Influence of Volatile Anesthetics on Thromboxane A²Signaling 

The study protocol was approved by the Animal Research Committee at the University of Virginia. Female Xenopus laevis toads were obtained from Xenopus I (Ann Arbor, MI), housed in an established frog colony, and fed regular frog brittle twice weekly. To remove the oocytes, a frog was anesthetized by immersion in 0.2% 3-amino-benzoic-methyl-ester until it was unresponsive to a painful stimulus (toe pinching). Animals were operated while positioned on ice. A 1-cm-long abdominal incision was made and a lobule of ovarian tissue, containing approximately 200 oocytes, was removed and placed in modified Barth's solution (containing 88 mM NaCl, 1 mM KCl, 2.4 mM NaHCO³, 0.41 mM CaCl², 0.82 mM MgSO⁴, 0.3 mM Ca²NO³, 0.1 mM gentamicin, and 15 mM HEPES, pH adjusted to 7.6). The wound was closed in two layers and the frog was allowed to recover from anesthesia. The oocytes were defolliculated by gentle shaking in a 1 mg/ml solution of collagenase type Ia in calcium-free OR2 solution (containing 82.5 mM NaCl, 2 mM KCl, 1 mM MgCl², 5 mM HEPES, pH adjusted to 7.5) for 2 h. After this process, the oocytes were returned to modified Barth's solution. Microscopic observation confirmed the absence of follicle cells.

The rat TXA²receptor clone was obtained from Dr. K. R. Lynch (Department of Pharmacology, University of Virginia, Charlottesville, VA) as a cDNA encoding a 343 amino acid protein in the pcDNAI vector (Invitrogen, San Diego, CA). The construct was linearized with the nuclease Xho I and transcribed in the presence of capping analog by T7 polymerase, using a commercial RNA preparation kit (mMessage mMachine; Ambion, Austin, TX). Oocytes were injected with 5 ng mRNA in 30.6 nl sterile water, using an automated microinjector (Nanoject; Drummond Scientific, Broomall, PA). Microinjector functioning was confirmed by noting a slight increase in cell size during injection. The cells were then cultured in modified Barth's solution for 72 h before study.

A single defolliculated oocyte was placed in a continuous-flow recording chamber (0.5 ml volume), perfused (3 ml/min) with Tyrode's solution (containing 150 mM NaCl, 5 mM KCl, 2 mM CaCl², 1 mM MgSO⁴, 10 mM dextrose, 10 mM HEPES, pH adjusted to 7.4). Microelectrodes were pulled in one stage from capillary glass (BBL with fiber; World Precision Instruments, Sarasota, FL) on a micropipette puller (model 700 C; David Kopf Instruments, Tujunga, CA). Electrode tips were broken to a diameter of approximately 10 micro meter, providing a resistance of 1–3 M Omega, and filled with 3 M KCl. The cell was voltage clamped using a two-microelectrode oocyte voltage-clamp amplifier (OC725A; Warner Corp., New Haven, CT), connected to an IBM-compatible computer for data acquisition (DAS-8A/D conversion board: Keithley-Metrabyte, Traunton, MA) and analysis (OoClamp software [8]). All measurements were performed at a holding potential of -70 mV. Only cells exhibiting stable holding currents < 1 micro A during a 1-min equilibration period were included in the analysis. Membrane current was sampled at 125 Hz and recorded for 5 s before and 55 s after administration of the agonist, which allowed sufficient time for currents to return to baseline levels. Agonist (the TXA²mimetic U-46619) was delivered as a 30-micro liter aliquot during a period of 1–2 s using a hand-held micropipettor positioned approximately 3 mm from the oocyte. Agonist binding to TXA²receptors activates one or more heterotrimeric G proteins, which in turn modulate the enzyme-regulated synthesis of second messengers (Figure 1(A)). Specifically, the G^qprotein activates phospholipase C-beta (PLC-beta), which cleaves membrane phosphatidylinositolbisphosphate to inositol-1–4-5-trisphosphate (IP³) and diacylglycerol. The IP³activates receptor-channel complexes on intracellular Ca²+ stores, resulting in an increase in intracellular Ca²+ concentration. In the Xenopus oocyte, intracellular Ca²+ opens endogenous membrane Cl sup - channels, resulting in a Cl sup - flux (I^Cl(Ca)) measured conveniently using the voltage-clamp technique (Figure 1). Responses were quantified by integrating the current trace (Figure 1(B)) and are reported in microcoulombs (micro C). All experiments were performed at room temperature.

