Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

Adenosine is an endogenous nucleoside that regulates many physiological functions in cardiovascular, respiratory, renal, immune, and central and peripheral nervous systems via G protein-coupled adenosine receptors (P1 receptors). Four receptor subtypes, A₁, A_2A, A_2B, and A₃, have been identified based on molecular structure, pharmacology, and mechanisms of G protein-mediated signaling mechanisms (for reviews, see Ralevic and Burnstock, 1998; Fredholm et al., 2000). The A₁ receptors are coupled to the inhibition of adenylate cyclase and can act through effector pathways such as stimulation of phospholipase C, activation of potassium channels, and inhibition of N-type calcium channels. The A₂ receptor subtypes, A_2A and A_2B, are coupled to activation of adenylate cyclase, whereas A₃ receptors has been shown to stimulate phospholipase C and D and to inhibit adenylate cyclase (Abbracchio et al., 1995; Baraldi et al., 2000).

In the past decade, considerable effort has been invested in elucidating the physiological roles of the various adenosine receptors in a variety of tissues, including cardiac, vascular, and nonvascular tissues. Adenosine and adenosine receptors are thought to play a critical role in regulating urinary bladder function, especially in conditions of bladder ischemia or during obstruction-induced changes in function that underlie lower urinary tract symptoms (Levin et al., 2000). Although evidence for inhibitory P1-type purinergic receptor-mediated effects by adenosine has been reported in the bladder (Nicholls et al., 1992; King et al., 1997), the receptor subtype(s) and their coupling to smooth muscle relaxation events remain poorly understood. mRNA analysis by Dixon et al. (1996) has revealed the presence of subunits corresponding to all adenosine receptor subtypes in the rat bladder, whereas Northern analysis showed relatively higher expression of the A_2B receptor subtype (Stehle et al., 1992). Based on the order of potency of adenosine agonists for inhibition of carbachol-evoked contractions in rat bladder, it was noted that the pharmacological profile was more consistent with the participation of adenosine A₂ receptor subtypes (Nicholls et al., 1992, 1996). On the other hand, pharmacological analysis of smooth muscle cells isolated from the circular muscle layer of the feline bladder showed that the adenosine receptor subtype mediating contraction resembled that of the A₁ subtype in this species (Yang et al., 2000). In vascular smooth muscle, adenosine interacts with both A₁ and A₂ receptor subtypes to activate K_ATP channels leading to relaxation. In the rabbit mesenteric artery, A₂ receptor subtypes are predominantly involved, whereas in pig coronary arteries, adenosine-evoked activation occurs via A₁ receptors (Quayle et al., 1997). However, it remains unclear which adenosine receptor subtype mediates functional responses in the bladder and whether functional coupling of these receptors to K_ATP channels might provide a mechanism underlying bladder smooth muscle relaxation.

In the present study, we have investigated the nature of the adenosine receptor subtype(s) involved in coupling to K_ATP channels in the guinea pig urinary bladder, a widely used model for studying the physiology of bladder function (Fujii et al., 1990; Herrera et al., 2000). Our studies provide evidence that hyperpolarization of membrane potential via K_ATP channel activation is a necessary component of adenosine receptor activation. Based on pharmacological analysis of membrane potential changes and tissue relaxation effects, it was found that adenosine-mediated relaxation of the guinea pig bladder involves the A_2A receptor subtype, which appears to be linked to K_ATP channel activation by activation of adenylate cyclase.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Materials. Cell culture reagents were purchased from Invitrogen (Carlsbad, CA). The fluorescent imaging plate reader (FLIPR) membrane potential kit was purchased from Molecular Devices (product number R-8034; Sunnyvale, CA). Glyburide (glibenclamide), adenosine, adenosine deaminase, and other adenosine agonists and antagonists, forskolin, and papaverine were purchased from Sigma Chemical (St. Louis, MO). ZM-241385 and 2Cl-IBMECA were purchased from Tocris Cookson Inc. (Ballwin, MO). N-Cyano-N'-(1,1-dimethylpropyl)-N"-3-pyridylguanidine (P1075) was synthesized in house. Compounds were prepared as 10 mM stocks in dimethyl sulfoxide and diluted in assay buffer just before use.

