INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

ADENOSINE MAY BE AN IMPORTANT regulator of coronary blood flow (6) and mediates dilation in a variety of vascular beds. Dilatory responses to adenosine show considerable tissue and species heterogeneity, suggesting roles for different receptor subtypes and/or effector mechanisms. Vascular adenosine receptors include the A₁ (4), A_2A (9, 10), A_2B (17, 33), and A₃ subtypes (12). Although an intact endothelium is not obligatory for the vasodilator action of adenosine (46), these cells may contribute to adenosine responses in coronary, aortic, and pulmonary vessels (14, 40, 45, 50, 51). Thus, within a tissue, adenosine may mediate vascular responses via endothelium-dependent and endothelium-independent mechanisms (20, 52) and via different receptor subtypes. The relative roles of these different components remain poorly understood. Indeed, there is evidence for (22, 28, 52) and against (26, 30, 31) a role for the endothelium [or nitric oxide (NO)] in coronary adenosine responses.

In addition to an incomplete understanding of the receptors and mechanisms involved in adenosine-mediated dilatory responses, contradictory observations exist regarding the impact of age on adenosine responses (7, 34, 35, 38, 48). Age-related changes in adenosine responses may occur as a result of alterations in the contributions of NO-dependent vs. NO-independent responses, altered receptor expression and coupling, or alterations in the ability of the tissue itself to respond to receptor stimulation. In the present study, we tested the hypothesis that NO-dependent and NO-independent mechanisms contribute to adenosine-mediated coronary and aortic dilation and that age selectively reduces the NO-dependent adenosine response. Previous studies provide evidence of age-related reductions in NO-mediated vascular responses (3, 24). Specifically, we 1) characterized maturational and aging-related changes in adenosine responses in the coronary vasculature and aorta, 2) assessed the NO dependence of these responses, and 3) tested whether age impairs the ability of adenosine receptors to induce vascular relaxation (i.e., reduces the efficiency of adenosine receptor-effector transduction).

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Isolated perfused rat hearts. The following studies conform with the Guide for the Care and Use of Laboratory Animals [DHHS Publication No. (NIH) 85-23, Revised 1996]. Coronary studies were performed in hearts isolated from immature (1-2 mo old), mature (3-4 mo old), and moderately aged (12-18 mo old) male Wistar rats. Mean wet heart weights were 0.63 ± 0.03, 1.26 ± 0.04, and 1.57 ± 0.05 g for immature, mature, and moderately aged age groups, respectively. Rats were anesthetized with pentobartitone sodium (50 mg/kg ip). A thoracotomy was performed, and hearts were rapidly excised and immersed in ice-cold perfusion fluid. The aorta was immediately cannulated, and the hearts were perfused in a retrograde fashion at a pressure of 100 mmHg with a Krebs-Henseleit solution containing (in mM) 119 NaCl, 25 NaHCO₃, 4.7 KCl, 1.2 KH₂PO₄, 2.55 CaCl₂, 1.2 Mg₂SO₄, 15 glucose, and 0.05 EDTA. Perfusate was equilibrated with 95% O₂-5% CO₂ at 37°C, giving a pH of 7.4. A small polyethylene tube inserted through the apex of the left ventricle to drain the cavity prevented intraventricular pressure development. Coronary perfusion pressure was measured using a Gould Statham P23XL pressure transducer (Viggo Spectramed, Oxnard, CA) connected to a sidearm of the aortic cannula and was continuously monitored and recorded on a Maclab data acquisition unit (AD Instruments, Castle Hill, Australia). Hearts were continuously bathed in buffer maintained at 37°C. Blood gas values were regularly monitored using a pH/blood gas analyzer (model 238, Ciba Corning Diagnostics, Essex, UK) to ensure an arterial pH of 7.40, PO₂ of 550 mmHg, and PCO₂ of 35 mmHg. Coronary flow rate was determined gravimetrically using a four-place balance. Perfusate was delivered to the heart using a peristaltic pump (Minipuls 2, Gilson, Middleton, WI). Before each individual experiment, the pump flow was calibrated to ensure accurate flow rate determination.

