The objective of this study was to characterize the effects of halothane on the agonist-induced nitric oxide-cyclic GMP (NO-cGMP) mechanisms by comparing the intracellular signal transduction mediating isoproterenol- and acetylcholine-induced nitric oxide formation.
Isoproterenol-induced relaxations of rat aortic rings with and without endothelia were examined in the absence and presence of halothane. Studies were also done in the presence of inhibitors of nitric oxide-synthase, adenylyl cyclase, calmodulin, protein kinase A, and intracellular Ca2+ release mechanism. The relaxations under some of these conditions were compared with those induced by acetylcholine. Cyclic nucleotide contents of the rings were also measured.
Isoproterenol relaxed aortic rings via the endothelium-dependent nitric oxide-cyclic GMP mechanism. Inhibition of adenylyl cyclase or of protein kinase A attenuated the isoproterenol-induced relaxations significantly but did not affect those induced by acetylcholine. Inhibition of intracellular Ca2+ release abolished the acetylcholine-induced relaxations but did not affect those induced by isoproterenol. Calmodulin inhibition attenuated both agonist-induced relaxations significantly. Unlike acetylcholine-induced relaxation, that induced by isoproterenol was not affected by halothane. Isoproterenol increased both the cyclic adenosine monophosphate and cGMP contents of rings significantly when endothelia were intact. Inhibition of nitric oxide synthase attenuated the isoproterenol-induced cGMP content increases significantly but did not affect the cyclic adenosine monophosphate content increases. Halothane (2%) did not affect isoproterenol-induced increases in nucleotide content.
Isoproterenol-induced nitric oxide formation requires the activation of constitutive nitric oxide synthase, but the Ca2+ release mechanism is not involved in activating this enzyme. Halothane can inhibit the nitric oxide-cyclic GMP mechanism only when Ca2+ release is greatly involved in the activation of constitutive nitric oxide synthase.
Key words: Animal: rat. Anesthetics, volatile: halothane. Artery: aorta. Enzyme: adenylyl cyclase; guanylyl cyclase. Pharmacy: nitric oxide. Sympathetic nervous system: beta-adrenergic receptors.
Acetylcholine binds to vascular endothelial muscarinic receptors to activate synthesis of nitric oxide (NO), which diffuses into vascular smooth muscle (VSM) to activate soluble guanylyl cyclase (sGC) to initiate the cyclic-3', 5' guanosine monophosphate (cGMP)-dependent relaxation mechanism. [1-3]Since this discovery, acetylcholine has been used commonly as an NO-producing agonist in many vascular pharmacologic investigations. Many investigators have reported the inhibitory effects of volatile anesthetics on the endothelium-dependent and NO-mediated vasorelaxation of rat aorta elicited by acetylcholine. [4-8]However, the mechanism underlying the inhibition of acetylcholine-induced vasorelaxation by anesthetics remains unclear.
Nitric oxide production elicited by acetylcholine is mediated via a constitutive type of NO synthase (cNOS), of which the enzymatic activity depends on the Calcium2+ -calmodulin complex. [9,10]Release of the intracellular store of Calcium2+ is triggered by inositol triphosphate, which is synthesized by phospholipase C linked to muscarinic receptors, and plays a major role in this enzymatic activation. [11-13]
Recently, two beta-adrenoceptor agonists, isoproterenol [14]and salbutamol, [15]were found to induce endothelium-dependent and NO-mediated vasorelaxation of rat aorta. Conventionally, stimulation by beta-adrenoceptor agonists is thought to evoke adenosine 3', 5'-cyclic adenosine monophosphate (cAMP) synthesis in VSM, which mediates vasorelaxation. [16-20]The mechanism responsible for the aortic NO formation elicited by isoproterenol, however, has not been clarified.
It is generally accepted in cell types other than the vascular endothelial that stimulation of beta-adrenoceptor evokes adenylyl cyclase activation by mediating G-protein to produce cAMP, which acts in conjunction with cAMP-dependent protein kinase (PKA) to induce Calcium2+ influxing through voltage-gated Calcium2+ channels, resulting in intracellular Calcium2+ increase. [21-25]Although it has been indicated that endothelial cells lack voltage-gated Calcium2+ channels, isoproterenol-elicited NO production may be mediated via cNOS activation due to Calcium2+ influx through other types of Calcium2+ channels. This hypothetical mechanism is clearly distinct from that responsible for acetylcholine-induced NO production.
