Direct Inhibitory Mechanisms of Halothane on Human Platelet Aggregation

This study was approved by the Sapporo Medical University Ethical Committee on Human Research. After informed consent was obtained, blood (approximately 15 ml) was drawn by antecubital venipuncture from 80 healthy volunteers who had not taken any medications for at least 14 days before the donation. The mean age of the volunteers was 26.8 yr (range 22-30 yr). Blood was collected into plastic syringes containing acid-citrate-dextrose solution. The acid-citrate-dextrose solution contained 85 mM sodium citrate, 70 mM citric acid, and 110 mM glucose. The ratio of blood to acid-citrate-dextrose solution was 17:3, and the final concentration of citrate in the whole blood was approximately 13 mM. Processing of blood samples was begun within 20 min of venipuncture.

Platelet-rich plasma and washed platelet suspension (WPS) were prepared by a modification of previously published methods. [19,20]Briefly, the whole blood was centrifuged at 150g for 10 min and the upper two thirds of the platelet-rich plasma was collected. The WPS was prepared by centrifuging the platelet-rich plasma at 600g for 10 min and resuspending the pellet in a modified Tyrode's solution. The Tyrode's solution contained (in mM) 140 NaCl, 2.7 KCl, 0.23 MgCl², 0.4 NaH²PO⁴, 12 NaHCO³, 5 glucose, and 10 N-(2-hydroxy-ethyl)piperazine-N'-2-ethanesulfonic acid; pH level was adjusted to 7.4 with NaOH. Platelets were counted by a Coulter Counter STKS (Coulter, Hialeah, FL) for adjustment of the concentrations of platelet-rich plasma and WPS. All procedures were performed at room temperature (22-24 degrees C). The experiments were started after preincubation with or without 1 mM Calcium²+ for 1 min at 37 degrees Celsius and were completed within 3 h after venipuncture. Siliconized glassware and plastic tools were used throughout.

To investigate the interruption of thrombin-receptor binding by halothane, we measured platelet-surface nonbinding GpIb, a part of the thrombin receptor, [21]by fluorescence flow cytometry. [22]After the preincubation with 1 mM Calcium²+ for 1 min, samples of WPS were incubated with halothane-containing solution (final concentration: 0, 0.5, 1.0, 1.5, or 2.0 mM) for 2 min and were then stimulated with or without 0.02 units/ml thrombin for 3 min at 37 degrees C. Platelet preparations were stirred at 1,000 rpm during this protocol. Each incubation was stopped by the addition of an equal volume of 1% (vol/vol) paraformaldehyde. After the fixation, the paraformaldehyde-treated platelets were washed twice by centrifugation (600g for 5 min) and resuspension in phosphate-buffered saline (composition in mM: 120 NaCl, 2.7 KCl, 8 Na²HPO⁴, and 2 KH²PO⁴; pH 7.4). The fixed platelets were then incubated with an isothiocyanate fluorescein-labeled mouse monoclonal immune globulin G antibody (SZ-2-FITC, 2 micro gram/5 x 10⁶platelets) for 30 min at room temperature, washed twice, and resuspended in phosphate-buffered saline at a final concentration of 5 x 10⁷platelets/milliliter. The GpIb antibody used in this study (SZ-2) binds specifically to the human platelet GpIb alpha-subunit with mean K sub d = 6.6 x 10 sup -10 M. [23]The labeled platelets were analyzed with an argon ion laser fluorescence flow cytometer (FACScan 440; Becton Dickinson, Mountain View, CA) using a 488-nm wavelength at 300 mW. Fluorescence was detected through a 530+/-15 nm band-pass filter. The fluorescence histograms were analyzed on an attached computer (Consort 30; Becton Dickinson). The determinations were made in duplicate of the channel-weighted mean fluorescence intensity for 10⁴platelets, and the results are expressed as the fluorescein intensity ratio to the control without thrombin or halothane.

