, 14]. These impairments are particularly detrimental when they affect the cerebral circulation and lead to cerebrovascular pathologies such as stroke or transient ischemic attack (TIA) due to the detrimental consequences associated with such events. However, cognitive impairments are not limited to individuals that have experienced an acute ischemic event since even in their absence MetS is strongly associated with impaired cognitive function and decreased quality of life [15, 16, 17, 18]. Therefore preventing their occurrence by protecting the cerebrovasculature from functional and structural decline is paramount. This chapter will present a description of the local mechanisms involved in the regulation of cerebral vascular tone, how they integrate with one another and how they can be compromised in disease. Although impairments to the regulation of cerebral vascular tone are not limited to conditions associated with MetS this discussion will focus on the impact of MetS and its associated risk factors.
2. Myogenic mechanism
The myogenic mechanism which was first described by Bayliss is an intrinsic property of the vascular smooth muscle to respond to changes in intravascular pressure which is independent of other mechanisms of tone regulation including neural, metabolic, and hormonal influences . The intrinsic nature of the myogenic response is supported by its existence in arteries and arterioles that have been sympathetically denervated and had their endothelium removed  thus leaving only the vessel itself to initiate and execute the response. The prototypical response of the vascular smooth muscle in response to an increase in intraluminal pressure is initial distension quickly followed by a constriction. The opposite can be said in situations of decreased intraluminal pressure; a fall in intraluminal pressure results in vessel collapse followed by dilation . The myogenic response has several physiological roles including the establishment and maintenance of basal vascular tone (some degree of constriction) so that resistance may be increased or decreased by metabolic vasoconstrictors or vasodilators respectively, in order to regulate tissue perfusion. The establishment and maintenance of this partially constricted state in a pressurized vessel is referred to as myogenic tone. Additionally, the myogenic tone has a role in flow and pressure regulation. It functions by constricting to drop the pressure that reaches the downstream capillaries and protect them from edema or vascular remodeling associated with hypertension in the capillaries [22, 23, 24, 25]. Equally important as protecting from hypertension is the ability of the vasculature to promote flow during low pressure by dilating. This ability to alter diameter over a range of pressures is referred to as myogenic reactivity. Beyond the local implications for the regulation of vascular tone, resistance to flow also has implications for systemic blood pressure since mean arterial pressure is the product of total peripheral resistance and cardiac output. This relationship illustrates system-wide implications of accurate vascular tone regulation. While the myogenic response is present throughout the body in a variety of vessels (arterioles, veins, lymphatic vessels)  the aforementioned functions of the myogenic response are particularly important in the cerebral circulation due to the catastrophic outcomes associated with under or over perfusion including unconsciousness and edema respectively. It is therefore not surprising that the most prominent myogenic response is found in the cerebral circulation with arterioles (resistance vessels) having the most pronounced response . It should also be noted that the large arteries feeding the brain have a greater contribution to regulating vascular resistance in the cerebral circulation compared to other vascular beds, again providing evidence for the importance of regulating tone in the cerebral circulation.
A phase model of arterial myogenic behavior is commonly used to describe the response over a range of pressure. In the first phase, there is an initial development of myogenic tone at approximately 40–60 mmHg with increasing pressure up to that point causing passive distension. There is then a phase of myogenic reactivity in the pressure range of 60–140 mmHg and finally a phase of forced dilation at transmural pressures greater than approximately 140 mmHg .
2.1 Myogenic tone
Myogenic tone develops at approximately 40–60 mmHg and is characterized by an increase in intracellular Ca2+, of about 200%, followed by a reduction in lumen diameter. This pressure causes cellular deformation, depolarization of the vascular smooth muscle cells (VSMC), and a significant increase in wall tension. Wall tension appears to be the controlled parameter in the myogenic response which is altered during an increase or decrease in transmural pressure. The myogenic response adjusts the diameter of the vessel in an attempt to restore or limit the change in basal wall tension through a negative feedback mechanism [28, 29]. The suggestion that wall tension is the controlled parameter is supported by its correlation with changes in cell calcium and myosin light-chain phosphorylation, a relationship that is not seen with vessel diameter [30, 31].
