Diabetes increases the risk of stroke and contributes to poor clinical outcomes in this patient population. Myogenic tone of the cerebral vasculature, including basilar arteries, plays a key role in controlling cerebral blood flow. Increased myogenic tone is ameliorated with ET receptor antagonism in Type 1 diabetes. However, the role of endothelin-1 (ET-1) and its receptors in cerebrovascular dysfunction in Type 2 diabetes, a common comorbidity in stroke patients, remains poorly elucidated. Therefore, we hypothesized that 1) cerebrovascular dysfunction occurs in the Goto-Kakizaki (GK) model of Type 2 diabetes, and 2) pharmacological antagonism of ETA receptors ameliorates, while ETB receptor blockade augments vascular dysfunction. GK or control rats were treated with antagonists to either ETA (atrasentan, 5 mg·kg−1·day−1) or ETB (A-192621, 15 or 30 mg·kg−1·day−1) receptors for 4 wk and vascular function of basilar arteries was assessed using a wire myograph. GK rats exhibited increased sensitivity to ET-1. ETA receptor antagonism caused a rightward shift, indicating decreased sensitivity in diabetes, while it increased sensitivity to ET-1 in control rats. Endothelium-dependent relaxation was impaired in diabetes. ETA receptor blockade restored relaxation to control values in the GK animals with no significant effect in Wistar rats and ETB blockade with 30 mg·kg−1·day−1 A-192621 caused paradoxical constriction in diabetes. These studies demonstrate that cerebrovascular dysfunction occurs and may contribute to altered regulation of myogenic tone and cerebral blood flow in diabetes. While ETA receptors mediate vascular dysfunction, ETB receptors display differential effects. These results underscore the importance of ETA/ETB receptor balance and interactions in cerebrovascular dysfunction in diabetes.
diabetes exists as an independent risk factor for cardiovascular disease (CVD) (29). Within the diabetic population, CVD is the leading cause of death, accounting for 65–75% of all deaths (4). Studies have demonstrated that there is a strong correlation between diabetes and cerebrovascular disorders, such as cerebral ischemia and stroke (5, 52). Diabetes not only increases the risk of stroke but also contributes to increased stroke mortality and impaired recovery after stroke (3, 15, 16).
Regulation of vascular tone is important for maintenance of proper blood flow. In the cerebral circulation, changes in blood flow are buffered by the myogenic response to maintain cerebral blood flow. However, alterations to this system may be detrimental and could contribute to cerebrovascular disease. Studies have demonstrated increased myogenic tone in experimental diabetes (8, 9, 54). In addition to increased basal tone, cerebral arteries from diabetic animals exhibit diminished endothelium-derived relaxation (2, 9, 34).
It is well established that the potent vasoconstrictor endothelin-1 (ET-1) is upregulated in the circulation in both clinical and experimental diabetes (7, 51). The effects of endothelin are mediated via two G protein-coupled receptors: ETA and ETB (17). ETA receptors reside on the smooth muscle cell (SMC) and produce vasoconstriction and mediate the proliferative effects of ET-1. Functional studies have suggested that different ETB receptor subtypes may exist (39, 40, 50). However, Mizuguchi et al. (38) demonstrated that the observed heterogeneous responses of the ETB receptor are abolished in ETB knockout mice, suggesting that one gene is responsible for these actions (38). Inasmuch, ETB receptors do elicit different responses. ETB receptors on endothelial cells promote vasodilation via cGMP, while vascular smooth muscle cells (VSMC) ETB receptors educe ETA-like responses. Mounting evidence suggests that ET-1 is involved in the pathology of cerebrovascular disease (26). Studies have demonstrated enhanced contractile responses to ET-1 (1, 33), as well as a reduction of increased myogenic tone after ET receptor antagonism (9) in Type 1 diabetes. However, the relative roles of ET-1 and its receptors in cerebrovascular dysfunction in Type 2 diabetes, which is a common comorbidity in stroke patients, are poorly elucidated. Thus, we hypothesized that 1) cerebrovascular dysfunction occurs in the Goto-Kakizaki (GK) model of Type 2 diabetes, and 2) pharmacological antagonism of ETA receptors ameliorates, whereas ETB receptor blockade augments vascular dysfunction.
