Cyclooxygenase-derived vasoconstriction restrains hypoxia-mediated cerebral vasodilation in young adults with metabolic syndrome

Recent insights into mechanisms of hypoxia-induced vasodilatation in the human brain

The cerebral vasculature manages oxygen delivery by adjusting arterial blood in-flow in the face of reductions in oxygen availability. Hypoxic cerebral vasodilatation, and the associated hypoxic cerebral blood flow reactivity, involve many vascular, erythrocytic and cerebral tissue mechanisms that mediate elevations in cerebral blood flow via micro- and macrovascular dilatation. This contemporary review focuses on in vivo human work - with reference to seminal preclinical work where necessary - on hypoxic cerebrovascular reactivity, particularly where recent advancements have been made. We provide updates with the following information: in humans, hypoxic cerebral vasodilatation is partially mediated via a - likely non-obligatory - combination of: (1) nitric oxide synthases, (2) deoxygenation-coupled S-nitrosothiols, (3) potassium channel-related vascular smooth muscle hyperpolarization, and (4) prostaglandin mechanisms with some contribution from an interrelationship with reactive oxygen species. And finally, we discuss the fact that, due to the engagement of deoxyhaemoglobin-related mechanisms, reductions in O2 content via haemoglobin per se seem to account for ∼50% of that seen with hypoxic cerebral vasodilatation during hypoxaemia. We further highlight the issue that methodological impediments challenge the complete elucidation of hypoxic cerebral reactivity mechanisms in vivo in healthy humans. Future research is needed to confirm recent advancements and to reconcile human and animal findings. Further investigations are also required to extend these findings to address questions of sex-, heredity-, age-, and disease-related differences. The final step is to then ultimately translate understanding of these mechanisms into actionable, targetable pathways for the prevention and treatment of cerebral vascular dysfunction and cerebral hypoxic brain injury.

Reduced basal macrovascular and microvascular cerebral blood flow in young adults with metabolic syndrome: potential mechanisms

Ninety-million Americans suffer metabolic syndrome (MetSyn), increasing the risk of diabetes and poor brain outcomes, including neuropathology linked to lower cerebral blood flow (CBF), predominantly in anterior regions. We tested the hypothesis that total and regional CBF is lower in MetSyn more so in the anterior brain and explored three potential mechanisms. Thirty-four controls (25 ± 5 yr) and 19 MetSyn (30 ± 9 yr), with no history of cardiovascular disease/medications, underwent four-dimensional flow magnetic resonance imaging (MRI) to quantify macrovascular CBF, whereas arterial spin labeling quantified brain perfusion in a subset (n = 38/53). Contributions of cyclooxygenase (COX; n = 14), nitric oxide synthase (NOS, n = 17), or endothelin receptor A signaling (n = 13) were tested with indomethacin, NG-monomethyl-L-arginine (L-NMMA), and Ambrisentan, respectively. Total CBF was 20 ± 16% lower in MetSyn (725 ± 116 vs. 582 ± 119 mL/min, P < 0.001). Anterior and posterior brain regions were 17 ± 18% and 30 ± 24% lower in MetSyn; reductions were not different between regions (P = 0.112). Global perfusion was 16 ± 14% lower in MetSyn (44 ± 7 vs. 36 ± 5 mL/100 g/min, P = 0.002) and regionally in frontal, occipital, parietal, and temporal lobes (range 15-22%). The decrease in CBF with L-NMMA (P = 0.004) was not different between groups (P = 0.244, n = 14, 3), and Ambrisentan had no effect on either group (P = 0.165, n = 9, 4). Interestingly, indomethacin reduced CBF more in Controls in the anterior brain (P = 0.041), but CBF decrease in posterior was not different between groups (P = 0.151, n = 8, 6). These data indicate that adults with MetSyn exhibit substantially reduced brain perfusion without regional differences. Moreover, this reduction is not due to loss of NOS or gain of ET-1 signaling but rather a loss of COX vasodilation.NEW & NOTEWORTHY We tested the impact of insulin resistance (IR) on resting cerebral blood flow (CBF) in adults with metabolic syndrome (MetSyn). Using MRI and research pharmaceuticals to study the role of NOS, ET-1, or COX signaling, we found that adults with MetSyn exhibit substantially lower CBF that is not explained by changes in NOS or ET-1 signaling. Interestingly, adults with MetSyn show a loss of COX-mediated vasodilation in the anterior but not posterior circulation.