The mechanism of the myogenic response is attributed to stretch-activated ion channels: including L-type calcium channels, voltage-activated calcium channels and calcium activated potassium channels along with enzymatic mechanisms. Specifically, increased intraluminal pressure causes depolarization of the VSMC membrane and calcium influx by the opening of voltage-gated calcium channels (VGCCs), with the most prominent involvement being from CaV1.2. This influx of Ca2+ leads to increased myosin light-chain (MLC) phosphorylation which promotes increased actin/myosin interaction followed by cross-bridge cycling and cell shortening (vasoconstriction) [7, 24, 32, 33, 34]. The importance of Ca2+ influx in the generation of myogenic tone supported by its complete abolishment under Ca2+ free conditions [32, 35]; this is also a technique frequently used to study the passive mechanical characteristics of vessels to determine the degree of vascular remodeling since the VMSC exhibit a passive response (no force production) its strain is only due to the applied stress applied and the composition of the vessel . Under physiological conditions the magnitude of the constrictor response to increased intraluminal pressure is limited by calcium-activated potassium channels that carry hyperpolarizing current proportional to the intracellular calcium concentration . This negative feedback mechanism is supported by enhanced myogenic constriction being observed following blockade of calcium-activated potassium channel by specific inhibitors of these channels [37, 38, 39]. Additionally, at this pressure of 40–60 mmHg there is an activation of enzymatic systems and a complex interaction between matrix metalloproteins, the extracellular matrix, integrins and the cytoskeleton [40, 41, 42] that contribute to the myogenic reactivity at higher intraluminal pressures within the range of 60–140 mmHg . This myogenic tone phase can also be characterized as the lower limit of autoregulation, which has important physiological implications. Below this pressure blood flow becomes dependent on blood pressure since the vessel cannot further dilate and begins to collapse as the pressure drops below this point . Having an appropriate lower limit becomes especially important in situations of cerebral ischemia to allow restoration of flow in the presence of hypotension.
2.2 Myogenic reactivity
In this range of intraluminal pressure of 60–140 mmHg where the myogenic tone has already been established increases in pressure generally result in mild constriction and decreased pressure leads to mild dilation. Just as previously discussed for the generation of myogenic tone, increased pressure within this range leads to stretch, depolarization, and constriction of the vascular smooth muscle. However, in the myogenic reactivity phase, there is little change in vessel diameter across the range of pressures along with relatively small increases in Ca2+ (<20%) despite sizable increases in force production . Multiple studies suggest an increased sensitivity to Ca2+ compared to the previous phase in the development of myogenic tone [20, 27, 44, 45, 46, 47, 48]. Increased sensitivity to Ca2+ is achieved by inhibition of myosin light chain phosphatase (MLCP) which promotes the accumulation of phosphorylated LC20 without an accompanying increase in calcium-induced myosin light chain kinase activity . The presence of a contractile mechanism that does not require large variation of calcium, such as altering Ca2+ sensitivity requires less storage and transmembrane shuttling and is therefore advantageous in terms of conserving Ca2+ .
There are several proposed mechanisms that regulate Ca2+ sensitivity within this phase including, activation of protein kinase C (PKC), RhoA/Rho kinase pathways, and reactive oxygen species (ROS) [27, 35, 44, 46, 50]. The following studies provide evidence for the aforementioned mechanisms of enhanced Ca2+ sensitivity in this phase of the myogenic response through the utilization of specific inhibitors or transgenic animal models. Inhibition of PKC stops myogenic vasoconstriction in middle cerebral arteries with no impact on pressure-induced Ca2+ elevation . Direct assessment of Ca2+ sensitivity by measuring the Ca2+-tone relationship has consistency found decreased sensitivity during Rho kinase inhibition [20, 45]. ROK has also been reported to trigger smooth muscle depolarization during myogenic constriction and limit the extent of depolarization by opening delayed rectifier potassium channels [51, 52]. Arteries from transgenic animals missing NADPH oxidase function show an absence of myogenic activity , while mice deficient in superoxide dismutase, an endogenous antioxidant enzyme that catalyzes the breakdown of superoxide radical to H2O2, acquired enhanced myogenic reactivity . Additional mechanisms that contribute to the myogenic reactivity phase independent of Ca2+ sensitivity include actin cytoskeleton reorganization and thin filament regulation [55, 56].
2.3 Forced dilation
Although the prototypical response of increased intraluminal pressure is a constriction, at excessively high pressures, beyond the autoregulatory range of approximately 140 mmHg, forced dilation often occurs . This process results in a loss of myogenic tone, and thus results in an increase in vessel diameter, rapid increase in wall tension and significant elevation in Ca2+ (>50%) . Although the name implies a degree of passiveness in the process, forced dilation is likely an active vasodilation involving KCa channels, nitric oxide (NO) and or ROS, which are activated to protect the arterial wall from damage [44, 57]. If the pressure is reduced to within the myogenic reactivity range reestablishment of tone and reduction in Ca2+ is observed.
