Circulating endothelin-1 alters critical mechanisms regulating the cerebral microcirculation

INTRODUCTION

Endothelin-1 (ET1), a 21-amino acid peptide produced mainly by endothelial cells, is one of the most potent vasoconstrictors known¹. ET1 acts on two distinct receptors (ET_A and ET_B), both coupled to phospholipase C via a GTP-binding protein¹. ET_A receptors (ET_AR) are expressed on vascular smooth muscle cells (VSMC) and, possibly, brain endothelial cells² and their activation leads to vasoconstriction¹. ET_BR are mainly located on endothelial cells and induce vasodilatation by promoting NO release¹.

ET1 has been implicated in a wide variety of neurological diseases³-5, and several studies in animal models have documented that ET_AR are involved in the CBF reduction observed in cerebral ischemia⁶, traumatic brain injury⁵, and subarachnoid hemorrhage⁷. Furthermore, increased plasma levels of ET1 have been observed in risk factors for stroke, such as hypertension, diabetes, obstructive sleep apnea, and cardiac failure⁸-11. The deleterious effects of ET1 have been attributed predominantly to its potent vascular actions¹². However, most studies have focused on the contractile effects of ET1 on vascular smooth muscle cell (VSMC) and little is known on the effect of ET1 on the mechanisms regulating the cerebral circulation¹³.

The brain is uniquely dependent on a well-regulated delivery of blood flow matched to its changing energy needs dictated by neural activity¹⁴. Complex cerebrovascular control mechanisms assure that the brain is adequately perfused at all times. For example, neural activity increases cerebral blood flow (CBF) (functional hyperemia) to support the increased energy requirements, whereas endothelial cells regulate the distribution of CBF within cerebrovascular networks by releasing vasoactive factors¹⁵. Alterations in these regulatory mechanisms have been implicated in a growing numbers of neurological conditions, ranging from ischemic stroke to neurodegenerative diseases¹⁶. ET1 is involved in the cerebrovascular dysfunction induced by angiotensin II “slow pressor” hypertension¹⁷, chronic intermittent hypoxia¹⁸, diabetes¹⁹, multiple sclerosis²⁰ and amyloid-β, a peptide involved in the pathogenesis of Alzheimer’s disease²¹. However, it remains to be established whether ET1 can reproduce the cerebrovascular alterations observed in these models.

In this study we investigated the effect of ET1 on critical regulatory mechanisms of the brain circulation. Using a combination of in vivo and in vitro approaches, we found that elevations in circulating ET1 comparable to those observed in cerebrovascular conditions profoundly alter the regulation of CBF by neural activity and by endothelial cells. Surprisingly, these effects, mediated by endothelial ET_AR, are not associated with a reduction in resting CBF and are not caused by oxidative stress. Rather, the dysfunction depends on endothelial rho kinase (ROCK) activity, which, in turn, leads to suppression of nitric oxide (NO) production by modulating the phosphorylation state of eNOS. These new findings demonstrate that circulating ET1 is able to impair vital regulatory mechanisms of the cerebral circulation. Such impairment in neurovascular regulation could render the brain more susceptible to injury by compromising the balance between energy demands and blood flow delivery.

METHODS

Materials and Methods pertaining to ROS detection, immunofluorescence, electron microscopy, cell cultures, western blotting, and biochemical measurements are described in the online-only Data Supplement.

General Surgical Procedures

All of the procedures were approved by the institutional animal care and use committee of Weill Cornell Medical College. Studies were conducted in 3-month–old C57Bl/6 male mice (weight: 25 to 30 g) (The Jackson Laboratory). Mice were anesthetized with isoflurane (2% maintenance) intubated and artificially ventilated (SAR-830, CWE Inc)¹⁸. Mean arterial pressure (MAP), rectal temperature, and blood gases were monitored and controlled¹⁸. After surgery, anesthesia was maintained with urethane (750 mg/kg ip) and chloralose (50 mg/kg ip)¹⁸.

Monitoring of CBF

CBF was monitored with a laser-Doppler probe (Periflux System 5010, Perimed AB) in a cranial window overlying the somatosensory cortex bathed with a modified Ringer’s solution¹⁸. CBF was expressed as percentage increases relative to the resting level.

