Endothelin-B Receptors in Cerebral Resistance Arterioles and their Functional Significance after Focal Cerebral Ischemia in Cats

Anesthesia and general preparations

The experiments were carried out on 11 (5 ischemic and 6 nonischemic) adult female cats weighing between 3 and 4 kg. The anesthesia regimen and the surgical procedure are fully described elsewhere (Patel et al., 1996b). Briefly, anesthesia was induced with alphaxalone/alphadolone (Saffan, Glaxo Harefield, U.K.) administered into the radial vein. The animals were intubated, and positive-pressure ventilation with nitrous oxide/oxygen (70%/30%) initiated. The femoral artery and vein were cannulated for monitoring arterial blood pressure and arterial blood gas status and also for administering anesthetic agent and fluids. Anesthesia was maintained using α-chloralose (60 mg/kg) and supplemented as required to prevent the return of the corneal reflex during the course of the experiment. The inspired gas mixture was altered to oxygen supplemented air (25% oxygen and arterial blood samples were taken at regular intervals to monitor blood gas status. The animals were maintained normocapnic by adjusting the stroke volume. Metabolic acidosis was controlled by administration of sodium bicarbonate (8.4% solution) when necessary. Core temperature was measured by a rectal thermometer and maintained at 37°C by means of a thermostatically controlled heating blanket. At the outset of the study, mean arterial pressure was 100 ± 18 mm Hg, arterial carbon dioxide tension was 29.1 ± 4.0 mm Hg, arterial pH was 7.44 ± 0.05, and arterial oxygen tension was 149 ± 39 mm Hg. These physiologic parameters did not vary significantly during the course of the microapplication study.

The animals were placed in a stereotactic frame and after midline incision, the scalp retracted and sutured onto a metal ring to form a well over the calvaria. The temporalis muscle was retracted and a rectangular craniectomy (∼2.5 × 1.5 cm) was made over the parietal cortex using a saline-cooled dental drill. The exposed dura was bathed continuously with mineral oil at 38°C. With the aid of a stereomicroscope (Bausch and Lomb), the dura was excised and then reflected laterally. The preparation was allowed to equilibrate for approximately 30 minutes before any adventitial microapplications were performed.

Measurement of cerebral resistance arteriolar caliber

Cerebral resistance arteriolar caliber was measured by the method originally developed by Beaz (1966) using an image splitter linked through a closed-circuit video display system. The individual cerebral resistance arterioles were viewed through a stereomicroscope and arteriolar diameter measured from the degree of shear applied to the image splitter to tangentially oppose the two images. The system was calibrated at ×40 and ×70 against threads of known diameter, which allows the direct measurement of the vascular diameter in absolute units (micrometers). Vessel diameter was measured before and after injection of substances and monitored over a period of 1 to 2 minutes. In this study, the size of the arterioles examined ranged between 40 and 290 μm.

Administration of drugs

Artificial CSF was prepared with the following composition: Na⁺ 156 mmol/L, K⁺ 3 mmol/L, Ca²⁺ 2.5 mmol/L, HCO₃⁻ 12 mmol/L, and Cl⁻ 152 mmol/L. The pH of the CSF was adjusted to 7.2 by aeration with 95% oxygen–5% carbon dioxide. The agents to be administered by adventitial microapplication were dissolved in CSF. All solutions were prepared freshly on the day of the study. Glass micropipettes (tip diameter 10 to 12 μm) were filled under vacuum with the solution to be studied and stored under mineral oil in CSF until required. Adventitial microapplication of substances was carried out using a micromanipulator to position the pipettes in the adventitial space around individual cerebral resistance arterioles. Small volumes (approximately 5 μL) were delivered by a pressure ejection system. Repeated application of the agents was performed in the same arteriolar site.

Focal cerebral ischemia studies

The middle cerebral artery (MCA) was exposed using a transorbital approach as previously described by Chen and colleagues (1991). Briefly, the contents of the left orbit were exenterated, and the optic foramen was enlarged to expose the dura and the MCA underlying it. The dura was excised, the MCA and its branches were occluded by means of bipolar diathermy, and the artery was sectioned. In each cat, perivascular microinjections of CSF (pH 7.2), BQ-3020 (1 μmol/L), and BQ-788 (0.1 μmol/L) were made at different sites after 30 to 210 minutes of cerebral ischemia. The arterioles studied were located in the ectosylvian and suprasylvian gyri (i.e., the regions supplied by the MCA and representing ischemic penumbra).