The aims of this study were to compare isoproterenol-induced cNOS activation with that induced by acetylcholine and to determine the effect of halothane on the isoproterenol-stimulated relaxation mechanism in the rat aorta.
Materials and Methods
With the approval of the Animal Use Committee of Wakayama Medical College, male Wistar rats weighing 300 to 400 g were anesthetized with intraperitoneally injected pentobarbital (30 mg/kg). Midline abdominal incisions were made, the abdominal aorta were exposed, and the rats were killed by exsanguination from the abdominal aorta. Each descending thoracic aorta was removed, cleared of fat and connective tissue, and cut into rings approximately 3 mm long. The endothelia were removed for some experiments by gently rubbing the intimal surfaces of the aortic rings with a cotton swab. The rings were mounted between two parallel tungsten wire hooks and placed in water-jacketed 10-ml organ baths containing modified Krebs' solution (composed of 119 mM NaCl, 4.7 mM KCl, 1.17 mM MgSO4, 25 mM NaHCO sub 3, 1.18 mM KH2PO4, 2.5 mM CaCl2, and 11 mM glucose), which contained indomethacin (10 sup -5 M). The solution was gassed with 5% v/v carbon dioxide in oxygen and maintained at 37 degrees Celsius. The lower hook was attached to a support leg and the upper one to a force transducer (Nihondenki-Sanei Co., Tokyo, Japan). Changes in isometric force were amplified (Nihondenki-Sanei Co., Tokyo, Japan) and displayed on an ink-writing recorder (Nihondenki-Sanei Co.). The aortic rings were placed under 1.0-g resting tension and allowed to equilibrate for 60 min, during which the bath fluids were exchanged every 15 min.
Submaximal contraction of the aortic rings in response to norepinephrine (10 sup -7 M) during a period of 20 min was confirmed. After rings were washed with fresh fluid they were contracted again and checked for the presence of endothelium by confirming that at least 60% relaxation occurred in response to acetylcholine (10 sup -6 M). Any rings showing less than this level of relaxation were discarded as having a partially damaged endothelium. Successful endothelial denudation of the aortic rings was confirmed by the loss of relaxation in response to acetylcholine (10 sup -6 M). To examine the mechanism of signal transduction mediating isoproterenol-induced rat aortic vasorelaxation, some rings were equilibrated with propranolol (a nonspecific beta-adrenoceptor antagonist, 5 x 10 sup -5 M), NG-nitro-L-arginine (L-NNA, an inhibitor of NOS, 5 x 10 sup -5 M), [26]2', 3'-dideoxyadenosine (2', 3'-DDA, an inhibitor of adenylyl cyclase, 10 sup -5 and 10 sup -4 M), [27]W-7 (a calmodulin inhibitor, 10 sup -5 and 5 x 10 sup -5 M), [28,29]H-89 (a PKA inhibitor, 10 sup -7, 10 sup -6 and 10 sup -5 M), [30,31]or 3, 4, 5-trimethoxybenzoic acid 8-(diethylamino) octyl ester hydrochloride (TMB-8, an inhibitor of Calcium2+ release from intracellular Calcium2+ store, 10 sup -5 M) [32,33]for 15 min. Control rings received no treatment for 15 min. The aortic rings were contracted again submaximally (approximately 70 to 80% of maximum contraction induced by norepinephrine 10 sup -5 M) with norepinephrine (10 sup -7 M), and cumulative concentration-effect curves to isoproterenol (10 sup -9 - 10 sup -6 M) and acetylcholine (10 sup -8 - 10 sup -5 M) were constructed. To examine the effects of halothane on isoproterenol and acetylcholine-induced relaxations, some rings were equilibrated with halothane (1, 2, or 3% v/v) for 15 min, and cumulative concentration-effect curves to isoproterenol (10 sup -9 - 10 sup -6 M) and acetylcholine (10 sup -8 - 10 sup -5 M), respectively, were constructed in the presence of halothane. To overcome the vasorelaxation effect of halothane, extra norepinephrine was added to produce the prehalothane level of contraction.