Experimental Protocol

After stabilization of MAP and blood gases (Table S2, please see http://hyper.ahajournals.org), baseline CBF responses were tested with i.v. infusion of vehicle (saline; rate: 200μL/h). Functional hyperemia was tested by repetitive deflection of the facial whiskers (3-5 Hz; 60 sec) contralateral to the window and recording the corresponding CBF changes. To test endothelial-dependent responses, ACh (10μmol/L) or {"type":"entrez-nucleotide","attrs":{"text":"A23187","term_id":"833253","term_text":"A23187"}}A23187 (5μmol/L) were topically applied for 3-5 min to the widow and the CBF changes recorded²². In the mouse cerebral microcirculation ACh acts by releasing NO from eNOS¹⁵, whereas {"type":"entrez-nucleotide","attrs":{"text":"A23187","term_id":"833253","term_text":"A23187"}}A23187 acts through cyclooxygenase reaction products²³. Adenosine (400μmol/L) or the NO donor SNAP (10-50 μmol/L) were used as smooth muscle relaxants. After obtaining baseline CBF responses, ET1 (Sigma) was administered intravenously (20, 35 and 50pmol/Kg/min; rate: 150-200μL/h) and CBF responses tested again 20 min later when a stable MAP was reached. In some experiments BQ123, BQ788, Y27632, or MnTBAP were applied to the cortex prior to i.v. infusion of ET1.

Data Analysis

Data are expressed as mean±SEM. Two group comparisons were evaluated using the Student’s t-test. Multiple comparisons were evaluated by the analysis of variance and Tukey’s test. Differences were considered statistically significant for p<0.05.

RESULTS

1. ET1 infusion does not affect resting CBF or the BBB, but impairs functional hyperemia and CBF response to ACh

ET1 infusion (20, 35, 50 pmol/Kg/min) increased MAP dose dependently reaching a stable increase after 15-20 min (fig. 1A). ET1 infusion did not reduce resting CBF, but attenuated the increase in CBF produced by whisker stimulation and ACh at 35 and 50pmol/Kg/min (fig. 2C-D). Unlike the MAP increase, the effect on CBF responses was already maximal at 35pmol/Kg/min. CBF responses to {"type":"entrez-nucleotide","attrs":{"text":"A23187","term_id":"833253","term_text":"A23187"}}A23187 or adenosine were not suppressed (fig.1E-F). Similarly, ET1 did not affect the increase in CBF induced by the NO donor SNAP (fig. S1A, please see http://hyper.ahajournals.org). To determine whether the effects of ET1 on CBF were secondary to entry of the peptide into the brain due to opening of the BBB we assessed the integrity of the BBB after 45 min of ET1 infusion. ET1 (35pmol/Kg/min) did not increase the BBB permeability to EB (fig. S1B, please see http://hyper.ahajournals.org), but the anticipated increase was observed after focal cerebral ischemia (fig. S1B, please see http://hyper.ahajournals.org), attesting to the validity of the method. In agreement with the BBB results, ET1 infusion increases ET1 levels in plasma (vehicle: 1.9±0.3; ET1: 15.3±3.9 pg/ml; p<0.05; n=5/group) and lung (vehicle: 245.4±90.5; ET1: 920.8±228.2 pg/ml; p<0.05; n=5/group), but not in brain (vehicle: 5.7±0.5; ET1: 5.0±0.4 pg/mg; p<0.05; n=5/group).

Figure 1

ET1 increases MAP (A) without altering resting CBF (B). ET1 attenuates the increase in CBF produced by whisker stimulation (C), ACh (D), but not {"type":"entrez-nucleotide","attrs":{"text":"A23187","term_id":"833253","term_text":"A23187"}}A23187 (E) and adenosine (F). *p<0.05 form respective vehicle group; Analysis of variance and Tukey’s test; n=8-10/group.

