INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

THE NUCLEOSIDE ADENOSINE plays an important role in many physiological processes. Its actions are mediated by specific cell surface receptors coupled to G proteins (7, 13). At least four adenosine receptors have been cloned and classified as A₁, A_2A, A_2B, and A₃ subtypes (7, 13). Adenosine is a potent dilator of cerebral vessels (9, 11) and has been implicated in the regulation of cerebral blood flow (16, 19). We (11) previously investigated the receptors involved in adenosine-induced dilation in cerebral resistance arterioles. On the basis of agonist potency orders, we concluded that adenosine acts primarily via A₂ receptors to elicit cerebral vasodilation (11). Since then, new and more specific pharmacological tools have been developed. The purpose of the present study was to use selective adenosine receptor agonists and antagonists to further identify the receptor subtypes involved in the cerebral vasodilator response to adenosine.

In the present study, we utilized a microvessel cannulation technique (3, 4) for the pressurization and study of parenchymal arterioles isolated from the rat brain. Our preparation has two important features. First, because of their unique origin, intracerebral arterioles are markedly sensitive to vasoactive substances released during functional and metabolic activity. Second, with our in vitro methodology, these vessels are capable of developing spontaneous myogenic tone without the aid of pharmacological constrictor agents, which may confound data interpretation.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Vessel isolation and cannulation. The experimental protocols in the present study were approved by the Animal Care Committee of the University of Washington. Male Sprague-Dawley rats weighing 350-400 g were initially anesthetized with pentobarbital sodium (50 mg/kg ip). The technique for the isolation and cannulation of brain-penetrating arterioles has been described in detail in previous publications (3, 11, 12). Briefly, a piece of cerebral cortex ~3 mm thick containing the first portion of the middle cerebral artery was dissected from the brain of Sprague-Dawley rats. The pia mater and its attached penetrating intracerebral arterioles were carefully separated from the parenchyma. A segment of intracerebral arteriole ~0.5-1.0 mm in length was excised and transferred to a temperature-controlled vessel chamber on the stage of an inverted microscope. The arteriole was cannulated at both ends with a system of concentric micropipettes consisting of a perfusion pipette within a holding pipette. After cannulation was completed, the vessel was pressurized to 60 mmHg and perfused intraluminally at a flow rate of 2 µl/min with buffered saline (pH 7.4) containing 1% albumin. Vessel diameter was measured with a video micrometer.

Protocol. The vessel bath contained buffered saline (pH 7.3, no albumin) with fresh buffer circulating at a rate of 0.75 ml/min. After the bath temperature was raised to 37°C, viable vessels gradually developed spontaneous tone and contracted to a stable baseline diameter. Vessel reactivity was assessed by replacing the bath fluid with acidic (pH 6.8) and alkaline (pH 7.6) buffer. Vessels with a poor pH response (<15% dilation or constriction) were excluded from data collection.

Drugs and solutions. The composition of the buffered saline solution was as follows (in mM): 144.0 NaCl, 3.0 KCl, 2.5 CaCl₂, 1.5 MgSO₄, 5.0 glucose, 2.0 pyruvate, 0.02 EDTA, 2.0 MOPS, and 1.2 NaH₂PO₄. Adenosine, CGS-21680, CGS15943, CPX, MRS-1191, and ATP were obtained from Sigma (St. Louis, MO). ZM-241385 was purchased from Tocris Cookson (Ballwin, MO). SCH-58261 was a generous gift from Schering-Plough Research Institute (Milan, Italy). Adenosine, ATP, and CGS-21680 (hydrochloride) were dissolved directly into the MOPS-buffered saline solution. CPX, CGS-15943, MRS-1191, and ZM-241385 were dissolved in DMSO to yield stock solutions of 1-10 mM. Subsequent dilutions were made with buffered saline, and pH was adjusted to 7.3.