Cell Culture. Urinary bladder smooth muscle cells were prepared as previously described (Gopalakrishnan et al., 1999; Whiteaker et al., 2001). Bladders were removed from anesthetized male guinea pigs (Hartley; Charles River Laboratories, Inc., Wilmington, MA) and cells were isolated by enzymatic digestion with collagenase (1 mg/ml) and pronase (0.2 mg/ml) at 37°C. The suspension was centrifuged at 1300g for 5 min and the cell pellet was resuspended in 5 ml of growth media (Dulbecco's modified Eagle's medium, 10% fetal bovine serum, 100 units/ml penicillin, 100 units/ml streptomycin, and 0.25 mg/ml amphotericin B). The suspension was then passed through a fine polypropylene mesh filter, seeded into 175-cm² cell culture flasks, and maintained in a tissue culture incubator at 37°C under an atmosphere of 10% CO₂. When confluent, cells were passed at least four times without loss of activity. For membrane potential studies, the cells were seeded into black 96-well clear-bottomed plates (VWR, West Chester, PA) at a density of 3 × 10⁴ cells/well.

Fluorescence Assays. Assays were carried out using the FLIPR membrane potential dye in the FLIPR (Molecular Devices). A stock solution of the dye was prepared by the addition of 10 ml of phosphate-buffered saline (catalog number 14287-072; Invitrogen; composition 138 mM NaCl, 2.68 mM KCl, 0.9 mM CaCl₂, 0.49 mM MgCl₂, 8.9 mM Na₂HPO₄, 1.47 mM KH₂PO₄, 0.33 mM sodium pyruvate, and 5.56 mM D-glucose) to a vial of the lyophilized dye stock. The stock solution was further diluted 1:20 in phosphate-buffered saline before use. After media were removed from the cell plate, 180 µl of the diluted FLIPR membrane potential dye was added to the wells and incubated at room temperature for 60 min. Assays were carried out in the FLIPR in a total assay volume of 200 µl at an excitation wavelength of 480 nm by using a 550-nm longpass emission filter (catalog number 0310-4027; Molecular Devices). After fluorescence baseline readout for 1 min, 20 µl of test compounds, at a stock concentration 10-fold higher than the final desired concentration in the assay, was added. Changes in fluorescence were measured for 9 min. In cases where glyburide sensitivities were assessed, 20 µl of 50 µM glyburide was added at the end of the 9-min time point to yield a final concentration of 5 µM. Data were collected for an additional 5-min period.

Isolated Tissue Relaxation Studies. Urinary bladders were removed from anesthetized male guinea pigs (Hartley; Charles River Laboratories, Inc.) weighing 250 to 300 g and immediately placed in Krebs-Ringer bicarbonate solution (composition 120 mM NaCl, 20 mM NaHCO₃, 11 mM dextrose, 4.7 mM KCl, 2.5 mM CaCl₂, mM 1.5 MgSO₄, 1.2 mM KH₂PO₄, 0.01 mM K₂EDTA, equilibrated with 5% CO₂, 95% O₂; pH 7.4 at 37°C). The trigonal and dome portions were discarded and strips 3 to 5 mm in width and 10 mm in length were prepared from the remaining tissue by cutting in a circular manner. One end of the strip was fixed to a stationary glass rod and the other to a Grass FT03 transducer at a basal preload of 1.0 g. This preload proved to be the best condition for a steady-state baseline and reproducible responses to K⁺ stimulation. As previously reported (Herrera et al., 2000; Buckner et al., 2002), the contractions evoked by mild depolarization (i.e., 20 mM K⁺) are largely myogenic in origin as revealed by the lack of sensitivity to the sodium channel blocker tetrodotoxin (0.1 µM). Tissues were allowed to equilibrate for at least 60 min before the assay. Concentration-response curves were generated in a noncumulative manner with a rinse between successive additions.