Figure 2

BQ123 (A, B) but not BQ788 (C, D), reverses the ET1-induced attenuation of the CBF responses to whisker stimulation and ACh. (E) ET_AR immunoreactivity is present in endothelial cells (arrow) and vascular smooth muscle cells (arrowheads). (F) ET_AR immunoreactivity (arrows) is in the cytoplasm and plasma membrane of endothelial cells. (G) ET_AR immunoreactivity in bEnd.3 cells and western blotting showing ET_AR expression in neocortex and bEnd.3 cells. Calibration bar: 10μm. *p<0.05 from vehicle group; n=5-8/group.

3. ROS are not involved in the cerebrovascular effects of circulating ET1

ET1 induces vascular oxidative stress ¹²,24, a well known cause of cerebrovascular dysfunction²⁵. Therefore, we examined whether ROS are involved in the cerebrovascular dysfunction induced by ET1 infusion. However, neocortical application of the ROS scavenger MnTBAP (100μmol/L), effective in suppressing AngII-induced cerebrovascular dysfunction in this model²⁶, did not abrogate the cerebrovascular effects of ET1 (fig. 3A-B). To confirm that ET1 did not induce oxidative stress, we assessed ROS production using hydroethidine fluoromicrography. Consistent with the findings with MnTBAP, ET1 infusion did not induce statistically significant ROS increases (fig. 3C). However, infusion of AngII (0.25μg/Kg/min; 200μL/h) markedly increased somatosensory cortex ROS production (fig. 3C).

Figure 3

The ROS scavenger MnTBAP does not reverse the ET1-induced suppression of the CBF response to whisker stimulation (A) or ACh (B). ET1 infusion does not increase ROS production in the somatosensory cortex (C). On the contrary, infusion of AngII markedly increased somatosensory cortex ROS production (C). *p<0.05 from respective vehicle group; n=5/group.

4. ROCK is involved in the cerebrovascular effects of circulating ET1

ROCK has been implicated in the vascular effects of ET1²⁷,28. Therefore, we used the ROCK inhibitor Y27632 to test the hypothesis that this enzyme is involved in the cerebrovascular dysfunction induced by circulating ET1. Y27632 superfusion did not affect resting CBF or baseline cerebrovascular responses (Table S1, fig. S2C, please see http://hyper.ahajournals.org), but counteracted the ET1-induced attenuation of the CBF response to whisker stimulation or ACh (fig. 4A-B). In contrast, Y27632 did not rescue the cerebrovascular dysfunction produced by systemic administration of AngII (fig. 4C-D), highlighting key differences in the mechanisms of the acute cerebrovascular effects of ET1 and AngII, the latter mediated by NADPH-derived ROS in this model²⁵.

Figure 4

Y27632 rescues the ET1-induced attenuation of the CBF response to whisker stimulation (A) or ACh (B), whereas Y27632 does not rescue the dysfunction induced by AngII (C, D). *p<0.05 from respective vehicle group; n=5-8/group.

5. ET1 attenuates acetylcholine-induced production of nitrite in brain endothelial cell cultures, an effect dependent on ET_AR and ROCK

The finding that ET1 attenuates the increase in CBF induced by ACh, an eNOS-dependent response, suggests that ET1 may affect the production of NO by endothelial cells. To address this issue, we used cultures of brain endothelial cells. As shown in figure 6A, ACh increased nitrite levels an effect blocked by the NOS inhibitor L-NNA (100μmol/L), confirming that ACh induces eNOS activation in these cells. The increase in nitrite production induced by ACh was attenuated by ET1, an effect counteracted by BQ123, but not BQ788 (fig. 5B), implicating ET_AR. Western blotting demonstrated ET_AR expression in the culture (fig. 2G). Our in vivo data implicate ROCK activation in the cerebrovascular effects of ET1. Therefore, we examined whether the ET1-induced attenuation of the nitrites produced by ACh in endothelial cultures was also dependent on ROCK activity. As anticipated, the ROCK inhibitor Y27632 (1μmol/L) counteracted the effects of ET1 on ACh-induced nitrite production (Figure 5B), suggesting that endothelial ROCK is involved in the effects of ET1 on NO production in vitro.