The method of Pollock and Rink [24]was followed. Platelet-rich plasma was incubated with Fura-2/AM (5 micro Meter) for 30 min at room temperature. After the Fura-2 loading, WPS was obtained as described earlier. Approximately 1 ml of the WPS (approximately 1.5 x 10⁷platelets per cuvette) was added to the stirred (1,000 rpm) cuvette attachment of a fluorometer (CAF-110; JASCO, Tokyo, Japan). After the WPS was preincubated with 1 mM Calcium²+ or 50 micro Meter EGTA for 1 min, the samples were incubated with various concentrations of halothane-containing solution (range of final concentration: 0-2.0 mM) for 2 min and then stimulated with 0.02 units/ml thrombin for 3 min. To further investigate the effect of other anesthetics on the increase of [Calcium²+]ⁱdue to Calcium²+ release from intracellular stores (DTS), we performed additional experiments using isoflurane (range 0-2.0 mM) in the absence of external Calcium²+ as well.

The changes of [Calcium²+]ⁱ, indicated by Fura-2 fluorescence, and of platelet aggregation were simultaneously measured with the CAF110 fluorometer. The platelet samples were illuminated alternately (128 Hz) at the excitation wavelengths of 340 and 380 nm. The intensities of 500-nm fluorescence induced by 340-nm excitation and by 380-nm excitation were monitored. At the end of each experiment, the cells were treated with 0.1% (vol/vol) Triton X-100 followed by the addition of 10 mM ethylene glycol-bis-(beta-amino ethylether)-N, N, N'-tetraacetic acid (EGTA) to obtain the maximum and minimum fluorescence, respectively. Absolute values of [Calcium²+]ⁱwere determined by the formula of Grynkiewicz et al. [25]using a Fura-2-Calcium²+ dissociation constant of 224 nM. The background fluorescence, measured by adding 1 mM MnCl², was negligible. Platelet aggregation was assessed using a spectrophotometer adjusted to 0% transmittance for control WPS and to 100% transmittance for distilled water.

The technique of Uemura et al. [26]was used to measure the intracellular IP³concentration ([IP³]ⁱ). After preincubation with 1 mM Calcium²+ for 1 min, WPS (0.5 ml, 1 x 10⁹platelets/ml) was first incubated with halothane-containing solution (final concentration: 0, 1, or 2 mM) for 2 min and then stimulated with thrombin (0.02 units/ml). The reactions were terminated after 0, 5, 10, 15, 30, 60, or 120 s of thrombin stimulation by the addition of 0.2 volume of ice-cold 20% (vol/vol) perchloric acid solution. The samples were kept in an ice bath for 20 min and were then centrifuged at 2,000g for 15 min to remove insoluble materials. The pH level of the supernatant was adjusted precisely to 7.5 with 10 N KOH and insoluble precipitates (primarily KClO⁴) were removed by centrifugation at 2,000g for 10 min. The resultant supernatant was lyophilized and stored at -20 degrees Celsius. The lyophilized samples were dissolved in 100 micro liter distilled water, and the amount of IP³was measured using the Amersham IP³assay system (code TRK 1000; Amersham Japan, Tokyo, Japan). This assay is based on competition between unlabeled IP³in the sample and a fixed quantity of tritium-labeled IP³for a limited number of high affinity binding sites on a specific IP³binding protein. [27]The determinations were made in duplicate and the results are expressed as pmol/5 x 10⁸platelets.