In the preliminary experiment, the responses of aortic rings to isoproterenol were compared between the first and the second trials. After the first construction of the concentration-effect curves of aortic rings with intact endothelia precontracted with norepinephrine (10 sup -7 M) to cumulatively applied isoproterenol, the rings were washed four times for 1 h. On the second trial, the norepinephrine-induced tensions were markedly attenuated and the relaxation responses of rings to isoproterenol were potentiated. This phenomenon was not observed in the rings without endothelia. Therefore, to assess the effects of treatments with drugs described above on the isoproterenol-induced relaxation correctly, eight aortic rings obtained from the same rat were randomly assigned to eight groups: control (without treatment); treatment with L-NNA; endothelial denudation; propranolol; W-7; 2', 3'-DDA; H-89; and TMB-8, respectively, and the first responses to isoproterenol were compared. To evaluate the effect of halothane, eight aortic rings from the same rat were also assigned to four groups: control and treatment with halothane 1, 2, and 3%, respectively, and the first responses to isoproterenol were compared. Because the response of aortic rings to acetylcholine were reproducible, the effects of four drugs (W-7; 2', 3'-DDA; H-89; and TMB-8) on acetylcholine-induced relaxation were examined between the first and second trials.
Some aortic rings, prepared and contracted as just described, were used for the cyclic nucleotide study. They were exposed to a single concentration of isoproterenol (3 x 10 sup -7 M) for 15 min, after which they were removed from the tissue baths and immersed immediately in liquid nitrogen. The following tissues were used: aortic rings with intact and no endothelia, those treated with and without L-NNA (5 x 10 sup -5 M), and those treated with and without halothane 2% v/v. In a preliminary study, incubation for 15 min was needed for the isoproterenol (3 x 10 sup -7 M) induced relaxations to reach a plateau. Finally, the cAMP and cGMP contents of the aortic rings were measured by radioimmunoassay (Yamasa cAMP and cGMP assay kits; Yamasa Shoyu Co., Chiba, Japan).
Halothane was delivered from a calibrated vaporizer (Fluotec 3; Cyprane Keighley, UK) to give appropriate concentrations in the gas mixture aerating the bathing fluid. The concentrations in the resulting gas mixture were measured and adjusted using an anesthetic gas monitor (Atom Co., Tokyo, Japan), which was calibrated using a standard halothane-calibrating gas mixture (Atom Co.). Concentrations of halothane in bathing fluid were confirmed by gas chromatography as previously described. [5]
Calculation and Statistical Analysis
The relaxations induced by isoproterenol and acetylcholine were expressed as percentages of the maximum relaxation induced by papaverine (10 sup -4 M). The data were expressed as means +/- standard error of the mean. Student's paired t test (two tailed) for the isometric tension and the cyclic nucleotides studies was used to determine whether the data differed significantly between groups. Differences at P < 0.05 were considered significant.
Drugs
The drugs used in this study were indomethacin; NG-L-nitro-arginine; 3, 4, 5-trimethoxybenzoic acid 8-(diethylamino) octyl ester hydrochloride; 2', 3'-dideoxyadenosine (Sigma Chemical Co., St. Louis, MO); acetylcholine (Dai-ichi Pharmaceutical Co., Osaka, Japan); papaverine hydrochloride (Dai-nippon Pharmaceutical Co.); DL-norepinephrine (Sankyo Pharmaceutical Co., Tokyo, Japan); DL-isoproterenol HCl (Nikken-kagaku Pharmaceutical Co., Tokyo, Japan); W-7; H-89 (Seikagaku-kogyo Co., Tokyo, Japan); and halothane (Takeda Pharmaceutical Co., Osaka, Japan). Indomethacin was dissolved in 99% v/v ethanol; W-7 and H-89 were dissolved in 50% v/v dimethyl sulfoxide solution, and all the other drugs were dissolved in distilled water.
Results
Norepinephrine (10 sup -7 M) increased the mean tension of rat aortic rings with and without endothelium to 1.41 +/- 0.24 and 1.63 +/- 0.12 g (n = 7 each), respectively. The difference between these developed tensions was not significant. Intact rings relaxed in response to acetylcholine (10 sup -6 M) by more than 60%, whereas endothelium-denuded rings displayed no such relaxation response. Isoproterenol (10 sup -9 - 10 sup -6 M) and acetylcholine (10 sup -8 - 10 sup -5 M) elicited concentration-dependent relaxations of rings with intact endothelia (Figure 1and Figure 3).
