Receptor subtypes mediating adenosine-induced dilation of cerebral arterioles.

The purpose of this study was to investigate the receptor subtypes that mediate the dilation of rat intracerebral arterioles elicited by adenosine. Penetrating arterioles were isolated from the rat brain, cannulated with the use of a micropipette system, and luminally pressurized to 60 mmHg. Both adenosine and the A2A receptor-selective agonist CGS-21680 induced dose-dependent vasodilation (-logEC(50): 6.5 +/- 0.2 and 8.6 +/- 0.3, respectively). However, adenosine, which is capable of activating both A2A and A2B receptors, caused a greater maximal dilation than CGS-21680. The A2A receptor-selective antagonist ZM-241385 (0.1 microM) only partially inhibited the dilation induced by adenosine but almost completely blocked CGS-21680-induced dilation. Neither 8-cyclopentyl-1,3-dipropylxanthine (0.1 microM), an A1 receptor-selective antagonist, nor MRS-1191 (0.1 microM), an A3 receptor-selective antagonist, attenuated adenosine dose responses. Moreover, ZM-241385 had no effect on the dilation induced by ATP (10 microM) or acidic (pH 6.8) buffer. We concluded that the A2A receptor subtype mediates adenosine-induced dilation of intracerebral arterioles in the rat brain. Furthermore, our results suggest that A2B receptors may also participate in the dilation response to adenosine.

Frequency-dependent changes in cerebral blood flow and evoked potentials during somatosensory stimulation in the rat.

Contrary to the concept of neuronal-vascular coupling, cortical evoked potentials do not always correlate with blood flow responses during somatosensory stimulation at changing stimulus rates. The goal of this study is to clarify the effects of stimulus frequency on the relationship between somatosensory evoked potentials (SEPs) and cerebral blood flow. In rats anesthetized with alpha-chloralose, we measured SEPs by signal-averaging field potentials recorded with an electrode placed on dura overlying the hindlimb somatosensory cortex. Regional blood flow was simultaneously assessed in the same region with a laser-Doppler flow (LDF) probe. The contralateral sciatic nerve was stimulated with 0.1 A pulses at the frequencies of 1, 2, 5, 10 and 20 Hz. SEPs (both P1 and N1 components) declined with increasing frequency regardless whether stimulus duration (20 s) or number (100) were kept constant, suggesting that frequency is an important determinant of neuronal activity. In contrast, LDF responses increased to a maximum at 5 Hz, and do not correlate with SEPs. Because CBF should reflect integrated neuronal activity, we computed the sum of SEPS (summation operatorSEP = SEP x stimulus frequency) as an index of total neuronal activity at each frequency. Summation operatorSEP indeed correlates positively (P<0.001) with LDF responses. Thus, during somatosensory stimulation at various frequencies, cerebral blood flow is coupled to integrated neuronal activity but not to averaged evoked potentials.

Suppression of somatosensory evoked potentials by nitric oxide synthase inhibition in rats: methodological differences.

We have previously shown that topically applied N(G)-nitro-L-arginine (L-NNA), a nitric oxide synthase (NOS) inhibitor, suppressed both somatosensory evoked potentials (SEPs) and vascular responses during sciatic nerve stimulation in rats. Due to the normal tight coupling between cerebral blood flow and neuronal activity, we surmise that the vascular response attenuation may be secondary to the SEP decrease. However, a recent study, in which SEPs were recorded with a 'non-contact' electrode placed longitudinally across the cranial window without touching the cortex, did not find a SEP decrease following NOS inhibition. In the present study, we compared SEPs recorded with 'contact' and 'non-contact' electrodes. Regardless of stimulation methods (sciatic nerve or hindpaw), an electrode in contact with the pial surface overlying the hindlimb somatosensory cortex recorded a steady SEP decline during I-NNA application. In contrast, a 'non-contact' electrode did not detect a significant SEP change in the presence of I-NNA. The present results thus confirm the attenuation of SEPs by NOS inhibition.

Estimation of shear and flow rates in pial arterioles during somatosensory stimulation.