Washed platelet suspension (1 ml, 5 x 10⁸platelets/milliliter) was incubated with 1 mM Calcium²+ for 1 min. Intracellular cAMP concentrations ([cAMP]ⁱ) were measured under four conditions 6 min after addition of Calcium²+: without exposure to halothane or thrombin; or with 5 min exposure to halothane (0, 1, and 2 mM, respectively) with additional exposure to 0.02 units/ml thrombin for the final 3 min. In experiments involving addition of forskolin, a potent adenylate cyclase activator, instead of halothane, concentrations of this drug were used that produced the same effect as halothane (1 and 2 mM) on the inhibition of the platelet aggregation. Measurement of the [cAMP]ⁱof platelets was done by a modification of previously published methods. [28]At the end of the incubations, the reactions were stopped by the addition of an equal volume of ice-cold 0.1 N HCl solution and the suspensions were immediately homogenized for 60 s at 4 degrees C with a VC-50 ultrasonic homogenizer (Sonic and Material, Danbury, CT). The homogenized samples were then centrifuged at 3,000g for 15 min at 4 degrees C and 100 micro liter of the supernatant was used for the measurement of the [cAMP]ⁱ. We analyzed [cAMP]ⁱby a sensitive radioimmunoassay method (Yamasa cyclic adenosine monophosphate assay system; Yamasa, Chiba, Japan). The determinations were made in duplicate and the results are expressed as pmol/10⁸platelets.

Because of the possibility that halothane or isoflurane bubbling per se affects platelet aggregation and its measurement, we introduced the anesthetic into the cuvette using a 50-micro liter microsyringe (MS-50P4AF; Ver Werk, Ilmenau, Germany) to deliver preconcentrated anesthetic in the Tyrode's solution. To prevent anesthetic volatilization from WPS in the cuvette, we sealed the top of the cuvette with mineral oil and circulated the expected concentration of the anesthetic over the cuvette (approximately 1 l/min). Anesthetic concentrations in the cuvette were analyzed in each experiment using a gas chromatograph (GC-12A; Shimadzu, Kyoto, Japan) equipped with a flame ionization detector (FTD-8; Shimadzu) and an integrator (Chromatopac C-R 3A; Shimadzu). Both halothane and isoflurane concentrations in the cuvette were constant during the experiment for more than 5 min (data not shown). In addition, we measured the halothane and isoflurane concentrations in WPS, which were bubbled extensively with known gas concentrations of the anesthetic, determined with a calibrated infrared anesthetic gas monitor (5250 RGM; Ohmeda, Madison, WI). After equilibration, the mean halothane concentrations in the solutions (1.0, 2.0, 3.0, and 4.0% halothane in the gas phase) were 0.8, 1.8, 2.5, and 3.5 mM, respectively; whereas the mean isoflurane concentrations in the solutions (2.0 and 4.0% isoflurane in the gas phase) were 1.2 and 2.2 mM, respectively. The concentration of the anesthetics in the solution had close linear correlation with the concentration of the agent in the gas phase.

The following drugs and chemicals were used in this study: halothane (ICI, Dighton, MA); isoflurane (Abbott Laboratories, North Chicago, IL); acetoxymethyl ester of Fura-2 (Fura-2/AM; Dojindo, Kumamoto, Japan); Triton X-100 and sodium citrate (Katayama Chemical, Tokyo, Japan); citric acid (Kanto Chemical, Tokyo, Japan); EGTA, thrombin, paraformaldehyde, Triton X-100 and mineral oil (Sigma Chemical, St. Louis, MO); isothiocyanate fluorescein-labeled mouse monoclonal IgG antibody (SZ-2-FITC; Immunotech, Marseilles, France) and phosphate-buffered saline (Life Technologies, Grand Island, NY).

Data are expressed in scatter diagrams or as means+/-SEM. The effects of halothane on the aggregation or on the [Calcium²+]ⁱof platelets were assessed by analysis of the variance of regression coefficients. Comparisons for all of the other data were analyzed with unpaired, two-tailed t test or one-factor analysis of variance with Fisher's a posteriori test. In all comparisons, P < 0.05 was considered significant.

We measured the level of thrombin-receptor binding indirectly by estimating the presence of platelet-surface nonbinding GpIb with a fluorescence technique. The control fluorescence intensity of isothiocyanate-labeled platelets without thrombin or halothane was 51.7 +/-4.8/10⁴platelets (n = 7). Table 1shows the effects of thrombin and halothane on the GpIb expression. Halothane (0-2 mM) did not alter the binding between GpIb and the fluorescein-labeled antibody SZ-2-FITC in the absence of thrombin (P = 0.6). Thrombin (0.02 units/ml) significantly decreased labeled nonbinding GpIb by 25+/-4% (P < 0.01). Halothane did not interfere with the binding between GpIb and thrombin in the range 0-2 mM (P = 0.7), but at a high concentration (5.5 mM), the anesthetic exhibited a small (21+/-5%) but significant inhibitory effect on the GpIb-thrombin binding (P < 0.05, data not shown in Table 1).