To study IP³- and guanosine 5'-0-(2thiodiphosphate)(GTP gamma S)-induced I^Cl(Ca), a third micropipette was inserted into the voltage-clamped oocyte. Tips of intracellular micropipettes were beveled with a microgrinder (Narishige EG-6 Glass Electrode microgrinder; Narishige Instrument Laboratories, Tokyo, Japan). The pipette was connected to an automated microinjector (Nanoject, Drummond Scientific). Under voltage clamp, 30 nl of 2 mM IP³or 100 mM GTP gamma S was injected, thereby activating the signaling pathway at the IP sub 3 receptor or the G protein, respectively (Figure 1(A)). Because the volume of an average Xenopus oocyte is approximately 500 nl, the injected volume approximated 5% of the oocyte volume. Therefore the estimated final concentrations were IP³100 micro Meter or GTP gamma S 5 mM. These concentrations were chosen to result in I^Cl(Ca) similar in size to those induced by TXA²agonist at its EC⁵⁰. Induced I^Cl(Ca) were recorded 5 s before and 55 s after intracellular injection and analyzed as described before.

To confirm that I^Cl(Ca) induced by TXA²indeed resulted from IP³signaling, we injected the IP³receptor antagonist heparin (MW, 3,000 g/mol). An automated microinjector was used (Nanoject, Drummond Scientific). Oocytes were injected with 50 nl heparin (2 ng/nl; approximate final concentration, 0.2 ng/nl) at least 30 min before agonist application.

To determine the effects of halothane, isoflurane, and sevoflurane on I^Cl(Ca) induced by U-46619 or intracellular mediators, anesthetic was bubbled through a reservoir filled with 40 ml Tyrode's solution for at least 10 min. Air at a flow rate of 500 ml/min was used as the carrier gas. After equilibration, the solution was perfused through the recording chamber, superfusing the oocyte at a flow rate of approximately 3 ml/min; measurements were obtained after 10 bath volumes had been exchanged. Anesthetic concentrations in the recording chamber were quantified by gas chromatography (Aerograph 940; Varian Analytical instruments, Walnut Creek, CA). Results were converted to concentration in liquid using partition coefficients in Tyrode's solution at 22 [degree sign] Celsius (halothane lambda = 1.31, isoflurane lambda = 1.08, sevoflurane lambda = 0.39), and to corresponding partial pressure (vol%) at room temperature. [9,10]Aqueous concentrations equivalent to 1 minimum alveolar concentration anesthetic in air were 0.43 mM for halothane, 0.44 mM for isoflurane, and 0.44 mM for sevoflurane. Although the halothane minimum alveolar concentration-equivalent concentration is approximately twice as high as that published by Franks and Lieb, [11]our values are similar to those reported by other investigators. [2,3,6,12]Each oocyte was exposed to a single concentration of anesthetic only.

Results are reported as means +/- SEM. Differences among treatment groups were analyzed using the Student's t test or the Mann-Whitney U test. If multiple comparisons were made, data were analyzed using analysis of variance followed by a t test corrected for multiple comparisons (Bonferroni). P < 0.05 was considered significant. Concentration-response curves were fit to the following logistic function, derived from the Hill equation:Equation 1where y^maxand y^minare the maximum and minimum responses obtained, respectively; n is the Hill coefficient; and X⁵⁰is the concentration at which the half-maximal response occurs (EC⁵⁰for agonist, IC⁵⁰for anesthetics).

The TXA²receptor agonist U-46619 (5-heptenoic acid, 7[[6-(3-hydroxy-1-octenyl)-2-oxabicyclo[2.2.1]-hept-5-yl]) was obtained from Cayman Chemical (Ann Arbor, MI) and diluted in 0.1% fatty acid-free bovine serum albumin (ICN Pharmaceuticals, Costa Mesa, CA) in Tyrode's solution to appropriate concentrations. The TXA²receptor antagonist Bay U 3405 ((3R)-3-(4-fluorophenylsulfonamido)-1,2,3,4 tetrahydro [4 alpha,4b-3H]-carbazolepropanoic acid) was a gift from Bayer AG (Wuppertal, Germany) and was dissolved in the same manner. Halothane was from Halocarbon Laboratories (River Edge, NJ), sevoflurane was from Abbott International (Abbott Park, IL), and isoflurane was from Ohmeda (Liberty Corner, NJ). All other chemicals were from Sigma Chemical Company (St. Louis, MO).