We tested the hypothesis that a shear stress-dependent mechanism is involved in the dilation of pial arterioles during somatosensory stimulation. In alpha-chloralose-anesthetized rats implanted with cranial windows, we simultaneously measured the diameter and flow velocity of pial arterioles with video and dual-slit methods. Stimulation (0.2-0.3 V, 5 Hz, 0.5 ms pulses for 20 s) of the contralateral sciatic nerve evoked consistent dilator responses in pial arterioles (36 +/- 1 micron diam) without affecting blood pressure. The dilator responses consisted of an initial transient peak dilation of 30 +/- 3%, followed by a sustained dilation of 13 +/- 1% (n = 11). Mean velocity increased by 16.4 +/- 5.7% at 5 s after stimulus onset. Wall shear rate and volume flow were calculated from diameter and velocity data by assuming a parabolic flow profile. There was no significant change in wall shear rate, whereas flow rate increased significantly during sciatic nerve stimulation. The present findings suggest that a flow (shear stress)-mediated mechanism does not play an important role in the dilator response of pial arterioles to sciatic nerve stimulation.

L-NNA suppresses cerebrovascular response and evoked potentials during somatosensory stimulation in rats.

We studied the local cerebrovascular response and somatosensory-evoked potentials (SEPs) to stimulation of the sciatic nerve during both short- (< 30 min) and long-term (90-150 min) exposure to topically applied NG-nitro-L-arginine (L-NNA). The pial circulation was visualized through a cranial window in alpha-chloralose-anesthetized rats. The diameter of pial arterioles (25-45 microns) and laser-Doppler flow (LDF) in the hindlimb sensory cortex were simultaneously measured during sciatic nerve stimulation. Short-term (< 30 min) treatment with L-NNA (1 mM) abolished the dilation of pial arterioles induced by acetylcholine, whereas the response to sciatic nerve stimulation was not affected. When applied for > 30 min, L-NNA induced severe vasomotion and attenuated the vascular responses to sciatic nerve stimulation. Long-term exposure to topically (1 mM) or systemically (10 mg/kg i.v.) applied L-NNA also attenuated cortical SEPs to sciatic nerve stimulation. Thus L-NNA-induced inhibition of vascular responses may be secondary to suppression of neuronal activity and an L-arginine metabolite, such as nitric oxide, may be involved in neurotransmission in the cerebral cortex during somatosensory activity.

Anesthetic-dependent pial arteriolar response to ethanol.

Anesthetic agents are often administered in the presence of ethyl alcohol, both in research and in the clinical setting. The authors tested the hypothesis that anesthetic agents may affect cerebrovascular responses to ethanol. A closed cranial window preparation in the rat was used to compare the response of pial arterioles to topically applied ethanol (0.01% to 1% vol/vol) in the presence of alpha-chloralose/urethane (50 and 600 mg/kg, respectively) or halothane (0.5% to 1%) anesthesia. Heart rate, mean arterial blood pressure, and blood gas levels were maintained stable and within the physiological range throughout each experiment. Ethanol induced significant vasoconstriction in alpha-chloralose/urethane-anesthetized animals (multivariate analysis of variance (MANOVA), p = 0.039); conversely, ethanol induced significant vasodilation of the pial arterioles in halothane-anesthetized animals (MANOVA, p = 0.017). These responses were significantly different from one another (MANOVA, p = 0.001). Thus, the choice of anesthetic agent alters the cerebrovascular response to ethanol, and care should be taken to ascertain the influence of anesthesia in both research and clinical settings.

Modulation of cerebral arteriolar diameter by intraluminal flow and pressure.

We determined whether cerebral arterioles in vitro adjust their diameters in response to changes in intraluminal flow rate and pressure. Intracerebral arterioles (38- to 55-microns diameter) were isolated from Sprague-Dawley rats and cannulated with a perfusion system that permitted separate control of intraluminal pressure and flow rates. Increasing pressure at 0 flow, in 20 mm Hg steps from 20 to 100 mm Hg, resulted in myogenic constriction, which was greatest at 60 mm Hg (approximately 20%). Increasing flow rate at a constant pressure of 60 mm Hg elicited a biphasic response. At flow rates of up to 10 microL/min, the arterioles dilated by up to 14.5 +/- 2.2% of their control diameter. At higher (> 10 microL/min) flow rates, however, a progressive restoration of resting diameter was observed. Application of the nitric oxide synthase inhibitor NG-mono-methyl-L-arginine (L-NMMA, 0.1 mmol/L) caused a 15.4 +/- 1.7% decrease in control diameter (at 60 mm Hg, zero flow). Although L-NMMA did not affect the responses to increases in pressure or to vasodilators (adenosine and pH 6.8 buffer), it abolished the dilator responses to flow rate increases and to acetylcholine. In contrast, inhibition of prostaglandin synthesis by indomethacin (10 mumol/L) had no effect on flow-induced dilation. These results show that changes in intraluminal flow rates and pressure can independently influence cerebral arteriolar tone and suggest that the flow-induced dilator responses of cerebral arterioles are mediated by an arginine metabolite, such as nitric oxide.