Statistical analysis

All results are expressed as means ± SD of the percentage change in arteriolar caliber from baseline immediately before microapplication of substances. The data were analyzed using a one-way analysis of variance followed by two-tailed Student's unpaired t test using a Bonferroni correction factor for multiple group comparisons.

Adventitial microapplication of BQ-3020

The application of CSF minimally altered pial arteriolar caliber (arteriolar caliber changed by 0.7 ± 4.8% from baseline). The first microapplication of BQ-3020 (1 μmol/L) mediated a marked and transient dilation of pial arterioles (caliber increased by 26.3 ± 15.1%). A second injection of the ET_B receptor antagonist (10 minutes later) at the same site failed to induce any arteriolar dilation. Subsequent (third and fourth) adventitial microapplications of BQ-3020 effected a significant (P < 0.001, analysis of variance) constriction of cerebral arterioles (Fig. 1). Potassium-induced marked dilation, subsequent to repeated injection of BQ-3020, indicated that the pial arterioles were still responsive to vasodilator agents (Fig. 1). In separate experiments, repeated adventitial microapplication of potassium (10 mmol/L) at the same site did not effect any tachyphylaxis of the dilator response (Fig. 2). Both CSF- and potassium-induced vascular effects were independent of the preinjection diameter (caliber range, 45 to 239 μm). However, the dilation induced by the first microapplication of BQ-3020 was significantly dependent on the preinjection caliber. Pial arterioles with diameter less than 100 μm demonstrated a greater vasodilator response to ET_B receptor agonist. The vasoconstriction, observed after repeated activation of ET_B receptors, also displayed some size dependency, with arterioles less than 100 μm displaying more pronounced vasoconstriction.

OPEN IN VIEWER

FIG. 1.

Vasomotor responses of pial arterioles to repeated perivascular microapplication of BQ-3020 (1 μmol/L) followed by adventitial application of potassium (10 mmol/L). Data are the percentage change from preinjection baseline in pial arteriolar caliber. **P < 0.001, ***P < 0.0001 for the comparison with CSF. Data are presented as mean ± SD. n, number of arterioles examined (n = 16).

OPEN IN VIEWER

FIG. 2.

Vasomotor responses of pial arterioles to repeated perivascular microapplication of potassium (10 mmol/L). Repeated application of potassium elicited reproducible dilation of the arterioles. Data are the percentage change from preinjection baseline in pial arteriolar caliber. Data are presented as mean ± SD. n, number of arterioles examined (n = 7).

Adventitial microapplication of BQ-788

The subarachnoid perivascular application of the ET_B receptor antagonist BQ-788 (0.001 to 1 μmol/L) had no significant effect on pial arteriolar caliber when compared with CSF injections (Fig. 3). This BQ-788–mediated response was similar, irrespective the size of the vessel examined (62 to 280 μm).

OPEN IN VIEWER

FIG. 3.

Vasomotor responses of pial arterioles to perivascular microapplication of BQ-788 per se. There are no significant alterations in pial arteriolar caliber at any concentration of BQ-788. Data are the percentage change from preinjection baseline in pial arteriolar caliber. Data are presented as mean ± SD. n, number of arterioles examined (n = 7 to 8 for each concentration).

Adventitial microapplication of BQ-788 and BQ-3020

Perivascular coapplication of BQ-788 with BQ-3020 (1 μmol/L) in vessels that have not been previously examined (preinjection caliber 114 ± 63 μm [44 to 286 μm]) produced a dose-dependent attenuation of BQ-3020–induced dilation (Fig. 4). The concentration of BQ-788 in the coinjectate, which produced half-maximal inhibition of dilatory response to the first application of BQ-3020, was estimated to be approximately 5 nmol/L. As shown earlier, the third microapplication of BQ-3020 effected a −15 ± 12% reduction in arteriolar caliber. In separate experiments, when BQ-788 (0.1 μmol/L) was coinjected with BQ-3020, after two previous applications of BQ-3020 alone, no change in arteriolar caliber was observed (0.2 ± 10.2% from baseline). This vasomotor response was significantly different from that elicited by the third application of BQ-3020 alone (P < 0.05).

OPEN IN VIEWER

FIG. 4.