3. Response to flow (shear stress)
An increase in flow leads to an increase in a frictional force known as shear that is detected by the endothelial cells lining the vessel lumen as blood moves through a vessel. As such, the magnitude of the shear force is proportional to blood flow. Shear catalyzes physiologically important responses in the cerebral vasculature such as encouraging reperfusion after ischemia, aiding in the hyperemic response to increased metabolic demand, and perhaps protection of downstream capillaries from edema and structural damage. Flow has been found to induce both constrictor and dilator pathways that act on the VSMC and result in a final level of tone taking into account the opposing processes. It is generally accepted that in the peripheral circulation flow leads to dilation; however, in the cerebral circulation the response is more controversial with both constriction and dilation being reported. These opposing observations may be because of a variety of factors including the area of the brain studied, the preparation used or because of interactions with other mechanisms affecting cerebral vascular tone. Within the cerebral circulation, the vertebrobasilar systems appear to elicit flow-mediated dilation, as measured in rats and mice  and humans . The increase in flow is sensed by the endothelium, which initiates a negative feedback mechanism in an attempt to decrease the shear stress by dilating. Shear-induced dilation is largely endothelium-dependent and is at least partially mediated by NO . The production of NO is controlled by the enzyme nitric oxide synthase (NOS), particularly endothelial NOS (eNOS), which catalyzes the formation of NO from L-arginine and is itself dependent on phosphorylation by Akt [61, 62]. The NO formed by eNOS then diffuses into the vascular smooth muscle where it activates soluble guanylyl cyclase (sGC) and increases cyclic guanine monophosphate (cGMP) and activating protein kinase G [63, 64]. Activation of protein kinase G opens large-conductance Ca2+-activated K+ (BKCa) channels reducing intracellular Ca2+ which leads to relaxation of the vascular smooth muscle and thus vasodilation . In addition to NO, eNOS can also lead to the formation of H2O2 which may also mediate flow-induced dilation . During enzymatic cycling, eNOS produces oxygen radicals [66, 67] which, in the presence of sufficient antioxidants, are converted into H2O2 that may then, like NO, activate sGC and lead to dilation . Other endothelium-dependent dilators such as prostaglandins do not appear to be involved in shear-induced dilation since the COX inhibitor indomethacin had no effect on the response to flow .
Constriction of cerebral vessels has also been reported in cats , rats  and human isolated cerebral arteries . Constriction appears to predominate within the carotid circulatory area  especially when studies using ex vivo isolated vessels. In contrast to the dilator response, flow-induced constriction occurs independent of the endothelium [72, 73]. Although the mechanics for constriction in response to increased flow is not fully understood it appears to be mediated in part by a combination of integrin signaling, free radicals, and tyrosine kinase. Giving integrin-binding peptides, scavenging ROS by superoxide dismutase (SOD) or inhibiting tyrosine  all significantly reduced or eliminated constriction in response to flow. Just as in flow-induced dilation shear stress appears to be the initiating parameter in flow induced constriction whose force can be transduced by integrin signaling . Additionally, Koller and Toth found that flow-induced constriction of cerebral vessels were blocked by inhibiting the synthesis of 20-HETE, thromboxane A2 receptor, COX activity, and scavenging ROS. From that, they proposed that increase flow activates an AA cascade with metabolism by CYP450 4A enzymes resulting in the production of 20-HETE and ROS which contribute to the constriction via thromboxane A2/prostaglandin H2 receptors .
The differences in observed responses may also be dependent on whether the study was conducted in vivo or ex vivo. Ex vivo studies typically use a physiologic salt solution that has a much lower viscosity than blood and may be inadequate to stimulate the endothelium to produce endothelium-dependent dilation . Consequently, the ex vivo responses to intraluminal shear have been shown to be endothelium-independent [69, 72]. Furthermore, ex vivo study generally lack pulsatile flow pumps but those that have been able to employ peristaltic or piston pump have indicated a role for shear-mediated regulation [68, 75, 76]. In contrast, in vivo study has produced more consistent shear-mediated dilation [62, 77]. Raignault et al. showed in a mouse model that the cerebrovascular endothelium optimally integrates shear stress to eNOS-mediated dilation under physiological pulse pressures, a phenomenon that was not seen in static flow conditions . It appears that cerebral arteries are more responsive to a pulsatile environment with a viscose fluid (blood) as is experienced in vivo and consequently able to dilate in response to shear stress through endothelium-dependent production of NO [68, 76, 78].
Although flow induced constriction is not usually seen in vivo at high flow rates it may play a protective role in the cerebral circulation. Just as low flow situations are dangerous, especially in the brain, so too are instances of extreme flow; therefore it is important to have multiple regulatory mechanisms that are able to attenuate an increase in flow as a protective mechanism for the cerebral circulation from high volumes [1, 79, 80]. It is advantageous for the response to flow to adapt in accordance with other inputs on the regulation of vascular tone. This may be the reason why characterizing the response to flow in the cerebral circulation has been variable. One variable that may alter the response to flow is the absolute rate of flow. In isolated branches of MCA from rats vessels dilated with flow up to 10 μL/min but their diameter constricted back to baseline at higher levels of flow . Shimoda et al. also found a biphasic flow-dependent response in anterior and middle cerebral arteries of neonatal pigs; flow rates between 0.077 and 0.212 mL/min constricted vessels while dilation occurred at higher flow rates up to 1.6 mL/min .