Data analysis. All data are expressed as means ± SE. Only one vessel was studied from each animal. For comparison of responses to vasoactive agents, internal vessel diameters were normalized as a percentage of control diameters (diameters measured after the development of steady tone at 60 mmHg of intraluminal pressure and 37°C). Data and statistical analyses were performed with GraphPad Prism 3.0 software (San Diego, CA). The agonist concentrations required to cause 25 and 50% of maximum dilation observed (EC₂₅ and EC₅₀ values, respectively) were obtained by averaging data from individual concentration-response curves. Comparisons between concentration-response curves were made using two-way analysis of variance (ANOVA), followed by the Bonferroni method for multiple group comparisons. A value of P < 0.05 was considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Intracerebral arterioles in this study (n = 94) spontaneously constricted to 63 ± 1% of their maximally dilated (passive) diameter as the bathing medium was warmed to 37°C. Average vessel diameter after spontaneous tone development was 47 ± 1 µm (range: 29-72 µm). As is characteristic of cerebral arterioles, these vessels are markedly sensitive to pH changes. A bath pH of 7.6 led to a constriction of 24 ± 1% from baseline diameter (pH 7.3), whereas decreasing pH to 6.8 dilated the vessels by 27 ± 1%. We compared the concentration-dependent dilations of intracerebral arterioles to adenosine and CGS-21680 (Fig. 1). The concentration-response curves showed significant (P < 0.05) differences, with logEC₅₀ values of 6.5 ± 0.2 for adenosine and 8.6 ± 0.3 for CGS-21680. Although CGS-21680 has a lower dilator threshold (0.1 nM) than adenosine (1 nM), its dilator effect flattened out at 1 µM and was surpassed by adenosine at a concentration of 10 µM. Adenosine concentration-response curves are reproducible when repeated in the same vessel (Fig. 1B) and do not degrade with time.

View larger version (19K): [in this window] [in a new window]   Fig. 1.   A: concentration-response curves comparing dilation of intracerebral arterioles to adenosine and the A2A receptor-selective agonist CGS-21680. To increase sample size, data were compiled from all experiments in which concentration responses to either of these agonists were studied. Dilation responses are expressed as a percentage of baseline diameter (vessel diameter after development of myogenic tone). For adenosine responses, n = 41 vessels, passive vessel diameter = 75 ± 2 mm, and tone = 63 ± 1%. For CGS-21680 responses, n = 35 vessels, passive diameter = 79 ± 2, and tone = 65 ± 1%. Values are means ± SE. *P < 0.05 vs. adenosine-induced dilation at the same concentration (10 µM). B: graphs showing the reproducibility of adenosine concentration-response curves. After a first adenosine concentration-response curve () was determined, vessel diameter was restored to baseline, and a second adenosine concentration-response curve () was constructed using the same vessel. Each symbol and error bar represent the mean and SE, respectively, of values from 4 vessels.

Figure 2 and Table 1 compare the effects of 0.1 µM ZM-241385 on CGS-21680 and adenosine dose-response curves. ZM-241385 almost completely inhibited CGS-21680-induced dilation, leading to an 156-fold increase of EC₂₅ values (Table 1). On the other hand, it only partially attenuated dilation induced by adenosine, increasing EC₂₅ values by only 16-fold. The almost complete inhibition of CGS-21680 suggests that ZM-241385 is an effective antagonist of the A_2A receptor. On the other hand, neither the A₁ receptor-selective antagonist CPX (0.1 µM) nor the A₃ receptor-selective antagonist MRS-1191 (0.1 µM) affected the dilator responses of intracerebral arterioles to adenosine (Fig. 3).

View larger version (22K): [in this window] [in a new window]   Fig. 2.   A: effect of 0.1 µM ZM-241385, a selective A2A receptor antagonist, on concentration responses to CGS-21680 (n = 12 vessels). B: effect of 0.1 µM ZM-241385 on concentration-responses to adenosine (n = 9 vessels). Dilation responses are expressed as a percentage of baseline diameter (vessel diameter after development of myogenic tone). Values are means ± SE.

View larger version (19K): [in this window] [in a new window]   Fig. 3.   Lack of effect of the A1 receptor-selective antagonist 8-cyclopentyl-1,3-dipropylxanthine (CPX; 0.1 µM, n = 4 vessels) and the A3 receptor-selective antagonist MRS-1191 (0.1 µM, n = 6 vessels) on dilator responses of intracerebral arterioles to adenosine (n = 10 vessels). Dilation responses are expressed as a percentage of baseline diameter (vessel diameter after development of myogenic tone). Values are means ± SE.

