Functional vascular contributions to cognitive impairment and dementia: mechanisms and consequences of cerebral autoregulatory dysfunction, endothelial impairment, and neurovascular uncoupling in aging

Abstract

Increasing evidence from epidemiological, clinical and experimental studies indicate that age-related cerebromicrovascular dysfunction and microcirculatory damage play critical roles in the pathogenesis of many types of dementia in the elderly, including Alzheimer’s disease. Understanding and targeting the age-related pathophysiological mechanisms that underlie vascular contributions to cognitive impairment and dementia (VCID) are expected to have a major role in preserving brain health in older individuals. Maintenance of cerebral perfusion, protecting the microcirculation from high pressure-induced damage and moment-to-moment adjustment of regional oxygen and nutrient supply to changes in demand are prerequisites for the prevention of cerebral ischemia and neuronal dysfunction. This overview discusses age-related alterations in three main regulatory paradigms involved in the regulation of cerebral blood flow (CBF): cerebral autoregulation/myogenic constriction, endothelium-dependent vasomotor function, and neurovascular coupling responses responsible for functional hyperemia. The pathophysiological consequences of cerebral microvascular dysregulation in aging are explored, including blood-brain barrier disruption, neuroinflammation, exacerbation of neurodegeneration, development of cerebral microhemorrhages, microvascular rarefaction, and ischemic neuronal dysfunction and damage. Due to the widespread attention that VCID has captured in recent years, the evidence for the causal role of cerebral microvascular dysregulation in cognitive decline is critically examined.

maintenance of adequate tissue perfusion through a dense cerebromicrovascular network is vital for the preservation of normal brain function (57, 119, 120, 260, 309). The total length of capillaries in the human brain is ~600 km and virtually every neuron is supplied by its own capillary. There is increasing evidence that aging elicits multifaceted functional impairment in the cerebral microcirculation, which plays a critical role in brain aging and the pathogenesis of age-related cognitive impairment (45, 54, 99, 123, 129, 166, 326). To recognize the contribution of cerebromicrovascular mechanisms to cognitive decline the phrase “vascular contributions to cognitive impairment and dementia (VCID)” was coined (46, 97, 237). The VCID concept implies that a spectrum of age-related vascular pathologies (including stroke, microinfarcts, microhemorrhages, leukoaraiosis, and cerebral amyloid angiopathy) can promote cognitive impairment in elderly patients. For the purpose of this review, we focus on the role of age-related dysregulation cerebral blood flow (CBF) in the development of cognitive decline.

Regulation of CBF has to comply with unique requirements, ensuring adequate delivery of nutrients and oxygen at all times, avoiding both hypoperfusion and hyperperfusion of the brain and enabling moment-to-moment adjustment of CBF. First, the brain has a very high metabolic demand for oxygen relative to other organs. Second, neurons do not have sufficient energy reserves. Third, metabolic demand rapidly changes with neuronal activation. Fourth, in the closed cranium the space is limited; thus regulation has to maintain normal blood flow and volume and thus intracranial pressure. Fifth, penetration of high pressure to the distal, vulnerable part of the cerebral arterial tree and consequential microvascular damage has to be prevented. To fulfill these requirements, regulation of CBF is exceedingly complex with multiple overlapping regulatory paradigms. There is increasing evidence that even mild impairment of CBF regulation has significant consequences on cerebral function, including impairment of cognition in the elderly. Furthermore, age-related alterations in homeostatic mechanisms also render the aged brain more susceptible to the damaging effects of the comorbid conditions (e.g., hypertension, obesity, neurodegenerative diseases) (50, 281, 287, 288). In this review, the effect of aging on key local vasoregulatory mechanisms acting in the cerebral circulation (myogenic autoregulation, endothelium-dependent pathways, and neurovascular coupling) is considered in terms of potential mechanisms involved in cerebrovascular dysfunction and its pathophysiological consequences.

Aging-Induced Changes in Autoregulation of CBF

Cellular mechanisms underlying autoregulation of CBF.

The integrated processes resulting in relatively constant CBF and microvascular pressure in the face of changing central arterial pressure are called autoregulation of CBF (153) (Fig. 1). Dynamic cerebral autoregulation refers to the ability to compensate fast changes in perfusion pressure by adjusting vascular resistance. Static cerebral autoregulation refers to adjustments of vascular resistance in response to larger steady-state changes in perfusion pressure. Dynamic and static cerebral autoregulations are not completely separate mechanistic entities and act on a continuum. The net result is that in healthy individuals CBF does not change in a linear manner with changes in systemic blood pressure and vascular resistance is readily adjusted to changes in perfusion pressure.

Fig. 1.

Aging impairs adaptation of cerebral blood flow (CBF) autoregulation to hypertension. A: scheme depicting that under normal conditions autoregulation of CBF maintains a nearly constant blood flow when perfusion pressure changes. This is ensured by pressure-induced myogenic constriction of the cerebral arteries (C), a homeostatic mechanism that rapidly adjust vascular resistance to changes in perfusion pressure. The significant increases in the resistance of proximal arteries also assure that increased arterial pressure does not penetrate the distal portion of the microcirculation and cause damage to the thin-walled arteriolar and capillary microvessels in the brain (103, 147). In young organisms in hypertension the myogenic constriction of cerebral arteries is enhanced (C) and the range of cerebrovascular autoregulation is extended (A), which represent functional adaptation of these vessels to higher systemic blood pressure, optimizing tissue perfusion and protecting the cerebral microcirculation. Aged cerebral arteries do not exhibit a hypertension-induced adaptive increase in myogenic constriction (D) and cerebrovascular autoregulatory dysfunction is manifested (B) (271, 281). E and F: proposed scheme showing that in young organisms activation of a 20-hydroxyeicosatrienoic-acid (20-HETE)/transient receptor potential cation channel (TRPC)-dependent pathway underlies functional adaptation of cerebral arteries to hypertension (blue arrows) and that this adaptive response is dysfunctional in aging (red arrows). Accordingly, in smooth muscle cells within the wall of young cerebral arteries (E), high pressure-induced mechanical stress leads to the activation of arachidonic acid metabolism (AA) by phospholipase A₂ (PLA₂), and upregulation of the 20-HETE producing CYP450 isoforms. The resulting increased production of the vasoconstrictor eicosanoid 20-HETE activates TRPC6 channels, resulting in increases in vascular smooth muscle Ca²⁺ concentration and subsequent sustained myogenic constriction (281). 20-HETE also blocks the activation of the hyperpolarizing Ca²⁺ activated potassium (BK_Ca) channels on vascular smooth muscle cells, which contributes to the increased pressure-induced activation of voltage-dependent L-type Ca²⁺ (L_Ca) channels and enhanced myogenic constriction. F: in aged cerebral arteries the functional adaptation to hypertension mediated by activation of the 20-HETE/TRPC-dependent pathway is impaired (red arrows).