RESEARCH DESIGN AND METHODS
All experiments were performed on male Wistar (Harlan, Indianapolis, IN) and GK (in-house bred, derived from the Tampa colony) rats (14, 49). The animals were housed at the Medical College of Georgia animal care facility that is approved by the American Association for Accreditation of Laboratory Animal Care. All protocols were approved by the Institutional Animal Care and Use Committee. Animals were fed standard rat chow and tap water ad libitum until death at 18 wk of age. Blood pressure was monitored either by telemetric method (as previously reported) (21) or via the tail cuff method, which we have previously validated on telemetry-implanted animals (10).
Endothelin receptor blockade.
GK rats develop hyperglycemia around 6–8 wk of age. We have previously shown that by 18 wk of age, this model displays significant cerebrovascular remodeling and ETA receptor antagonism started after 8 wk of diabetes at week 14 can ameliorate this response (19). Using the same treatment paradigm, at 14 wk of age, animals were divided into groups and treated for 4 wk with highly selective antagonists for each receptor as follows: ETA receptor blockade, atrasentan (Abbott Laboratories) 5 mg·kg−1·day−1 in drinking water and ETB receptor blockade, A-192621 (Abbott Laboratories) 15 or 30 mg·kg−1·day−1 by oral gavage split into two daily doses or vehicle (10). Vehicle for the A-192621 consisted of 83% deionized water, 10% polyethylene glycol-400, 5% ethanol, and 2% cremophor EL. Tap water was used as vehicle for the atrasentan.
Animals were anesthetized with pentobarbital sodium and exsanguinated via cardiac puncture. The brain was quickly excised for isolation of basilar arteries.
Determination of vascular function.
Isometric tension exerted by the vessels was recorded via a force transducer using the wire-myograph technique (Danish Myo Technologies). The myograph chambers were filled with Krebs buffer (NaCl, 118.3; NaHCO3, 25; KCl, 4.7; MgSO4, 1.2; KH2PO4, 1.2; CaCl2, 1.5; and dextrose, 11.1 mM), gassed with 95% O2 and 5% CO2 and maintained at 37°C. Vessel segments were mounted in the chamber using 40-μm-thin wires and adjusted to a baseline tension of 0.4 g. Cumulative dose-response curves to ET-1 (0.1–100 nM) were generated, and the force generated was expressed as % change of baseline. In a subset of vehicle-treated animals, basilar arteries were challenged with sarafotoxin-6-c (S6c, 1–100 nM) with or without preincubation (30 min) with 1 μM BQ-123. In additional vehicle-treated rats, cumulative dose-response (1 nM-1 μM) curves to serotonin (5-HT) were generated to determine whether hyperreactivity is a general response. Endothelium-dependent relaxation to acetylcholine (ACh; 1 nM to 1 μM) was assessed after vessels were constricted to 60% of the baseline tension with 5-HT or directly after ET-1 dose-response. Sensitivity (EC50) and maximum response (Rmax) values were calculated from the respective dose-response equations.
Results are given as means ± SE. For EC50 and Rmax values, a two-way ANOVA was done to analyze disease and treatment effects with a post hoc Bonferroni test. A repeated-measures ANOVA was used to determine group differences (diabetic vs. control) across the ET-1 or ACh concentrations. Post hoc group comparisons at each concentration used a Tukey's adjustment for the multiple comparisons. GraphPad Prism 4.0 was used for all statistical tests performed.