Differential effects of alcohols on intracerebral arterioles. Ethanol alone causes vasoconstriction.

We compared the effect of the acute application of ethanol, methanol, 1-propanol, 1-butanol, urea, and mannitol (1-100 mM) on the basal tone of isolated-cannulated rat intracerebral arterioles to determine if the response of these arterioles to ethanol could be attributed to alteration of membrane fluidity or changes in osmolality. These arterioles spontaneously developed tone to 62.0 +/- 8.4% of passive diameter (44.2 +/- 11.9 vs. 70.9 +/- 14.7 microns). Ethanol caused a dose-dependent reduction in arteriolar diameter starting at 3 mM (p = 0.03), reaching a diameter of 81.4 +/- 3.0% of basal tone at 100 mM. In comparison, all other agents tested caused the arterioles to dilate, with the exception of 1-propanol, which produced inconsistent vessel responses. At 100 mM concentration, methanol, 1-butanol, urea, and mannitol dilated intracerebral arterioles by 116.1 +/- 12.7, 151.5 +/- 12.4, 131.1 +/- 17.0, and 149.8 +/- 6.6%, respectively. Thus, in a concentration range associated with acute intoxication, ethanol causes constriction of isolated intracerebral arterioles. The mechanism of action of ethanol cannot be accounted for solely based upon its physicochemical characteristics of osmolality or lipid solubility, but rather may reflect a more specific action on one or more cellular mechanisms responsible for determining basal intracerebral arteriolar tone. The characterization of the response of intracerebral arterioles to ethanol is important in view of epidemiologic links between ethanol consumption and cerebrovascular disease.

Simultaneous measurements of pial arteriolar diameter and laser-Doppler flow during somatosensory stimulation.

We simultaneously measured pial arteriolar diameter and changes in cortical blood flow during activation of the somatosensory cortex by sciatic nerve stimulation. The pial vasculature was visualized with a closed-cranial window technique in chloralose-anesthetized rats (n = 13). Local blood flow was monitored with laser-Doppler flowmetry. During stimulation of the sciatic nerve (0.2 V, 5 Hz, 20 s), vascular diameter and laser-Doppler flow consistently displayed similar response profiles. With 0.5-ms stimulation pulses, the responses showed an initial peak followed by a smaller but sustained plateau dilation. In contrast, 5-ms pulses evoked a monotonically rising response. Our results support the concept that pial arteriolar diameter changes reflect cortical blood flow responses during somatosensory stimulation.

Effects of adenosine and its analogues on isolated intracerebral arterioles. Extraluminal and intraluminal application.

We evaluated the responses of brain parenchymal arterioles to intraluminal and extraluminal application of adenosine and its analogues. Intracerebral arterioles (28.4- to 60.3-microns diameter) were isolated from Sprague-Dawley rats, cannulated with micropipettes, and perfused in vitro. Both extraluminal and intraluminal adenosine, 5'-(N-ethylcarboxamido)adenosine (NECA), R-N6-(phenylisopropyl)adenosine (R-PIA), and S-N6-(phenylisopropyl)adenosine (S-PIA) elicited concentration-dependent dilation of these arterioles, but intraluminal application was less potent and efficacious than extraluminal application. Inosine was not vasoactive. A common order of agonist potency (NECA > adenosine > R-PIA > or = S-PIA) was determined for both extraluminal and intraluminal application. Theophylline (10 microM) caused a rightward shift of the adenosine concentration-response curve and a 50-fold reduction in potency. Intraluminal theophylline was one sixth as effective as extraluminal theophylline in antagonizing the extraluminal adenosine response, whereas intraluminal 8-sulfophenyltheophylline, a polar theophylline derivative, was ineffective. Polyadenylic acid (PolyA, 1 microM), an adenosine polymer that does not penetrate the endothelium, induced a dilation of 44.2 +/- 5.3% when applied extraluminally but had no effect when infused intraluminally. The dilator effect of PolyA was antagonized by theophylline. We conclude that: (1) intraluminal adenosine and its analogues are effective dilators of intracerebral arterioles, (2) the dilator effects of both intraluminally and extraluminally applied adenosine are predominantly mediated by A2-type receptors, and (3) adenosine receptors mediating vasodilation are not present on the luminal surface of the endothelium.