Data Analysis. In FLIPR studies, the fluorescence responses were corrected for any background changes in the negative control wells, and data were normalized to those observed with 10 µM adenosine (100% assigned to maximum decrease in fluorescence units). In tissue bath assays, responses were measured as decreases in tension and expressed as percentage of tension evoked by 20 mM K⁺. Data are expressed as mean ± S.E.M. The EC₅₀ values were calculated from the concentration-response curve generated by nonlinear regression analysis (GraphPad Prism; GraphPad Software, San Diego, CA). Significant differences between group means were assessed by analysis of variance followed by Student-Newman-Keuls test. Significance was accepted at the P < 0.05 level.

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

Effect on Membrane Potential Responses in Bladder Smooth Muscle Cells

Adenosine evoked concentration-dependent decreases in fluorescence responses in guinea pig urinary bladder smooth muscle cells (Fig. 1). Significant changes in membrane potential responses were noted at concentrations as low as 10 nM and maximal responses were obtained with 10 µM adenosine. The EC₅₀ value was calculated to be 467 nM (log EC₅₀ = 6.33 ± 0.11; n = 9). Adenosine responses were reversed by the addition of 10 µM glyburide or attenuated by glyburide pretreatment (see below), suggesting that these effects could be mediated by activation of K_ATP channels. Glyburide alone did not alter the basal fluorescence response (data not shown). The maximal efficacy of adenosine (94.9 ± 2.5%) was comparable to that of a prototypical K_ATP channel opener such as the cyanoguanidine, P1075 under similar conditions.

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Concentration-dependent changes in fluorescence responses evoked by adenosine in guinea pig bladder smooth muscle cells. Data are normalized to the maximal responses evoked by adenosine (10 µM). Inset, representative traces showing decreases in responses (fluorescence units) to varying concentrations of adenosine (0.003, 0.03, 0.3, 3, 10, and 30 µM) and its reversal upon addition of the KATP channel blocker glyburide.

Effect of Glyburide and Adenosine Deaminase Treatments

To further investigate the involvement of K_ATP channels, smooth muscle cells were pretreated with glyburide for 30 min before addition of adenosine. As shown in Fig. 2, glyburide pretreatment significantly attenuated the response of adenosine in a concentration-dependent manner. To examine whether the effects of adenosine are directly mediated, experiments were performed in the presence of adenosine deaminase, which catalyzes deamination of adenosine to inosine. Addition of varying concentrations of adenosine (0.03-10 µM) along with adenosine deaminase (0.03 units/ml) completely abolished the responses of adenosine, which demonstrates that adenosine activates K_ATP channels by itself, and not via its catabolic products.

View larger version (27K): [in this window] [in a new window]   Fig. 2.   Inhibition of adenosine-mediated responses in bladder smooth muscle cells by glyburide and adenosine deaminase. Shown are responses to adenosine in the absence and presence of glyburide (3 and 30 µM) or adenosine deaminase. Cells are pretreated with glyburide for 30 min and then challenged with adenosine. In studies with adenosine deaminase, adenosine at various concentrations was added in the presence of adenosine deaminase (final concentration of 0.03 units/ml). Shown are means ± S.E.M. of at least three separate determinations.

Effect of Adenosine Receptor Agonists

To address the nature of the adenosine receptor(s) involved in this phenomenon, the effects of various adenosine receptor agonists (Jacobsen et al., 1992; Jacobsen and Knutsen, 2001) were evaluated. In each case, agonist-evoked reductions in membrane potential effects were attenuated by glyburide.