Figure 5

ET1 reduces ACh-induced nitrite production in brain endothelial cells (A). BQ123 and Y27632 but not BQ788 prevent the effect of ET1 on ACh-induced nitrite production (B). ACh increase p-eNOS (Ser^{) followed by an increase of} p-eNOS (Thr^{) (C). ET1 attenuates the} p-Ser ^{and increases the} p-Thr ^{induced by ACh (D). Optical densities of p-Ser and p-Thr bands were measured and changes in phosphorylation levels were expressed relative to vehicle treated cells. *p<0.05 from vehicle; † p<0.05 from ACh; n=8-10/group.}

Figure 6

BQ123 and Y27632 prevent the effect of ET1 on eNOS phosphorylation state (A-D). Optical densities of p-Ser ^{and p-Thr bands were measured and changes in phosphorylation levels were expressed relative to vehicle treated cells. *p<0.05 from vehicle and ACh + ET1 + BQ123 or Y27632; t-test.}

1. ET1 infusion does not affect resting CBF or the BBB, but impairs functional hyperemia and CBF response to ACh

ET1 infusion (20, 35, 50 pmol/Kg/min) increased MAP dose dependently reaching a stable increase after 15-20 min (fig. 1A). ET1 infusion did not reduce resting CBF, but attenuated the increase in CBF produced by whisker stimulation and ACh at 35 and 50pmol/Kg/min (fig. 2C-D). Unlike the MAP increase, the effect on CBF responses was already maximal at 35pmol/Kg/min. CBF responses to {"type":"entrez-nucleotide","attrs":{"text":"A23187","term_id":"833253","term_text":"A23187"}}A23187 or adenosine were not suppressed (fig.1E-F). Similarly, ET1 did not affect the increase in CBF induced by the NO donor SNAP (fig. S1A, please see http://hyper.ahajournals.org). To determine whether the effects of ET1 on CBF were secondary to entry of the peptide into the brain due to opening of the BBB we assessed the integrity of the BBB after 45 min of ET1 infusion. ET1 (35pmol/Kg/min) did not increase the BBB permeability to EB (fig. S1B, please see http://hyper.ahajournals.org), but the anticipated increase was observed after focal cerebral ischemia (fig. S1B, please see http://hyper.ahajournals.org), attesting to the validity of the method. In agreement with the BBB results, ET1 infusion increases ET1 levels in plasma (vehicle: 1.9±0.3; ET1: 15.3±3.9 pg/ml; p<0.05; n=5/group) and lung (vehicle: 245.4±90.5; ET1: 920.8±228.2 pg/ml; p<0.05; n=5/group), but not in brain (vehicle: 5.7±0.5; ET1: 5.0±0.4 pg/mg; p<0.05; n=5/group).

Figure 1

ET1 increases MAP (A) without altering resting CBF (B). ET1 attenuates the increase in CBF produced by whisker stimulation (C), ACh (D), but not {"type":"entrez-nucleotide","attrs":{"text":"A23187","term_id":"833253","term_text":"A23187"}}A23187 (E) and adenosine (F). *p<0.05 form respective vehicle group; Analysis of variance and Tukey’s test; n=8-10/group.

Figure 2

BQ123 (A, B) but not BQ788 (C, D), reverses the ET1-induced attenuation of the CBF responses to whisker stimulation and ACh. (E) ET_AR immunoreactivity is present in endothelial cells (arrow) and vascular smooth muscle cells (arrowheads). (F) ET_AR immunoreactivity (arrows) is in the cytoplasm and plasma membrane of endothelial cells. (G) ET_AR immunoreactivity in bEnd.3 cells and western blotting showing ET_AR expression in neocortex and bEnd.3 cells. Calibration bar: 10μm. *p<0.05 from vehicle group; n=5-8/group.