Functional responses in beating hearts. To examine functional characteristics of the coronary vasculature in beating hearts from the three age groups, hearts were isolated as described above (n = 6/group), perfused at a constant pressure of 100 mmHg, and electrically paced at 5 Hz for the duration of the experiment. Baseline measurements were made after a 60-min stabilization period. Maximal vasodilation responses to adenosine and reactive hyperemic responses were then determined randomly in each heart. Hyperemic flows were measured in hearts subjected to 60 s of total coronary occlusion followed by reperfusion. Maximal adenosine responses were acquired during infusion of 3 mM adenosine until stable maximal dilation was achieved. The order of treatment was randomized, and a 15-min recovery period was allowed between stimuli. To examine the role of endogenously released adenosine in reactive hyperemia, a second series of hearts (n = 7/group) was stabilized for 50 min and then switched to perfusion with buffer containing 100 µM 8-p-sulfophenyltheophylline and 10 IU/ml adenosine deaminase. After an additional 10 min of stabilization, responses to transient occlusion were acquired as described above.

Concentration-response curves for KCl and vasodilator drugs in perfused hearts. Cumulative concentration-response curves for KCl were acquired in immature (n = 3), mature (n = 6), and moderately aged hearts (n = 5) to determine contractile sensitivity and concentrations giving 85% or 95% of maximal response (EC₈₅ or EC₉₅) for the constrictor in each age group and to determine the minimal KCl concentration required to induce complete cardiac arrest (no detectable cardiac contractions). After a 30-min stabilization period, hearts were treated with cumulative concentrations of KCl ranging from 5 to 200 mM. Coronary vasoconstriction was allowed to plateau at each concentration before the level of KCl was further increased.

Aortic ring experiments. To obtain additional information regarding the impact of age on vascular sensitivity and responses to adenosine, studies were performed in aortic rings obtained from immature (n = 34), mature (n = 34), and moderately aged (n = 33) male Sprague-Dawley rats. Rings were prepared as described by us in detail for rats and other species (18, 20, 34, 35). Specifically, arterial segments were placed in the perfusion fluid used for perfused hearts and flushed gently to remove adherent blood cells. Segments were cleaned of connective tissue and cut into 2- to 4-mm transverse rings. Two ring segments from each animal were examined. Rings were vertically mounted on stainless steel wires passed through the lumen; one wire was attached to a Grass FT03C strain gauge and the other was fixed in place. The rings were placed in 20-ml tissue chambers and bathed in a modified Krebs-Henseleit solution containing (in mM) 118 NaCl, 4.7 KCl, 2.5 CaCl₂, 1.2 MgCl₂, 1.2 KH₂PO₄, 25.0 NaHCO₃, 11.0 glucose, and 0.03 EDTA. The solution was maintained at 37°C and constantly bubbled with 95% O₂-5% CO₂, giving a pH of 7.4. Rings were equilibrated for 60 min and then stretched to optimal tensions, determined to be 0.8 g in immature, 1.0 g in mature, and 1.5 g in moderately aged rings. These resting tension values were obtained in preliminary experiments (n = 6/group) by progressive stretch of rings in 0.1-g increments until maximal contractions to 65 mM KCl were observed. After 60 min of stabilization at optimal tensions, rings were maximally contracted with 65 mM KCl, washed three times, and allowed to recover for 30 min. In the first series of experiments, concentration-response curves for norepinephrine-induced contractions were acquired in quiescent rings from immature (n = 7), mature (n = 7), and moderately aged rats (n = 8). Concentration-response curves for adenosine were subsequently acquired in one of two ways. In one group, responses were obtained via the conventional cumulative method: adenosine concentration-response curves were acquired in vessels precontracted with a single EC₈₅ of norepinephrine (n = 7 for immature, n = 7 for mature, and n = 8 for moderately aged groups). This conventional cumulative method yields information regarding tissue sensitivity to adenosine but constrains response magnitude to the degree of preconstriction. Thus it does not provide information regarding unconstrained receptor transduction maxima. Therefore, we also acquired extended concentration-response curves for adenosine using functional antagonism, as described in detail by Lew (29). This method generates concentration-response curves unconstrained by tissue maximum response and permits comparison of occupancy-response coupling ranges for receptor systems (such as adenosine receptors) that maximally activate the tissue under study (29). Specifically, norepinephrine was applied cumulatively until tension increased to >50% of the tissue maximum response to 65 mM KCl. Adenosine was then applied cumulatively until vessels relaxed to <50% of maximal KCl tone (n = 7 for immature, n = 7 for mature, and n = 6 for moderately aged groups). This process was repeated until no further response to adenosine was observed or the adenosine concentration approached 10 mM (nearing maximal solubility). Extended concentration-response curves for adenosine were plotted by summation of relaxations to adenosine. The above procedure was repeated in aortic rings incubated with 50 µM L-NAME to inhibit NO synthase (n = 7 for immature, n = 7 for mature, and n = 6 for moderately aged groups). To test the ability of 50 µM L-NAME to inhibit NO-dependent relaxations, rings were precontracted with an EC₈₅ of norepinephrine, and relaxations in response to 10⁸-10⁶ M ACh, the endothelium-dependent dilator, were studied in the absence (n = 7 for all age groups) and presence of 50 µM L-NAME (n = 7 for immature, n = 8 for mature, and n = 6 for moderately aged groups). Additionally, non-endothelium-dependent responses to sodium nitroprusside were examined in immature (n = 8), mature (n = 6), and moderately aged (n = 7) rings.