Adenosine A₁ Agonists. The N⁶-substituted adenosine derivatives, including N⁶-cyclopentyladenosine (CPA, A₁ K_i = 2.3 nM) and its analog 2-chloro-N⁶ cyclopentyladenosine (CCPA) and N⁶-(R)-phenylisopropyladenosine (R-PIA; A₁ K_i = 2 nM) are selective agonists at A₁ receptors (Jacobson and Knutsen, 2001). The kinetic data of CCPA are shown in Fig. 3A. The concentration-response curves for changes in fluorescence responses by CCPA and other A₁ receptor agonists are provided in Fig. 3, B and C. Although the efficacies of these compounds were comparable to that of adenosine, the EC₅₀ values were generally in the micromolar range (Table 1). 2-Chloro-adenosine, a stable analog of adenosine that is only about 7-fold more selective for the A₁ receptor (K_i = 9 nM) compared with A_2A/A_2B receptor subtypes (Fredholm et al., 1994), also elicited concentration-dependent changes in membrane potential responses.

View larger version (19K): [in this window] [in a new window]   Fig. 3.   Effects of adenosine A1 receptor agonists in bladder smooth muscle cells. A, typical concentration-dependent reductions in fluorescence responses evoked by A1-selective agonist CCPA and its reversal after the addition of glyburide. Concentration-response profile of adenosine receptor agonists CPA, R-PIA, and CCPA (B) and phenyl ethyl adenosine, 2-chloroadenosine, and adenosine amine congener (ADAC; C). Shown are means ± S.E.M. of at least three separate determinations.

Adenosine A_2A/A_2B Agonists. Next, the effects of CGS-21680, a compound that is approximately 170-fold selective for the A_2A versus A₁ and (A_2A K_i = 15 nM; A₁ K_i = 2600 nM) with weak affinity at A_2B receptors was examined together with effects of the nonselective agonist NECA (Hutchinson et al., 1990; Jacobson and Knutsen, 2001). Both CGS-21680 and NECA reduced fluorescence responses (Fig. 4, A and B) with comparable EC₅₀ values of 22 nM (log EC₅₀, 7.65 ± 0.17) and 11 nM (log EC₅₀, 7.97 ± 0.23), respectively. Other relatively selective agonists, including 2-phenyl amino adenosine, 5'(N-cyclopropyl)-carboxamidoadenosine (CPCA), and methyl carboxamido adenosine were also active, with EC₅₀ values in the nanomolar range (Table 1).

View larger version (28K): [in this window] [in a new window]   Fig. 4.   Effects of adenosine A2 receptor agonists in bladder smooth muscle cells. Shown are concentration-dependent reductions in glyburide-sensitive fluorescence responses evoked by A2A-selective agonist CGS-21680 (A) and A1/A2 agonist NECA (B). C, concentration-response profile of CGS-21680, NECA, CPCA, MECA, and phenylaminoadenosine. Shown are means ± S.E.M. of at least three separate determinations.

Adenosine A₃ Agonists. The effect of A₃ receptor-selective agonist 2Cl-IBMECA, which is about 2500- and 1400-fold selective for rat A_2A and A₁ receptors, respectively (A₃ K_i = 0.3 nM; A_2A K_i = 470 nM; A₁ K_i = 820 nM; Jacobson and Knutsen, 2001), was evaluated. As shown in Table 1 and depicted in Fig. 5A, the EC₅₀ value of 2Cl-IBMECA was 16.6 µM (log EC₅₀ = 4.78 ± 0.23), although the affinity of this compound at the A₃ receptor itself is in the low nanomolar range. Other moderately selective A₃ agonists, including AB-MECA (K_i = 1.39 nM; Jacobson et al., 1993) and N⁶-benzyl NECA (K_i = 6.8 nM; van Galen et al., 1994) were effective, but considerably less potent compared with their affinities reported at the A₃ receptor subtype (Fig. 5B; Table 1).

View larger version (18K): [in this window] [in a new window]   Fig. 5.   A, effects of adenosine A3 receptor agonists in bladder smooth muscle cells. Shown are concentration-dependent reductions in fluorescence responses evoked by the A3-selective agonist 2Cl-IBMECA and its reversal by glyburide. B, concentration-response profile of 2Cl-IBMECA, ABMECA, and N6-benzyl NECA. Shown are means ± S.E.M. of at least three separate determinations.