(Figure 2) shows the effects of halothane on [Calcium²+]ⁱand on the aggregation of thrombin-stimulated human platelets in the presence of 1 mM extracellular Calcium²+. The resting [Calcium²+]ⁱwas 106+/-6 nM (n = 5), and halothane did not change [Calcium²+]ⁱwhen tested at concentrations up to 2 mM. [Calcium sup 2+]ⁱwas rapidly increased by 0.02 units/ml thrombin with a biphasic response: a first peak was seen at approximately 15 s and a second at approximately 2 min after exposure to thrombin. The [Calcium²+]ⁱat the first and second peaks were 520+/-18 and 453 +/-22 nM, respectively (n = 5). The thrombin-stimulated platelets were rapidly aggregated, reaching a peak (approximately 60% aggregating ratio) at approximately 3 min. We show representative effects of halothane at 0.9 and 1.8 mM on the platelet aggregation and [Calcium²+]ⁱin Figure 2. In this case, both peaks of [Calcium²+]ⁱinduced by thrombin were substantially decreased by halothane, but the time courses of changes in [Calcium²+]ⁱand of aggregation did not appear to change.

The relationships between halothane concentration (mM) and changes in (A) percentage of aggregation at 3 min after exposure to thrombin and (B) [Calcium²+]ⁱof the first peak in the presence of 1 mM external Calcium²+ are shown in Figure 3. Halothane (0-2 mM) significantly inhibited aggregation and caused a dose-dependent decrease in [Calcium²+]ⁱ. Linear relationships were observed both between halothane concentrations and percent aggregation (r = -0.94, n = 50, P < 0.01) and between halothane concentrations and [Calcium²+]ⁱ(r = -0.92, n = 50, P < 0.01).

We also investigated the effects of halothane and isoflurane on [Calcium²+]ⁱand the aggregation of thrombin-stimulated human platelets in the absence of external Calcium²+ (solution containing 50 micro Meter EGTA instead). As shown in Figure 4(A), the resting [Calcium²+]ⁱwas 73+/-5 nM (n = 5) and was significantly less than that obtained in the presence of 1 mM external Calcium²+ (P < 0.05). Halothane did not change the resting [Calcium sup 2+]ⁱin the concentration range of 0-2 mM. As observed in the presence of external Calcium²+, [Calcium²+]ⁱwas rapidly increased by exposure to 0.02 units/ml thrombin. However, in contrast to the observation with external Calcium²+, this initial increase was followed by a substantial reduction and there was no second peak. The time course of the initial change in [Calcium²+]ⁱwas similar to that seen in the presence of 1 mM external Calcium²+ with a peak occurring approximately 15 s after exposure to thrombin. However, the peak [Calcium sup 2+]ⁱ(293+/-14 nM) was significantly less than that of the first peak obtained with extracellular Calcium²+ (P < 0.01, n = 5). Platelets smoothly aggregated, reaching their respective peaks (approximately 30% aggregating ratio) at approximately 3 min. Halothane (1.0 and 2.0 mM) decreased the induced changes in [Calcium²+]ⁱand inhibited aggregation in the absence of external Calcium²+. To determine whether these effects were induced by other anesthetics as well, we performed additional experiments using isoflurane in the absence of external Calcium²+ (Figure 4(B)). Isoflurane at concentrations up to approximately 2 mM (= approximately 3.6% in gas phase) had no apparent effect on either platelet aggregation (inhibited by approximately 5 +/-5% at 1 mM and approximately 8+/-6% at 2 mM, respectively) or the increase in [Calcium²+]ⁱ(inhibited by approximately 3+/-5% at 1 mM and approximately 6+/-8% at 2 mM, respectively) induced by thrombin (n = 5 at each point).

