Halothane, Enflurane, and Isoflurane Do Not Affect the Basal or Agonist-stimulated Activity of Partially Isolated Soluble and Particulate Guanylyl Cyclases of Rat Brain

This research protocol was approved by the Research and Animal Welfare Committee of the University of Virginia.

NO solutions were freshly prepared immediately before use, under anaerobic conditions by the method of Rengasamy and Johns. [25]2 ml deoxygenated 50 mM tris(hydroxymethyl)aminomethane (Tris)-HCl (pH = 7.4) contained in a 7-ml Vacutainer tube (Becton Dickinson, Rutherford, NJ) at room temperature was saturated with NO by bubbling 50 ml pure NO gas, which had passed through a column of KOH pellets. This produced an NO stock solution of 1 mM as determined by the chemiluminescence method as described previously. [26]Serial dilutions were made from this stock solution by removing aliquots with gas-tight syringes. These aliquots were added to deoxygenated 50 mM Tris-HCl (pH = 7.4) solution.

Whole brain (including both cerebrum and cerebellum) and olfactory bulb were removed from halothane anesthetized Sprague-Dawley rats (250-300 g) after extensive in situ perfusion with ice-cold heparinized normal saline (2,000 U/l saline, 80 ml/rat cardiac perfusion with right atrial transection) to minimize residual hemoglobin. Whole brain and olfactory bulb were used as the source of soluble and particulate fraction of guanylyl cyclases, respectively, because whole brain is rich in soluble guanylyl cyclase [27]and olfactory bulb contains high quantities of atrial natriuretic peptide sub (1-28) (ANP sub (1-28))-sensitive particulate guanylyl cyclase, [28]which can bind to ANP sub (1-28) when partially isolated. [29]Brain and olfactory bulb were washed in three changes of ice-cold homogenization buffer containing 50 mM Tris-HCl (pH = 7.4), 250 mM sucrose, and 0.2 mM benzamidine. They were then homogenized in ice-cold homogenization buffer with three Polytron homogenizer bursts of 30 s. The homogenate was centrifuged at 1,000g for 10 min at 2 degrees Celsius. The pellet was discarded and the supernatant was then centrifuged again at 105,000g for 60 min at 4 degrees Celsius. The supernatant was harvested and used in the soluble guanylyl cyclase assay. The pellet was washed three times with homogenization buffer and then stored in this buffer at -80 degrees Celsius until it was used in particulate guanylyl cyclase measurements. [30]Protein concentrations were measured by the Bio-Rad (Hercules, CA) protein assay method. [31].

The reaction mixture solution (250 micro liter) contained in a 7-ml scintillation glass vial without a cap was gassed with humidified carrier gas (95% O², 5% CO²) in the presence or absence of inhalational anesthetics. Carrier gas was delivered at 3 l/min through or not through a halothane, enflurane, or isoflurane vaporizer and subsequently passed through a warmed (37 degrees Celsius) humidification chamber. Gas entry into each glass vial was valve regulated and maintained at 50 ml/min. Because of the humidification of the gas, no significant evaporation of the reaction mixture solution occurred during the course of the experiment. Reaction mixture solution was equilibrated with anesthetics for 5 min before the addition of enzyme (in a 10-micro liter volume) to start the reaction. Preliminary experiments were designed to determine the time course of equilibration of anesthetics in the reaction mixture solution. The concentration of anesthetics in the solution was measured by using standard gas chromatography methods. [9]The concentration of anesthetics in the carrier gas after passing through the vaporizer was also confirmed by the same methods. [9].

The guanylyl cyclase activity assay was performed as described previously. [30,32]The reaction mixture solution was prepared in a final volume of 250 micro liter containing 50 mM Tris-HCl (pH 7.4), 1 mM isobutylmethylxanthine, 4 mM MnCl², 0.5 mM adenosine triphosphate (disodium salt, grade I: from yeast), 1 mM guanosine triphosphate (sodium salt, type III), 15 mM creatine phosphate, and 100 micro gram creatine phosphokinase. The reaction mixture solution was gassed with carrier gas in the presence or absence of anesthetics for 5 min in a 37 degrees Celsius water bath before the addition of enzyme (15-25 micro gram protein in 10 micro liter volume) to initiate the reaction. Rat ANP sub (1-28) (in 10 micro liter) was added just before the addition of enzyme. NO or SNP stock solution (in 10 micro liter) was added immediately after the addition of enzyme to the reaction mixture. The reaction continued at 37 degrees Celsius with or without guanylyl cyclase activating agents with continuous gassing by carrier gas in the presence or absence of inhalational anesthetics. The reaction was terminated 5 min later by adding 250 micro liter ice-cold 0.2 N HCl. cGMP concentrations were quantified by radioimmunoassay as previously described. [32,33]Preliminary experiments were carried out to determine the time course of guanylyl cyclase activity in the presence or absence of stimulators.