2. The cerebrovascular effects of ET1 are mediated by ET_A receptors

Next, we examined the role of ET_AR and ET_BR in the cerebrovascular effects of ET1. Neocortical superfusion with the ET_AR antagonist BQ123 (1 μmol/L) reversed the attenuation in CBF responses produced by ET1 (35pmol/Kg/min), whereas the ET_BR antagonist BQ788 (100 nmol/L) did not (fig. 2B). Neither antagonist influenced resting CBF or baseline CBF responses (Table S1, fig. S2A-B, please see http://hyper.ahajournals.org). Then, we used immunocytochemistry to examine the localization of ET_A R in pial arterioles (diameter 80-100μM) of the somatosensory cortex. As anticipated¹, patchy ET_AR immunoreactivity was observed deep in the vessel wall consistent with smooth muscle cell localization (fig. 2E). However, ET_AR immunoreactivity was also observed in cerebral endothelial cells and endothelial cell cultures (fig. 2E-G). To provide further evidence for ET_AR in endothelial cells we used immunogold electron microscopy and demonstrated that the receptor was present in the membrane and the cytoplasm of cerebral endothelial cells (fig. 2F).

3. ROS are not involved in the cerebrovascular effects of circulating ET1

ET1 induces vascular oxidative stress ¹²,24, a well known cause of cerebrovascular dysfunction²⁵. Therefore, we examined whether ROS are involved in the cerebrovascular dysfunction induced by ET1 infusion. However, neocortical application of the ROS scavenger MnTBAP (100μmol/L), effective in suppressing AngII-induced cerebrovascular dysfunction in this model²⁶, did not abrogate the cerebrovascular effects of ET1 (fig. 3A-B). To confirm that ET1 did not induce oxidative stress, we assessed ROS production using hydroethidine fluoromicrography. Consistent with the findings with MnTBAP, ET1 infusion did not induce statistically significant ROS increases (fig. 3C). However, infusion of AngII (0.25μg/Kg/min; 200μL/h) markedly increased somatosensory cortex ROS production (fig. 3C).

Figure 3

The ROS scavenger MnTBAP does not reverse the ET1-induced suppression of the CBF response to whisker stimulation (A) or ACh (B). ET1 infusion does not increase ROS production in the somatosensory cortex (C). On the contrary, infusion of AngII markedly increased somatosensory cortex ROS production (C). *p<0.05 from respective vehicle group; n=5/group.

4. ROCK is involved in the cerebrovascular effects of circulating ET1

ROCK has been implicated in the vascular effects of ET1²⁷,28. Therefore, we used the ROCK inhibitor Y27632 to test the hypothesis that this enzyme is involved in the cerebrovascular dysfunction induced by circulating ET1. Y27632 superfusion did not affect resting CBF or baseline cerebrovascular responses (Table S1, fig. S2C, please see http://hyper.ahajournals.org), but counteracted the ET1-induced attenuation of the CBF response to whisker stimulation or ACh (fig. 4A-B). In contrast, Y27632 did not rescue the cerebrovascular dysfunction produced by systemic administration of AngII (fig. 4C-D), highlighting key differences in the mechanisms of the acute cerebrovascular effects of ET1 and AngII, the latter mediated by NADPH-derived ROS in this model²⁵.

Figure 4

Y27632 rescues the ET1-induced attenuation of the CBF response to whisker stimulation (A) or ACh (B), whereas Y27632 does not rescue the dysfunction induced by AngII (C, D). *p<0.05 from respective vehicle group; n=5-8/group.

5. ET1 attenuates acetylcholine-induced production of nitrite in brain endothelial cell cultures, an effect dependent on ET_AR and ROCK

The finding that ET1 attenuates the increase in CBF induced by ACh, an eNOS-dependent response, suggests that ET1 may affect the production of NO by endothelial cells. To address this issue, we used cultures of brain endothelial cells. As shown in figure 6A, ACh increased nitrite levels an effect blocked by the NOS inhibitor L-NNA (100μmol/L), confirming that ACh induces eNOS activation in these cells. The increase in nitrite production induced by ACh was attenuated by ET1, an effect counteracted by BQ123, but not BQ788 (fig. 5B), implicating ET_AR. Western blotting demonstrated ET_AR expression in the culture (fig. 2G). Our in vivo data implicate ROCK activation in the cerebrovascular effects of ET1. Therefore, we examined whether the ET1-induced attenuation of the nitrites produced by ACh in endothelial cultures was also dependent on ROCK activity. As anticipated, the ROCK inhibitor Y27632 (1μmol/L) counteracted the effects of ET1 on ACh-induced nitrite production (Figure 5B), suggesting that endothelial ROCK is involved in the effects of ET1 on NO production in vitro.