3.1 Flow and pressure
In a physiological setting multiple inputs are being processed by the cerebral vasculature leading to the generation of a certain level of tone. Pressure and shear stress exerted by flowing blood are two mechanical stimuli that have been described to play a major role in the regulation of vascular tone [83, 84]. It is therefore important to consider their interaction when determining the resultant effect on vascular tone. At high pressures (around >80 mmHg) cerebral vessels tend to constrict in response to flow [72, 85, 86, 87]. This biphasic response is further supported by findings from Garcia-Roldan and Bevan in isolated rabbit pial arterioles with flow rates from 0 to 20 μL/min at 90 mmHg but did not with the same flow at 60 mmHg . It is interesting that the tipping points for “high” and “low” pressures observed in previous studies are around 80 mmHg since, in resistance arterioles such as the MCA this pressure is commonly measured under physiological conditions. If 80 mmHg is indeed the point where higher pressures lead to constriction and lower to dilation in response to shear the resistance arterioles that are resting around this pressure would have the opportunity to tightly regulate the response to flow and shear thus allowing for precise control of cerebral blood flow. Please see Figure 1 for a schematic representation of how flow and intravascular pressure can impact cerebral vascular tone.
4. Metabolic control
The mechanical stimuli of pressure and flow are generally thought to be important in setting the basal vascular tone so that metabolic influences are able to cause dilation or constriction depending on the needs of the cerebral tissue . Metabolic control of vascular resistance is of particular importance in the cerebral circulation since cerebral tissue is extremely intolerant to ischemia . As such, the cerebral circulation has a precise and highly localized coupling between the metabolic requirements of cerebral tissue and the magnitude of blood flow by controlling vascular resistance. There are numerous vasoactive metabolites that contribute to the control of cerebral vascular tone including adenosine, CO2, H+, O2 and K+. Increasing concentrations of adenosine, CO2, H+, and K+ and decreased concentration of O2 result in relaxation of VSM and dilation of cerebral resistance vessels. Each metabolite is associated with a cascade of events that ultimately either alters intracellular Ca2+ concentration or Ca2+ sensitivity of the VSM and results in a change in vessel diameter. Although the effects of each metabolite have been well characterized the relative importance of each along with its interaction with each other and other parameters of tone remains an area of further investigation. The discussion below is summarized in Figure 1.
Adenosine has been proposed as being the primary metabolite controlling metabolic regulation of cerebral vascular tone. It is a naturally occurring nucleoside produced as a byproduct of ATP metabolism; thus, its accumulation signals a need for increased blood flow to match the metabolic activity. This relationship with metabolism has widely implicated adenosine in local regulation of cerebral vascular tone during functional hyperemia, ischemia, or whenever PO2 becomes limited [89, 90, 91]. Adenosine has direct effects on the vasculature [92, 93] that can both vasodilate and hyperpolarize VSM and is therefore considered an EDHF . There are four distinct subtypes of adenosine receptors; however, the A2A receptor appears to be of high importance in mediating vasodilation [95, 96, 97]. The A2A receptor is a purinergic P1 receptor that has been confirmed to be present in cerebral microvessels [98, 99, 100, 101] on the VSM cells [102, 103]. It causes dilation in a concentration-dependent manner [101, 104] once bound to the A2A receptor by activation of adenylate cyclase  and therefore cAMP [106, 107, 108] which reduces cytosolic calcium and leads to vasodilation. The opening of KATP channels also occurs secondary to the increase in cAMP levels as a result of adenosine binding to its receptors on the cell membrane [109, 110, 111, 112]. The contribution of opening KATP channels to dilation is likely substantial since during blockade of KATP channels with glibenclamide, adenosine-induced dilation was reduced by approximately 50% . Although A2A receptors are generally considered the most important mediators of the effects of adenosine on vascular tone the A2B receptors are also proposed to cause dilation through similar mechanisms as A2A in addition to coupling to Gq proteins to produce Ca2+ mobilization by activation of phospholipase C and mitogen-activated protein kinase activation [98, 114, 115]. In addition to dilating the cerebral vasculature adenosine may also block vasoconstrictive signals in the parenchyma as evidenced by in vitro data from Gordon et al. . When adenosine receptors are blocked with theophylline dilation was attenuated to arterial hypoxia . Similarly the competitive adenosine receptor antagonist aminophylline causes a 20–30% decrease in CBF and cerebral oxygen delivery in normoxia .