We also evaluated the effect of the nonselective adenosine antagonist CGS-15943 on the dilator responses of intracerebral arterioles to adenosine. CGS-15943 attenuated adenosine-induced dilation in a concentration-dependent manner (Fig. 4). The inhibitory effect of 0.1 µM ZM-241385 appears to be intermediate between that of 0.1 and 1 µM CGS-15943, although there were no significant statistical differences between adenosine concentration-response curves in the presence of either 0.1 µM CGS-15943 or 0.1 µM ZM-241385. On the other hand, 1 µM CGS-15943 caused significantly (P > 0.05) more pronounced inhibition of the adenosine concentration-response curve than 0.1 µM ZM-241385 (Fig. 4). Coapplication of CGS-15943 (0.1 or 1 µM) and 0.1 µM ZM-241385 did not lead to further inhibition of adenosine-induced dilation compared with inhibition by CGS-15943 alone. Moreover, the attenuation of the dilation response to adenosine by CGS-15943 was not concentration specific but occurred throughout the entire range of adenosine concentrations. The present finding thus contrasts with a previous study (8) in piglet coronary arterioles in which CGS-15943 (0.1 µM) attenuated dilation only to higher (1 µM) concentrations of adenosine, presumably by selectively antagonizing A_2B receptors.

View larger version (26K): [in this window] [in a new window]   Fig. 4.   Effect of the nonselective adenosine antagonist CGS-15943 (CGS) on dilator responses of intracerebral arterioles to adenosine. Adenosine concentration responses in the presence of 0.1 and 1 µM CGS-15943 are compared with those in the presence of 0.1 µM ZM-24138 (ZM). Also shown is the effect of coapplication of 0.1 µM ZM-241385 and CGS-15943 (0.1 or 1 µM) on adenosine-induced dilation of cerebral arterioles. Values are means ± SE. *P < 0.01 vs. responses in the presence of antagonists; #P < 0.05 vs. 1 µM CGS-15943 and vs. 1 µM CGS-15943 + 0.1 µM ZM-24138.

To assess the specificity of the inhibitory action of ZM-241385, we determined its effects on dilation induced by ATP (intraluminal application) and acidic (pH 6.8) buffer. As shown in Fig. 5, 0.1 µM ZM-241385 significantly attenuated adenosine-induced dilation of intracerebral arterioles but did not affect the dilator responses to ATP and acidic buffer. Because DMSO was used as a solvent in some of the above experiments, we evaluated the effect of various concentrations of DMSO on intracerebral arterioles. As depicted in Fig. 6, buffer containing as little as 0.5% DMSO caused pronounced dilation of intracerebral arterioles. In the present study, the DMSO concentration in all test solutions did not exceed 0.1%.

View larger version (36K): [in this window] [in a new window]   Fig. 5.   The inhibitory effect of ZM-241385 (ZM) is specific to adenosine (ADO)-induced dilation. ZM-241385 (0.1 µM) significantly attenuated the dilation response of intracerebral arterioles to 10 µM adenosine (n = 9 vessels) but had no effect on dilator responses to 10 µM ATP (n = 6 vessels) and acidic (pH 6.8) buffer (n = 11 vessels). ATP was applied by intraluminal perfusion, whereas both adenosine and acidic buffer were applied adventitially (into the vessel chamber). Values are means ± SE. *P < 0.001 vs. adenosine.

View larger version (33K): [in this window] [in a new window]   Fig. 6.   Effect of DMSO concentration on intracerebral arterioles (n = 5). Buffer containing as little as 0.5% DMSO caused pronounced dilation of intracerebral arterioles. In the present study, the concentration of DMSO in all test solutions did not exceed 0.1%. Values are means ± SE.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

We (11) have previously shown that adenosine-induced dilation of brain parenchymal arterioles occurred mainly by activation of A₂ adenosine receptors. The present study extends these earlier findings by providing the first evidence that both A_2A and A_2B receptor subtypes mediate adenosine-induced dilation in cerebral arterioles in vitro.