Dietary nitrate supplementation effect on dynamic cerebral autoregulation in normoxia and acute hypoxia

We tested the hypothesis that increasing the nitric oxide (NO) bioavailability by dietary nitrate would recover the hypoxia-induced reduction in dynamic cerebral autoregulation (CA). Twelve healthy males (age 21 ± 2 years) completed four days of dietary supplementation with a placebo or inorganic nitrate drink (140-ml beetroot juice per day) followed by 60-min of normoxia or hypoxia (fraction of inspired oxygen [FiO₂] = 13%). Duplex ultrasonography was used to perform volumetric change-based assessment of dynamic CA in the internal carotid artery (ICA). Dynamic CA was assessed by rate of regulation (RoR) of vascular conductance using the thigh-cuff method. Four days of beetroot supplementation increased circulating nitrate by 208 [171,245] μM (mean difference [95% confidence interval]) compared with placebo. Dynamic CA was lower in hypoxia than normoxia (RoR Δ-0.085 [-0.116, -0.054]). Compared with placebo, nitrate did not alter dynamic CA in normoxia (RoR Δ-0.022 [-0.060, 0.016]) or hypoxia (RoR Δ0.017 [-0.019, 0.053]). Further, nitrate did not affect ICA vessel diameter, blood velocity or flow in either normoxia or hypoxia. Increased bioavailability of NO through dietary nitrate supplementation did not recover the hypoxia-induced reduction in dynamic CA. This suggests the mechanism of hypoxia-induced reduction in dynamic CA does not relate to the availability of NO.

Nitric oxide synthase inhibition in healthy adults reduces regional and total cerebral macrovascular blood flow and microvascular perfusion

The importance of nitric oxide (NO) in regulating cerebral blood flow (CBF) remains unresolved, due in part to methodological approaches, which lack a comprehensive assessment of both global and regional effects. Importantly, NO synthase (NOS) expression and activity appear greater in some anterior brain regions, suggesting region-specific NOS influence on CBF. We hypothesized that NO contributes to basal CBF in healthy adults, in a regionally distinct pattern that predominates in the anterior circulation. Fourteen healthy adults (7 females; 24 ± 5 years) underwent two magnetic resonance imaging (MRI) study visits with saline (placebo) or the NOS inhibitor, L-NMMA, administered in a randomized, single-blind approach. 4D flow MRI quantified total and regional macrovascular CBF, whereas arterial spin labelling (ASL) MRI quantified total and regional microvascular perfusion. L-NMMA (or volume-matched saline) was infused intravenously for 5 min prior to imaging. L-NMMA reduced CBF (L-NMMA: 722 ± 100 vs. placebo: 771 ± 121 ml/min, P = 0.01) with similar relative reductions (5-7%) in anterior and posterior cerebral circulations, due in part to the reduced cross-sectional area of 9 of 11 large cerebral arteries. Global microvascular perfusion (ASL) was reduced by L-NMMA (L-NMMA: 42 ± 7 vs. placebo: 47 ± 8 ml/100g/min, P = 0.02), with 7-11% reductions in both hemispheres of the frontal, parietal and temporal lobes, and in the left occipital lobe. We conclude that NO contributes to macrovascular and microvascular regulation including larger artery resting diameter. Contrary to our hypothesis, the influence of NO on cerebral perfusion appears regionally uniform in healthy young adults. KEY POINTS: Cerebral blood flow (CBF) is vital for brain health, but the signals that are key to regulating CBF remain unclear. Nitric oxide (NO) is produced in the brain, but its importance in regulating CBF remains controversial since prior studies have not studied all regions of the brain simultaneously. Using modern MRI approaches, a drug that inhibits the enzymes that make NO (L-NMMA) reduced CBF by up to 11% in different brain regions. NO helps maintain proper CBF in healthy adults. These data will help us understand whether the reductions in CBF that occur during ageing or cardiovascular disease are related to shifts in NO signalling.