Data analysis. Values are means ± SE. Statistical comparisons between groups were made using a multiway ANOVA followed by the Newman-Keuls post hoc test for individual comparisons when significant effects were detected. In all tests, significance was accepted at the 95% confidence level (P < 0.05). One-site (3 parameter) and two-site (4 parameter) concentration-response relationships were fit to coronary or aortic data using the following equations
for one-site relationships and

for two-site relationships, where A is the response at infinite dose, [agonist] is the adenosine or NECA concentration, B is the percent contribution from the first site, and pEC and pEC are apparent pEC₅₀ values for the first and second sites, respectively. To determine whether the two-site model provided a statistically improved description of the data relative to the one-site model, an F-test was employed to compare regressions
where SS is the sum of squares, df is degrees of freedom for each model, and the subscripts 1 and 2 refer to the one- and two-site models, respectively. P < 0.05 was considered evidence of a statistically improved fit. Reported pEC₅₀ values and maximum responses represent means of individual determinations ± SE. For coronary responses not reaching a clear maximal relaxation (response plateau), data were constrained to the KCl-induced tone to extrapolate curves to reasonable limits. For extended aortic concentration-response curves, the maximal observed tissue response and the maximal response extrapolated from individual curve fits are reported.

Materials. Adenosine, ACh, KCl, norepinephrine, sodium nitroprusside, and adenosine deaminase were purchased from Sigma Chemical (Castle Hill, Australia). NECA, L-NAME, and 8-p-sulfophenyltheophylline were purchased from Research Biochemicals (Natick, MA).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Effects of maturation and aging on basal and stimulated coronary flow in beating hearts. Resting (or basal) coronary flow was highest in hearts from immature animals and lowest in hearts from moderately aged animals (Table 1). The age-related decline in coronary flow was associated with reduced peak dilation in response to adenosine and a parallel decline in the peak reactive hyperemic response to 60 s of occlusion. Vasodilatory reserve (maximal/basal flow), therefore, declines significantly with age (Table 1). Treatment with 8-p-sulfophenyltheophylline and adenosine deaminase reduced basal coronary flow in immature (but not mature and moderately aged) hearts and significantly reduced reactive hyperemia in all age groups (Table 1). An age-related decline in hyperemic flow was still evident in the presence of adenosine inhibition.

Effects of KCl and adenosine in the coronary vasculature. KCl elevated coronary vascular resistance in a concentration-dependent manner (Fig. 1). The pEC₅₀ for KCl declined modestly with maturation (from 1.98 ± 0.07 to 1.73 ± 0.06, P < 0.05) and did not change further with aging (1.64 ± 0.08). The magnitude of the KCl response increased significantly with maturation and aging (Fig. 1).

View larger version (24K): [in this window] [in a new window]   Fig. 1.   Effect of KCl on coronary resistance in perfused hearts obtained from immature (n = 6), mature (n = 6), and moderately aged (n = 6) rats. Values are means ± SE. *P < 0.05 vs. immature hearts; P < 0.05 vs. mature hearts.

1

1

1

5

4

View larger version (26K): [in this window] [in a new window]   Fig. 2.   Concentration-response curves for adenosine-mediated coronary vasodilation in KCl-arrested perfused hearts from immature (1-2 mo, n = 7), mature (3-4 mo, n = 6), and moderately aged (12-18 and 12-20 mo, n = 7) rats. Responses are expressed as absolute resistance units (A) and percentage of KCl-induced tone (B). Values are means ± SE. *P < 0.05 vs. immature hearts; P < 0.05 vs. mature hearts.