Effects of Endothelial Denudation, L-NNA, and Propranolol on Isoproterenol-induced Relaxation
Pretreatment with L-NNA (5 x 10 sup -5 M) and propranolol (5 x 10 sup -5 M) did not augment the tension increase induced by norepinephrine (10 sup -7 M; data not shown). Endothelial denudation shifted the concentration-relaxation response curve for isoproterenol to the right and upward significantly (P < 0.05 and 0.001; n = 7) and reduced the maximum response from 99 to 15% (Figure 1(a)). Treatment with L-NNA (5 x 10 sup -5 M) shifted the concentration-relaxation response curve for isoproterenol to the right and upward significantly (P < 0.05 or 0.001; n = 7) and reduced the maximum response from 99% to 26% (Figure 1(b)).
Treatment with propranolol (5 x 10 sup -5 M) shifted the isoproterenol-induced relaxation curve to the right significantly in a parallel manner (P < 0.05; n = 7; Figure 1(c)).
Effects of Endothelial Denudation and L-NNA on Isoproterenol-induced Accumulation of Cyclic Nucleotides
The cAMP and cGMP contents of rat aortic rings with intact endothelia (n = 5) contracted with norepinephrine (10 sup -7 M) were 256.6 +/- 21 and 88.6 +/- 12.9 fmol/mg wet tissue, respectively (Figure 2(a-1 and b-1)). Isoproterenol (3 x 10 sup -7 M) increased both the cAMP and cGMP contents significantly to 334.1 +/- 21.9 and 110.9 +/- 16.8 fmol/mg wet tissue, respectively (Figure 2(a-1 and b-1)). The cAMP and cGMP contents of L-NNA-treated endothelium-intact rings exposed to isoproterenol (n = 5) were 322.6 +/- 20.3 and 68.3 +/- 8.5 fmol/mg wet tissue, respectively (Figure 2(a-2 and b-2)). Treatment with L-NNA inhibited the isoproterenol-induced cGMP content increase significantly (P < 0.05), but not that of the cAMP content (Figure 2(a-2 and b-2)). The cAMP and cGMP contents of endothelium-free rings to isoproterenol cAMP and cGMP contents were 239.9 +/- 9.9 and 70.9 +/- 13.5 fmol/mg wet tissue, respectively (Figure 2(a-3 and b-3)). Endothelial denudation inhibited the isoproterenol-induced increases in both cyclic nucleotide contents significantly (P < 0.05).
Comparison of the Effects of 2', 3'-DDA, W-7, H-89, and TMB-8 on Isoproterenol- and Acetylcholine-induced Relaxation
Norepinephrine (10 sup -7 M)-induced tension was not affected significantly by treatment with 2', 3'-DDA (10 sup -5 and 10 sup -4 M), W-7 (10 sup -5 and 5 x 10 sup -5 M), H-89 (10 sup -7, 10 sup -6 M and 10 sup -5 M), or TMB-8 (10 sup -5 M).
Effects of 2', 3'-DDA on Isoproterenol- and Acetylcholine-induced Relaxation.
Adenylyl cyclase inhibition by 2', 3'-DDA (10 sup -5 and 10 sup -4 M) shifted the concentration-relaxation response curve for isoproterenol to the right and upward significantly (P < 0.05 and 0.001, respectively; n = 7 each) and reduced the maximum response from 99% to 78% and 24%, respectively (Figure 3(a)), whereas 2', 3'-DDA (10 sup -6 M) did not affect acetylcholine-induced relaxations (n = 7; Figure 3(b)).
Effects of W-7 on Isoproterenol- and Acetylcholine-induced Relaxation.
Calmodulin inhibition by W-7 (10 sup -5 and 5 x 10 sup -5 M) shifted the concentration-relaxation response curve for both isoproterenol (P < 0.05 and 0.001, respectively; n = 7 each; Figure 4(a)) and acetylcholine to the right and upward significantly (P <0.001; n = 7; Figure 4(b)) and reduced the respective maximum responses from 99% to 81% and 37% (isoproterenol) and from 80% to 35% (acetylcholine).