Results are presented as mean plus/minus standard error of the mean. Each data point represents a mean of 2-4 experiments with a n (number of repetitions) > 6. Statistical analysis was performed by one-way analysis of variance with multiple-range testing (the Newman-Keuls test) when the effects of inhalational anesthetics on the basal or agonist-stimulated guanylyl cyclase activity were evaluated. Comparisons of guanylyl cyclase activity in the presence or absence of NO, SNP, or rat ANP sub (1-28) and comparisons of soluble guanylyl cyclase activity at each dose of NO in the presence and absence of anesthetics were made by paired t test. P < 0.05 was accepted as significant.

Adenosine triphosphate, isobutylmethylxanthine, Tris-HCl, MnCl², guanosine triphosphate, creatine phosphate, creatine phosphokinase, SNP, sucrose, and benzamidine were obtained from Sigma Chemical (St. Louis, MO). Rat ANP sub (1-28) was obtained from Peninsula Laboratories (Belmont, CA). Halothane was obtained from Halocarbon Laboratories (Hackensack, NJ), enflurane from Airco (Madison, WI), and isoflurane from Ohmeda Caribe (Liberty Corner, NJ). NO gas was obtained from Liquid Carbonic (Chicago, IL) and 95% O², 5% CO²gas from Roberts Oxygen (Waynesboro, VA).

The concentrations of halothane, enflurane, and isoflurane in the reaction mixture solution reached a plateau by 4-5 min over the concentration range of inhalational anesthetics used in this study (Figure 1). The final concentrations reached in the solution were within 10.1 plus/minus 8.0% of the predicted value, calculated as previously described. [33-35]The concentrations of halothane, enflurane, and isoflurane in the carrier gas were confirmed by gas chromatography to be accurate to within -1.9 plus/minus 4.7% of the concentrations delivered by the vaporizer.

cGMP concentrations produced by soluble guanylyl cyclase progressively increased to 5 min of incubation time; thereafter the cGMP contents in the reaction mixture solution were stable throughout 30 min incubation. The presence of 0.1 mM of SNP increased the cGMP content at each incubation time but did not alter the time course pattern (Figure 2(A)).

Soluble guanylyl cyclase activity was significantly stimulated after exposure to 0.1 mM SNP for 5 min. However 0.44-4.4% halothane, or 1.32-6.7% enflurane, or 0.6-5.0% isoflurane did not significantly affect either the basal or 0.1 mM SNP-stimulated guanylyl cyclase activity (Figure 3).

NO (10 sup -9 - 10 sup -6 M) produced a concentration-dependent increase of soluble guanylyl cyclase activity; this increase was not affected by 4.4% halothane, 6.7% enflurane, or 5.0% isoflurane (Figure 4). Moreover, the soluble guanylyl cyclase activity stimulated by 1 micro Meter of NO was not inhibited by 0.44-4.4% halothane, or 1.32-6.7% enflurane, or 0.6-5.0% isoflurane (Figure 5).

The time course of cGMP production of particulate guanylyl cyclase in the presence or absence of 1 micro Meter rat ANP sub (1-28), an agonist of natriuretic peptide receptor A, [36]progressively increased up to the incubation time of 30 min. One micromolar rat ANP sub (1-28) significantly increased the cGMP concentrations in the reaction mixture solution after 5 min of incubation time (Figure 2(B) and Figure 6).

Both basal and 1 micro Meter of rat ANP sub (1-28) -stimulated particulate guanylyl cyclase activity were not significantly affected by 0.44-4.4% halothane, or 1.32-6.7% enflurane, or 0.6-5.0% isoflurane (Figure 7). To confirm these results, the time course experiments investigating particulate guanylyl cyclase activity with or without 1 micro Meter rat ANP sub (1-28) were performed in the presence or absence of 4.4% halothane. Halothane did not affect either basal or 1 micro Meter rat ANP sub (1-28) -stimulated cGMP formation produced by particulate guanylyl cyclase (Figure 6).