Similar to many of the other metabolites discussed CO2 tends to increase under conditions of increased metabolism without adequate flow to eliminate it from the area of production and thus its accumulation leads to dilation of the vasculature. High sensitivity to PCO2 is unique to the cerebral circulation  causing approximately 3–6% increase and 1–3% decrease in flow per mmHg change in PaCO2 above or below eupnoeic PaCO2 respectively. This high sensitivity is seen throughout the arterial side of the vascular network including the large arteries in the neck  and large intracranial arteries [121, 122, 123] to the smallest pial arterioles  and parenchymal vessels [125, 126, 127, 128]. There are likely several redundant mechanistic contributors to the sensitivity of the vasculature to PCO2 which may contribute to the debate as to whether the dilation is triggered by increased PCO2 or rather the accompanying increase H+ concentration from the carbonic anhydrase reaction. There is strong support that the change in PCO2 mediates at least in part alterations in cerebral vascular tone locally by changes in perivascular pH [107, 129, 130, 131] as evidenced by acidic and alkaline perfusate administered through an intracranial window. Experiments that have been able to alter pH and PCO2 independently have provided evidence of the dependence of altered pH to initiate a change in cerebral vascular tone . In a cat pial arteriole cranial window preparation, lowering pH along with hypercapnia resulted in no difference in the magnitude of vasodilation when compared to acidic isocapnia . Interestingly, vessel tone was unaltered in response to intraluminal CO2 change, suggesting that a change in superfusate pH is necessary to evoke a change in cerebral vascular tone [107, 133]. This is supported by the findings of unchanged CBF in humans  and animals  in response to changes in arterial pH, and highly localized pial arteriolar diameter changes in response to the application of acidic/basic solution into the perivascular space . Additionally, when pH is maintained as seen in an experiment with artificial CSF pretreated with sodium bicarbonate, pial vessel dilation is eliminated in response to intraluminal hypercapnia [107, 134] further supporting that a change in superfusate pH is necessary to alter the cerebral vascular tone. In addition to its link with CO2 and H+ through the carbonic anhydrase reaction, there is some evidence that bicarbonate ion may independently influence vascular tone. Having the ability to sense and react to multiple parameters associated with acidosis allows for control of tone in response to not only the pH but also the cause of the disturbance (accumulation of CO2 for instance). In isolated rat basilar arteries reduced [HCO3−]o has been shown to directly increase cerebral vascular tone (with pH maintained at 7.4 and CO2 kept constant at 5%) through the binding of receptor protein tyrosine phosphatase through an endothelium-dependent response . HCO3− may, therefore, stimulate soluble adenylate cyclase activity through a pH-independent mechanism . If this is correct then the modest increase in [HCO3−] during hypercapnia may aid in the dilation response to CO2 but the opposing vasocontractile response to decreases in [HCO3−]o may limits the vasorelaxation caused by a reduction in pH . Limiting vasorelaxation during acidosis is important to lessen the increase in capillary pressure associated with vasorelaxation of upstream arterioles. Hyper-relaxation may overload the capillaries leading to edema and damage thereby worsening the consequences of local inadequate perfusion.
The effects of oxygen are unique in that its availability is required for aerobic metabolism rather than a byproduct like some of the metabolic factors discussed. Therefore it is not surprising that its abundance leads to vasoconstriction and its relative shortage leads to dilation. Although its availability is tightly linked to that of PCO2 and H+ and other metabolic byproducts in a physiological setting, studies have been able to discern its independent effect in the presence of otherwise constant conditions. Data from isolated arteries/arterioles suggest there is in fact an oxygen sensor independent of other vasoactive metabolic byproducts within the vascular wall itself [137, 138, 139, 140, 141, 142]. Once the change in oxygen is sensed there are various mediators of hypoxic dilation including endothelial-derived NO [138, 141], prostanoids [137, 140, 141, 143], 20-HETE  and EDHF ; however, the contribution of each factor appears to be dependent on the severity of hypoxia . In skeletal muscle, dilation from mild hypoxia (15% O2) was mostly NO-dependent, while moderate (10% O2) was mediated by a combination of increased PGI2 and decreased 20-HETE, and severe (0% O2) was almost entirely accounted for by an increase in PGI2 . In all cases, there appears to be significant involvement by the endothelium to mediate the dilation, which is further supported by a reduction in hypoxic dilation when isolated vessels were exposed to indomethacin (an inhibitor of AA metabolism and thus the production of PGI2) [137, 144, 145] and to a lesser degree by L-NAME (an inhibitor of NO production from NOS) . Therefore, PGI2 is likely a substantial contributor to hypoxic dilation with a lesser but likely still significant role for NO. Human studies measuring CBF with pcMRI, instead of isolated vessel diameter as in the aforementioned studies, with and without administration of L-NAME suggest that hypoxic dilation is highly dependent on NO, with no change from baseline observed in hypoxia when L-NAME was administered . These differences may be because of a species-specific response or differences in ex vivo and in vivo conditions. Additional mechanisms of hypoxic dilation include adenosine whose action has previously been discussed. In hyperoxic conditions constriction is favored, mediated by a greater conversion of AA to 20-HETE as opposed to dilators such as PGI2 .