Effect of Adenosine Receptor Antagonists

To further investigate the adenosine receptor subtype involved in K_ATP channel activation, the effects of selective receptor antagonists were examined.

Inhibition of Adenosine-Mediated Responses by Selective Antagonists. DPCPX, an adenosine antagonist with about 500-fold greater selectivity for the rodent A₁ (K_i = 0.46 nM) than A₂ receptors (Halleen et al., 1987), was used to assess the involvement of the A₁ receptor. As shown in Fig. 6A, 100 nM DPCPX did not alter adenosine-evoked changes in membrane potential. CGS-15943, another compound that is about 100-fold more selective in antagonizing A_2A receptor compared with A_2B, but only about 10-fold selective versus the A₁ subtype (Ongini et al., 1999) also inhibited responses of adenosine (0.3 and 1 µM). The effects of ZM-241385, an A_2A receptor-selective compound with 30- to 80-fold greater selectivity for the A_2A receptor subtype compared with the A_2B receptor and 1000-fold selectivity compared with the A₁ receptor (Poucher et al., 1995), was also examined. As shown in Fig. 7A, ZM-241385 (0.1, 1, and 10 µM) significantly inhibited adenosine-evoked responses.

View larger version (24K): [in this window] [in a new window]   Fig. 6.   Effect of adenosine receptor antagonists. Responses of varying concentrations of adenosine (0.3, 1, and 3 µM) in the absence and presence of the A1 receptor antagonist DPCPX (100 nM) (A) or in presence of an A2 receptor antagonist CGS-15943 (50 nM) (B). Shown are means ± S.E.M. of at least three separate determinations. , values significantly different from responses evoked by adenosine alone.

View larger version (25K): [in this window] [in a new window]   Fig. 7.   Effect of A2 receptor antagonists on adenosine and NECA-mediated effects. A, responses of 3 µM adenosine (denoted by C) or with varying concentrations of ZM-241385 (0.1, 1, and 10 µM indicated along the x-axis) or alloxazine (1 and 10 µM). B, responses of 1 µM NECA alone (denoted by C) or with varying concentrations of ZM-241385 (0.1, 1, and 10 µM indicated along the x-axis) or alloxazine (1 and 10 µM). Shown are means ± S.E.M. of at least three separate determinations. , values significantly different from responses evoked by adenosine alone.

Inhibition of NECA-Mediated Responses by Selective Antagonists. The nonselective adenosine receptor agonist NECA was also used in conjunction with the A_2A-selective antagonist ZM-241385 to further clarify the receptor subtype(s) involved (Fig. 7B). ZM-241385 inhibited responses evoked by NECA in a concentration-dependent manner, suggesting involvement of the A_2A subtype. Again, similar to the effects observed with adenosine, alloxazine at 10 µM, but not at 1 µM, significantly inhibited NECA-evoked responses. Collectively, these observations confirm the involvement of A_2A receptors in activating K_ATP channels in the bladder smooth muscle.

Coupling of Adenosine A_2A Receptors to K_ATP Channels

The observation that adenosine activates glyburide-sensitive changes in membrane potential by activation of A_2A receptors suggests a role for the involvement of intracellular cAMP and subsequent activation of protein kinase A. To assess the nature of coupling of A_2A receptors to K_ATP channels, agents that modulate adenylate cyclase and/or cAMP levels were examined. First, the effect of MDL-12330A, a known inhibitor of adenylate cyclase, was evaluated. Addition of 10 µM MDL-12330A (Fukushima et al., 2001) shifted the concentration-response curve of adenosine to the right, whereas at 100 µM, the effects were completely abolished. These findings provide evidence that adenosine activates K_ATP channels in bladder smooth muscle cells by elevation of intracellular cAMP and stimulation of protein kinase A. Consistent with these observations, direct activation of cAMP levels by forskolin also resulted in concentration-dependent decreases in fluorescence responses that were reversed by glyburide (Fig. 8B). In addition to forskolin, papaverine, a phosphodiesterase inhibitor that elevates cellular cAMP levels by preventing its breakdown, was also found to be effective in activating K_ATP channels.