Effects of H-89 on Isoproterenol- and Acetylcholine-induced Relaxation.
Protein kinase inhibition by H-89 (10 sup -7, 10 sup -6, and 10 sup -5 M) significantly shifted the concentration-relaxation response curve for isoproterenol to the right and upward (P < 0.05 and 0.001, respectively; n = 7 each; Figure 5(a)) in a concentration-dependent manner and reduced the maximum responses from 99% to 93%, 81% and 59%, respectively, whereas H-89 (10 sup -6 M) did not affect acetylcholine-induced relaxations (n = 7 each; Figure 5(b)).
Effects of TMB-8 on Isoproterenol- and Acetylcholine-induced Relaxation.
Effect of Halothane on Isoproterenol- and Acetylcholine-induced Relaxation
Halothane (1%, 2%, and 3% v/v) treatment reduced the norepinephrine (10 sup -7 M)-induced tensions by approximately 16%, 20%, and 27%, respectively, of those in its absence. The addition of extra norepinephrine (1 to 2 x 10 sup -7 M) restored the tension levels. After restoring the tension levels, the effects of halothane on isoproterenol- and acetylcholine-induced relaxation were analyzed. Halothane, at all three concentrations, had no effect on the relaxation responses to isoproterenol (n = 7 each; Figure 7(a, b, and c)), whereas it significantly inhibited the responses to acetylcholine (n = 7 each; Figure 8(a, b, c)).
Effect of Halothane on Isoproterenol-induced Cyclic Nucleotides Accumulation
Halothane (2% v/v) treatment reduced the mean norepinephrine (10 sup -7 M)-induced tension of the rings used for cyclic nucleotide measurement by approximately 14% of those in its absence. The addition of extra norepinephrine (1 to 2 x 10 sup -7 M) restored the tensions. The amounts of isoproterenol (3 x 10 sup -7 M)-induced cAMP and cGMP accumulations in the absence and presence of halothane were 334.1 +/- 21.9 and 341.28 +/- 30.33 fmol/mg wet tissue, respectively (n = 5 each; Figure 9(a)), and 110.9 +/- 16.8 and 93.6 +/- 11.6 fmol/mg wet tissue, respectively (n = 5 each; Figure 9(B)).
Discussion
Our findings show that isoproterenol, in an endothelium-dependent manner, induced vasorelaxation of rat aortic rings that were completely inhibited by L-NNA, and that the cAMP and cGMP contents of endothelium-intact aortic rings precontracted with norepinephrine exposed to isoproterenol increased significantly, whereas no such increases were demonstrated in endothelium-denuded rings. These results confirm that isoproterenol-induced relaxations of rat aortic rings are confined to the mediation with the NO-cGMP relaxing mechanism, as shown previously. [14,34-38]
Isoproterenol, a beta-adrenoceptor agonist, has long been regarded as an endothelium-independent vasodilator, and its effects are mediated by a cAMP-dependent relaxation mechanism within VSM. [16-20]However, recent investigations showed that isoproterenol-induced relaxations of rat aorta were inhibited by oxyhemoglobin, [37]methylene blue, [37]and NG-1-nitro-arginine methyl ester, [14]suggesting that isoproterenol-induced rat aortic relaxation is attributed to NO released by the endothelium. Our results strongly support these findings. Furthermore, our study showed that isoproterenol-induced relaxation was inhibited markedly by the calmodulin inhibitor, W-7, indicating that isoproterenol-induced NO production is mediated via the activation of cNOS, of which the enzymatic activity depends on the Calcium2+ -calmodulin complex, [11-13]as it is well known to be in acetylcholine-induced NO production. [9,10]
The evidence that isoproterenol-induced cAMP accumulation was seen only in rings with intact endothelia suggests that isoproterenol induces cAMP accumulation in the endothelium, not in VSM cells. Although inhibition by cGMP of cAMP phosphodiesterase may be involved, [39]the evidence that 2', 3'-DDA, an inhibitor of adenylyl cyclase, markedly inhibited the isoproterenol-induced endothelium-dependent relaxation greatly supports the most likely explanation that the stimulation of endothelial adenylyl cyclase is involved in isoproterenol-induced cAMP production. A beta-adrenoceptor antagonist, propranolol, strongly and competitively inhibited the isoproterenol-inducedendothelium-dependent relaxation, suggesting that isoproterenol binds to endothelial beta-adrenoceptors to evoke the NO formation. That beta-adrenoceptors exist on arterial endothelium has already been demonstrated. [40-43]Because beta-adrenoceptors are known to be linked to adenylyl cyclase via the stimulating guanine nucleotide-binding protein, [44,45]the first stage of signal transduction in isoproterenol-induced cAMP production is probably the activation of endothelial beta-adrenoceptor-adenylyl cyclase coupling. Therefore, our data indicate that isoproterenol-induced cAMP production by the rat aorta is an event that occurs within endothelium, where beta-adrenoceptor-adenylyl cyclase coupling initiates the signal transduction pathway that mediates cAMP production.