Oocytes injected with 5 ng mRNA encoding the TXA²receptor responded to U-46619 with transient inward currents. The current developed after a delay of 3–5 s and consisted of a fast inward component followed by a fluctuating relaxation over several seconds (Figure 1(B)). These responses are typical for I^Cl(Ca) induced by Ca²+-signaling G protein-coupled receptors expressed in Xenopus oocytes. [3]The responses were concentration dependent (Figure 1(C)). Curve fitting using the Hill equation revealed a half-maximal effect concentration (EC⁵⁰) of 3.2 x 10 sup -7 +/- 1.1 x 10 sup -7 M and a Hill coefficient of 0.8 +/- 0.2.

To confirm that the I^Cl(Ca) induced by U-46619 is indeed mediated by recombinantly expressed TXA²receptors, we applied the agonist (1) to uninjected cells to determine if endogenous TXA²receptors were present and (2) to cells expressing the TXA²receptor, 1–2 min after application of the specific TXA²receptor antagonist Bay U 3405 (10 sup -5 M). U-46619 (10 sup -4 M)(had no effect in uninjected cells (data not shown). Treatment with Bay U 3405 suppressed the responses to U-46619 (10 sup -6 M) to 15% of control (Figure 2). Together these findings confirm that the agonist indeed signaled through recombinantly expressed TXA²receptors.

To determine if the expressed TXA²receptors signaled through IP³release (Figure 1(A)), we injected 100 ng heparin into oocytes expressing the TXA²receptor. Heparin selectively blocks IP³receptors. [13]U-46619 (10 sup -6 M), applied 30 min later, induced only minute currents in heparin-injected cells (Figure 3): Responses were suppressed to 10% of control. In contrast, control injections with sterile water or KCl (150 mM) did not affect TXA²signaling (data not shown). Thus intracellular Ca²+ release after agonist application is mediated almost completely by IP³.

After confirming that the receptor was functional, that the agonist was acting selectively on expressed receptors, and that the intracellular signaling pathway was as described for this receptor (Figure 1(A)), we tested the ability of halothane to interfere with TXA²signaling. Halothane at clinically relevant concentrations inhibited I^Cl(Ca) induced by U-46619 (10 sup -6 M)(Figure 4(A)). Curve fitting using the Hill equation revealed a half-maximal inhibitory concentration (IC⁵⁰) of 0.47 +/- 0.05 mM (0.83%) and a Hill coefficient of 1.6 +/- 0.2 (Figure 4(B)). The highest halothane concentration tested, 1.07 mM (2.5%), suppressed signaling to 17% of control. To determine if the effect of halothane was reversible, we studied U-46619-induced currents under three conditions: in the absence and presence of halothane and after washout of halothane with anesthetic-free solution (Figure 4(C)). All three measurements were performed in the same oocyte, separated by washes of at least 10 min. A high concentration of halothane (1.07 mM, 2.5%) was used because it was considered more likely to cause irreversible effects. Although U-46619-induced responses were depressed more than 80% in the presence of anesthetic, I^Cl(Ca) obtained in response to U-46619 after wash with anesthetic-free solution were similar in size to control responses (Figure 4(C)), indicating that halothane's effect is reversible.

Next we investigated isoflurane's effect on TXA²signaling. Isoflurane, at clinically relevant concentrations, inhibited U-46619 (10 sup -6 M)-induced I^Cl(Ca)(Figure 5(A)). Curve fitting revealed an IC⁵⁰of 0.69 +/- 0.12 mM (1.7%, Figure 5(B)) and a Hill coefficient of 1.3 +/- 0.3. The highest isoflurane concentration tested (1.32 mM, 3.3%) suppressed signaling to 29.5% of control. The inhibitory effect was reversible (Figure 5(C)).

If anesthetic inhibition of TXA²signaling results from hydrophobic interactions at the ligand-binding pocket, we would expect that the effect would be competitive (that is, fully reversible by high agonist concentrations) in nature. In contrast, a site of action elsewhere in the signaling pathway would likely be noncompetitive (that is, not fully reversible by high agonist concentrations). To determine if halothane and isoflurane act as competitive or noncompetitive antagonists on TXA²signaling, the concentration-response relation for U-46619 was determined in the presence of halothane 0.56 mM (1%, close to halothane's IC⁵⁰) or isoflurane 0.7 mM (1.75%, close to isoflurane's IC⁵⁰). Halothane acted in a competitive manner: it did not affect E^maxbut caused a parallel shift of the concentration-response relation to the right. This increased the U-46619 EC⁵⁰by approximately two orders of magnitude to 1.1 x 10 sup -5 M (Table 1). In contrast, isoflurane acted in a noncompetitive manner: It did not affect the U-46619 EC⁵⁰(5.3 x 10 sup -7 M) but decreased E^maxto 70% of control (Table 1, Figure 6).