The myogenic response, which is intrinsic to the vascular smooth muscle cells, is a key mechanism contributing to autoregulation of CBF (26, 27, 72, 93, 101, 102, 173, 180, 281, 306) (Fig. 1). Accordingly, cerebral arterial vessels actively dilate and constrict in response to decreases and increases in blood pressure, respectively (5, 43, 83, 84, 170, 171, 192). Importantly, in the cerebral circulation large proximal arteries represent a significant part (up to 40%) of total cerebrovascular resistance (73, 104, 147, 171, 258) and their myogenic response is critical for preventing high pressure from reaching the distal part of the cerebral circulation (236, 281). The myogenic reactivity of serially connected cerebral arteries and arterioles effectively protects the microcirculation against the harmful effects of rapid changes in blood pressure, exemplified by the maintenance of steady capillary perfusion pressure during changes in arterial pressure. In addition, a pressure-induced myogenic mechanism maintains intrinsic basal tone of the arterial microvessels, thus enabling optimization of tissue perfusion in the heterogeneous capillary network by neurometabolic and neurovascular/gliovascular coupling mechanisms.

Remarkable progress has been made in the past two decades to elucidate the cellular and molecular mechanisms underlying pressure-induced myogenic constriction of cerebral arterial vessels (96, 142–144, 312). In the search for a soluble mediator and a receptor-mediated signaling pathway previous studies demonstrated that vascular smooth muscle cells located in the wall of cerebral arteries express cytochrome P (CYP)450 4A enzymes that catalyze the formation of the potent vasoconstrictor arachidonic acid metabolite 20-hydroxyeicosatrienoic-acid (20-HETE) and that production of 20-HETE significantly increases in response to elevations in intravascular pressure (71, 92, 93) (Fig. 1). Previous studies showed that 20-HETE lead to activation of protein kinase C, inhibition of Ca²⁺-activated K⁺ channels, and activation of L-type Ca²⁺ (L_Ca) and transient receptor potential cation channel 6 (TRPC6) channels, which promote depolarization of vascular smooth muscle cells, increasing intracellular Ca²⁺ levels and promoting vasoconstriction (102). The concept that production of 20-HETE plays a role in myogenic response is supported by the observations that inhibition of 20-HETE formation attenuates pressure-induced arterial myogenic constriction in vitro and impairs the autoregulation of CBF in vivo (102). Moreover, there is evidence that upregulated production of 20-HETE underlies increased myogenic response and autoregulatory adaptation to hypertension (68, 71, 271, 281). In addition to the role of 20-HETE synthesis (158) other pathways, including other stretch-activated TRP channels (TRPM4) (214) and chloride channels (183), integrins, and other cytoskeletal elements (42, 53) and pathways governing smooth muscle cell Ca²⁺ sensitivity (27, 230, 231) also contribute to pressure-induced depolarization and consequent increase in intracellular Ca²⁺ concentration in vascular smooth muscle cells and the development of myogenic constriction of cerebral arteries.

In cerebral arteries pressure-induced myogenic constriction also appears to be augmented by a unique mechanism: flow-induced vasoconstriction. Since the original observations of Schretzenmayr in 1933 (229), there have been hundreds of reports documenting that arteries from virtually all vascular beds in the peripheral circulation (including brachial, femoral, mesenteric, and coronary arteries) dilate in response to increases in blood flow (145). The cerebral circulation is an important exception. While basilar arteries were reported to dilate in response to increases in flow (86) similar to peripheral vessels, isolated middle cerebral arteries of the rat (33, 273), mouse (281) and cat (163) and fronto-temporal small arteries isolated from the human brain (273) exhibit significant constriction in responses to increases of intraluminal flow/shear stress. Flow-induced constriction of both human and rodent cerebral arteries was shown to be mediated by 20-HETE acting via thromboxane/endoperoxide receptors (273). The initial observation that isolated rabbit pial resistance arteries dilate in response to increases in flow when intraluminal pressure is low, but they constrict in response to the same increases in flow when pressure is high (90, 91), led to the formulation of the hypothesis that flow-induced constriction may play a role in autoregulation of CBF (145). Interestingly, like the myogenic response flow-induced constriction is enhanced in hypertension, as well, probably representing another adaptive vasomotor mechanism to high blood pressure (90, 281). In theory, if cerebral arteries dilated to flow, it would reduce the magnitude of myogenic constriction, counteracting myogenic autoregulation of CBF. In contrast, flow-induced constriction is predicted to act as a negative feedback mechanism to autoregulate CBF in concert with pressure-induced myogenic constriction. Further in vivo studies should provide direct experimental evidence to support or reject this hypothesis (308).

Role of autoregulation in cerebromicrovascular protection.

The myogenic response of proximal cerebral arteries plays a critical role in neuroprotection, by preventing the penetration of high pressure to the thin-walled distal portion of the microcirculation and protecting the microcirculation from high pressure-induced damage (41, 281, 283). Direct measurements of cerebromicrovascular pressure demonstrate that approximately half of the total vascular resistance in brain depends on changes in the segmental resistance of vessels upstream from the penetrating arteries (for an excellent review see Ref. 57). As a result of the significant resistance of larger proximal arteries, the high central systolic pressure cannot penetrate the microcirculation under steady-state conditions (Fig. 2). In healthy young individuals increases in blood pressure, episodic or sustained, result in proportionate increases in cerebral vascular resistance such that, due to a larger pressure drop along the proximal resistance arteries, the increased pressure does not penetrate the thin-walled microvessels (Fig. 2). Studies in young experimental animals show that during chronic hypertension resistance of both larger and smaller resistance arteries increases (168, 313). Because of these adaptive changes in resistance, capillary pressure is maintained relatively constant (Fig. 2). Thus the thin-walled cerebral microvessels are protected from barotrauma as long as the autoregulatory protective mechanisms are intact and the blood pressure remains within the physiological autoregulatory range. It is believed that alterations in the cerebral autoregulatory capacity in different pathological conditions significantly contribute to cerebromicrovascular damage (281).

Fig. 2.