Effects of NECA ± L-NAME in the coronary vasculature. Results for NECA were qualitatively similar to those for adenosine, although NECA was substantially more potent. NECA induced a biphasic concentration-dependent dilation in immature and moderately aged hearts, with a two-site model providing a statistically superior fit to the data (Fig. 3). There were no significant age-related differences in pEC₅₀ values for NECA at the high- or low-sensitivity sites. However, dilatory responses for 10⁷-10⁶ M NECA were significantly reduced with age (Fig. 3), and the magnitude of the high-sensitivity response was reduced by ~50% with aging (Table 2).

View larger version (27K): [in this window] [in a new window]   Fig. 3.   Concentration-response curves for coronary vasodilation in KCl-arrested perfused hearts in response to 5'-N-ethylcarboxamidoadenosine (NECA) alone (n = 12 for immature hearts, n = 11 for moderately aged hearts) and in the presence of NG-nitro-L-arginine methyl ester (L-NAME; n = 8 for immature hearts, n = 6 for moderately aged hearts). Responses are expressed as absolute resistance units (A) and percentage of KCl-induced tone (B). Values are means ± SE. *P < 0.05 vs. immature hearts; P < 0.05 vs. NECA alone.

7

6

Responses to adenosine in aortic rings. Aortic rings from the three age groups responded differently to KCl. Contractions to 65 mM KCl, normalized to tissue cross-sectional area, were significantly greater in mature rings (10.5 ± 0.5 mN/mm²) than in rings from immature (8.0 ± 0.3 mN/mm²) and moderately aged (7.5 ± 0.8 mN/mm²) rats. Additionally, aortic sensitivity to norepinephrine declined with maturation and aging, with pEC₅₀ values of 8.1 ± 0.1 in immature rings (n = 6) vs. 7.3 ± 0.1 in mature rings (n = 7) and 7.2 ± 0.1 in moderately aged rings (n = 6, P < 0.05).

View larger version (27K): [in this window] [in a new window]   Fig. 4.   Adenosine-mediated relaxation of norepinephrine-contracted aortas from immature, mature, and moderately aged rats. Data were acquired for conventional cumulative concentration-response curves with response magnitude expressed relative to norepinephrine-induced tone (A; n = 6 for immature, n = 7 for mature, n = 6 for moderately aged groups) and extended concentration-response curves with response magnitude expressed as mN normalized to vascular cross-sectional area (B; n = 7 for immature, n = 7 for mature, n = 6 for moderately aged groups). Extended concentration-response curves for adenosine were acquired in the absence and presence of 50 µM L-NAME. Values are means ± SE. L-NAME had no effect in moderately aged tissue. Although not indicated (for clarity), L-NAME significantly reduced responses to 104-102 M adenosine in mature rings and to 106-102 M adenosine in immature tissue (P < 0.05). *P < 0.05 vs. immature rings; P < 0.05 vs. mature rings.

Responses to ACh and sodium nitroprusside in aortic rings. ACh potently relaxed norepinephrine-contracted rings from all age groups (Fig. 5A). Response magnitude and sensitivity to ACh were not markedly altered by maturation but were significantly reduced with aging. pEC₅₀ values for ACh were 7.6 ± 0.2 and 7.4 ± 0.1 in immature and mature rings, respectively, and 7.1 ± 0.1 in moderately aged rings (P < 0.05). At 50 µM, L-NAME almost completely abolished responses to ACh in all age groups (Fig. 5A). Sodium nitroprusside concentration dependently relaxed aortic rings from all age groups, and sensitivity was independent of age (Fig. 5B). However, there was a modest but significant aging-related decline in response magnitude for the lowest sodium nitroprusside concentration examined (3 nM).

View larger version (22K): [in this window] [in a new window]   Fig. 5.   Relaxations in response to ACh (A) and sodium nitroprusside (B) in norepinephrine-contracted aortas from immature and moderately aged rats. Responses to ACh were obtained in the absence (n = 7 for immature, n = 7 for mature, n = 6 for moderately aged groups) and presence (n = 7 for all age groups) of 50 µM L-NAME. Responses to sodium nitroprusside were acquired in untreated hearts (n = 7 for immature, n = 6 for mature, n = 6 for moderately aged groups). Values are means ± SE. All responses in L-NAME-treated rings differed significantly from responses in untreated rings (A). *P < 0.05 vs. immature rings; P < 0.05 vs. mature rings.