4.4 Potassium and neurovascular coupling
Although K+ is not directly a byproduct of a metabolite pathway it tends to increase in concentration when the frequency of neuronal depolarization is increased, and is therefore indicative of increased metabolism. K+ channels are present in cerebrovascular smooth muscle cells and are important regulators of tone because of their ability to alter membrane potential [148, 149]. Although K+ channels are present through the peripheral vasculature, the cerebral circulation has a unique anatomical feature that allows for intimate interaction of astrocytic endfeet and cerebral vessels. This tight interaction between the astrocytic endfeet and vasculature allows for precise localized changes in blood flow to match the site-specific neural activity in the brain [150, 151]. K+ can therefore be thought of as a direct link between neuronal activity and blood flow. This pairing of neuronal metabolism with appropriate blood flow and is termed neurovascular coupling (NVC). NVC forms the mechanistic basis for neuroimaging techniques that are able to map changes in neuronal activity based on vascular responses such as changes in blood flow or oxygen saturation . As such, an understanding of NVC is not only crucial to understand the regulation of cerebral vascular tone but is also needed to interpret these neuroimaging techniques including functional magnetic resonance imaging (fMRI), positron emission tomography (PET), and near-infrared spectroscopy .
NVC is dependent on the interactions of the neurovascular unit which is made up of three major components: the vascular smooth muscle, neuron, and astrocyte glial cell. Somewhat surprisingly, the initiation site of neurovascular coupling is at the level of capillaries which then leads to changes upstream; however, once the close anatomical locations of neurons to the capillaries (8–20 μm) are considered this becomes a logical point of initiation . Not only does NVC rely on interactions apart from the vasculature but it also requires several structural components to facilitate cell-to-cell interactions of these different cell types. These components include gap junctions [155, 156], anchoring proteins  and specialized ion channels  expressed on cell–cell interface membranes. Neurons initiate NVC by generating direct signals that act on the vasculature and indirect signals that are transmitted through astrocytes and lead to increases in intracellular Ca2+ within the astrocyte. This is achieved by the glutamatergic synaptic activity initiating post-synaptic N-methyl-D-receptors (NMDA) and a-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors. The increase in intracellular Ca2+ activates Ca2+-dependent enzymes to produce vasodilators. Some of the enzymes activated include neuronal NOS (nNOS) and cyclooxygenase 2 (COX-2) [159, 160]. Glutamate also acts on metabotropic glutamate receptors in astrocytes increasing Ca2+ in these cells and leading to the production of vasoactive metabolites including adenosine, ATP, and K+ that act of the VSMCs .
Astrocytes are anatomically well positioned to transmit signals from neurons to the vasculature because of their close proximity to the capillaries; however, in a vascular network both the downstream and upstream vasculature must work in concert in order to effectively regulate blood flow delivery. It appears that endothelial cells provide the crucial role of retrograde propagation of the vasomotor response allowing for coordination between up and downstream vasculature. Endothelial cells can produce numerous metabolites including NO, prostanoids, and endothelin to alter VSMCs constriction. Endothelial intermediate K+ channels (KIR) channels and small conductance K+ channels (KSK) have been implicated as a mechanistic contributor to the propagation of the vasodilatory signal upstream for a synchronized hemodynamic response between the capillaries and arterioles via myoendothelial gap junctions . Secondary to the movement of hyperpolarizing current through gap junctions, K+ efflux from KIR and KSK channels generates what has been described as a “K+ cloud”  that also hyperpolarizes neighboring VSMC through both the inducible Na+-K+ electrogenic pump and stimulation of inward rectifying potassium channels in the VSMC of cerebral arteries and arterioles. In both cases, there is a net efflux of K+ from VSMC [164, 165, 166, 167] in conjunction with closed voltage-gated calcium channels and therefore results in the relaxation of the VSMC. The importance of K+ channels in dilating the cerebral vasculature is supported by reduced K+ conductance, depolarization and ultimately constriction seen in experimental inhibition of the channels . For example, 6–10 mM K+ applied to capillaries generated a hyperpolarizing response in endothelial cells which was transmitted upstream to penetrating arterioles hyperpolarizing and relaxing VSMCs . This propagation of the vasodilation signal was blocked with inhibition of KIR with barium or endothelial deletion of KIR1.2 channels, supporting its role transmitting the hyperpolarizing current. Interestingly the capillaries themselves did not dilate to K+ suggesting they act as a sensor detecting the signal for the upstream arterioles to be the effector of the response .