View larger version (16K): [in this window] [in a new window]   Fig. 8.   Effects of adenylate cyclase modulation on adenosine-evoked membrane potential responses. A, concentration-dependent changes in adenosine responses in the absence and presence of MDL-12330A (10 and 100 µM). Cells were treated with the adenylate cyclase inhibitor MDL-12330A before addition of adenosine. B, concentration-dependent changes in fluorescence responses evoked by forskolin and papaverine. Data are normalized to the decrease in fluorescence responses evoked by 10 µM adenosine.

Adenosine Agonist-Evoked Relaxation of Bladder Strips

The inhibitory effects of adenosine receptor ligands were evaluated in bladder smooth muscle strips precontracted with 20 mM K⁺. It is known that spontaneous phasic contractility of the bladder smooth muscle are largely myogenic in nature, insensitive to suppression by 100 nM tetrodotoxin, but inhibited by L-type calcium channel blockers and K_ATP channel openers (Fujii et al., 1990; Herrera et al., 2000; Buckner et al., 2002). As shown in Fig. 9, the nonselective agonist NECA and other A_2A receptor-selective agonists, CGS-21680, CPCA, and 2-phenylaminoadenosine, inhibited bladder contractions with log IC₅₀ values of 6.41 ± 0.05, 6.38 ± 0.22, 6.91 ± 0.15, and 5.53 ± 0.21, respectively (Table 2). In contrast, the A₁-selective agonist CCPA and the A₃-selective agonist 2Cl-IBMECA did not inhibit contractions, even at the highest concentrations (100 µM) tested. 2-Chloroadenosine also suppressed phasic contractility (log IC₅₀ = 5.92 ± 0.32), which could be attributed to an effect at the A₂ receptor subtypes. The rank order potencies of adenosine agonists in suppressing phasic contractility of the bladder strips and for evoking membrane potential effects were comparable.

View larger version (20K): [in this window] [in a new window]   Fig. 9.   Inhibition of 20 mM K+-stimulated guinea pig bladder strips by adenosine agonists. Shown are concentration-dependent suppression of phasic contractile activity evoked by A2-selective agonists (NECA, CGS-21680, CPCA, phenylaminoadenosine) (A) and A1-selective agonists (CCPA), A3-selective agonist (2Cl-IBMECA), and 2-chloro-adenosine, a stable analog of adenosine (B).

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

The present study provides evidence that adenosine evokes membrane hyperpolarization in guinea pig bladder smooth muscle cells by activation of K_ATP channels. Pharmacological characterization with a variety of subtype-selective agonists demonstrates that these effects are mediated predominantly by activation of the adenosine A_2A receptor. This is further supported by inhibition of adenosine and NECA-evoked hyperpolarization by the A_2A-selective antagonist ZM-241385. Relaxation of myogenic contractions of bladder smooth muscle strips also show a rank order potency for adenosine receptor agonists that further supports the participation of the A_2A receptor subtype. Our studies also demonstrate that, like those previously reported in the vasculature (Kleppisch and Nelson, 1995), adenosine receptor-mediated activation of K_ATP channels in bladder smooth muscle involves activation of adenylate cyclase, leading to an elevation of intracellular cAMP and stimulation of protein kinase A pathways.

Adenosine Receptor-Mediated Activation of K_ATP Channels. By using fluorescence-based readout of membrane potential changes, the present study shows that adenosine can hyperpolarize guinea pig bladder smooth muscle cells in a concentration-dependent manner. Fluorescence-based membrane potential assays of K_ATP channel function serves as a sensitive method to evaluate changes in membrane potential with a response time comparable to ligand-gated channel activation measured by patch-clamp studies. Previous studies have shown that K_ATP channels in bladder myocytes can be activated by P1075 and other structurally divergent K_ATP channel openers with a rank order of potency that correlates well with relaxation of bladder smooth muscle strips (Gopalakrishnan et al., 1999; Whiteaker et al., 2001). The reversal of membrane potential effects by adenosine and subtype-selective agonists by glyburide demonstrates their activation of K_ATP channels in the bladder smooth muscle.