Metabolic parameters of animals are listed in Table 1. Diabetic animals were significantly smaller than control rats in all treatment groups (P < 0.001 vs. control). Plasma ET-1 levels indicated a disease and drug interaction, such that ET-1 was higher in diabetic rats for all treatment groups except for the A-192621 30 mg·kg−1·day−1 group, which was similar in control and diabetic rats. In addition, A-192621 (30 mg·kg−1·day−1) treatment caused a significant increase in plasma ET-1 levels in both control and diabetic rats compared with vehicle, atrasentan or the A-192621 (15 mg·kg−1·day−1) group.
Basilar arteries of GK rats were hypersensitive to ET-1 (EC50 2.8 ± 1.6 vs. 11.3 ± 1.7 nM in controls, P < 0.05). (Fig. 1A, Table 2). Dose-response curves in both control and diabetic animals showed significant differences between treatment protocols (Fig. 2, A and B, Table 2). In control animals, atrasentan treatment caused a paradoxical leftward shift that was statistically significant compared with vehicle or ETB receptor blockade groups. While A-192621 at 15 mg/kg dose had no effect on responses to ET-1, blockade with 30 mg/kg dose caused a rightward shift of the dose-response curve. On the other hand, in the diabetic animals, blockade of the ETA receptor with atrasentan attenuated the contractile response to ET-1, as indicated by the rightward shift. Interestingly, low-dose ETB receptor blockade with 15 mg/kg A-192621 rendered basilar arteries more sensitive to ET-1. A dose-response curve to 5-HT in vehicle-treated control and diabetic animals showed no difference between Rmax (102 ± 8 in control vs. 103 ± 10 in GK rats) or EC50 values (15.3 ± 4.2 control vs. 13.2 ± 2.5) in GK rats.
Basilar arteries of GK rats exhibited impaired endothelium relaxation following preconstriction with 5-HT. Basilar arteries of GKs relaxed only 26 ± 4% in response to ACh, whereas the control rat arteries relaxed 56 ± 5% (Fig. 1B, Table 2). In the control group, dose-response curves indicated significant differences between treatment arms (Fig. 3A). While atrasentan had no effect on relaxation response, ETB receptor antagonism with A-192621 30 mg/kg significantly improved relaxation, and A-192621 15 mg/kg group displayed about 40% relaxation across the ACh concentration range. In diabetic rats, however, atrasentan, significantly improved relaxation. Interestingly, treatment with A-192621 30 mg/kg caused paradoxical constriction, while treatment with A-192621 15 mg/kg had no effect (Fig. 3B).
To determine whether there is an up-regulation of VSMC ETB receptors, contractile responses to S6c, an ETB receptor-selective ligand, were determined (Fig. 4). Because previous studies suggested that a possible ETA-ETB receptor interaction prevents ETB-receptor responses to S6c, these studies were repeated in basilar artery segments preincubated with 1 μM BQ-123. There was a small but significant increase in S6c-mediated constriction in the presence of BQ-123 in diabetic animals, suggesting a possible involvement of VSMC ETB receptors in the contractile response.
There are three major findings in the current study: first, basilar arteries were hypersensitive to ET-1 in a Type 2 diabetic rat model; second, ACh-induced relaxation of basilar arteries was impaired in diabetes; lastly, these effects are primarily mediated by the ETA receptor. These results build upon the past reports that showed cerebrovascular dysfunction in different models of diabetes and provide evidence for the involvement of ET-1 in cerebrovascular dysfunction in a nonobese, spontaneous model of Type 2 diabetes.
While there are many models of diabetes, most are either chemically induced Type 1 models or have comorbid conditions, such as hypertension, obesity, and hyperlipidemia. The spontaneously diabetic GK rat is a nonobese, normotensive rat model originally developed from selective inbreeding of glucose-intolerant Wistar rats (18). Previous studies have demonstrated that GK rats retain greater than 40% of their beta cell mass and have a reduction of postprandial glucose when given the insulin secretagogue nateglinide (28, 30, 47, 48). In addition, our laboratory has previously shown that these rats have impaired glucose tolerance compared with control Wistar rats (21). Therefore, the GK rat serves as an excellent model for studying the effects of hyperglycemia alone on cerebrovascular function in Type 2 diabetes.