Although in a physiological setting K+ tends to lead to vasodilation, the response to K+ may be dependent on the [K+]. In isolated cerebral arterioles low (<7 mM) and moderate (8–15 mM) increases in [K+] result in endothelium-independent dilations and sustained dilation respectively through mechanisms previously discussed . Higher K+ concentrations cause constriction; however, under normal physiological conditions this concentration is not reached and is only seen under pathological processes such as spreading depression and stroke [169, 170]. This response seen in isolated vessel has also been confirmed in vivo . There is a combination of mechanisms at work in response to K+ that may explain the opposing responses to different concentrations. Alone an increase in [K+]o actually tends to depolarize neighboring VSMC resulting in constriction; this response predominates at high [K+]. At low concentrations, the minor increase in [K+]o would by itself lead to constriction but it also stimulates both KIR channels [150, 171, 172, 173, 174] and Na+/K+ ATPase [175, 176, 177] promoting K+ efflux as previously described. This description of the electrophysiology at work in the VSMC is supported by Em measurement in VSMCs when a solution containing [K+] (<20 mM) is applied and produces VSM hyperpolarization, whereas at higher K+ concentrations depolarization predominates [166, 172, 178]. Ultimately the signals generated by neurons, astrocytes, and endothelial cells must be received and integrated into a final level of tone by the VSMCs. A schematic representation of the impact of neurovascular coupling on cerebral vascular tone is presented in Figure 1.
5. Effect of diseased states
Each component implicated in the regulation of tone has a multitude of signals being produce that is intended to affect a vasomotor response from the VSMCs. Signals from myogenic, shear, metabolic, and neurovascular influences may be additive or opposing in their effect on VSMCs and further combine with one another creating intricate and precise regulation of cerebral vascular tone. It is this intricacy however, that also introduces many possible steps in the pathway for an abnormal response to occur. As such changes to the regulation of tone may form the basis of several pathologies including MetS. The contributing risk factors associated with MetS alter the local regulation of cerebral vascular tone by inducing changes in both the structure and function of the vessels. These risk factors include hypertension, T2DM, and obesity which promote a pro-inflammatory pro-oxidant state. Functionally, MetS is highly linked to increased smooth muscle activation and endothelial dysfunction which has important implications for the ability of a vessel to dilate in response to a multitude of stimuli previously discussed including hypotension, shear, hypoxia, and other metabolic stimuli. Increased myogenic properties are consistently overserved in MetS in multiple vascular beds [179, 180] including in the cerebral circulation [181, 182, 183]. This increase in constriction may be due to both a decrease in buffering capacity from endothelial dysfunction and alterations in the vascular smooth muscle itself [183, 184]. Endothelial dysfunction also has implication for shear-induced dilation since it is highly dependent on NO bioavailability which has consistently been shown to be reduced in MetS. The chronic inflammatory state seen in MetS likely contributes to the reduction in NO bioavailability due to increased scavenging to the produced NO by reactive oxygen species since this reduction in NO bioavailability is consistently reported to evolve in parallel with oxidant stress and the development of a chronic inflammatory state [185, 186]. This is supported by improved dilatory reactivity of MCA with the pretreatment of the cell-permeable superoxide dismutase mimetic TEMPOL in a model of T2DM . Interestingly some studies have actually shown an increase in eNOS expression which may be an attempt to compensate for the increased scavenging; however, they too continue to find reduced dilator reactivity .
Aside from reduced NO bioavailability, a shift in arachidonic acid metabolism toward constrictors and away from dilators that are highly responsible for hypoxic dilation has been demonstrated . Thus not only does endothelial dysfunction seen in MetS increase cerebrovascular resistance by decreased dilator metabolite production it may also promote the production of constrictors that exacerbate the impaired dilation capacity of cerebral vessels [184, 188]. A change in sensitivity to various metabolites in addition to their differential production may also contribute to differential vasomotor responses. For example, studies using SNP, an exogenous NO donor, while blocking endogenous NO production by eNOS using L-NAME, found smaller relaxation of the MCA in spontaneously hypertensive rats which was attributed to a decreased expression of soluble guanylate cyclase [189, 190, 191, 192]. In a model of T2DM decreased sensitivity of MCA to the PGI2 analog iloprost was also found suggesting that both decrease production and sensitivity of dilators may be contributing to the impaired dilation of the cerebral circulation in MetS .