Age-related autoregulatory dysfunction exacerbates hypertension-induced cerebromicrovascular injury. Shown is a schematic illustration of the likely consequences of autoregulatory dysfunction in the aging brain. The model proposed implies that in healthy young organisms pressure-induced myogenic constriction of the proximal cerebral arteries acts as a critical homeostatic mechanism that assures that increased arterial pressure does not penetrate the distal portion of the microcirculation and cause damage to the thin-walled arteriolar and capillary microvessels in the brain (103, 147). In aging, proximal resistance arteries lose their capability to adapt to hypertension with an enhanced pressure-induced constriction, which leads to a mismatch in perfusion pressure and segmental vascular resistance (resistance is inversely related to the 4th power of vessel radius). Lack of proper autoregulatory protection in aging likely allows high blood pressure to penetrate the vulnerable downstream portion of the cerebral microcirculation. The hemodynamic burden exacerbates age-related disruption of the blood-brain barrier (BBB), leading to extravasation of plasma factors, which promote neuroinflammation (e.g., activation of microglia by IgG via the IgG Fc receptors). Microglia-derived proinflammatory cytokines, chemokines, proteases [i.e., matrix metalloproteinase (MMP)] and reactive oxygen species (ROS) promote neuronal damage (273, 281). In addition, the increased microvascular pressure activates matrix metalloproteinases in the vascular wall in a redox-sensitive manner, contributing to the development of microhemorrhages (276). The age-related autoregulatory dysfunction and its consequences may also contribute to the dysfunction of the glymphatic system (128, 148), and the development of age-related vascular rarefaction (281). We posit that exacerbation of neuroinflammation, cerebral microhemorrhages, glymphatics dysfunction and/or microvascular rarefaction are causally linked to hypertension-induced cognitive impairment in aging (85, 210, 285) and contribute to the increased prevalence of Alzheimer’s disease in hypertensive elderly individuals. Bottom: representative images showing cerebral microhemorrhages (brown lesions after diaminobenzidine-hematoxylin staining, scale bar = 200 μm) in the brain of aged (24 mo old) hypertensive mice, which associate with autoregulatory dysfunction. Note that most hypertension-induced microhemorrhages are located in the cortical and subcortical region. Hypertension was induced in the mice by treatment with angiotensin II and the nitric oxide synthase inhibitor nitro-l-arginine methyl ester (l-NAME) (279).

Age-related changes in dynamic and static components of cerebral autoregulation.

Recent studies suggest that age-related alterations in the cerebral autoregulatory capacity may play an important role in the pathophysiology of brain aging. In mice aging impairs the dynamic component of the myogenic response of isolated cerebral arteries induced by a sudden increase in pressure (281) and impairs myogenic adaptation to pulsatile pressure (250). In contrast, the static component of the myogenic response and static autoregulation of CBF are largely unaffected in the autoregulated range (269, 281). In aged rodents the upper limit of CBF autoregulation appears to be unchanged (281), whereas the lower limit of CBF autoregulation increases by ~20 mmHg (152, 284). The diminished compensatory dilatation of aged cerebral resistance arteries during hypotension likely increases the risk of ischemia of the brain during hypotensive conditions. Analysis of the available human data yielded mixed results (216). In elderly patients Lipsitz et al. (154) found retained dynamic autoregulation by transfer function analysis during standing and sit-to-stand challenges. Other studies also reported retained dynamic autoregulatory function in aged patients challenged by negative pressure release, Valsalva maneuver, thigh cuff test, or sit-to-stand maneuver indicated by normal autoregulatory index and unaffected transfer function in lower frequencies of oscillation in blood pressure (36, 181). Studies of Yam at al. (319) found also no differences between dynamic autoregulatory response of younger and older groups of patients. The findings of studies investigating aging-induced changes in response to sudden hypotension in humans are also controversial. Aging is associated with a higher incidence of postural symptoms (such as syncope), a common condition of sudden blood pressure drop (35), which can cause temporal hypoperfusion in the brain in case of ineffective compensatory decrease in cerebrovascular resistance. Larger postural reduction in cerebral cortical oxygenation (by near-infrared spectroscopy) and in mean blood flow velocity in middle cerebral arteries was found in elderly patients compared with young controls (160, 172). Preliminary studies also show that in the elderly impaired dynamic autoregulation (assessed using gain and phase) predicts development of symptoms during orthostatic tolerance test (226). In contrast, in another study autoregulatory response to hypotension during orthostatic stress was found to be unaffected by aging (140). Wollner et al. (318) investigating aged individuals with postural hypotension demonstrated that patients with clinical signs of cerebral ischemia exhibited autoregulatory failure, meanwhile a similar pressure drop did not cause any symptoms when autoregulatory function was intact. During ergomotor exercise dynamic autoregulation was reported to be intact; however, onset of autoregulatory correction of CBF was found to be delayed in older patients (110). Sorond et al. (244) found regional differences in changes of cerebrovascular resistance: there appears to be a significantly greater decline in blood flow velocity in the area of posterior cerebral artery than in the territory of middle cerebral artery indicating regional differences in dynamic autoregulatory function. Interestingly, the Lipsitz laboratory (232) reported an even more effective autoregulatory response to hypotension and during spontaneous oscillations in blood pressure both in treated and untreated hypertensive elderly patients, which was associated with a disturbed vasoreactivity to changes in CO₂ levels. In summary, studies investigating dynamic autoregulation in elderly patients reported variable results, mostly depending on the measured parameter of dynamic features of autoregulation. Further studies are evidently needed to resolve these controversies. When interpreting the aforementioned data, several limitations should also be considered, including methodological limitations (e.g., lack of direct measurement of cerebrovascular pressure and/or volumetric flow in most studies); potential confounding effects of lifestyle factors and medications, small sample size used in many of the studies relative to the age range of the participants, especially that of older subjects; the substantial interindividual variability in many of the parameters assessed; and the cross-sectional nature of most of the studies. In contrast to findings obtained under steady-state conditions, age-related autoregulatory dysfunction is more evident under conditions of hypertension (281) and increased pressure pulsatility (251) (see below).

Impaired autoregulatory adaptation to hypertension in aging.

In industrialized societies, there is a consistent age-related increase in systolic blood pressure (80). In healthy young individuals, the elastic conduit arteries (including the aorta and carotid artery) provide a Windkessel effect to dampen hemodynamic pulsatility and facilitate a continuous blood flow into the cerebral microvessels (23, 267). Due to the age-related stiffening and impaired Windkessel function of conduit arteries, the amplitude of systolic pressure in the aorta significantly increases with age (63, 206). The existing evidence suggests that such an increase in central pulse pressure is transmitted into organs that are characterized by low resistance and high blood flow (108, 175, 311). Arterial wave reflections returning from the peripheral resistance vessels may augment pressure pulsatility in the aged cerebral microcirculation. Hypertension in the elderly is a major risk factor for both large hemorrhagic strokes and microvascular injury (capillary damage, blood-brain barrier disruption, and microhemorrhages) contributing to the development of vascular cognitive impairment (97).