Integrated respiratory chemoreflex-mediated regulation of cerebral blood flow in hypoxia: Implications for oxygen delivery and acute mountain sickness

NEW FINDINGS

What is the central question of this study? To what extent do hypoxia-induced changes in the peripheral and central respiratory chemoreflex modulate anterior and posterior cerebral oxygen delivery, with corresponding implications for susceptibility to acute mountain sickness? What is the main finding and its importance? We provide evidence for site-specific regulation of cerebral blood flow in hypoxia that preserves oxygen delivery in the posterior but not the anterior cerebral circulation, with minimal contribution from the central respiratory chemoreflex. External carotid artery vasodilatation might prove to be an alternative haemodynamic risk factor that predisposes to acute mountain sickness.

ABSTRACT

The aim of the present study was to determine the extent to which hypoxia-induced changes in the peripheral and central respiratory chemoreflex modulate anterior and posterior cerebral blood flow (CBF) and oxygen delivery (CDO₂ ), with corresponding implications for the pathophysiology of the neurological syndrome, acute mountain sickness (AMS). Eight healthy men were randomly assigned single blind to 7 h of passive exposure to both normoxia (21% O₂ ) and hypoxia (12% O₂ ). The peripheral and central respiratory chemoreflex, internal carotid artery, external carotid artery (ECA) and vertebral artery blood flow (duplex ultrasound) and AMS scores (questionnaires) were measured throughout. A reduction in internal carotid artery CDO₂ was observed during hypoxia despite a compensatory elevation in perfusion. In contrast, vertebral artery and ECA CDO₂ were preserved, and the former was attributable to a more marked increase in perfusion. Hypoxia was associated with progressive activation of the peripheral respiratory chemoreflex (P < 0.001), whereas the central respiratory chemoreflex remained unchanged (P > 0.05). Symptom severity in participants who developed clinical AMS was positively related to ECA blood flow (Lake Louise score, r = 0.546-0.709, P = 0.004-0.043; Environmental Symptoms Questionnaires-Cerebral symptoms score, r = 0.587-0.771, P = 0.001-0.027, n = 4). Collectively, these findings highlight the site-specific regulation of CBF in hypoxia that maintains CDO₂ selectively in the posterior but not the anterior cerebral circulation, with minimal contribution from the central respiratory chemoreflex. Furthermore, ECA vasodilatation might represent a hitherto unexplored haemodynamic risk factor implicated in the pathophysiology of AMS.

Differential contribution of cyclooxygenase to basal cerebral blood flow and hypoxic cerebral vasodilation

Cyclooxygenase (COX) is proposed to regulate cerebral blood flow (CBF); however, accurate regional contributions of COX are relatively unknown at baseline and particularly during hypoxia. We hypothesized that COX contributes to both basal and hypoxic cerebral vasodilation, but COX-mediated vasodilation is greater in the posterior versus anterior cerebral circulation. CBF was measured in 9 healthy adults (28 ± 4 yr) during normoxia and isocapnic hypoxia (fraction of inspired oxygen = 0.11), with COX inhibition (oral indomethacin, 100mg) or placebo. Four-dimensional flow magnetic resonance imaging measured cross-sectional area (CSA) and blood velocity to quantify CBF in 11 cerebral arteries. Cerebrovascular conductance (CVC) was calculated (CVC = CBF × 100/mean arterial blood pressure) and hypoxic reactivity was expressed as absolute and relative change in CVC [ΔCVC/Δ pulse oximetry oxygen saturation (SpO2)]. At normoxic baseline, indomethacin reduced CVC by 44 ± 5% (P < 0.001) and artery CSA (P < 0.001), which was similar across arteries. Hypoxia (SpO2 80%-83%) increased CVC (P < 0.01), reflected as a similar relative increase in reactivity (% ΔCVC/-ΔSpO2) across arteries (P < 0.05), in part because of increases in CSA (P < 0.05). Indomethacin did not alter ΔCVC or ΔCVC/ΔSpO2 to hypoxia. These findings indicate that 1) COX contributes, in a largely uniform fashion, to cerebrovascular tone during normoxia and 2) COX is not obligatory for hypoxic vasodilation in any regions supplied by large extracranial or intracranial arteries.