Impaired dilation in response to exogenous dilator metabolites may also be due to vascular remodeling. Structural changes are an important consideration since even if the smooth muscle of a cerebral vessel is able to relax due to metabolic influences resulting in hyperpolarization, remodeling may prevent an increase in lumen diameter which is ultimately the major contributor to acute changes in resistance and thus the regulator of flow. Hypertension is largely implicated in the thickening of the vascular smooth muscle as well as increasing the ratio of collagen to elastin in the vessel. High intraluminal pressure increases the shear stress exerted on the vascular endothelium which normally could be restored to baseline by NO-induced vasodilation [193, 194, 195]; however, in a disease state with impaired NO production, there is a reduced ability to dilate resulting in endothelial damage and upregulation of atherogenic genes [193, 194, 195]. As a means of protection from chronic increased shear stress and wall tension that may lead to downstream edema cerebral vessels tend to hypertrophy with chronic hypertension, but this protective hypertrophy is also detrimental . Since wall tension is equal to intraluminal pressure X radius and wall stress is wall tension/wall thickness, hypertension-induced hypertrophy and inward remodeling resulting in a decrease in radius and increase wall thickness can normalize both the wall tension and wall stress . Although this remodeling may be protective in regards to increases in pressures and protecting the downstream capillaries from edema it increases the cerebrovascular resistance and limits the dilation reserve during hypotension and therefore presents itself as a right-shift in the autoregulatory range [196, 197]. The development of myogenic tone at higher pressure implies an increased lower limit of autoregulation. This predisposes cerebral tissue to reduced blood flow during hypotension. When pressure drops below the lower limit flow becomes dependent on the passive diameter of the vessel. Not only does an impaired lower limit predispose hypertensive individuals to ischemia but the reduced passive diameter from vascular remodeling further compromises flow to cerebral tissue resulting in hypoxic areas .
In addition to hypertrophy and inward remodeling, there is substantial arterial stiffening commonly seen in MetS. The pro-oxidant stress of ROS may interact with components of the perivascular matrix and initiate collagen cross-linking and deposition as well as the breakdown of elastin making the vessel less distensible . This is measured by a left shift in the stress-strain curve of isolated cerebral vessels under passive conditions achieved by using a Ca2+ free solution preventing the development of tone . The stiffening vessel from increased collagen to elastin ratio is made worse by the thickening of vessel walls previously discussed. In a model of T2DM with hypertension significant collagen deposition in addition to medial hypertrophy increasing the wall the lumen ratio and stiffness of the rats MCA  was demonstrated while T2DM in the absence of hypertension does not appear to induce structural changes to the cerebral vasculature . This along with data that suggests the increase in arterial stiffness seems to follow a time course similar to that of the onset of hypertension suggests a strong relationship between hypertension and vascular remodeling . Chronic uncontrolled hyperglycemia and inflammation do tend to lead to the development of hypertension and thus contribute to changes in the composition of cerebral vessels and it is likely the combination of both seen in MetS increases the degree to which remodeling occurs. A summary of the impact of disease states on the regulation of cerebral vascular tone is presented in Figure 2.
In conclusion, the regulation of vascular tone involves a complex set of pathways with myogenic, shear, and metabolic control. The mechanical influences of pressure and flow serve as a stimulus for the myogenic and shear responses to set a basal level of tone over a wide range so that metabolic factors have room to produce vasoactive responses on the vasculature. Due to the paramount importance of precise cerebral blood flow control these mechanisms are particularly pronounced and redundant in the cerebral circulation allowing for greater protection against insufficient perfusion or edema and capillary damage in situations of hypotension and hypertension respectively. However, due to the complexity of these homeostatic blood flow mechanisms there is the potential for the development of a pathological state. MetS presents a constellation of cardiovascular risk factors that are highly linked to the development of such cerebrovascular pathologies increasing the risk of stroke, TIA, and vascular dementia. The risk factors associated with MetS result in vascular remodeling which decreases the lumen size and increases stiffness and when paired with endothelial dysfunction and increased activation of the vascular smooth muscle it promotes increased cerebrovascular resistance. This right shifts the autoregulatory zone of myogenic regulation allowing for enhanced protection from hypertension but leaves cerebral tissue vulnerable to underperfusion. Controlling these risk factors and well as implementing targeted therapeutic strategies aimed at ameliorating the regulation of cerebrovascular tone has the potential to restore function in the cerebral circulation and improve current negative outcomes associated with MetS and cerebrovascular dysfunction.
This work was supported by Canadian Institutes of Health Research #389769 and Natural Sciences and Engineering Research Council (Canada) RGPIN-2018-05450. All figures were created with