Figure 5

ET1 reduces ACh-induced nitrite production in brain endothelial cells (A). BQ123 and Y27632 but not BQ788 prevent the effect of ET1 on ACh-induced nitrite production (B). ACh increase p-eNOS (Ser^{) followed by an increase of} p-eNOS (Thr^{) (C). ET1 attenuates the} p-Ser ^{and increases the} p-Thr ^{induced by ACh (D). Optical densities of p-Ser and p-Thr bands were measured and changes in phosphorylation levels were expressed relative to vehicle treated cells. *p<0.05 from vehicle; † p<0.05 from ACh; n=8-10/group.}

Figure 6

BQ123 and Y27632 prevent the effect of ET1 on eNOS phosphorylation state (A-D). Optical densities of p-Ser ^{and p-Thr bands were measured and changes in phosphorylation levels were expressed relative to vehicle treated cells. *p<0.05 from vehicle and ACh + ET1 + BQ123 or Y27632; t-test.}

6. ET1 suppresses NO production by altering eNOS phosphorylation through ROCK

eNOS phosphorylation is a key regulator of eNOS activity²⁹. Thr ^{phosphorylation is associated with reductions and Ser phosphorylation with increases in eNOS catalytic activity30}. Therefore, we examined whether the effects of ET1 on eNOS activity are related to eNOS phosphorylation. In agreement with previous reports³⁰, ACh elicited a relatively rapid (1-5 min) and transient increase in p-Ser ^{eNOS, followed by a delayed and slow rise in p-Thr eNOS (}fig. 5C, E, F). ET1 suppressed ACh-induced eNOS activation by reducing p-Ser ^{and enhancing p-Thr (}fig. 5D-F), an effect prevented by BQ123 (fig. 6A-B). ROCK can inhibit eNOS by increasing p-Thr ^{and reducing p-Ser31},32,33. Therefore, we examined the effect of the ROCK inhibitor Y27632 on the attenuation of ACh-induced eNOS phosphorylation by ET1. As illustrated in fig. 6C-D, Y27632 prevented the effects of ET1 on ACh-induced eNOS phosphorylation implicating ROCK in the mechanisms of ACh-induced eNOS phosphorylation.

DISCUSSION

Novel findings of the study

This study has several novel aspects. First, we demonstrated that ET1 impairs the regulation of the cerebral circulation by endothelial cells and neural activity. Most studies on the vascular actions of ET1 in brain and other organs have focused on the ability of this peptide to constrict arteries and reduce flow, and little was known on the effects of this peptide on the dynamic regulation of organ blood flow¹²,13. Second, we found that the endothelial dysfunction was not mediated by vascular oxidative stress, but was dependent on activation of endothelial ROCK leading to changes in the phosphorylation state of eNOS. Third, we found that the effect of ET1 on endothelial NO production was mediated by endothelial ET_AR, which we localized to endothelial cells both in vivo and in vitro. These observations demonstrate for the first time that increases in ET1 plasma levels are associated with profound alterations of vital mechanisms regulating the cerebral circulation, mediated by a previously unrecognized effect of ET1 on endothelial ROCK activity and eNOS phosphorylation.

The cerebrovascular effects of circulating ET1 are mediated by ET_AR

The cerebrovascular dysfunction induced by ET1 was dependent on ET_AR. Using immunocytochemistry to determine the cellular localization of ET_AR, we found that ET_AR immunoreactivity is present both in VSMC and in brain endothelial cells. The endothelial localization of ET­_AR was also confirmed by immunogold electron microscopy and by western blotting in brain endothelial cell cultures. ET_AR are abundant in VSMC³⁴, but the present results cannot be attributed to these receptors since ET1 did not reduce resting CBF and did not impair the CBF response evoked by smooth muscle relaxants. Therefore, our data suggest that endothelial ET_AR are the target of plasma ET1 and mediate the deleterious vascular effects of the peptide.