Two findings in the present study strongly suggest that A_2A receptors are involved in adenosine-induced dilation of rat intracerebral arterioles. First, the selective A_2A receptor agonist CGS-21680 induced marked concentration-dependent dilation in these arterioles. Second, the relatively selective A_2A receptor antagonist ZM-241385 inhibited the dilation responses to both CGS-21680 and adenosine. CGS-21680 is a highly selective agonist that is virtually ineffective in A_2B receptors as well as other adenosine receptor subtypes (6, 7). On the other hand, ZM-241385 is only modestly selective for A_2A over A_2B receptors, with a reported A_2A versus A_2B receptor selectivity of 30- to 80-fold in guinea pig functional preparations (17) and a ~60-fold selectivity in human receptor functional assays (15). In the present study, however, ZM-241385 almost completely abolished the vasodilation induced by CGS-21680 without affecting the vasodilation induced by intraluminal ATP and pH changes in the extraluminal bath, suggesting that it is both an effective and specific antagonist for the A_2A receptor in rat intracerebral arterioles. A robust response to CGS-21680, together with inhibition of vasodilation by ZM-241385, suggests that A_2A receptors at least partially mediate the dilator response to adenosine.

Although no selective receptor antagonists are currently available to unequivocally assess the role of A_2B receptors in adenosine-induced dilation, our results nevertheless suggest that such receptors may play a role in the dilation response to adenosine, at least at concentrations > 1 µM. Two lines of evidence support the notion that both A_2A and A_2B receptor subtypes are coupled to adenosine-induced dilation in cerebral arterioles in the rat. First, adenosine, which is capable of activating both receptor subtypes, caused a greater maximal dilation than the specific agonist CGS-21680. Second, ZM-241385, the relatively selective A_2A receptor antagonist, only partially inhibited dilation induced by adenosine but almost completely blocked CGS-21680-induced dilation. The significantly greater dilation by adenosine compared with that caused by CGS-21680 occurred at concentrations > 1 µM. We therefore speculate that at lower concentrations, both adenosine and CGS-21680 primarily activate the A_2A receptor, whereas at higher concentrations (>1 µM), adenosine but not CGS-21680 also activates the A_2B receptor to cause greater maximal dilator effects. Such reasoning is consistent with the concept of high (A_2A) and low (A_2B) affinity adenosine A₂ receptors (1, 5).

To further assess the role of A_2B receptors, we determined the effect of the adenosine antagonist CGS-15943 on adenosine-induced dilation in intracerebral arterioles. CGS-15943 has recently been reported to be a potent and relatively selective inhibitor of A_2B receptor-induced dilation in piglet coronary arterioles (8). In that study, 0.1 µM CGS-15943 attenuated adenosine- and 5'-N-ethylcarboxamideadenosine (NECA)-induced vasodilation only at high agonist concentrations (1 µM) but not at low concentrations, suggesting that CGS-15943 preferentially inhibited the low-affinity (A_2B) receptors. In contrast, the inhibitory effect of CGS-15943 in this study did not appear to be restricted to high adenosine concentrations but was apparent throughout the entire concentration range of adenosine studied. CGS-15943 (1 µM) abolished the adenosine-induced dilation up to an adenosine concentration of 1 µM and potently attenuated dilation responses to higher concentrations of adenosine. Such an observation is consistent with the nonselectivity of CGS-15943 for the A_2A and A_2B receptor subtypes in most tissues (6, 7). Parenthetically, at human receptors expressed in cell lines, CGS-15943 has been reported to have a higher A_2A versus A_2B receptor selectivity than ZM-241385. We concluded that CGS-15943 did not selectively attenuate the A_2B receptor-mediated dilation in this study and thus is not useful for identifying A₂ receptor subtypes in cerebral arterioles.

In the study by Hein et al. (8), 1 µM ZM-241385 completely abolished dilation responses to 1 µM adenosine and markedly attenuated dilation to a higher (10 µM) adenosine concentration. The residual dilation to 10 µM adenosine was abolished by adding 0.1 µM CGS-15943, presumably because of selective blockade of the high-affinity A_2B receptors by CGS-15943. However, because of the relatively modest A_2A versus A_2B receptor selectivity of ZM-241385, we elected to use a lower (0.1 µM) concentration of ZM-241385. Although 0.1 µM ZM-24138 effectively antagonized A_2A receptor (CGS-21680)-induced dilation in this study, it caused a less pronounced attenuation of dilation to high (1 µM) concentrations of adenosine (Fig. 2). Moreover, addition of 0.1 µM ZM-241385 to CGS-15943 in this study did not lead to further inhibition of the dilation responses that remained after blockade by CGS-15943. Such data is consistent with the notion that 0.1 µM ZM-241385 is relatively A_2A receptor selective and has minimal effects on A_2B receptors in cerebral arterioles.