Vasomotor responses of pial arterioles to perivascular microapplication of BQ-3020 (1 μmol/L) and BQ-788. BQ-788 significantly reduces the dilation induced by BQ-3020. **P < 0.001 for the comparison with BQ-3020 alone (open circle). Data are the percentage change from preinjection baseline in pial arteriolar caliber. Data are presented as mean ± SD. n, number of arterioles examined (n = 11−13 for each concentration).

Middle cerebral artery occlusion–induced alterations in arteriolar diameter

The occlusion of the MCA effected an immediate dilation of all pial arterioles overlying the parasagittal, suprasylvian, and the ectosylvian gyri. This vasodilation was sustained in the parasagittal gyrus (zone irrigated by the anterior cerebral artery) over the entire experimental period. However, in the ectosylvian and the suprasylvian, few arterioles displayed a transient vasoconstriction. In two cats, a circumscribed zone of marked and sustained vasoconstriction was developed in the ectosylvian gyrus during the 30 minutes after MCA occlusion. Because of the limited availability of vasoconstricted arterioles in our preparation, we elected to investigate the effect of the ET_B receptor agonist and antagonist only on dilated pial arterioles located within the suprasylvian and the ectosylvian gyri (i.e., the region supplied by the MCA and representing ischemic penumbra).

Adventitial microapplication of BQ-3020 after middle cerebral artery occlusion

The microapplication of CSF (pH = 7.2) on postischemic dilated arterioles elicited a small vasoconstriction (−17 ± 7%) compared with preinjection caliber (Fig. 5). The magnitude of this response did not alter significantly during the period of 30 to 210 minutes subsequent to induction of ischemia (Fig. 6). The perivascular microapplication of BQ-3020 (1 μmol/L) induced a vasoconstriction of a similar magnitude to that observed after injection of CSF (Fig. 5). Furthermore, the magnitude of the response to BQ-3020 did not vary with time after MCA occlusion (Fig. 6). No relation between the preinjection diameter, which ranged between 49 and 288 μm, and the BQ-3020–induced response was found.

OPEN IN VIEWER

FIG. 5.

Vasomotor responses of postocclusion dilated pial arterioles to perivascular microapplication of CSF, BQ-3020 (1 μmol/L), and BQ-788 (0.1 μmol/L). Microinjections were performed 30 to 210 minutes after MCA occlusion. ***P < 0.001 for the comparison with CSF and BQ-3020. Data are the percentage change from preinjection baseline in pial arteriolar caliber. Data are presented as mean ± SD. n, number of arterioles examined (n = 34–43).

OPEN IN VIEWER

FIG. 6.

Vasomotor responses of postocclusion dilated pial arterioles to perivascular microapplication of CSF and BQ-3020 (1 μmol/L) at varying times after MCA occlusion. Each point represents the response of a single pial arteriole. The MCA was occluded at time zero. Data are the percentage change from preinjection baseline in pial arteriolar caliber.

Adventitial microapplication of BQ-788 after middle cerebral artery occlusion

The adventitial microapplication of the ET_B receptor antagonist (BQ-788, 0.1 μmol/L) on postocclusion dilated pial arterioles elicited no change in the arteriolar caliber when compared with preinjection baseline (Fig. 5). However, this BQ-788–induced response was significantly different from that induced by perivascular microinjection of CSF (P < 0.001, analysis of variance). The vasomotor response to perivascular application of BQ-788 did not exhibit any marked dependency with respect to the time that had elapsed after occlusion (Fig. 7) or with respect to the size of the arterioles analyzed (diameter ranged between 41 and 277 μm).

OPEN IN VIEWER

FIG. 7.

Vasomotor responses of postocclusion dilated pial arterioles to perivascular microapplication of CSF and BQ-788 (0.1 μmol/L) at varying times after MCA occlusion. The responses to CSF are identical to those in Fig. 6. Each point represents the response of a single pial arteriole. The MCA was occluded at time zero. Data are the percentage change from preinjection baseline in pial arteriolar caliber.