Reactive oxygen species and cyclooxygenase products explain the majority of hypoxic cerebral vasodilation in healthy humans

AIM

The role of reactive oxygen species (ROS) in human cerebral blood flow (CBF) during hypoxia is largely unknown. Additionally, it is unknown whether ROS interact with cyclooxygenase-derived signals during hypoxia to increase CBF. We hypothesized ROS inhibition would reduce hypoxic CBF, and combined inhibition of cyclooxygenase (COX) and ROS would decrease hypoxic CBF more than ROS suppression alone.

METHODS

We measured middle cerebral artery velocity with transcranial Doppler ultrasound in 12 healthy adults during normoxia and 2 isocapnic hypoxia trials. Intravenous ascorbic acid infusion during the first hypoxia trial suppressed ROS. Oral indomethacin inhibited COX between hypoxia trials. The second bout of hypoxia tested the combined effects of ROS and COX inhibition. Middle cerebral artery velocity was normalized for blood pressure as cerebrovascular conductance index.

RESULTS

Hypoxia increased cerebrovascular conductance index in both trials (P < 0.05). Ascorbic acid infusion did not alter cerebrovascular conductance index during hypoxia. Combined ascorbic acid and indomethacin significantly reduced hypoxia-mediated increases in cerebrovascular conductance index from 17 ± 2 to 4 ± 1 cm s^-1 100 mm Hg^-1 (P < 0.05).

CONCLUSION

ROS are not obligatory for hypoxic cerebral vasodilation. Current data indicate ROS and COX together may account for the majority of the increase in CBF through the middle cerebral artery during hypoxia. These data are the first to demonstrate compensatory hypoxic vasodilatory signalling in human cerebral circulation.

Insulin resistance is associated with lower arterial blood flow and reduced cortical perfusion in cognitively asymptomatic middle-aged adults

Insulin resistance (IR) is associated with poor cerebrovascular health and increased risk for dementia. Little is known about the unique effect of IR on both micro- and macrovascular flow particularly in midlife when interventions against dementia may be most effective. We examined the effect of IR as indexed by the Homeostatic Model Assessment of Insulin Resistance (HOMA-IR) on cerebral blood flow in macro- and microvessels utilizing magnetic resonance imaging (MRI) among cognitively asymptomatic middle-aged individuals. We hypothesized that higher HOMA-IR would be associated with reduced flow in macrovessels and lower cortical perfusion. One hundred and twenty cognitively asymptomatic middle-aged adults (57 ± 5 yrs) underwent fasting blood draw, phase contrast-vastly undersampled isotropic projection reconstruction (PC VIPR) MRI, and arterial spin labeling (ASL) perfusion. Higher HOMA-IR was associated with lower arterial blood flow, particularly within the internal carotid arteries (ICAs), and lower cerebral perfusion in several brain regions including frontal and temporal lobe regions. Higher blood flow in bilateral ICAs predicted greater cortical perfusion in individuals with lower HOMA-IR, a relationship not observed among those with higher HOMA-IR. Findings provide novel evidence for an uncoupling of macrovascular blood flow and microvascular perfusion among individuals with higher IR in midlife.

Adenosine receptor-dependent signaling is not obligatory for normobaric and hypobaric hypoxia-induced cerebral vasodilation in humans