The role of cAMP as a second messenger for receptor-operated intracellular signal transduction is to be involved in relaxation of the VSM cells of other arteries: cAMP activates cAMP-dependent PKA, which subsequently activates myosin light chain kinase, resulting in VSM relaxation. Our study showed that a PKA inhibitor, H-89, markedly inhibited the isoproterenol-induced endothelium-dependent relaxation, indicating that PKA activation is involved in it. Recently, MacDonell and Diamond [46]showed that the application of a membrane-permeable cAMP, 8-bromo cAMP, to rat aortic rings without endothelia actually stimulated PKA activity but failed to induce VSM relaxation, indicating that, in conjunction with concomitant reports, [47-49]protein kinase activation within rat aortic VSM cells does not correlate with relaxation. Therefore, isoproterenol appears to activate PKA via cAMP production within the endothelium, and this is responsible for VSM relaxation. Protein kinase has been shown to modulate intracellular Calcium2+ -dynamics, [25,26]and influx of Calcium2+ into the cell is one of the primary means whereby it does so. [25,26]It is, however, unfortunate that the particular channel responsible for this type of Calcium2+ entry has yet to be isolated and defined in detail, and other specific blockers are not known.
Because isoproterenol-induced NO production is mediated by cNOS, in which the Calcium2+ -calmodulin complex is necessary for NO production, isoproterenol probably evokes increases in the Calcium sup 2+ within the endothelium. It is generally accepted that acetylcholine induces increases in the vascular endothelial Calcium2+ levels, thereby activating cNOS through two mechanisms, Calcium2+ release and subsequent Calcium2+ influx. There are at least four pharmacologically defined muscarinic receptor subtypes in primary tissues, and five subtypes, (m1, m2, m3, m4, and m sub 5) have been cloned. [50]Muscarinic receptors mediate their effects by activating G-proteins, which in turn may directly modulate an ion channel or modulate an enzyme involved in synthesis of second messengers. Examples of second messenger modulation by muscarinic receptors include inhibition of adenylyl cyclase by m2and m4, resulting in reduced cAMP levels within the cell. m1, m3, and m5stimulate phospholipase C, which hydrolyzes phosphatidyl inositol to various inositol phosphates, such as inositol triphosphate and diacylglycerol. [50]Studies using selective antagonists for muscarinic receptor subtypes revealed that m1and m3, but not m2, are actively involved in acetylcholine-induced endothelium-dependent relaxation of rat aorta, indicating that the major action of acetylcholine on rat aorta is to initiate inositol triphosphate generation in their endothelia. [51,52]Inositol triphosphate binds to its own receptors on intracellular Calcium2+ stores and releases Calcium2+. [53]Recently researchers showed that this inositol triphosphate-induced Calcium2+ release depleted the endoplasmic reticular Calcium2+ store, and this depletion facilitated Calcium2+ influx, resulting in a sustained increase in the intracellular Calcium2+ levels (Calcium2+ depletion-induced Calcium2+ influx). [54]Such Calcium2+ depletion-induced Calcium2+ influx was also demonstrated in other nonexcitable cells, such as platelets, [55]thymocytes, [56]and neutrophils. [57]To examine the contribution of "Calcium2+ release" to isoproterenol-elicited NO production, the effects of TMB-8 on acetylcholine- and isoproterenol-induced relaxation were tested. TMB-8 has been called an "intracellular Calcium2+ antagonist" and is characterized as a strong inhibitor of Calcium2+ release from intracellular Calcium2+ stores. [31,32]At 10 sup -5 M, TMB-8 completely blocked Calcium2+ release from intracellular Calcium2+ stores of guinea pig coronary arterial cultured endothelial cells. [32]We found that TMB-8 (10 sup -5 M) abolished acetylcholine-induced relaxations, whereas it did not affect those induced by isoproterenol. These findings indicate that "Calcium2+ release from intracellular Calcium2+ stores" is