EDRF/NO is now recognized as the transduction mechanism for the activation of soluble guanylyl cyclase. This NO-guanylyl cyclase pathway has been demonstrated in a wide variety of tissue types, including brain. [2,5,6]The interaction between inhalational anesthetics and the NO-guanylyl cyclase signaling pathway has been dynamically studied and it is generally accepted that inhalational anesthetics inhibit the NO-guanylyl cyclase pathway, although the actual site of this inhibition has been highly controversial. [7-10,20-24,37].

We used partially isolated guanylyl cyclases from rat brain to specifically address the question of whether inhalational anesthetics directly affect particulate or soluble guanylyl cyclase activity. The use of partially isolated enzymes in the current study allows us to clearly distinguish the activity of soluble guanylyl cyclase from that of particulate guanylyl cyclase and also to avoid many of the potentially complicating factors associated with cultured cells or intact freshly isolated cells because multiple factors inside cells, such as oxidation-reduction state and endogenous inhibitors, may affect guanylyl cyclase activity. [38-40]We investigated the changes of both basal and agonist-stimulated guanylyl cyclase activity in the presence or absence of inhalational anesthetics. Wide-ranging concentrations of halothane, enflurane, and isoflurane did not stimulate or inhibit basal soluble or particulate guanylyl cyclase activity, suggesting that these three inhalational anesthetics do not influence cGMP production through either of these two types of guanylyl cyclases. The same concentrations of halothane, enflurane, and isoflurane also did not affect the SNP- or NO-stimulated soluble guanylyl cyclase activity nor the rat ANP sub (1-28) -stimulated particulate guanylyl cyclase activity. However, there was significant interassay variation of apparent guanylyl cyclase activity, probably the result of the use of different batches of enzyme preparation with different concentrations of guanylyl cyclase for the different experiments.

Initial studies by several groups [20-22]using arterial rings suggested that the site of inhibition of inhalational anesthetics on the NO-guanylyl cyclase signaling pathway may be at the synthesis, release, or transport of EDRF/NO and may not be at the point of guanylyl cyclase activation. The results supporting this conclusion were that acetylcholine- and bradykinin-induced vessel relaxation, which is receptor mediated and NO dependent, was inhibited by inhalational anesthetics, whereas nitroglycerin- and SNP-induced relaxation, which is caused by the direct activation of soluble guanylyl cyclase, was not affected. [9,20,21]However, recently Hart and colleagues [7]demonstrated that halothane (2 MAC) significantly attenuated NO-induced vessel relaxation (NO concentrations to 1 micro Meter) and low concentrations (to 3 x 10 sup -8 M) of nitroglycerin-induced relaxation of rat aortic rings. Isoflurane, on the other hand, was unable to inhibit NO-induced relaxation.* Furthermore, NO-stimulated cGMP content (NO concentrations to 5 x 10 sup -6 M) was also significantly attenuated by halothane. These results suggest that halothane may affect the activation of guanylyl cyclase by NO. One study by Su and Zhang, [41]which was cited by Hart et al. [7]to support their suggestion, also showed that halothane increased the tension of rabbit intact aortic rings relaxed by 10 sup -7 M SNP, although in the Su and Zhang study there was no indication that the increased tension of these rings was attributable to the inhibitory effect of halothane on guanylyl cyclase activity stimulated by SNP. Also, a study by Nakamura and associates [8]demonstrated that SNP-induced relaxation of rat aorta was reduced by halothane (only at 10 sup -8 M but not lower or higher concentrations of SNP) and that the increase in cGMP of aorta stimulated by NO or SNP was inhibited by halothane, although similar inhibitions did not occur using isoflurane or sevoflurane. These studies suggest that the interaction of NO and soluble guanylyl cyclase may be a site of inhibition of the NO-guanylyl cyclase signaling pathway by halothane but not by the other inhalational anesthetics studied. However, this suggestion was not supported by other recent studies, [9,23,24]which are more consistent with the early studies by Muldoon et al. [20]and Toda et al., [21]which failed to show an effect of halothane on nitrovasodilator-induced vasodilation.