There is increasing evidence in support of the concept that age-related impairment of autoregulatory adaptation to hypertension contributes to the increased susceptibility of the elderly to hypertension-induced microvascular damage and cognitive decline (262, 287). Recently, we demonstrated that in mice aging is associated with impaired myogenic adaptation of cerebral arteries to pulsatile pressure (250). If arteries of elderly hypertensive patients also show impaired myogenic constriction when exposed to pulsatile pressure, this is probably associated with a significant hydrodynamic resistance decrease in the proximal larger resistance arteries, imposing a significantly larger burden on the downstream portion of the cerebral microcirculation. Importantly, a recent study demonstrated that in elderly individuals, higher pulse pressure led to increased CBF pulsatility (262). This finding supports the idea that with aging the cerebral microcirculation lacks protection against increases in pulsatile pressure (250). Recent findings provide important evidence to support the concept that pressure pulsatility and, consequently, CBF pulsatility increase due to age-related increases in central arterial stiffness and wave reflection in elderly patients (265). Cerebromicrovascular damage has long been hypothesized to result from the penetration of increased pulsatile pressure into the vulnerable distal portion of the microcirculation in the elderly (reviewed in Refs. 188, 247). Importantly, in aged individuals increased central arterial stiffness and higher pressure/CBF pulsatility are associated with increased incidence and volume of white matter damage (287). In the elderly activities that result in significant transient increases in blood pressure also represent a dynamic challenge to the autoregulatory mechanisms of the cerebral circulation. For example, the Valsalva maneuver, which causes a significant transient rise in arterial pressure for a short period of time, inadvertently occurs during daily activities in which straining is present (266). Heavy-weight lifting, defecation straining, playing of wind instruments, nose blowing, heavy coughing, and vomiting are all events that cause a sudden increase in arterial pressure, which simulates the Valsalva maneuver. Anger, startling, sexual intercourse, and vigorous physical exercise, all of which are documented trigger factors for intracerebral hemorrhage, are also characterized by significant transient increases in blood pressure, posing a challenge to the autoregulatory mechanisms of the cerebral circulation. There are also studies showing an association of early morning increases in blood pressure, which is coincident with arousal and arising from overnight sleep (131), with cerebrovascular events in elderly patients.

Previous studies provide evidence that in young organisms cerebral arteries exhibit functional and structural adaptation to hypertension, which protect the injury-prone distal portion of the cerebral microcirculation from pressure overload (179, 203, 222, 223, 254–256, 271, 281). Among these physiological adaptive responses the increased pressure-induced myogenic constriction of cerebral arteries is of great significance (103, 147, 191). Previous studies demonstrated that in young hypertensive animals increased pressure-induced myogenic constriction leads to an increased resistance at the level of the larger pial arteries (269, 281). With the manifestation of this adaptive vascular response, the protective CBF autoregulatory range extends to higher pressure values in hypertensive patients as well as in laboratory animals with pharmacologically induced hypertension (203, 254–256, 281) (Fig. 1). Recently, we provided evidence that cerebral arteries of aged mice do not exhibit a hypertension-induced adaptive increase in myogenic tone observed in young mice and aged-hypertensive animals do not show extension of CBF autoregulation to high pressure values (269, 281). The mechanisms responsible for the age-dependent loss of myogenic protection in hypertension likely involve dysregulation of the pressure-induced activation of the 20-HETE/TRPC6 pathway (271, 281). In theory, dysregulation of potassium channels, including BK_Ca channels, may contribute to functional maladaptation of resistance arteries to high pressure (29, 143). 20-HETE inhibits BK_Ca channels, which are known to be activated in the high pressure range (200) in the vascular smooth muscle cells in cerebral arteries. Yet, pharmacological inhibition of BK_Ca channels does not appear to significantly increase myogenic tone in cerebral arteries isolated from hypertensive aged mice (271). Future studies are warranted to elucidate the role of other mechanisms, including other TRP channels potentially involved in the mediation of myogenic mechanisms (TRPM4 etc.) in age-related functional maladaptation of cerebral arteries to hypertension.

In recent years a growing amount of evidence has provided support to the view that endocrine mechanisms play a crucial role in cerebrovascular alterations associated with advanced aging (241, 294). In particular, the age-related decline in circulating insulin-like growth factor-1 (IGF-1) levels appears to contribute significantly to vascular aging and age-related cerebrovascular alterations (11, 241, 275, 281, 294). Low circulating IGF-1 levels in humans are also associated with an increased risk for hypertension-induced microvascular brain damage (3) and stroke (130, 151), findings that have been also replicated in laboratory animals (240). Using a novel mouse model of endocrine IGF-1 deficiency (adeno-associated viral knockdown of IGF-1 specifically in the mouse liver using Cre-lox technology; Igf1^f/f + TBG-iCre-AAV8) (11), we showed that low circulating IGF-1 levels lead to impaired autoregulatory protection in the brain of hypertensive mice, potentially exacerbating cerebromicrovascular injury and neuroinflammation (281). Importantly, in IGF-1-deficient mice hypertension fails to upregulate TRPC6 expression and the TRPC-dependent component of the myogenic constriction (281), mimicking the aging phenotype. Experimental IGF-1 deficiency also mimics other aspects of cerebromicrovascular aging (11, 12, 241, 262, 275). For example, hypertension in rodent models of both aging (281) and IGF-1 deficiency (262) promotes cerebromicrovascular rarefaction. It should be noted that in response to hypertension cerebral arteries also exhibit structural adaptation. Vascular hypertrophy reduces wall stress. Inward remodeling contributes to adaptive increases in segmental vascular resistance, protecting the microcirculation (17–19, 57). Thus it is significant that both aging and IGF-1 deficiency are associated with impaired structural adaptation of cerebral microvessels to hypertension (unpublished observations), which likely exacerbates microvascular injury. There may be a cross talk between IGF-1 and insulin signaling pathways in the smooth muscle cells. There is initial evidence that both insulin resistance (132, 133) and IGF-1 deficiency (48) may impact mitochondrial function and mitochondrial reactive oxygen species (ROS) production in vascular smooth muscle cells, which may affect mechanotransduction of pressure, myogenic constriction, and vasomotor responses. Given the incidence of insulin resistance in aging, further studies are evidently needed to test these possibilities.

In elderly hypertensive patients, the lower limit of autoregulation of CBF is shifted to the right. Previous studies in spontaneously hypertensive rats also demonstrated an age-related shift in the lower limit of autoregulation, which results in significant reduction in CBF in response to experimentally induced hypotension (87, 115). The age- and hypertension-related mechanisms, which impair dilation of cerebral vessel in response to decreases in blood pressure, are presently poorly understood. Arterial morphological changes with aging, including thickening, stiffening, and eccentric remodeling (150), might contribute to the decreased capability of the cerebral vessels to dilate when intraluminal pressure is decreasing. An interesting consideration of the possible mechanisms is the role of endothelium. Recently, Bagher et al. (10) demonstrated that decreasing intraluminal pressure activates TRPV4 channels in endothelial cells of pressurized arterioles leading to increased frequency of spontaneous endothelial calcium events and activation of calcium-activated K⁺ channels, which then lead to dilation of the vessel. Age-associated endothelial dysfunction of cerebral vessels (109, 116, 166, 199, 281) might impair these mechanisms. These possibilities should be experimentally tested in the future.

Downstream consequences of cerebrovascular autoregulatory dysfunction.