We and others have observed elevated plasma ET-1 levels in both clinical and experimental diabetes (13, 21, 32, 45). Several studies have demonstrated enhanced contractile responses to ET-1 in aorta and peripheral arteries in diabetes (22, 27, 31). In addition, McIntyre et al. (37) reported increased sensitivity to ET-1 in subcutaneous resistance arteries from patients with Type 1 diabetes. In the present study, we have demonstrated an exacerbated contractile response of the rat basilar artery to ET-1 in diabetes. Matsumoto et al. (33) as well as Alabadi et al. (1) have both demonstrated similar results in the rat and rabbit basilar arteries, respectively. In contrast, Mayhan (35) reported no differences in contractile responses to ET-1 in isolated rat basilar arteries. However, these discrepancies may be attributable to the differences in methodologies (e.g., in vivo vs. in vitro). Collectively, these previous studies were all performed in chemically induced Type 1 models of diabetes. To our knowledge, the present study is the first demonstration of enhanced cerebrovascular contractile response to ET-1 in a model of Type 2 diabetes.
Previous studies using models of Type 2 diabetes have demonstrated that endothelium-derived relaxation is impaired in aortas and the peripheral vasculature (6, 33, 42, 53). However, Type 1 models, such as the streptozotocin (STZ) rat and alloxan rabbit, have predominated in studies of the cerebral vessels (24, 34, 36). Only in recent years have investigators begun to study the effects of Type 2 diabetes on the cerebral circulation, and thus far, there are but a handful of these studies (8, 11, 12, 25, 46). Schwaninger et al. (46) and Karagiannis et al. (25) both reported diminished vasodilation in response to ACh in obese Zucker rats (OZR). However, the OZR is a model of Type 2 diabetes known to have comorbid conditions such as hypertension and hyperlipidemia, which may contribute to altered vascular states. More recently, diabetic db/db mice, characterized by hyperinsulinemia, severe hyperglycemia and obesity, were shown to exhibit similar reductions in endothelium-dependent relaxation (8). In the present study, we found that Type 2 diabetes significantly impairs ACh-induced relaxation in serotonin-preconstricted basilar arteries. ETA receptor antagonism completely restored ACh-induced relaxation in diabetic basilar arteries, indicating that an activated endothelin system contributes to this impairment. These data, in accordance with previous studies done in varying models of diabetes, suggest that diabetes impairs vascular relaxation without regard to the etiology of the disease or other comorbid conditions.
The vascular effects of ET-1 are mediated by two distinct receptor subtypes: ETA and ETB. In the present study, we have demonstrated hypersensitivity of the rat basilar artery to ET-1 in diabetes. ETA receptor blockade induced a significant rightward shift of the dose-response curve, indicating a decrease in sensitivity to ET-1. Interestingly, in control animals, ETA receptor antagonism increased sensitivity to ET-1. Schilling et al. (44) previously reported decreased sensitivity to ET-1 upon acute blockade of this receptor subtype, suggesting that acute vs. chronic blockade of the receptors yield different responses. Unexpectedly, a higher-dose blockade of the ETB receptor also produced a rightward shift of the dose-response curve in a manner similar to ETA blockade. Previous studies in diabetic rabbit and rat basilar arteries have produced similar results using in vitro inhibition of ETB receptors (1, 31, 33). Interestingly, Matsumoto et al. (33) showed that endothelial denudation of the diabetic basilar artery produced a significant leftward shift of the ET-1 dose-response curve. In the current study, lower-dose ETB blockade also caused a leftward shift, possibly suggesting that removal of endothelial ETB receptors alone affects the dilatory actions of ET in diabetes, while complete blockade (both endothelial and VSMC ETB) alters constriction. It has been suggested that ET-1 mediated contraction in the cerebrovasculature may be, in part, mediated through crosstalk of the ET receptor subtypes. Zuccarello et al. (55) previously demonstrated that ETB-mediated vasoconstriction relies on activation of the smooth muscle, as well as the endothelial ETB receptors.