Changes in pial arteriolar diameter and CSF adenosine concentrations during hypoxia.

We measured the changes in pial arteriolar diameter and CSF concentrations of adenosine, inosine, and hypoxanthine during hypoxia in the absence and presence of topically applied dipyridamole (10(-6) M) and erythro-9-(2-hydroxy-3-nonyl)adenine (EHNA; 10(-5) M). Closed cranial windows were implanted in halothane-anesthetized adult male Sprague-Dawley rats for the observation of the pial circulation and collection of CSF. The mean resting arteriolar diameter in mock CSF was 31.2 +/- 5.9 microns. Topically applied dipyridamole and EHNA, in combination, caused a slight but significant (p < 0.05) increase in resting arteriolar diameter (33.8 +/- 4.3 microns). With mock CSF, moderate hypoxia caused a 22.1 +/- 9.7% increase in pial vessel diameter. Topically applied dipyridamole and EHNA significantly (p < 0.01) potentiated pial arteriolar vasodilation in response to hypoxia. Moreover, the potentiating effects of dipyridamole and EHNA during hypoxia were completely abolished by theophylline (0.20 mumol/g, i.p.; p < 0.05), an adenosine receptor antagonist. Resting concentrations of adenosine, inosine, and hypoxanthine in the subwindow CSF were 0.18 +/- 0.09, 0.35 +/- 0.21, and 0.62 +/- 0.12 microM, respectively. In the absence of dipyridamole and EHNA, these levels were not affected by sustained moderate hypoxia (PaO2 = 36 +/- 6 mm Hg). However, in the presence of dipyridamole and EHNA, the concentration of adenosine in the CSF during hypoxia was significantly (p < 0.05) increased. Our data indicate that dipyridamole and EHNA potentiate hypoxic vasodilation of pial arterioles while simultaneously increasing extracellular adenosine levels, thus supporting the hypothesis that adenosine is involved in the regulation of cerebral blood flow.

Adenosine release and changes in pial arteriolar diameter during transient cerebral ischemia and reperfusion.

We utilized the closed cranial window technique in the anesthetized rat to determine changes in CSF concentrations of adenosine, inosine, and hypoxanthine and pial arteriolar diameter during transient (20 min) forebrain ischemia and reperfusion. After mock CSF under the cranial window was allowed to equilibrate with cerebral interstitial fluid, endogenous adenosine concentration was found to be 0.16 +/- 0.05 microM, while inosine and hypoxanthine were 0.35 +/- 0.17 and 1.23 +/- 0.47 microM, respectively. The concentration of adenosine in CSF increased 4.2-fold during ischemia and 13.8-fold during the first 5 min of reperfusion. Inosine and hypoxanthine concentrations were also significantly increased during ischemia and reperfusion. After 1 h of reperfusion, CSF adenosine and inosine levels had decreased from peak value but remained significantly above preischemic values. In contrast, hypoxanthine remained at peak concentrations even after 60 min of reperfusion. Preischemic arteriolar diameter was 42.6 +/- 11.3 microns and was not significantly changed after 20 min of ischemia. However, during the first 5 min of reperfusion, arteriolar diameter increased significantly (p less than 0.05), coincident with peak adenosine concentrations. By 60 min of reperfusion, arteriolar diameter had returned to baseline. These results indicate that during the postischemic period, adenine nucleosides and hypoxanthine in CSF are elevated and could affect reperfusion.

Lack of sympathetic and cholinergic influences on cerebral vasodilation caused by sciatic nerve stimulation in the rat.