The functional maladaptation of aged cerebral arteries to hypertension is likely responsible for the loss of autoregulatory protection in the aging brain, which likely allows high blood pressure to penetrate the distal, injury-prone portion of the cerebral microcirculation (Fig. 1). It is likely that when in elderly patients blood pressure exceeds the threshold for vascular injury and the autoregulatory ability of the resistance arteries to protect the cerebral capillaries is breached, microvascular damage also ensues (Fig. 2). Potential downstream consequences of cerebrovascular autoregulatory dysfunction (in the high pressure range) and pressure/volume overload include exacerbated disruption of the blood-brain barrier, neuroinflammation and neurodegeneration, structural damage to capillaries and capillary rarefaction, and increased propensity for intracerebral hemorrhages (Fig. 2). The existing data support this concept showing that in mice aging exacerbates hypertension-induced cerebromicrovascular damage and increases the incidence of cerebral microhemorrhages (276, 281). In aged mice increased blood-brain barrier permeability is exacerbated by hypertension, which associates with increased presence of activated microglia (281). The exacerbation of microvascular damage [including blood-brain barrier disruption (325, 326)] in aged hypertensive subjects is likely causally linked to increased neuroinflammation and cognitive decline (281) and is likely to contribute the known association between hypertension and Alzheimer’s disease in aging (50, 56, 59, 60, 64, 75, 77, 94, 97, 114, 118, 122, 124, 125, 138, 169, 178, 205, 215, 218, 235, 238, 239, 248). In that regard it is interesting that a high-impact recent study from the Zlokovic laboratory demonstrates that the level of blood-brain barrier disruption in the aged human hippocampus predicts cognitive impairment in elderly patients (177).

Although a direct cause-and-effect relationship is difficult to prove experimentally, the available clinical evidence strongly support the concept that cerebrovascular autoregulatory dysfunction is causally linked to downstream microcirculatory damage (185, 212, 228, 257, 264). Critical proof of concept was provided recently by the studies of Fan at al. (71) showing that experimental disruption of the myogenic machinery in cerebral arteries (by genetic inhibition of 20-HETE synthesis) results in significant microvascular damage, including blood-brain barrier disruption.

Hypertension in the elderly is often associated with small vessel disease (detected as white matter hyperintensities on MRI images) (reviewed in Ref. 194), which leads to gait disturbances and a decline in cognitive performance, executive function, and processing speed (139, 207, 310). The pathogenesis of with diffuse white matter disease is thought to involve microvascular injury, blood-brain barrier disruption, and consequential demyelination. There is growing evidence suggesting a causal relationship between cerebral autoregulatory dysfunction and brain white matter hyperintensity in older adults (30, 156, 212). The concept that age-related impairment of myogenic autoregulatory protection promotes hypertension-induced downstream microvascular damage and barrier disruption is supported by the observations that in the renal circulation of older hypertensive Faw-Hooded rats impairment of myogenic constriction of the afferent arterioles is associated with increased proteinuria, an indicator of downstream renal microvascular damage (302). The renal circulation features a prominent autoregulatory function similar to the cerebral circulation and previous studies show that in hypertensive humans renal microvascular injury often associates with clinical markers of cerebral microvascular damage (40, 272).

Aging-induced autoregulatory failure is also likely to contribute to increased prevalence of hypertension-induced intracerebral hemorrhages, especially cerebral microhemorrhages (261) (Fig. 2). Cerebral microhemorrhages are small (<5 mm in humans) vascular lesions associated with rupture of small intracerebral vessels and are considered of emerging importance as a contributing factor to the progressive impairment of neuronal function in aging. Epidemiological studies demonstrate that hypertension in aging is the major risk factor for the development of cerebral microhemorrhages (208). Recent data from animal models extend the clinical findings, showing that impaired functional adaptation of the aged cerebral arteries to hypertension exacerbates the development of cerebral microhemorrhages (279). Importantly, aging not only promotes the penetration of high pressure in the microcirculation but also alters pressure-induced mechanosensitive cellular and molecular pathways in the vascular wall, which render aged cerebral vessels vulnerable to the deleterious effects of hypertension (250, 279). Among other factors, aging was shown to exacerbate pressure-induced oxidative stress and promote activation of matrix metalloproteinases, compromising the structural integrity of cerebral arteries (250, 279).

In addition to prevention of high pressure-induced microcirculatory damage, autoregulation has also to avoid hypoperfusion of cerebral tissue. Due to dysfunction of cerebral autoregulation in hypertensive aged subjects (in whom the lower limit of autoregulation is shifted to higher pressures), inadequate dilation in response to hypotension may cause hypoperfusion and thus ischemic neuronal damage (74).

Aging-Induced Endothelial Dysfunction

The endothelial layer of cerebral vessels is capable of producing a variety of vasoactive substances [nitric oxide (NO), eicosanoid mediators, endothelium-derived hyperpolarizing factors (EDHFs), and endothelins] and it is in direct contact with blood flow making sensitive to changes in hemodynamic forces and various hormones present in the sera. The microvascular endothelium is involved in many aspects of the regulation of CBF. Endothelial NO contributes to setting resting CBF demonstrated by studies showing that acute blockade of NO synthases attenuates basal CBF and leads to hypoperfusion (117). Also, systemic administration of the NO precursor l-arginine increased CBF velocity in humans (88). Aging is associated with endothelial dysfunction in the cerebral circulation, similar to other vascular beds (31, 166, 176). The mechanisms underlying age-related endothelial dysfunction are multifaceted and involve oxidative stress. Accordingly, aging is associated with increased production of ROS in the vasculature of the brain and other organs (52, 195) in part due to an increased activity/expression of NADPH oxidases (195, 279). Aging also leads to increased mitochondrial production of superoxide (250) and impairment of Nrf2-dependent cellular antioxidative pathways (11, 47, 292, 293). Increased levels of superoxide readily react with NO to form peroxynitrite, decreasing the bioavailability of NO and leading to endothelial dysfunction (25, 281). Previous studies suggest that decreased endothelium-dependent vasodilation in aging is a universal phenomenon (89, 149, 198, 220) and may be exacerbated by upregulation of arginase (24, 134), which decreases cellular l-arginine supply, uncoupling of endothelial nitric oxide synthase, increases in assymetric dimethylarginine (ADMA) levels, endocrine changes (58), and age-related upregulation of angiotensin signaling and chronic vascular inflammation (55, 82, 184). Furthermore, aging-induced endothelial dysfunction is likely exacerbated by comorbid conditions, including metabolic diseases and hypertension (132, 281, 288, 290). Age-related endothelial dysfunction likely contributes to the chronic cerebral hypoperfusion observed in aging and consequent cerebral dysfunction, including cognitive decline (221, 309). Endothelium-dependent NO production also contributes to neurovascular coupling responses (38, 276, 279). Accordingly, recent studies demonstrate that endothelial dysfunction plays a critical role in aging-induced impairment of moment-to-moment adjustment of regional CBF to changes in neuronal activity (195). Endothelium-derived NO is also an important inhibitor of platelet aggregation, smooth muscle cell proliferation, and leukocyte adhesion and exerts potent anti-inflammatory, antiapoptotic, and proangiogenic effects (recently reviewed in Ref. 136). It also modulates cellular metabolism, mitochondrial function, and synaptic transmission (135, 213, 234, 238). Age-related decline in microvascular NO production, therefore, is likely to exert multifaceted detrimental effects on cerebrovascular, neuronal, astrocytic and microglial functions. Age-related impairment of microvascular endothelial cells also impairs angiogenic processes (15, 47, 298, 299), promoting microvascular rarefaction (290). Moreover, there is growing evidence implicating endothelial dysfunction in the pathogenesis of Alzheimer’s disease (61). Experimental studies also demonstrate that impaired endothelial NO production increases amyloid precursor protein, Aβ levels, promotes microglial activation, and exacerbates Aβ-induced impairment of cognitive function (6). For further reading on the effects of age-related endothelial dysfunction on the blood-brain barrier and its relation to neurodegenerative diseases (e.g., Alzheimer’s disease) we refer to the excellent recent review of Di Marco et al. (61). Aging may also modulate the endothelial production of arachidonic acid metabolites. For example, soluble epoxide-hydrolase [which catalyzes the hydrolysis of the dilator epoxyeicosatrienoic acids (EETs) into their inactive metabolites] was reported to be enhanced in the microvascular endothelium of older patients with cerebral small vessel disease and vascular cognitive impairment (182). There are also studies suggesting that the balance of constrictor and dilator eicosanoid metabolites produced in the microcirculation is altered by aging (16).