Vasomotor effects of endothelin-B receptor agonist in normal vessels

In the cerebral circulation, the in vitro or in vivo activation of ET_B receptors in rat basilar artery or in feline pial arteriole produces a marked vasodilation, which is dependent on the production of nitric oxide (Feger et al., 1994; Kitazono et al., 1995a; Schilling et al., 1995; Patel et al., 1996c). In contrast, in some peripheral vascular beds, an ET_B receptor also has been linked to the contractile properties of endothelins. Indeed, it was observed that ET-1–mediated vasoconstriction was resistant to the actions of ET_A receptor antagonists, and therefore could not be solely attributed to ET_A receptor activation (Sumner et al., 1992; McMurdo et al., 1993; Shetty et al., 1993; Seo et al., 1994; Gellai et al., 1996; Awane-Igata et al., 1997). The current study provides the first definitive evidence of ET_B–mediated constriction. Our data demonstrate that a single perivascular microapplication of selective ET_B receptor agonist (BQ-3020, 1 μmol/L) induces vasodilation in cerebral resistance arterioles in the same magnitude to that reported by Patel and colleagues (1996c) in anesthetized cats. This ET_B dilator response displays tachyphylaxis, as described by Kitazono and colleagues (1995a). In addition to this, we demonstrate, for the first time, that subsequent (third and fourth) activations of the ET_B receptors (10-minute interval) induce significant constriction of pial arterioles (Fig. 1). The fact that repeated perivascular application of potassium revealed a reproducible degree of dilation of pial cortical arterioles and that vasoconstriction was inhibited by selective ET_B receptor antagonist (BQ-788, 0.1 μmol/L) further implies that both responses (i.e., tachyphylaxis and vasoconstriction) were induced by specific ET_B receptor activation. Evidence from in vivo and in vitro studies has indicated the susceptibility of endothelin receptors to desensitization. In the rat, it has been shown that in the basilar artery, vasodilation was markedly attenuated in response to a second application of IRL 1620 (ET_B receptor agonist), suggesting desensitization to vasodilator effect of the ET_B receptor agonist (Kitazono et al., 1995a). Nevertheless, no ET_B receptor–mediated vasoconstriction was reported in this study, perhaps because of the lack of repeated activation of the ET_B receptors. In awake rats, ET-1 induced depressor responses, which are probably mediated by ET_B receptors and also display pronounced tachyphylaxis (Le Monnier de Gouville et al., 1996). Furthermore, an in vitro study by Clozel and Gray (1995) reports that in epithelium-denuded rat tracheal rings, the ET_B receptor exhibited tachyphylaxis, since incubation of the rings for 45 minutes with sarafotoxin S6c (selective ET_B agonist) induced desensitization to sarafotoxin S6c. In contrast, tachyphylaxis of the vasomotor response has not been shown to occur after application of some neuropeptides (i.e., calcitonin gene-related peptide, adrenomedullin) known to induce vasodilation in cerebral vessels (Wahl et al., 1994; Lang et al., 1997). An alternate explanation for the apparent tachyphylaxis to the vasodilation mediated by ET_B receptor agonism is a disruption in vascular endothelial cell function. Repeated challenges with the ET_B receptor agonist could have depleted the supply of the endothelial vasorelaxant substances that mediate the response (nitric oxide or prostayclin) or damage the endothelial cells. This mechanism also may be consistent with the lack of vasodilation with BQ-3020 after ischemia, since endothelial cell integrity could have been compromised with this insult. Nevertheless, Kitazono and coworkers (1995a) report that pretreatment of the basilar artery with an ET_B receptor agonist did not affect acetylcholine-induced vasodilation (which is nitric oxide mediated). This may indicate that tachyphylaxis of the vasodilator response, because of repeated activation of ET_B receptors, could not be ascribed to the depletion of nitric oxide.

The varied effect of the endothelins at the ET_B receptor have led to the subdivision of the ET_B receptor into ET_B1 receptors (on endothelial cells), which mediate vasodilation, and ET_B2 receptors (on smooth muscle cells), which mediate vasoconstriction (Warner et al., 1993; Masaki et al., 1994). The exact mechanisms underlying the contraction of pial arterioles, found in the current study, after repeated activation of ET_B receptors are unknown. One hypothesis is that repeated perivascular microapplication of BQ-3020 induces desensitization of endothelial ET_B (ET_B1) receptors that are vasodilator and thus unmask vasoconstriction mediated by ET_B2 receptors located on smooth muscle. These data argue for a potential role of ET_B receptors in pathophysiologic situations where release of endogenous agonist is increased. The identification and characterization of ET_B receptor subtypes, if any, would be of great importance to elucidate the function of ET_B receptors in the cerebral vasculature, especially in resistance arterioles, which play an important role in the regulation of cerebral blood flow.