Circulating ET1 does not alter resting CBF

In some vascular districts ET1 has been reported to produce vasodilatation at low concentrations, an effect attributed to ET_BR, and vasoconstriction at higher concentrations via ET_AR³⁵. We found that circulating ET1 did not affect resting CBF. This finding is in agreement with previous observations demonstrating that circulating ET1 does not reduce brain blood flow except in areas outside the BBB, such as choroid plexus and dura mater³⁶. Considering that ET1 in our model did not breech the BBB in the somatosensory cortex, the data support the hypothesis that ET1 needs to cross the BBB in order to reduce CBF. Therefore, circulating ET1 could induce a more severe disruption of cerebrovascular function, involving also VSMC, if the BBB is disrupted.

Selectivity of the cerebrovascular effects of circulating ET1

The cerebrovascular dysfunction induced by ET1 did not result from an indiscriminate impairment of cerebrovascular reactivity. Rather, only CBF responses evoked by neural activity and ACh were attenuated, whereas those evoked by {"type":"entrez-nucleotide","attrs":{"text":"A23187","term_id":"833253","term_text":"A23187"}}A23187, adenosine, and SNAP were preserved. These findings have important mechanistic implications for the cerebrovascular effects of circulating ET1. First, they suggest that ET1 does not impair the ability of smooth muscle cells to relax, but only the capacity of neural activity and selected endothelial agonists to increase CBF. Second, the fact that the CBF increase produced by ACh, a response mediated by eNOS, was attenuated, while the response to {"type":"entrez-nucleotide","attrs":{"text":"A23187","term_id":"833253","term_text":"A23187"}}A23187, which is mediated by cyclooxygenase²³ was not affected, suggests that ET1 interferes only with NO-mediated endothelial responses. Finally, the finding that ET1 did not inhibit the response to the NO donor SNAP suggests that ET1 does not act by impairing the ability of NO-cGMP to relax vascular smooth muscles.

Based on these observations we hypothesized that ET1 impairs the ability of endothelial cells to generate NO. This hypothesis was supported by the observation that ET1 suppresses ACh-induced NO production in endothelial cell cultures. Consistent with the results in vivo, the effect was dependent on ET_AR and ROCK activity. Since ROCK can inhibit eNOS activity by modulating its activatory (Ser^{) and inhibitory (Thr) phosphorylation sites31}-33, we investigated the effect of ET1 on eNOS phosphorylation. We found that ET1 leads to Thr ^{phosphorylation and Ser de-phosphorylation resulting in a reduction in eNOS-dependent NO production. Our findings, taken together, implicate ET1 in the regulation of eNOS phosphorylation via endothelial ROCK. Although ET1 could also inhibit the NO response to ACh by downregulating endothelial muscarinic receptors, that fact that ET1 does not inhibit muscarinic receptor signaling37} argues against this possibility.

Mechanisms of the effects of ET1 on functional hyperemia

We also found that ET1 attenuates the increase in CBF produced by neural activity. ET1 could attenuate functional hyperemia by suppressing neuronal NOS activity, which plays a role in the CBF increase¹⁴. Since we did not detect an increase in ET1 in the brain after iv infusion, we doubt that the peptide crossed the BBB. Although small increases in brain ET1 below the sensitivity of the assay cannot be ruled out, a direct neuronal action of the peptide is unlikely. Furthermore, the fact that ET1 did not alter resting CBF, which is closely linked to basal neural activity¹⁴, supports the hypothesis that ET1 did not act by suppressing neural activity. Another possibility is that plasma ET1 exerts its effects via the circumventricular organs (CVOs), structures outside the BBB that could affect cerebrovascular function broadly through neurohumoral mechanisms. Indeed, we recently reported that chronic administration of AngII alters neurovascular coupling by acting on the subfornical organ, one of the CVOs, and inducing cerebrovascular oxidative stress¹⁷. However, this possibility seems unlikely because the effects of circulating ET1 on functional hyperemia are not mediated by ROS. Therefore, additional studies are required to elucidate how ET1 affects neurovascular coupling.