Aging-Induced Impairment of Neurovascular Coupling Responses

Cellular and molecular mechanisms of neurovascular coupling.

The energetic demand of neurons is very high, but the brain has very little reserve capacity. During neuronal activity there is a requirement for rapid increases in nutrients delivery, as well as washing-out of toxic metabolic by-products. Fulfilling this requirement regional CBF is closely adjusted to neuronal activation in a spatially and temporally well-regulated manner (69, 165). This is ensured by neurovascular coupling responses (“functional hyperemia”; Fig. 3), which maintain the optimal microenvironment for normal neuronal function (4). Neurovascular coupling responses depend on a coordinated interaction of neurons, astrocytes, endothelial cells and smooth muscle cells of cerebral arterioles (4). Recent findings also implicate pericytes and capillary dilation in the initial phase of the CBF response (100). Based on current models of neurovascular coupling, interaction of several parallel processes ensure that neuronal activity is coupled to localized vasodilation and increases in regional CBF. Upon neuronal act ivation, neuronal nitric oxide synthase-derived NO (34, 162) and/or neuronal prostaglandin release (37) can contribute to dilation of cerebral arterioles both indirectly, through modulating astrocytic mechanisms, and directly, acting on the arteriolar smooth muscle cells. Since astrocytes are positioned between neurons and vascular cells they are in ideal position to transform neuronal activation into blood vessel responses. It appears that the main astrocytic responses that contribute to increases in CBF during neuronal activation are triggered by glutamate released from synapses (209). Glutamate activates metabotropic glutamate receptors (mGluR) and NMDA receptors (253) on astrocytes, leading to increased Ca²⁺ influx, which activates the metabolism of arachidonic acid by cyclooxygenases to prostaglandins (PGE₂) and by epoxygenases to EETs. These mediators can cause dilation of cerebral blood vessels via mechanisms that involve activation of BK_Ca channels and TRPV4 channels on vascular smooth muscle cells (79, 187). Under pathological conditions arachidonic acid can be converted into 20-HETE in the neurovascular unit, which elicits constriction of cerebral arterioles counteracting the dilatory stimuli mediated by EETs, prostaglandins, and NO (39, 157). The current view is that the balance between production of dilator and constrictor metabolites of arachidonic acid is influenced by the preceding tone of cerebral vessels, the O₂ level and the availability of NO, among other factors. One of the most important signaling molecules by which astrocytes communicate with each other and with other cells is ATP and its metabolites, adenosine, and ADP. Since ATP is directly linked to astrocytic metabolism, it is logical to assume that purinergic pathways are involved in neurovascular coupling. Indeed, astrocyte-derived ATP, after hydrolyzed to adenosine, contributes to cerebral reactive hyperemia via A2A purinergic receptors on vascular smooth muscle cell. In addition to this pathway, astrocytic ATP released in response to neuronal activation may also act on endothelial P2Y1 receptors triggering the production of endothelial NO and subsequent vasodilation. Indeed, most studies (95, 323) (95, 253, 280), but not all (9), suggest that endothelium-derived NO, released in response to astrocyte-derived signals, contributes importantly to neurovascular coupling. Another hypothesis concerning astrocytic mechanisms of arteriolar dilation during neurovascular coupling centers on the potential dilator role of extracellular K⁺. According to the K⁺ syphoning theory (202, 317), after neuronal activation astrocytes take up excess extracellular K⁺ and transport it to the arterioles where they release it. K⁺ in the perivascular space is believed to activate K_ir in smooth muscle cells. Since the membrane potential of smooth muscle cells is higher than the reversal potential of K_ir, the resulting outward K⁺ flux leads to hyperpolarization, decreasing smooth muscle cell intracellular Ca²⁺ concentration ([Ca²⁺]_i), and dilating cerebral arterioles (317). The glutamate-induced [Ca²⁺]_i increase can open BK_Ca channels on astrocyte endfeet and release astrocytic K⁺ onto blood vessels, as well (81). An important recent study suggested that pericytes may also be involved in activity-evoked increase in CBF by dilating capillaries before arterioles dilate via a PGE₂-dependent pathway, which requires NO to suppress the production of constrictor 20-HETE (100).

Fig. 3.

Aging impairs neurovascular coupling responses: potential role of insulin-like growth factor-1 (IGF-1). Shown is a schematic illustration of age-related alterations in glio-endothelial coupling mechanisms, which are responsible for impaired functional hyperemia in the elderly. Accordingly, under normal conditions astrocytes mediate the interaction between neurons and vascular cells by physically connecting neuronal synapses to cerebrovascular smooth muscle wall. Glutamate released from active excitatory synapses triggers a calcium wave that travels through the astrocyte and reaches the end-feet wrapped around the vessel wall. The glutamate-induced calcium surge activates CYP450- and cyclooxygenase (COX)-mediated production of vasodilator eicosanoids [epoxyeicosatrienoic acids (EETs) and prostaglandins, respectively] and promotes activation of ATP release machinery. Astrocyte-derived ATP promotes endothelial release of vasodilator nitric oxide (NO) via activation of P2Y1 receptors (276). The model predicts that aging impairs all of these mechanisms involved in glio-vascular coupling responses. Of particular importance is the purinergic endothelial NO-mediated pathway, which may be affected by both endothelial oxidative stress [increased ROS production by NOX oxidases (195) and mitochondrial sources] and astrocyte-derived ROS production. The known age-related changes are showed using red arrows. Age-related decreases in levels of circulating IGF-1 is one of the most important endocrine changes accompanying aging. On the basis of evidence obtained in IGF-1-deficient mouse models of aging (275) the model predicts that age-related decline in IGF-1 impairs both astrocyte function and endothelium-mediated mechanisms of functional hyperemia. Note, that the scheme does not include IGF-1 deficiency-induced potential alterations in neuronal release of vasodilator substances and/or the role of IGF-1-related changes in astrocyte-mediated capillary dilation.