(Figure 5) shows the relationships obtained in the absence of external Calcium²+ between halothane concentration and changes in (A) percentage of aggregation at 3 min and (B) peak [Calcium²+]ⁱ. Halothane (0-2 mM) significantly inhibited aggregation and caused a dose-dependent decrease of [Calcium²+]ⁱ. Linear relationships were observed both between halothane concentrations and percent aggregation (r = -0.95, n = 40, P < 0.01) and between halothane concentrations and [Calcium²+]ⁱ(r = -0.93, n = 40, P < 0.01). Halothane, at approximately 2 mM, almost completely inhibited thrombin-induced platelet aggregation and decreased the [Calcium²+] sub i to the resting value.

The time course and effects of 2 mM halothane on intracellular IP³concentrations ([IP³]ⁱ) in thrombin-stimulated human platelets are shown in Figure 6. The [IP³]ⁱat time 0 was 1.4+/- 0.2 pmol/5 x 10⁸platelets (n = 6) and did not change with the addition of halothane (1.3+/-0.3 pmol/5 x 10⁸platelets at 1 mM and 1.2+/-0.2 pmol/5 x 10⁸platelets at 2 mM halothane). Thrombin (0.02 units/ml) produced a rapid increase in the [IP sub 3]ⁱ, reaching its maximum (5.9+/-0.9 pmol/5 x 10⁸platelets) at 10 s after the stimulation. The rapid increase in [IP³] sub i induced by thrombin was followed by a rapid and substantial decrease to a concentration of approximately 2 pmol/5 x 10⁸platelets. Halothane (2 mM) significantly inhibited the increase of [IP³]ⁱinduced by thrombin at 5-15 s after thrombin stimulation without an apparent change in the time course of [IP³]ⁱ. The inset of Figure 6summarizes the effects of various concentrations of halothane (0, 1, and 2 mM) on the peak [IP³]ⁱat 10 s after thrombin stimulation. Halothane significantly inhibited in a dose-dependent manner the increase in [IP³]ⁱinduced by thrombin.

The effect of various concentrations of forskolin on thrombin (0.02 units/ml)-induced platelet aggregation and on intracellular cAMP concentrations ([cAMP]ⁱ) were compared with those of halothane. As shown in Figure 3(A), halothane at 1 and 2 mM decreased thrombin-induced platelet aggregation to approximately 36% and approximately 4% of control, respectively. Forskolin at 10 and 21 micro Meter concentrations decreased thrombin (0.02 units/ml)-induced platelet aggregation to 35+/-4% and to 6+/-3%, respectively (n = 5). There was no significant difference between halothane and forskolin with respect to the extent of inhibition of platelet aggregation.

The effects of halothane and forskolin at these concentrations on [cAMP]ⁱof thrombin-stimulated human platelets are shown in Figure 7. The [cAMP]ⁱin the resting state without thrombin was 1.6 +/-0.2 pmol/10⁸platelets (n = 6). After stimulation with 0.02 units/ml thrombin, [cAMP]ⁱtended to decrease to 1.4+/- 0.1 pmol/10⁸platelets, but this change was not significant. Halothane at 1 and 2 mM moderately but significantly increased [cAMP]ⁱby approximately 20% and approximately 40%, respectively. In contrast, forskolin at 10 and 21 micro Meter caused significantly greater increases in [cAMP]ⁱ(P < 0.05, n = 6) than did halothane at concentrations equieffective for inhibition of aggregation.