ROS do not contribute to the cerebrovascular effects of circulating ET1

ROS play a major role in cerebrovascular dysfunction induced by different agent, including AngII and β-amyloid²⁵,38. ET1 has also been reported to induce oxidative stress in blood vessels, endothelial cells and VSMC, effects attributed to mitochondrial ROS, eNOS uncoupling or activation of the superoxide producing enzyme NADPH oxidase¹²,24. In contrast, we found that circulating ET1 does not increase ROS in the neocortex and that a ROS scavenger is not able to reverse the cerebrovascular dysfunction, suggesting that ROS do not play a primary role in the cerebrovascular dysfunction. The reasons for this discrepancy remain unclear, but the difference could be due to the fact that the ET1 concentrations previously used were greater than those of the present experiments²⁴. On the other hand, increases in endogenous ET1 in the wall of cerebral blood vessels have been reported to contribute to the cerebrovascular oxidative stress induced by administration of subpressor doses of AngII¹⁷, chronic intermittent hypoxia¹⁸, or diabetes¹⁹. Therefore, the present study, in which ET1 had access mainly to the endothelium, may indicate that the cellular localization of the peptide and its receptors may be critical for inducing vascular oxidative stress. Alternatively, the ability of ET1 to increase ROS may be context dependent, requiring the presence of other mediators, and/or longer exposure times.

Role of ET1 in conditions associated with cerebrovascular dysfunction

ET1 has been implicated in pathogenesis of several conditions associated with cerebrovascular dysfunction and damage, including subarachnoid hemorrhage, ischemic stroke, and brain trauma³,4,7. As pointed out above these effect have been attributed to the vasoconstrictor effects of ET1 or to its ability to promote ROS production¹². The present results provide evidence that ET1 can also alter cerebrovascular regulation independently of vasoconstriction or vascular oxidative stress. Our observations are relevant to conditions in which circulating levels of ET1 are elevated, such as risk factors for ischemic stroke⁸-11. Although in these conditions ET1 is likely to act in concert with other vascular mechanisms, the data suggest that ET1 is sufficient to induce cerebrovascular dysfunction. The dependence of the cerebrovascular effect of circulating ET1 on ET_AR and ROCK strengthen the argument for targeting ET_AR and ROCK for therapeutic purposes³⁹. Clinical trials using ET_AR inhibitors in subarachnoid hemorrhage have not been conclusive⁴⁰, but there is increasing interest in targeting these receptors in human disease⁴¹.

Perspectives

We have demonstrated that elevations in circulating levels of ET1, comparable to those observed in cerebrovascular conditions, suppress functional hyperemia and impair selected endothelium-dependent responses, critical factors in CBF regulation. The effect is mediated by endothelial ET_AR and is independent of vascular oxidative stress, requiring instead activation of endothelial ROCK. In vitro studies suggest that endothelial ROCK acts by modulating eNOS phosphorylation and attenuating eNOS activity. These novel observations, collectively, suggest that circulating levels of ET1 are sufficient to alter cerebrovascular regulation, which, alone or in combination with other pathogenic mechanisms, may play a role in the cerebrovascular dysfunction associated with acute and chronic brain injury.

1. What Is New?

• Endothelin-1 (ET1) alters critical cerebrovascular regulatory mechanisms

• The effect is mediated by endothelial ET_AR and ROCK

• ET1 suppresses NO production by altering eNOS phophorylation status.

2. What Is Relevant?

• Elevated ET1 plasma levels have been observed in risk factors for stroke

• ET1-induced cerebrovascular dysfunction may increase the susceptibility of the brain to ischemic injury by reducing cerebrovascular reserves

• ET_AR antagonists and ROCK inhibitor are potentially useful to reverse the cerebrovascular abnormalities.

3. Summary

• ET1 alters cerebrovascular function and might be involved in the increased susceptibility to ischemic injury observed in risk factors for stroke. The involvement of ET_AR and ROCK suggests new therapeutic approaches to counteract the deleterious cerebrovascular effects of ET-1.

Supplementary Material

1

Acknowledgments