Hypoxia increases cerebral blood flow (CBF) with the underlying signaling processes potentially including adenosine. A randomized, double-blinded, and placebo-controlled design, was implemented to determine if adenosine receptor antagonism (theophylline, 3.75 mg/Kg) would reduce the CBF response to normobaric and hypobaric hypoxia. In 12 participants the partial pressures of end-tidal oxygen ([Formula: see text]) and carbon dioxide ([Formula: see text]), ventilation (pneumotachography), blood pressure (finger photoplethysmography), heart rate (electrocardiogram), CBF (duplex ultrasound), and intracranial blood velocities (transcranial Doppler ultrasound) were measured during 5-min stages of isocapnic hypoxia at sea level (98, 90, 80, and 70% [Formula: see text]). Ventilation, [Formula: see text] and [Formula: see text], blood pressure, heart rate, and CBF were also measured upon exposure (128 ± 31 min following arrival) to high altitude (3,800 m) and 6 h following theophylline administration. At sea level, although the CBF response to hypoxia was unaltered pre- and postplacebo, it was reduced following theophylline (P < 0.01), a finding explained by a lower [Formula: see text] (P < 0.01). Upon mathematical correction for [Formula: see text], the CBF response to hypoxia was unaltered following theophylline. Cerebrovascular reactivity to hypoxia (i.e., response slope) was not different between trials, irrespective of [Formula: see text] At high altitude, theophylline (n = 6) had no effect on CBF compared with placebo (n = 6) when end-tidal gases were comparable (P > 0.05). We conclude that adenosine receptor-dependent signaling is not obligatory for cerebral hypoxic vasodilation in humans.NEW & NOTEWORTHY The signaling pathways that regulate human cerebral blood flow in hypoxia remain poorly understood. Using a randomized, double-blinded, and placebo-controlled study design, we determined that adenosine receptor-dependent signaling is not obligatory for the regulation of human cerebral blood flow at sea level; these findings also extend to high altitude.

Cerebral blood flow regulation in women across menstrual phase: differential contribution of cyclooxygenase to basal, hypoxic, and hypercapnic vascular tone

In healthy young women, basal cerebral blood flow (CBF) and cerebrovascular reactivity may change across the menstrual cycle, but mechanisms remain untested. When compared with the early follicular phase of the menstrual cycle, we hypothesized women in late follicular phase would exhibit: 1) greater basal CBF, 2) greater hypercapnic increases in CBF, 3) greater hypoxic increases in CBF, and 4) increased cyclooxygenase (COX) signaling. We measured middle cerebral artery velocity (MCAv, transcranial Doppler ultrasound) in 11 healthy women (23 ± 1 yr) during rest, hypoxia, and hypercapnia. Subjects completed four visits: two during the early follicular (∼day 3) and two during the late follicular (∼day 14) phases of the menstrual cycle, with and without COX inhibition (oral indomethacin). Isocapnic hypoxia elicited an SPO2 = 90% and SPO2 = 80% for 5 min each. Separately, hypercapnia increased end-tidal CO2 10 mmHg above baseline. Cerebral vascular conductance index (CVCi = MCAv/MABP·100, where MABP is mean arterial blood pressure) was calculated and a positive change reflected vasodilation (ΔCVCi). Basal CVCi was greater in the late follicular phase (P < 0.001). Indomethacin decreased basal CVCi (∼37%) and abolished the phase difference (P < 0.001). Hypoxic ΔCVCi was similar between phases and unaffected by indomethacin. Hypercapnic ΔCVCi was similar between phases, and indomethacin decreased hypercapnic ΔCVCi (∼68%; P < 0.001) similarly between phases. In summary, while neither hypercapnic nor hypoxic vasodilation is altered by menstrual phase, increased basal CBF in the late follicular phase is fully explained by a greater contribution of COX. These data provide new mechanistic insight into anterior CBF regulation across menstrual phases and contribute to our understanding of CBF regulation in women.

Hypoxemia, oxygen content, and the regulation of cerebral blood flow

This review highlights the influence of oxygen (O2) availability on cerebral blood flow (CBF). Evidence for reductions in O2 content (CaO2 ) rather than arterial O2 tension (PaO2 ) as the chief regulator of cerebral vasodilation, with deoxyhemoglobin as the primary O2 sensor and upstream response effector, is discussed. We review in vitro and in vivo data to summarize the molecular mechanisms underpinning CBF responses during changes in CaO2 . We surmise that 1) during hypoxemic hypoxia in healthy humans (e.g., conditions of acute and chronic exposure to normobaric and hypobaric hypoxia), elevations in CBF compensate for reductions in CaO2 and thus maintain cerebral O2 delivery; 2) evidence from studies implementing iso- and hypervolumic hemodilution, anemia, and polycythemia indicate that CaO2 has an independent influence on CBF; however, the increase in CBF does not fully compensate for the lower CaO2 during hemodilution, and delivery is reduced; and 3) the mechanisms underpinning CBF regulation during changes in O2 content are multifactorial, involving deoxyhemoglobin-mediated release of nitric oxide metabolites and ATP, deoxyhemoglobin nitrite reductase activity, and the downstream interplay of several vasoactive factors including adenosine and epoxyeicosatrienoic acids. The emerging picture supports the role of deoxyhemoglobin (associated with changes in CaO2 ) as the primary biological regulator of CBF. The mechanisms for vasodilation therefore appear more robust during hypoxemic hypoxia than during changes in CaO2 via hemodilution. Clinical implications (e.g., disorders associated with anemia and polycythemia) and future study directions are considered.