essential for the acetylcholine-induced, but not for the isoproterenol-induced NO production. Therefore, these data suggest strongly that isoproterenol-induced cNOS activation is attributable to stimulation of the Calcium2+ influx mechanism, possibly evoked by PKA. Figure 10summarizes our proposed signal transduction pathway for mediating endothelial isoproterenol-induced NO-production, compared with that induced by acetylcholine. Considered with all the evidence obtained in the current study, we came to the following conclusions regarding isoproterenol-induced rat aortic relaxation: (1) Isoproterenol acts on endothelial beta-adrenoceptors to evoke cAMP production, which is followed by PKA activation within the endothelium; (2) activation of cAMP-PKA pathway potentiates Calcium2+ -calmodulin-dependent NOS activity to evoke NO-cGMP-mediated relaxation, as occurs in response to acetylcholine; (3) Calcium2+ release from endothelial Calcium2+ store plays an important role in cNOS activation in acetylcholine-induced, but not in isoproterenol-induced NO production; (4) the cAMP-PKA line probably contributes to cNOS activation via an intracellular Calcium2+ -increase mechanism, possibly sustained Calcium2+ influx, rather than Calcium2+ release. The findings of Toms and Roberts [58]strongly support this last conclusion. They showed that N-methyl-D-aspartate receptor-mediated NO formation in rat cerebellum is regulated by the cAMP-PKA pathway.
It has been generally accepted that halothane inhibits the muscarinic or bradykinin receptor-operated endothelium-dependent vasorelaxation mediated via the NO-cGMP relaxation mechanism. However, our study showed that halothane did not affect the isoproterenol-induced NO-cGMP relaxation mechanism. The inhibitory effects of halothane on the NO-cGMP pathway stimulated by these agonists have been suggested to be attributable to events occurring in the multifarious sites: impairment of agonist-receptor coupling, impairment of endothelial Calcium2+ mobilization, inactivation of other cofactors for cNOS activation (i.e., calmodulin), direct inhibition of NOS, inactivation of NO, and inhibition of sGC within VSM cells (reviewed by Johns RA). [59]However, the major site of its inhibitory action, whether within endothelium or VSM, remains controversial. Our study yielded three key findings that may identify the major site of action of halothane in the NO-mediated relaxation mechanism evoked by acetylcholine: (1) Both acetylcholine- and isoproterenol-induced cNOS activation were followed by activation of the cGMP relaxation pathway; (2) Calcium2+ release from endothelial Calcium2+ stores is necessary to activate cNOS in acetylcholine-induced, but not in isoproterenol-induced relaxation; and (3) halothane did not inhibit the NO-cGMP relaxation mechanism evoked by isoproterenol. These findings suggest strongly that the major target site for the inhibitory action of halothane on acetylcholine-induced relaxation is not the NO-cGMP relaxation pathway within VSM but rather the pathway from muscarinic receptor stimulation to cNOS activation within the endothelium.
A signal-transduction system mediating the NO-cGMP pathway elicited by isoproterenol in the rat aorta was investigated: beta-adrenoceptor-adenylyl cyclase coupling-cAMP production-PKA activation-accelerated Calcium2+ influx (independent of Calcium2+ release from intracellular Calcium2+ store)-activation of cNOS-NO production-cGMP-mediated VSM relaxation. The hypothesis that this agonist-induced NO formation is distinct from that induced by acetylcholine, with respect to Calcium2+ dynamics, was confirmed as reasonable. Furthermore, halothane was shown not to inhibit this relaxation mechanism. These results suggest that halothane inhibits the NO-cGMP relaxation mechanism only when cNOS is activated by Calcium2+ release from intracellular Calcium2+ stores. Our findings should provide further insight into the effects of volatile anesthetics on the mechanisms responsible for Calcium2+ release from Calcium2+ storing organelles and subsequent cellular reactions.