Vasomotor of effects endothelin-B receptor antagonist in normal vessels. The contribution of the ET_B receptor in the regulation of normal vessel tone has not been the subject of wide attention. In the current study, perivascular microapplication of the selective ET_B receptor antagonist (BQ-788) had no effect on pial arteriolar caliber (Fig. 3). This finding suggests that at least at the level of the adventitial surface, there is minimal tone from activation of ET_B receptors by endogenous endothelins. These data are in agreement with the reports of Patel and colleagues (1994, 1996c) who demonstrated that microapplication of both ET_A selective and mixed ET_A/ET_B receptor antagonist did not induce any change in feline cerebral resistance arterioles caliber. However, these observations contrast with those reported by Yoshimoto et al. (1990) and Haynes and Webb (1994), which suggest that there is continuous production of endothelins by cerebrovascular endothelium and that there is basal endothelin-mediated tone in the systemic circulation generally. In the current study and for the first time, the application of BQ-788 onto the cerebral resistance arterioles has been shown to attenuate the vasodilation induced by the ET_B receptor agonist. Moreover, BQ-788 was able to block vasoconstriction mediated by repeated perivascular injection of BQ-3020. This is in agreement with the study of Ishikawa and colleagues (1994) in which BQ-788 has been shown to antagonize, with dose-dependent manner, the constriction of isolated rabbit pulmonary arteries elicited by BQ-3020.

Endothelin-B receptor–mediated vasomotor effects in postischemic vessels

Several lines of evidence indicate the implication of endothelins in the pathogenesis of focal cerebral ischemia, although the precise definition of their pathophysiologic role has not been fully elucidated (Barone et al., 1995b; Patel, 1996). Endothelin receptor antagonists have been shown in some studies to reduce brain damage after induced focal cerebral ischemia (Barone et al., 1995a; Patel and Wilson, 1995; McAuley et al., 1996; Patel et al., 1996a,b). Critical evaluation of these studies is difficult because of differences in experimental design, species, and receptor antagonists used. Most of the studies that have intervened with endothelin receptors antagonists after induced focal ischemia have focused on the role of ET_A receptors as mediator of brain damage. The role of ET_B receptors in ischemic disease, however, is still to be elucidated. Endothelin ET_B receptors do in part mediate vasospasm in large vessels after subarachnoid hemorrhage, since therapeutic intervention with a mixed endothelin ET_A/ET_B receptor antagonist was shown to be more effective than with an endothelin ET_A receptor antagonist alone (Zuccarello et al., 1994). Moreover, it has been shown that there is an increase in ET_B receptor agonist binding sites in basilar arteries after experimental subarachnoid hemorrhage, which may indicate a potential role of ET_B receptors in these pathologic conditions (Roux et al., 1995; Hino et al., 1996). In our investigation, we addressed the implication of ET_B receptor in vasomotor disturbances associated with acute focal cerebral ischemia. Perivascular microinjection of CSF (pH 7.2) onto dilated pial arterioles within the ischemic territory elicited small reduction in arteriolar caliber. This is most likely caused by the acidotic milieu within the MCA territory rendering the exogenously applied CSF alkalotic by comparison. The direct adventitial application of BQ-3020 onto dilated arterioles after MCA occlusion showed that ET_B receptor–induced dilation, in normal pial arterioles, is lost in postischemic vessels. This indicates that the ET_B receptor–induced vascular effects may vary in pathologic situations such as acute stroke. More importantly, the BQ-788–mediated effects in dilated arterioles, compared with the responses induced by CSF (vehicle), suggest that ET_B receptors participate significantly in restricting postischemic dilation of cortical arterioles in the ischemic penumbra (Fig. 6). Nonetheless, this contribution of ET_B receptors in limiting vasodilation is less pronounced compared with that of ET_A receptors described by Patel et al. (1996b). These findings suggest that the use of broad-spectrum endothelin receptor antagonist (i.e., mixed endothelin ET_A/ET_B receptor antagonist) would be of greater benefit in reversing the increased cerebrovascular tone resulting from the release of endogenous endothelins in acute stroke.

Acknowledgment: The authors thank the staff at the Wellcome Surgical Institute and Hugh Fraser Neuroscience Laboratories, University of Glasgow, Glasgow, Scotland, U.K., for their skilled technical assistance.