Cerebrovascular regulation in men and women: stimulus-specific role of cyclooxygenase

Greater cerebral artery vasodilation mediated by cyclooxygenase (COX) in female animals is unexplored in humans. We hypothesized that young, healthy women would exhibit greater basal cerebral blood flow (CBF) and greater vasodilation during hypoxia or hypercapnia compared to men, mediated by a larger contribution of COX. We measured middle cerebral artery velocity (MCAv, transcranial Doppler ultrasound) in 42 adults (24 women, 18 men; 24 ± 1 years) during two visits, in a double-blind, placebo-controlled design (COX inhibition, 100 mg oral indomethacin, Indo). Women were studied early in the follicular phase of the menstrual cycle (days 1–5). Two levels of isocapnic hypoxia (S_PO₂ = 90% and 80%) were induced for 5-min each. Separately, hypercapnia was induced by increasing end-tidal carbon dioxide (PET_CO2) 10 mmHg above baseline. A positive change in MCAv (ΔMCAv) reflected vasodilation. Basal MCAv was greater in women compared to men (P < 0.01) across all conditions. Indo decreased baseline MCAv (P < 0.01) similarly between sexes. Hypoxia increased MCAv (P < 0.01), but ΔMCAv was not different between sexes. Indo did not alter hypoxic vasodilation in either sex. Hypercapnia increased MCAv (P < 0.01), but ΔMCAv was not different between sexes. Indo elicited a large decrease in hypercapnic vasodilation (P < 0.01) that was similar between sexes. During the early follicular phase, women exhibit greater basal CBF than men, but similar vasodilatory responses to hypoxia and hypercapnia. Moreover, COX is not obligatory for hypoxic vasodilation, but plays a vital and similar role in the regulation of basal CBF (∼30%) and hypercapnic response (∼55%) between sexes.

Effect of acute hypoxia on regional cerebral blood flow: effect of sympathetic nerve activity

We examined 1) whether global cerebral blood flow (CBF) would increase across a 6-h bout of normobaric poikilocapnic hypoxia and be mediated by a larger increase in blood flow in the vertebral artery (VA) than in the internal carotid artery (ICA); and 2) whether additional increases in global CBF would be evident following an α1-adrenergic blockade via further dilation of the ICA and VA. In 11 young normotensive individuals, ultrasound measures of ICA and VA flow were obtained in normoxia (baseline) and following 60, 210, and 330 min of hypoxia (FiO2 = 0.11). Ninety minutes prior to final assessment, participants received an α1-adrenoreceptor blocker (prazosin, 1 mg/20 kg body mass) or placebo. Compared with baseline, following 60, 220, and 330 min of hypoxia, global CBF [(ICAFlow + VAFlow) ∗ 2] increased by 160 ± 52 ml/min (+28%; P = 0.05), 134 ± 23 ml/min (+23%; P = 0.02), and 113 ± 51 (+19%; P = 0.27), respectively. Compared with baseline, ICAFlow increased by 23% following 60 min of hypoxia (P = 0.06), after which it progressively declined. The percentage increase in VA flow was consistently larger than ICA flow during hypoxia by ∼20% (P = 0.002). Compared with baseline, ICA and VA diameters increased during hypoxia by ∼9% and ∼12%, respectively (P ≤ 0.05), and were correlated with reductions in SaO2. Flow and diameters were unaltered following α1 blockade (P ≥ 0.10). In conclusion, elevations in global CBF during acute hypoxia are partly mediated via greater increases in VA flow compared with ICA flow; this regional difference was unaltered following α1 blockade, indicating that a heightened sympathetic nerve activity with hypoxia does not constrain further dilation of larger extracranial blood vessels.