We also tested the ability of sevoflurane to inhibit TXA²signaling. In contrast to the other anesthetics tested, sevoflurane, at concentrations up to 5%(0–1.3 mM, 2.92 minimum alveolar concentration), [14]did not affect TXA²signaling (Figure 7).

To determine the site of action of halothane on the TXA²signaling pathway, we activated segments of the intracellular pathway directly by intracellular microinjection of second messengers. IP³directly activates its receptor-channel on intracellular Ca²+ stores; GTP gamma S irreversibly activates G proteins (Figure 1(A)). Therefore, a lack of anesthetic effects on I^Cl(Ca) induced by these mediators would indicate the receptor or the coupling between G protein and receptor as a site of anesthetic action. IP³and GTP gamma S induced I^Cl(Ca) in uninjected oocytes. Average responses were 3.9 +/- 0.2 micro C for IP³-induced I^Cl(Ca) and 4.4 +/- 0.4 micro C for GTP gamma S-induced I^Cl(Ca). Neither halothane, isoflurane, nor sevoflurane affected these currents: IP³-induced responses in the presence of 1.07 mM halothane (2.5%), 1.32 mM isoflurane (3.3%), or 1.3 mM sevoflurane (5%) were 101 +/- 19.6%, 89 +/- 22.4%, and 93 +/- 14.07% of control, respectively. GTP gamma S-induced responses were 131 +/- 25.4% of control in the presence of halothane, 122 +/- 18.6% in the presence of isoflurane, and 120 +/- 32.5% of control in the presence of sevoflurane (not significantly different from control;Figure 8). When 10-fold greater or 10-fold smaller concentrations of IP³and GTP gamma S were injected, the anesthetics were similarly without effect (data not shown).

Our findings show that clinically relevant concentrations of halothane and isoflurane inhibit functioning of recombinantly expressed TXA²receptors. Halothane acts in a competitive, and isoflurane in a noncompetitive manner. In contrast, sevoflurane does not inhibit TXA sub 2 signaling. Neither anesthetic affects intracellular signaling pathways.

Thromboxane A²is a potent stimulator of platelet aggregation [5,6,15]and a constrictor of vascular [5,16]and respiratory smooth muscle. [17]Its function is counterbalanced with that of prostacyclin, which inhibits platelet aggregation and elicits vasorelaxation. Disruption of this balance in favor of TXA²has been suggested to play a role in thrombosis, asthma, and unstable angina, as well as in myocardial infarction. [18]Thus TXA²receptor antagonists are of considerable therapeutic importance. [17]Therefore acute inhibitory effects of anesthetics on TXA²signaling may be beneficial; in addition, short-term modulation of TXA²signaling during anesthesia may result in long-term beneficial effects as well. [19] 

Hirakata et al. [20]determined the amino acid sequence of the human TXA²receptor by cloning it from a human placental cDNA library. Nilsing et al. [12]recently cloned the gene for the human TXA²receptor and found no evidence for additional genes. Thus only one TXA²receptor type appears to exist. This is unusual, because for most other members of the G protein-coupled receptor superfamily the existence of several receptor subtypes has been demonstrated. The present study had the advantage that our findings are not subtype dependent.