Aging-induced alterations in neurovascular coupling.

There is increasing evidence that neurovascular coupling is impaired in aging both in humans and laboratory animals (14, 195, 269, 322). Previous studies demonstrate that in healthy aged persons occipital blood flow responses to visual evoked potentials, measured by transcranial Doppler flowmetry, are significantly decreased compared with young ones (269, 322). Although another study using a similar approach did not report an age-related decline in flow response to visual signals, in this study all the participants were under 60 yr of age (219), which makes the interpretation of the findings in the context of aging difficult. Similar to the findings obtained in elderly humans laboratory rodents also exhibit age-related neurovascular uncoupling (195, 280). There are also studies extant suggesting that aging-induced changes in blood flow response to neuronal activation may show regional differences. For example, Sorond et al. found that aged individuals, unlike young subjects, during word stem completion cognitive task show decreased frontal responses (a brain region supplied by the anterior cerebral artery) as compared with occipital responses (a brain region supplied by posterior cerebral artery) (245).

The mechanisms by which aging impairs neurovascular coupling mechanisms are likely multifaceted (Fig. 3). Although attenuation of the underlying neuronal activity may theoretically contribute to impaired functional hyperemia, age-related neurovascular uncoupling appears to be reversible by interventions that improve cerebrovascular reactivity. The existing evidence suggests that increased production of ROS plays a central role in cerebromicrovascular impairment and neurovascular uncoupling in aging. This concept is supported by experimental findings showing that acute inhibition of NADPH oxidases is able to rescue neurovascular coupling in aged mice (195). Furthermore, aging is associated with increased ROS production in microvessels both in the brain and other vascular beds (52, 195) due, at least in part, to an increased activity/expression of NADPH oxidases (195). Interestingly, in mouse models of age-related AD-type neurodegeneration, enhanced generation of NADPH-derived ROS was also reported to contribute to neurovascular uncoupling (196, 197). There are multiple pathways through which increased ROS production may impair CBF responses induced by neuronal activation. An increasing body of literature supports the concept that endothelial NO production has an important role in neurovascular coupling (95, 159, 253). In aged rodents increased $O_2^{-}$ reacts with NO produced by the endothelial cells of cerebral microvessels forming ONOO⁻. The resulting decreases in bioavailability of NO likely contributes to the impaired neurovascular responses (195). Supporting this concept is the finding that in young rodents inhibition of endothelial NO synthesis using l-NAME significantly attenuates CBF responses to both whisker stimulation and the endothelium-dependent dilator acetylcholine, whereas it does not affect neurovascular responses and endothelium-dependent dilations in aged animals (279). Interestingly, in APP transgenic mice the antioxidants N-acetyl-l-cysteine and tempol also restored neurovascular function (186), suggesting that oxidative stress associated with the pathological processes of Alzheimer’s disease plays a causal role in neurovascular uncoupling in this age-related neurodegenerative diseases as well. Age-related endothelial dysfunction is reversible, which offers a potential target for therapeutic interventions for improvement of neurovascular coupling and, consequently, higher brain function in the elderly. Recent studies demonstrate that treatment of aged mice with resveratrol, which decreases cellular ROS production and downregulates NADPH oxidases (49, 204, 291, 295–297), restores microvascular endothelial function and neurovascular coupling in the brain of aged mice (279). It is possible that rescued neurovascular coupling contributes to the beneficial effects of resveratrol treatment on cognitive function reported in aged rodents (155, 190, 324). It is plausible to hypothesize that aging and age-related increased oxidative stress also impair other mechanisms of neurovascular signaling, such as the neuronal production of nitric oxide (321), astrocytic Ca²⁺-dependent signaling (164), and metabolism of arachidonic acid (137), the K⁺-dependent mediation of vascular dilation and the metabolism and action of glutamate (2). Since ischemic injury of pericytes has been causally linked to impaired neurovascular response (100) and the number pericytes and pericyte coverage of microvessels tend to decrease with age (288); it is also possible that age-related alterations of pericytes contribute to neurovascular uncoupling. Future studies are warranted to experimentally test the aforementioned hypotheses.

As discussed above, age-related IGF-1 deficiency was shown to significantly contribute to cerebromicrovascular alterations associated with aging (241, 294). Each cell type involved in neurovascular coupling (i.e., neurons, astrocytes, endothelial cells) abundantly expresses IGF-1 receptors and is a known target of IGF-1 (241). Importantly, we recently found that experimentally induced circulating IGF-1 deficiency impairs neurovascular coupling in the mouse somatosensory cortex via dysregulation of astrocytic glutamate-signaling, impairing production of astrocyte-derived EETs and increasing production of 20-HETE (275). In addition, IGF-1 deficiency also promotes cerebral oxidative stress and endothelial dysfunction, which also contribute to neurovascular uncoupling (273). Future studies are warranted to determine whether IGF-1 deficiency also predicts impaired functional hyperemia in elderly patients and to assess the efficacy of IGF-1 treatment to improve neurovascular coupling responses in animal models of aging.

Potential consequences of neurovascular uncoupling in aging.

Because the increased energy consumption has to be fueled by adequate amount of nutrients during neuronal activation (126), the attenuated increase in CBF to neuronal activation most likely disrupts the balance between the metabolic demand of the functioning cerebral tissue and the supply of nutrients (201). Therefore, age-related neurovascular uncoupling is expected to impair neuronal function in the brain. This concept is supported by studies conducted in aged laboratory animals and elderly humans (70, 252, 269, 322) showing that dysfunction of neurovascular coupling is associated with cognitive decline (243). The direct link between neurovascular uncoupling and cognitive decline is supported by recent studies showing that pharmacological treatment of young mice with inhibitors of the mediators of neurovascular coupling impairs functional hyperemia, which is associated with impairment of spatial and recognition memory (260). Importantly, elderly patients living with AD exhibit exacerbated impairment of neurovascular coupling responses (113, 127, 217), which may be contributing to worsening cognitive outcomes over time. Factors impairing neurovascular signaling, such as hypertension and accumulation of Aβ peptide in Alzheimer models (120–122, 141, 186, 268), are also associated with cognitive impairment in laboratory animals, which seem to be reversible by pharmacological treatments that improve neurovascular coupling (186, 268). On the basis of these observations, the neurovascular unit could be considered as a target for pharmacological intervention to reverse/delay cognitive decline associated with both aging and age-related neurodegenerative diseases.