NO-dependent and NO-independent responses to adenosine. Assuming that vascular responses in the presence of L-NAME represent NO-independent relaxations (verified by abolition of ACh-mediated dilation with 50 µM L-NAME), we calculated the L-NAME-insensitive (NO-independent) response in coronary and aortic tissues by subtraction of the NO-independent response from control responses (Fig. 6). We previously applied this analysis in studies of vascular responses in the aorta and rabbit ear arteries (20, 21). The results indicate that age-related changes in coronary and aortic sensitivities to NECA and adenosine involve reductions in the magnitude of the NO-dependent and NO-independent relaxations (Fig. 6). Figure 6A shows that the NO-dependent coronary response (occurring at 10⁸-3 × 10⁷ M NECA) is reduced by >70% with aging. Similarly, the NO-independent response (in the presence of L-NAME) is reduced by >70% with aging. Figure 6B depicts similar changes in aortic tissue responses with maturation and shows that there is no detectable NO-dependent response in moderately aged tissue.

View larger version (26K): [in this window] [in a new window]   Fig. 6.   NO-independent (L-NAME insensitive) and NO-dependent (L-NAME sensitive) responses for NECA in coronary vessels (A) and adenosine (B) in aortas from different age groups. NO-independent responses are those acquired in the presence of 50 µM L-NAME. NO-dependent responses were determined by subtraction of NO-independent responses from control responses, as described in METHODS.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Adenosine may play a key role in coronary vasoregulation during elevations in energy demand and during and after ischemia or hypoxia (6, 41). We previously showed that maturation is associated with alterations in vascular sensitivity to adenosine (35-37) and observed changes in adenosine levels during postnatal development (37). However, mechanisms underlying these age-related changes and their physiological relevance remain unclear. Moreover, no data exist regarding the relative roles of NO-dependent vs. NO-independent responses in different age groups or regarding the impact of age on the adenosine receptor transduction maxima in vascular tissue. Here we show that adenosine and NECA induce biphasic concentration-dependent coronary dilation in immature, mature, and moderately aged hearts, supporting the existence of multiple adenosine "sites" or receptors. The coronary responses are partially NO dependent, and the age-related decline in response magnitude (but not sensitivity) involves reductions in NO-dependent and NO-independent components. Data for a large conduit vessel (the aorta) demonstrate a single adenosine site that is also partially NO dependent and reveals that sensitivity to adenosine declines significantly with age. This age-related change involves a reduction in the relative contribution of NO-dependent dilation with age, together with a decline in the adenosine receptor transduction maximum.

Aging and adenosine responses in functioning hearts. Much has been made of morphological changes contributing to age-related reductions in coronary flow and vasodilator responses (1, 2, 11). Aging reduces myocardial arteriolar density and leads to luminal narrowing, which may impair perfusion and dilatory responses (1, 2). The importance of these structural changes vs. changes at the level of smooth muscle and endothelium remains unclear. As shown in Table 1, basal and maximal hyperemic flows decline with maturation and aging, and the peak reactive hyperemic and adenosine-mediated responses change in parallel. Moreover, simultaneous adenosine antagonism and enzymatic degradation significantly reduce the age-related difference in basal flow and reduce reactive hyperemia, supporting a role for endogenous adenosine in this response. These data are consistent with previous studies supporting a role for adenosine in coronary hyperemic responses in the rat (47), guinea pig (15), dog (32), and pig (27). Collectively, these data support a regulatory role for endogenous adenosine and suggest that age-related changes in the vascular adenosine response could contribute to age-related alterations in coronary flow and hyperemia in the intact heart.

Multiple adenosine sites in coronary vessels. Adenosine appears to mediate coronary dilation by interacting with two sites or mechanisms possessing quite different sensitivities (Fig. 2, Table 1). A straightforward explanation is that two receptor subtypes with different sensitivities exist in coronary vessels, an interpretation consistent with other recent studies (16, 43). All age groups displayed high- and low-sensitivity responses to adenosine, with pEC₅₀ values of ~5.5 and 2.4, respectively. The pEC₅₀ values for adenosine and NECA at the high-sensitivity site are compatible with adenosine A₂ receptors. A_2A and A_2B subtypes have been localized to coronary vessels, with A_2A receptors mediating coronary relaxation in the dog (13), pig (30), and guinea pig (5) and A_2B receptors mediating coronary relaxation in humans (26) and in the rat (31). It is therefore likely that the high-sensitivity site reflects activation of adenosine A_2B receptors. In contrast, the pEC₅₀ values for adenosine and NECA at the low-sensitivity site are inconsistent with known affinities at A₂ or A₃ receptors. A low-affinity P₃ receptor such as that reported in aortic smooth muscle (8) could contribute to this response, and there is also evidence supporting an intracellular adenosine receptor in vascular tissue (42). An intracellular receptor would explain low sensitivity to extracellular agonists and would also explain the continued responsiveness of coronary vessels at high agonist concentrations (i.e., nonsaturation of the functional response). The physiological significance of this response seems questionable, since the pEC₅₀ corresponds to extracellular adenosine concentrations >0.5 mM, and interstitial or vascular adenosine levels rarely exceed 0.01 mM, except during global myocardial ischemia (19, 49). However, an intracellular receptor could respond effectively to cytosolic adenosine, potentially stimulating relaxation during periods of vascular deenergization.