Pharmacological Characterization of Adenosine Receptors. Functional analysis with subtype-selective agonists of K_ATP channel-mediated membrane potential changes and relaxation responses support the participation of adenosine A₂ receptor subtypes (Table 1). In particular, CGS-21680 that is selective for the A_2A versus A_2B receptor (Hutchinson et al., 1989; Jacobson and Knutsen, 2001) was found to be as potent as NECA in evoking membrane potential effects (Table 1). The rank order of potencies, NECA (log EC₅₀ = 7.97) ~ CGS-21680 (7.65) > 2-chloro adenosine (5.90) ~ CCPA ~ CPA ~ R-PIA > 2Cl-IBMECA (4.78) is consistent with that defining the A_2A receptor subtype (Feoktistov and Biaggioni, 1997). As shown in Figs. 3 and 5, agonists with reportedly nanomolar affinity at the A₁ and A₃ receptor subtype were found to affect membrane potential responses only at relatively higher concentrations (EC₅₀ values in the micromolar range). Similarly, in tissue relaxation studies, NECA and CGS-21680 were equipotent in suppressing myogenic contractions of the bladder strips, whereas the A₁ receptor-selective agonist (CCPA) and A₃-selective agonist (2Cl-IBMECA) were inactive. Taken together, the data indicate that the A_2A receptor subtype mediates adenosine-evoked relaxation of urinary bladder smooth muscle.

Mechanism of Adenosine A_2A Receptor Signaling. The inhibition of adenosine effects by the adenylate cyclase inhibitor MDL-123390A together with the observation that the effects can be mimicked by direct activation of adenylate cyclase by using forskolin or after inhibition of cAMP-dependent phosphodiesterase, in a glyburide-reversible manner, implicates that the coupling to K_ATP channels involves elevation of cAMP. Although both A_2A and A_2B receptors are coupled through G_s protein to the activation of adenylate cyclase and elevation of intracellular cAMP (Fredholm et al., 1994), the present study provide evidence that it is the A_2A receptor that is linked to the activation of K_ATP channels in guinea pig bladder smooth muscle. This situation is comparable to that reported in the vasculature, as for example, in arterial myocytes where adenosine acts via the A_2A receptor subtype and protein kinase A to stimulate K_ATP currents (Kleppisch and Nelson, 1995). Adenosine-induced vasodilation in other vascular beds, including the microvasculature of rat diaphragm and afferent arterioles in rat kidney, have also been shown to involve the A_2A receptor, which are coupled to adenylate cyclase activation and opening of the K_ATP channel (Tang et al., 1999; Chen et al., 2000).

A_2A Receptor-K_ATP Channel Interactions. K_ATP channels have been shown to be hetero-octomeric complexes composed of four inward rectifying K⁺ channels belonging to the Kir 6.0 subfamily that forms the K⁺-selective pore and four regulatory proteins, the sulfonylurea receptors. The latter serves to catalyze nucleotide binding and hydrolysis and hosts binding sites for K_ATP channel openers and sulfonylurea blockers (Bryan and Aguilar-Bryan, 1999; Seino, 1999). On the basis of reverse transcription-polymerase chain reaction and pharmacological studies, it has been suggested that K_ATP channels in guinea pig bladder smooth muscle may be derived from combinations of sulfonylurea receptor 2B and Kir6.2. Although our studies show that adenosine A_2A receptor elevates cAMP and possibly activates protein kinase A-mediated phosphorylation processes, the mechanism by which this leads to activation of K_ATP channels remains to be elucidated by future coexpression studies of the A_2A receptor with K_ATP channel subunits.