It is now well accepted that A₂ receptors mediate vasodilation to adenosine in the cerebral vasculature (16) as well as in a variety of other vascular beds (14). The present study, in addition, provides evidence against the involvement of both A₁ and A₃ receptors by showing that the adenosine-induced dilation of cerebral arterioles was neither affected by the A₁ receptor-selective antagonist CPX nor by the A₃ receptor-selective antagonist MRS-1191. Few studies have attempted to identify the A₂ receptor subtypes (A_2A or A_2B) involved in cerebral vasodilation to adenosine, presumably due to a lack of selective pharmacological tools. The present study represents the only in vitro study thus far designed to address this issue. With the use of an in vivo open cranial window preparation, Coney and Marshall (2) reported that topical application of CGS-21680 increased cortical blood flow in rats and that ZM-241385 attenuated the cortical response to topically applied adenosine. However, at the very high concentration used (150 µM), ZM-241385 lacks selectivity. Moreover, no concentration-response curves were obtained. Thus the results of the study by Coney and Marshall (2) cannot discriminate between A_2A and A_2B receptors. In another in vivo study in rats (18), dilations of pial arterioles (closed cranial window preparation) to adenosine and NECA were attenuated by topical ZM-241385 (1 µM). Because alloxazine (1 µM) caused a greater suppression of adenosine and NECA dose responses than ZM-241385, the authors concluded that the A_2B subtype mediates adenosine-induced dilation in cerebral vessels. Such an interpretation is questionable because of the lack of selectivity of alloxazine, which is only about nine times more potent at A_2B receptors than at A_2A receptors (6).

We (11, 12) as well as other investigators (8) have discussed in detail the inherent advantages of the isolated vessel preparation over in vivo methods. Those include the precise control of physical and chemical factors and relative immunity to confounding problems such as tissue uptake and hormonal changes. With the pressure myograph approach used in the present study, vessels developed spontaneous tone without the need for pharmacological vasoconstrictors. As in previous studies (11, 12), we were careful to preserve the viability and responsiveness of isolated cerebral arterioles with the use of criteria such as myogenic tone development at 37°C, rhythmic vasomotion, and reactivity to agents such as pH changes. In the present study, intraluminal application of 10 µM ATP, an endothelium-dependent vasodilator (20), caused a 35% dilation of intracerebral arterioles, confirming the functional integrity of the endothelium in these cannulated vessels.

However, the cannulated and perfused cerebral arterioles in this preparation tended to lose basal tone and/or pH reactivity after repeated dose-response determinations or after prolonged exposure to high agonist concentrations. A similar fragility has also been described for coronary arterioles (8). As stated in METHODS, we took precautions to assure the reliability of data collection. For example, experiments were stopped if vessel diameter did not recover to ±10% of control values after washout. In addition, vessels that lost pH reactivity (<15%) at the end of an experiment were also excluded from data analysis. In our hands, multiple exposures to adenosine did not degrade vascular responses, as evidenced by our time control experiments (Fig. 1B), which revealed that agonist concentration-response relationships were reproducible when repeated in the same experiment. Nevertheless, because of the inability of these vessels to regain tone after exposure to very high agonist concentrations (>0.1 mM), we did not perform classical pharmacological analyses such as Schild regressions (10). In addition, the vessels in this study were highly sensitive to organic solvents such as DMSO. As shown in Fig. 5, even a 0.5% concentration of DMSO led to a marked dilation of ~26%. This sensitivity to solvent concentration thus precluded us from studying drugs with low organic solvent solubility. For example, the A_2A receptor-selective antagonist SCH-58261 has a maximal solubility of 0.5 µM in 10% DMSO solution (solubility data supplied by Schering-Plough Research Institute, Milan, Italy). Thus the maximum concentration of SCH-58261 that can be used in the present study without significant effect on basal diameter is 5 nM. Because this concentration of SCH-58261 only had a weak inhibitory effect on CGS-21680 dose responses (data not shown), we were unable to compare its effect in a meaningful manner with ZM-241385.

In summary, the results of this study provide functional evidence that A_2A receptors mediate dilation of cerebral arterioles to adenosine. Despite the current lack of highly selective A_2B receptor antagonists, our results also suggest that A_2B receptors are probably involved in adenosine-induced dilation. At lower concentrations of adenosine, the response is likely to be mediated primarily by the A_2A receptor, whereas at higher concentrations (>1 µM), A_2B receptors may also be activated to contribute to the dilation response.