NO-dependent and NO-independent adenosine responses. Although the existence of multiple receptor subtypes is a straightforward interpretation of our data (Fig. 2), tissue sensitivity is a function of receptor affinity and the ability of receptors to generate measurable tissue responses (i.e., the functional efficiency of the receptor effector). Thus a biphasic response could result from activation of a single class of receptors interacting with multiple effector mechanisms. A biphasic response could also occur with activation of a single receptor subtype within different vascular compartments or cell types (21). We tested the involvement of multiple effector mechanisms by studying effects of NO synthase inhibition. L-NAME was shown to effectively abolish NO-dependent responses to ACh (Fig. 5), verifying its inhibitory properties. Importantly, L-NAME inhibited adenosine responses in immature and moderately aged hearts, eliminating the biphasic nature of the concentration-response curves. However, the age-related decline in response magnitude was preserved. Conversion of the biphasic concentration-response curve to a monophasic curve, with a pEC₅₀ between the pEC₅₀ values for high- and low-sensitivity sites in untreated hearts, indicates that L-NAME inhibits or antagonizes the high-sensitivity site (since it is unlikely that the low-sensitivity site is sensitized by an inhibitor). Moreover, the absence of a detectable low-sensitivity response indicates antagonism by L-NAME, with a rightward shift to concentrations beyond the NECA range examined. The data shown in Fig. 3, coupled with the demonstrated ability of L-NAME to abolish NO-dependent vasodilation (Fig. 5), support mixed NO-dependent and NO-independent responses to adenosine in coronary vessels, with a major role for the NO-dependent response at lower agonist concentrations (Fig. 6). The NO-independent adenosine response is less sensitive than the NO-dependent response (by >10-fold).

Mechanisms contributing to age-related reductions in the vascular adenosine response. There are three basic ways by which age might reduce the vascular response to adenosine. First, recruitment of different receptor subtypes may extend the response range if these receptors have different affinities. This first possibility is consistent with the apparent existence of two coronary adenosine receptors (a high-affinity A_2B receptor and a low-affinity receptor of unknown identity), as discussed above, and with the observation that age significantly reduces the magnitude of the high-sensitivity response (Table 2). A second possible mechanism involves a change in the relative roles of different effector mechanisms. Effects of L-NAME in coronary and aortic tissues support mixed NO-dependent and NO-independent adenosine responses (Figs. 3 and 4) and demonstrate that the magnitude of both components declines with age (Fig. 6). A decline in NO-dependent relaxation is consistent with previous studies demonstrating reduced NO-dependent vasodilation with aging (3, 24) and is consistent with our observation of an age-related decline in the response to ACh (Fig. 5A). Thus NO-mediated vasodilation appears to be generally impaired with aging. The age-related decline in the L-NAME-insensitive (i.e., NO-independent) adenosine response contrasts with observations of unaltered or increased responses to endothelium-independent dilators (23, 43) and with our observation of an unaltered response to sodium nitroprusside (Fig. 5B). Our data therefore support a selective age-related inhibition of the NO-independent adenosine response together with an age-related decline in the NO-dependent response (Fig. 6).

Conclusions. The present study reveals the presence of multiple sites or receptors mediating vasodilation to adenosine in coronary vessels and demonstrates significant maturational and aging-related reductions in the adenosine response. The high-sensitivity adenosine response may be mediated by A_2B receptors, whereas the identity of the low-sensitivity site remains obscure. Effects of L-NAME indicate that adenosine mediates vasodilation via NO-dependent and NO-independent mechanisms in coronary and aortic tissues. Age-related reductions in adenosine-mediated vasodilation involve changes in both of these responses and may also involve an age-related reduction in the intrinsic ability of vascular tissue to respond to adenosine receptor activation. These alterations in adenosine receptor-mediated responses may impact significantly on vascular function and coronary flow regulation with age.