We studied the influences of sympathetic and cholinergic mechanisms on pial arteriolar responses during cortical activation in the rat. Adult male Sprague-Dawley rats were anesthetized with alpha-chloralose and urethane and mechanically ventilated. Pial arterioles on the somatosensory cortex were visualized on a video monitor through a closed cranial window. Changes in arteriolar diameter induced by sciatic nerve stimulation (0.2 V, 5 Hz, 5 ms, for 20 s) were measured before and after (a) ipsilateral superior cervical ganglionectomy (n = 5), (b) intravenous (0.5 mg/kg) administration and topical (10(-5) M) application of atropine (n = 5), and (c) lesion of the nucleus basalis magnocellularis (the major source of intracerebral acetylcholine neurons, n = 7). Unilateral nucleus basalis magnocellularis lesions were performed stereotactically by injection of ibotenic acid (25 nmol/microliter). Sensory cortex cholinergic denervation was confirmed histologically. These treatments had no significant effect on arteriolar responses to sciatic nerve stimulation. Thus, the present results suggest that neither sympathetic nor cholinergic mechanisms play a significant role in somatosensory evoked cerebral vasodilation.

Effects of topical adenosine analogs and forskolin on rat pial arterioles in vivo.

We utilized the closed window technique to study the in vivo responses of rat pial arterioles to superfused adenosine agonists. Adenosine and its analogs dilated pial arterioles and exhibited the following order of potency: 5'N-ethylcarboxamide adenosine (NECA) greater than 2-chloroadenosine (2-CADO) greater than adenosine = R-N6-phenylisopropyladenosine (R-PIA) = S-PIA greater than N6-cyclohexyladenosine (CHA). This potency profile suggests that cerebral vasodilation is mediated through the A2 receptor. Forskolin (10(-9) M) potentiated the vasodilation caused by 10(-6) M NECA, thus implicating adenylate cyclase activation during NECA-induced vasodilation and providing further support for involvement of the A2 receptor.

Role of adenosine in regulation of regional cerebral blood flow in sensory cortex.

We have previously demonstrated that rat pial arterioles located on the somatosensory cortex dilated in response to contralateral sciatic nerve stimulation (SNS). We hypothesized that the vasodilation was mediated by adenosine, released as a result of somatosensory cortex activation. To test this hypothesis, we examined the effects of SNS (0.15-0.2 V, 5 ms, 5 Hz for 20 s) on pial arterioles under conditions of altered adenosine availability. Cerebrospinal fluid (CSF) adenosine was altered by perfusing mock CSF, under a cranial window in anesthetized rats, containing either an adenosine uptake competitor (dipyridamole or inosine) or an adenosine receptor blocker (theophylline). With CSF only, SNS caused pial arterioles (resting diam, 29 +/- 1 micron) to dilate by 38 +/- 10% (peak magnitude) for 32 +/- 2 s. Dipyridamole (10(-6) M) significantly (P less than 0.02) enhanced both the magnitude (to 62 +/- 12%) and duration (to 68 +/- 10 s) of the response. Similarly, inosine (10(-3) M) significantly (P less than 0.02) potentiated the vasodilative response from resting values of 27 +/- 5% and 34.8 +/- 4.1 s to 37 +/- 6% and 89.6 +/- 14.1 s. In contrast, theophylline (5 x 10(-5) M) significantly (P less than 0.001) attenuated arteriolar vasodilation from resting values of 38 +/- 5% and 29.3 +/- 1.2 s to 18 +/- 3% and 22.0 +/- 0.9 s. Neither dipyridamole nor theophylline had a significant effect on neuronal response (sensory-evoked response) recorded from the somatosensory cortex. These results suggest that adenosine is involved in the regulation of pial vasodilation during cerebral cortical activation.

Effect of inosine on pial arterioles: potentiation of adenosine-induced vasodilation.

We examined the cerebral vasoactivity of inosine and its effect on pial arteriolar vasodilation induced by adenosine. Pial circulation was observed through cranial windows implanted in rats anesthetized with halothane. No significant change in venous or arteriolar diameter was apparent when inosine (10(-6) to 10(-3) M) was superfused. In contrast, adenosine in concentrations of 10(-7) and 10(-6) M dilated pial arterioles by 9.0 +/- 1.2 and 17.7 +/- 1.7%, respectively. Addition of 10(-5) M inosine had no effect, whereas 10(-4) M inosine enhanced the vasodilation induced by 10(-7) M adenosine to 19.4 +/- 1.7% and that by 10(-6) M adenosine to 23.3 +/- 2.3%. When theophylline (5 X 10(-5) M) was perfused together with 10(-7) M adenosine and 10(-4) M inosine, the vasodilation was almost completely abolished. The present study indicates that inosine alone does not affect pial vessel diameter but potentiates the response of pial arterioles to exogenous adenosine. This potentiating action of inosine may be attributed to an increase in perivascular adenosine and may be explained by the action of inosine as an adenosine uptake inhibitor.