Gait dysfunction of varying severity is present in a significant portion of elderly patients (189, 305) and is a major contributor to falls and predicts increased risk of institutionalization and death (1, 303, 304). Frontal executive functions play an important role in the cortical control of gait and there are studies extant suggesting that an association exists between neurovascular coupling and gait speed (242, 245). Future studies should determine whether pharmacological treatments, which improve neurovascular coupling are also effective in improving gait in the elderly.

Vasomotor responses associated with spreading depolarization highlight additional potential pathophysiological roles for neurovascular coupling responses. Cortical spreading depolarization is an intense depolarization wave that propagates in the cortex, triggering rapid vasoconstriction, followed by a pronounced hyperemic response and then a long-lasting oligemic phase (often called post-spreading depolarization oligemia) (7, 8, 21, 22, 44, 78, 111, 174, 300). There is strong clinical and experimental evidence that cortical spreading depolarizations occur after intracerebral hemorrhage, ischemic stroke, subarachnoid hemorrhage, as well as traumatic brain injury (7, 8, 13, 65–67, 105, 107, 112, 161, 193, 225, 227, 307, 316, 318). The clinical significance of these observations lies in the fact that injury-induced depolarizations (e.g., in stroke) propagate along ischemic, but viable, areas adjacent to the damaged core areas, exacerbating the mismatch between blood supply and metabolic demand and thereby worsening the clinical outcome (76, 106). Importantly, age-related alterations in neurovascular coupling pathways significantly increase both the incidence of spreading depolarizations and exacerbate their functional consequences (44, 78, 111, 174).

Perspectives

In the present review we have outlined the current understanding of mechanisms and consequences of age-related impairment of autoregulation of CBF, endothelial dysfunction, and neurovascular uncoupling (Fig. 4). Better understanding the specific age-related cellular and molecular mechanisms that underlie cerebromicrovascular aging is imperative to create usable tools for preventive and therapeutic interventions for age-related cognitive impairment. Future studies should provide answers to a number of critical questions about microvascular dysregulation in VCID. What is the functional role of mitochondrial oxidative stress in microvascular aging? What are the vascular effects of newly discovered mechanisms of aging? How astrocyte phenotype and function are altered in aging? What is the role of dysregulation of glymphatic flow in VCID? What therapeutic interventions are effective to protect the aging glymphatic system? Under what circumstances do perivascular macrophages contribute to cerebromicrovascular dysregulation and the pathogenesis of VCID? In the past decade significant progress has been made to understand the role of pericyte dysfunction in the pathogenesis of VCID (20, 224, 301, 314, 315). Treatments that target pericytes, preventing/reversing microvascular dysregulation in aging, are needed to be tested. An important area of future research is the link between age-related cerebromicrovascular dysregulation and its role in diffuse white matter disease. The vast majority of previous studies have focused on mainly the somatosensory cortex, whereas age-related microcirculatory alterations in the white matter are less understood. Furthermore, mechanisms involved in autoregulation of CBF and neurovascular coupling are potentially affected by medications used in the elderly. For example, recent studies indicate that widely used, non-steroid anti-inflammatory drugs (NSAIDs), given orally in usual therapeutic doses, inhibit neurovascular coupling in humans (32, 259). A number of drugs (including calcium antagonists) have also the potential to interfere with cerebral autoregulation. Potential cerebrovascular effects of these drugs should be considered in elderly patients, especially in those with advanced atherosclerosis of the arteries supplying the brain, as they may increase the risk of cerebral ischemia. Furthermore, the complex effects of multiple comorbidities and aging (e.g., co-occurrence of hypertension and obesity in geriatric patients with the metabolic syndrome) must be studied simultaneously. Most extant studies investigate the effects of only one disease state in young animal models. To mimic real-life conditions, animal models of aging have to be utilized and interaction of various risk factors has to be elucidated. A critical area for future research will be to develop therapeutic interventions that improve autoregulatory protection of the microcirculation and prevent microhemorrhages. Finally, future studies should determine whether interventions that target the microvasculature can prevent and/or reverse cognitive decline associated with aging and age-related pathologies.

Fig. 4.

Functional vascular contributions to cognitive impairment and dementia in aging. The schematic representation illustrates the interrelated microvascular mechanisms that contribute to age-related cognitive decline. The model highlights that age-related IGF-1 deficiency compromises the neurovascular unit, impairing the function of astrocytes, endothelial cells and smooth muscle cells. The resulting endothelial dysfunction and decreased NO bioavailability, increased oxidative stress, and/or dysregulation of astrocytic mediators contribute to neurovascular uncoupling, which impairs cognitive function due to inadequate supply of oxygen and nutrients to active brain regions. Age-related impairment of microvascular homeostasis, including alterations of myogenic autoregulatory mechanisms, renders the aged brain more susceptible to damage induced by comorbid conditions such as hypertension. In particular, the model predicts that impaired myogenic adaptation to hypertension promotes both the pathogenesis of cerebral microhemorrhages and blood-brain-barrier disruption, contributing to neuronal damage and cognitive decline. Aging and age-related IGF-1 deficiency also promote structural remodeling of the cerebral microcirculation, including microvascular rarefaction, contributing to an age-related decline in cerebral blood flow. They also promote structural maladaptation to hypertension, increasing microvascular fragility. Additionally, age-related microvascular proinflammatory alterations, impairment of vascular clearance of toxic waste products (such as Aβ) and metabolic by-products from the brain parenchyma and impaired trophic function of the microvascular endothelium that regulate stem cell self-renewal and differentiation in neurogenic niches could be implicated in impaired cognitive function.

GRANTS

This work was supported by an American Heart Association grant (to P. Toth, S. Tarantini, A. Csiszar, and Z. Ungvari), Bolyai Research Scholarship of the Hungarian Academy of Sciences Grant BO/00634/15 (to P. Toth), National Center for Complementary and Alternative Medicine Grant R01-AT-006526 (to Z. Ungvari), National Institute on Aging Grants R01-AG-047879 and R01-AG-038747, National Institute of Neurological Disorders and Stroke Grant R01-NS-056218 (to A. Csiszar), Arkansas Claude Pepper Older Americans Independence Center at University of Arkansas Medical Center Grant P30-AG-028718 (to Z. Ungvari), an Oklahoma Center for the Advancement of Science and Technology grant (to A. Csiszar and Z. Ungvari), an Oklahoma IDeA Network for Biomedical Research Excellence grant (to A. Csiszar), and a Reynolds Foundation grant (to A. Csiszar and Z. Ungvari).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

P.T., S.T., A.C., and Z.I.U. conception and design of research; P.T., S.T., A.C., and Z.I.U. prepared figures; P.T., S.T., A.C., and Z.I.U. drafted manuscript; P.T., S.T., A.C., and Z.I.U. edited and revised manuscript; P.T., S.T., A.C., and Z.I.U. approved final version of manuscript.