The intact artery ring may be too complicated to define and evaluate clearly the effects of inhalational anesthetics on specific sites of the NO-guanylyl cyclase signaling pathway because, for example, endothelial cells can release several other vasodilators and vasoconstrictors such as prostacyclin, endothelin, endothelium-derived hyperpolarizing factor [42]as well as C-type natriuretic peptides, which may also activate guanylyl cyclase. [43]Therefore we chose to use the partially isolated guanylyl cyclase preparation to clarify the interaction of inhalational anesthetics and guanylyl cyclase.

Our results, using guanylyl cyclases partially isolated from rat brain, clearly show that inhibition of the NO-guanylyl cyclase pathway by inhalational anesthetics occurs proximal to the activation of both soluble and particulate guanylyl cyclases. In agreement with our study, Eskinder et al. [23]studied the effect of halothane on partially isolated soluble and particulate guanylyl cyclase from canine cerebral arteries and found that 1 mM halothane did not affect the basal soluble guanylyl cyclase activity. In addition, a preliminary study by Van Dyke and colleagues [44]showed that halothane had no effect on basal activity of soluble guanylyl cyclase from rat liver. However, the latter study did show that halothane and isoflurane dose-dependently inhibited the NO-stimulated soluble guanylyl cyclase activity, which disagreed with our results. The reason for this discrepancy is not known; however, they did not include any phosphodiesterase inhibitor in their reaction mixture and they did not add guanosine triphosphate-regenerating system in their reaction mixture, which makes their results difficult to judge.

Although it has been well demonstrated that inhalational anesthetics inhibit the NO-guanylyl cyclase pathway and therefore decrease the cGMP concentrations in tissues, it has also been reported that halothane increases cGMP concentrations in mouse myocardium [45]and cortex, [46]in rat aortas, [47]and in canine cerebral arteries. [23]In support of this, Eskinder et al. [23]did demonstrate that a single concentration (3.3%) of halothane stimulated basal particulate guanylyl cyclase activity in their partially isolated enzyme system. However, using a similar system, we could not demonstrate any direct inhibitory or stimulatory effects of inhalational anesthetics on basal or agonist-stimulated particulate guanylyl cyclase activity. The reason for this difference is not apparent, although our work examined a wide concentration range for halothane, enflurane, and isoflurane as well as time dependence of the cyclase assays, and Eskinder et al. [23]studied a single halothane concentration and did not provide any time course data for their enzyme preparations. The disagreement between their findings and ours may also reflect species differences in the action of inhalational anesthetics, although cyclase preparations from a variety of tissue beds and species behave quite similarly and the cyclase amino acid sequence is highly homologous across species. [48].

Blaise et al., [24]by using rabbit abdominal aorta and cultured bovine aortic endothelial cells, concluded that halothane may modify either EDRF/NO half-life or its activated oxidation-reduction form, as a means of inhibiting the NO-guanylyl cyclase signaling pathway. However, our data do not agree with Blaise et al.'s conclusion because wide-ranging concentrations of halothane, enflurane, or isoflurane did not inhibit the stimulatory effects of 1 micro Meter of NO on soluble guanylyl cyclase activity and the guanylyl cyclase activity stimulated by different doses of NO was not affected by halothane, enflurane, or isoflurane either. It has been suggested that inhalational anesthetics can affect arterial ring tone by other mechanisms, such as the Calcium²+ -dependent Potassium sup + channel, [49,50]apart from the NO-guanylyl cyclase signaling pathway. Therefore it is difficult to isolate the mechanisms that contributed to the effect of halothane on vessel tone in Blaise et al.'s [24]experiments. It is also important to note that the vessel tension in the presence of halothane was always higher than that in the absence of halothane in Blaise et al.'s experiment, no matter whether it was in the presence or absence of NO.

In summary, various doses of halothane, enflurane, and isoflurane did not affect the basal or agonist-stimulated soluble or particulate guanylyl cyclase activity. These results suggest that the activation of guanylyl cyclase by agonists and guanylyl cyclase itself are not the sites of the inhibition of the NO-guanylyl cyclase signaling pathway by inhalational anesthetics. In agreement with the previous studies from our laboratory, [9,33]inhalational anesthetics appear to inhibit the NO-guanylyl cyclase signaling pathway proximal to activation of guanylyl cyclase.

The authors thank T. N. Pajewski, MD, Ph.D., for a critical reading of the manuscript.

* Muldoon SM: Personal communication. 1994.