The effects of dipyridamole and theophylline on rat pial vessels during hypocarbia.

Hypocarbia results in an increase in brain adenosine concentrations, presumably because of brain hypoxia associated with hypocarbic vasoconstriction. It was hypothesized that adenosine limits the degree of hypocarbic vasoconstriction. To test this hypothesis, the effects of dipyridamole and theophylline on CO2 reactivity during hypocarbia were investigated in anesthetized rats. Dipyridamole should reduce the vasoconstriction by potentiating adenosine action, whereas theophylline should increase the vasoconstriction by blocking adenosine receptors. Cortical pial arterioles of mechanically ventilated and anesthetized rats were displayed on a video monitor system through a closed cranial window. Arterial blood pressure and oxygen tension were stable. CO2 reactivity, formulated as 100 X [delta diameter (micron)/resting diameter (micron)]/delta PaCO2 (mmHg), in the hypocarbic phase was calculated before and after topical superfusion of dipyridamole (10(-6) M; n = 7) and theophylline (5 X 10(-5) M; n = 6). CO2 reactivity was significantly decreased after superfusion of dipyridamole (0.57 +/- 0.08; mean +/- SEM) as compared with mock cerebrospinal fluid (CSF) (0.97 +/- 0.17, p less than 0.05, n = 7). On the other hand, CO2 reactivity after superfusion of theophylline was increased (1.63 +/- 0.28) as compared with mock CSF (1.00 +/- 0.20, p less than 0.05, n = 6), indicating that adenosine is involved in hypocarbic vasoconstriction.

Effect of sciatic nerve stimulation on pial arterioles in rats.

The present study documents the microvascular response of the pial circulation in sensory hindlimb cortex to sciatic nerve stimulation. Rats, anesthetized with alpha-chloralose and urethan, were equipped with closed cranial windows, and pial arteriolar diameter was measured during stimulation of the contralateral sciatic nerve. The effects of varying stimulus frequency, intensity, and duration were examined. Optimal stimulus frequency was 5 Hz, but response diminished significantly beyond 10 Hz. Optimal stimulus intensity was 0.2 V. At higher stimulus strength, arteriolar dilation was reduced, but systemic blood pressure rose significantly. At low stimulus frequency and intensities, pial arterioles responded to stimulation with a consistent pattern: initial delay of 1.4 s followed by abrupt dilation to a peak magnitude, subsequent decline to a lesser but still dilated state, and recovery to a resting diameter after the cessation of stimulation. No consistent response profile was discernible at high stimulus intensity and/or frequency. This vasodilatory response was discretely restricted to a limited number of arterioles, confined to the hindlimb somatosensory cortex as confirmed by sensory evoked response. The response of the pial circulation provides a well-characterized model for analysis of brain microcirculation, which presumably is linked to cerebral metabolism.

Role of adenosine in regulation of cerebral blood flow: effects of theophylline during normoxia and hypoxia.

We studied the effects of the methylxanthine theophylline, an adenosine receptor blocker, on cerebral circulation. Cerebral blood flow (CBF) was measured by the retroglenoid outflow and microsphere techniques, and pial circulation changes were observed through a closed cranial window. Intraperitoneal administration of theophylline in normoxic animals resulted in a biphasic response of pial vessels and CBF. At low concentrations (0.05 mumol/g) of theophylline, pial vessel diameter and CBF decreased, whereas vasodilatation and hyperemia were observed at higher levels. After intraperitoneal administration of either 0.05 or 0.2 mumol/g, hypoxic hyperemia was attenuated both during short (c. 30 s) and sustained (c. 2-3 min) hypoxia, as was hypoxic pial arteriolar vasodilatation. These actions of theophylline appear to be due to adenosine receptor blockade, since micromolar concentrations were achieved in cerebrospinal fluid (CSF), and no increases in adenosine 3',5'-cyclic monophosphate concentrations in brain were noted. Moreover, theophylline (either intraperitoneal or topical) blocked pial vasodilatation caused by topically applied adenosine, but had little effect on hypercarbic hyperemia or pial vasodilatation induced by topically applied acetylcholine. The results of these studies suggest that adenosine is involved in the maintenance of resting cerebral vascular tone and has a paramount role in the regulation of CBF during hypoxia.