Vascular reactivity comprises both constriction and relaxation of blood vessels. In this study, ETA receptor blockade with Atrasentan caused a significant improvement of endothelium-dependent dilation in diabetic rats and no significant effect in control animals, suggesting a greater involvement of ET-1 in the regulation of vascular relaxation in diabetes. More interestingly, ETB receptor antagonism at 30 mg/kg dose resulted in completely opposite results in control vs. diabetic rats; that is, it improved relaxation in control animals and caused paradoxical constriction in diabetic animals. Complete reversal of the relaxation response in A-192621(30 mg/kg)-treated GK rats suggests that endothelial ETB receptors are upregulated as a compensatory mechanism to offset impaired relaxation in diabetes and that blockade of these receptors ultimately results in decreased relaxation. Alternatively, these receptors are involved to a greater extent in the regulation of ACh-mediated dilatation in diabetes and again antagonism of this receptor subtype causes paradoxical constriction to ACh.
Collectively, these studies strongly suggest involvement of both endothelial and VSMC ETB receptors in the regulation of vascular function in diabetes. Indeed, it has been demonstrated that diabetes does upregulate vascular ETB receptor expression. Ikeda et al. (23) found increased ETB receptor density in the diabetic rabbit urinary bladder, while Ortmann et al. (41) showed a significantly increased ETB gene expression in STZ diabetic rat adrenal glands. In 10-wk-old NOD mice, a Type 1 diabetes model, aortic ETB gene expression was significantly increased while ETA expression was unchanged (41). Conceivably, blockade of ETB receptors in a system that may be shifted more toward contractile responses, would produce effects similar to ETA blockade. It is also possible that ETA-ETB heterodimerization can affect the reactivity studies. Both homo and heterodimerization of the two ET receptors have been reported by functional, as well as fluorescence resonance studies (19, 20, 43). It is suggested that when heterodimerized, the ETA receptor overrides ETB receptor activation and thus ETB receptor antagonism provides an ETA blockade-like effect. Harada et al. (20a) reported that because of ETA and ETB receptor heterodimerization, ETB receptors do not independently recognize ligands such as ET-1 and S6c, unless ETA receptors are blocked with BQ-123. To test this hypothesis, we examined vascular responses to the ETB-selective agonist S6c in the presence and absence of ETA blockade with BQ-123. These studies demonstrated that S6c can induce vasoconstriction in the diabetes group when ETA receptors are antagonized, thus indicating increased presence of VSMC ETB receptors.
Though our studies correlate with previously published data, certain limitations must be addressed. First, we examined reactivity of the basilar artery to ET-1 in all groups and to 5-HT only vehicle-treated control and GK rats but not in ET receptor antagonist treatment groups. Others have found similar responses to ET-1 in diabetic rat and rabbit basilar arteries (1, 33). Second, because of the limited availability of ETB receptor antagonist, as well as the basilar artery segments that can be obtained from one animal, especially in the low-dose A-192621 group, we had a limited number of animals. We treated five animals in this group, but in two animals, vessels were completely unresponsive to any stimulus. For the same reasons, reactivity experiments could not be performed with endothelium-denuded vessels.
PERSPECTIVES AND SIGNIFICANCE
Diabetes induces vascular dysfunction in isolated rat basilar arteries in the form of increased sensitivity to ET-1 and diminished relaxation capacity to ACh. Inasmuch, these effects appear to be, at least in part, due to an increased contribution of the ETA and smooth muscle ETB receptor activation. These alterations of cerebrovascular function may potentially underlie the increased propensity for cerebrovascular disease in diabetes.
This work was supported by grants from National Institutes of Health (HL-076236, DK-074385), American Heart Association Established Investigator Award and Philip Morris Research Award to Adviye Ergul and American Heart Association Southeast Affiliate Predoctoral Fellowship to Alex Harris.
↵* These two authors contributed equally to this work.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2008 the American Physiological Society