Data on the interactions between volatile anesthetics and TXA sub 2 signaling are contradictory. Hirakata et al. [6]demonstrated interference of halothane and sevoflurane with the TXA²synthetic pathway in human platelets. As as result, halothane (0.49–1.25 mM) and sevoflurane (0.13–1.3 mM) blocked secondary aggregation induced by epinephrine or adenosine diphosphate. Isoflurane (0.28–0.84 mM) had no significant effect. Whereas sevoflurane (3 mM) and isoflurane (2.5 mM) had minimal effects on platelet binding of the TXA²-receptor antagonist [sup 3 H]S145 (5Z-7-(3-endo([ring-4-sup 3 H]phenyl)sulphonylamino-[2.2.1.]bicyclohept-2-exo-yl)heptenoic acid), halothane (3.3 mM) suppressed binding. It should be noted that the halothane concentration required to interfere with binding was much greater than that required to block aggregation. Scatchard analysis of [sup 3 H]S145 binding showed that sevoflurane affected neither B^maxnor K^d, whereas halothane (3.3 mM) markedly increased K^dwithout significantly altering B^max. The authors concluded that (1) halothane interfered both with TXA²-receptor interaction and with cyclo-oxygenase function, (2) sevoflurane inhibited only cyclo-oxygenase and (3) isoflurane seemed to affect neither cyclo-oxygenase nor TXA²receptor binding. In contrast, Blaise et al., [5]while observing similar effects of halothane, reported different results for isoflurane: In their model, isoflurane inhibited platelet aggregation induced by adenosine diphosphate, collagen, epinephrine, and arachidonic acid. These findings were confirmed in a second model: vascular smooth muscle contraction induced by U-46619 and potassium chloride.

Because of these inconsistencies, and because functioning of the TXA²receptor per se had not been studied, we did this investigation. We found no interference of sevoflurane with TXA²signaling, in agreement with the lack of receptor binding observed by Hirakata et al. [6]Halothane did inhibit receptor functioning in a competitive manner, consistent with its effects on receptor binding (although the functional effects are notable at much lower concentrations). The noncompetitive interaction with receptor functioning observed for isoflurane is consistent with its lack of effect on receptor binding. However, the lack of effect of isoflurane on platelet aggregation as reported by Hirakata et al. [6]cannot be reconciled with either Blaise et al.'s [5]or our own data. Although the special properties of our model (such as use of room temperature, amphibian membrane, and G protein) should be kept in mind, previous studies show that the anesthetic effects on receptors expressed in Xenopus oocytes are similar to those observed in other models. [2,3,21,22]In addition, our data are consistent with those of Blaise et al. [5] 

Our experiments with microinjected intermediates show that the intracellular signaling pathway is unaffected by the anesthetics. Therefore, the receptor itself is the most likely site of action. Because halothane's effect is competitive and it interferes with ligand binding, [6]its action is probably at the ligand-binding site. Because halothane is the most lipid soluble of the three anesthetics tested (oil-water partition coefficient 310 [23]), we hypothesize that the site of halothane's action is hydrophobic. Yamamoto et al. [15]modeled the structure of the TXA²receptor and identified several amino acid residues likely involved in ligand binding, as well as a large hydrophobic pocket among these amino acids. Based on this combination of data, we postulate that halothane interacts with this hydrophobic pocket.

In contrast, inhibition by isoflurane is noncompetitive, and the anesthetic does not interfere with ligand binding. [6]Therefore its (allosteric) site of action is unlikely to be at the ligand-binding pocket. Its lipophilicity (oil-water partition coefficient 170 [23]) is less than that of halothane, making it a less suitable candidate for interaction with the lipophilic pocket. In agreement, sevoflurane (oil-water partition coefficient 32 [23]) was completely without effect. The exact site of action of isoflurane cannot be determined from the present studies. However, mutation analysis might make localization of this site possible.

These findings reinforce the concept that different anesthetics may have different sites of action within the same molecular structure. In other words, even if anesthetics have similar actions on a molecule, it does not follow that they act on the same site. Eckenhoff [24]recently demonstrated the existence of multiple binding domains for inhalational anesthetics in the nicotinic acetylcholine receptor. Similarly, different binding domains may exist within the gamma-aminobutyric acid^Areceptor complex. Halothane and isoflurane enhance the ligand binding of alpha¹gamma²gamma-aminobutyric acid^Areceptors. However, cotransfection with the beta²subunit reduced the efficacy of both isoflurane and halothane, whereas cotransfection with the beta³subunit increased the efficacy of isoflurane but not halothane. [25]Thus at least several signaling molecules may have multiple, separate sites of action for volatile anesthetics.

Whereas halothane and isoflurane inhibit TXA²receptor functioning, halothane acts in a competitive and isoflurane acts in a noncompetitive manner. In contrast, sevoflurane has no effect. The site of action appears to be the receptor molecule itself. The site of halothane's action is most likely the hydrophobic pocket in the ligand-binding domain. In contrast, isoflurane most likely acts at an allosteric site.

The authors thank Dr. C. Lynch, III, (University of Virginia) for insightful comments on the manuscript and Bayer AG (Wuppertal, Germany) for providing Bay U 3405.