Physiologic platelet activation or inhibition involves interaction of an extracellular signaling molecule with the platelet surface via ligand-receptor coupling. Based on the interactions of thrombin with the glycocalicin component of GpIb, [29]GpIb is proposed to be a functionally significant thrombin receptor (Figure 1). [21]As reported previously, [30]we found using flow cytometry with fluorescent labeling of the platelet surface proteins [22]that thrombin (0.02 units/ml) significantly reduced the amount of antibody binding to GpIb. Halothane did not affect the thrombin-induced decrease in antibody labeling of GpIb, suggesting that halothane within the range of 0-2 mM (0-2.4% in gas phase) does not interfere with the interaction of thrombin with its receptor. This is in agreement with the general observation that volatile anesthetics, at clinically used concentrations, have little or no effect on agonist binding. [31]Although Hirakata et al. [18]demonstrated that halothane reduced the receptor-binding affinity of a potent aggregating agonist thromboxane A²in human platelets, this effect likely required the extremely high concentration of halothane (14 mM [nearly equal] to 16% in gas phase) that they used.

As is established in other tissues, Calcium²+ is thought to be an important regulator of platelet activation (Figure 1). [9,32]Results from the current study also support a central role for Calcium²+ in platelet aggregation (Figure 2and Figure 4). Our results are similar to those observed by others with platelet stimulation by thrombin, [19,33]collagen, [33]or adenosine diphosphate. [34]An increase in [Calcium²+]ⁱactivates myosin light chain kinase, which initiates platelet aggregation through increased phosphorylation of myosin light chain. [10,34]In platelets, the increase in [Calcium²+]ⁱinvolves the release of Calcium²+ from intracellular stores, especially DTS, and later Calcium²+ influx from the extracellular space. [35]As shown in Figure 2, [Calcium²+]ⁱinduced by aggregating agonists (e.g., thrombin) usually induces a biphasic response. The first peak of [Calcium²+]ⁱis thought to result mainly from Calcium²+ release from DTS and the second peak of [Calcium²+] sub i is thought to reflect Calcium²+ influx from the extracellular space. [24,36]Our measurement of [Calcium²+]ⁱwith or without external Calcium²+ support this model of Calcium²+ regulation and its role as a regulator of platelet activation.

Halothane decreased [Calcium²+]ⁱin parallel with its inhibition of platelet aggregation in the presence of external Calcium²+ (Figure 2and Figure 3). Our data support the hypothesis that halothane inhibits platelet aggregation mainly by suppressing the increase of [Calcium²+]ⁱinduced by aggregating agonists. Halothane also decreased the peak [Calcium²+]ⁱin the absence of external Calcium²+. There is some evidence that Calcium²+ release from intracellular stores may be more essential for platelet activation than is Calcium²+ influx from the extracellular space. [20,24,36]Therefore, it is possible that the inhibition of the increase in [Calcium sup 2+]ⁱdue to intracellular stores is the main mechanism by which halothane inhibits platelet aggregation. Interestingly, isoflurane had little effect on the [Calcium²+]ⁱin the absence of external Calcium²+. This observation parallels the clinical observation that halothane is more effective than other anesthetics at inhibiting platelet aggregation at clinically used concentrations. [5].

Hossain and Evers [37]reported that in clonal (GH³) pituitary cells halothane increased the resting [Calcium²+]ⁱthrough Calcium²+ release from IP³-gated intracellular stores. Although we did not observe any increase in [Calcium²+]ⁱby halothane in the absence of external Calcium²+, our data do not exclude the possibility that the decrease in thrombin-induced increase of [Calcium²+]ⁱby halothane results partly from the depletion of the Calcium²+ stores by preincubation with the anesthetic. [38,39]Our data do indicate that halothane inhibits Calcium²+ influx through the platelet membrane, but the role of Calcium²+ influx from the extracellular space in the aggregation response of platelets to activation remains uncertain. [40]Further studies are required to clarify the significance of this effect.

Stimulation of the thrombin receptor activates the G protein-linked phospholipase C, resulting in hydrolysis of phosphatidylinositol 4,5-bisphosphate to the two potent stimulatory second messengers IP³and 1,2-diacylglycerol (Figure 1). [12,13,41]While 1,2-diacylglycerol activates Calcium²+ /phospholipid-dependent protein kinase at the membrane, [41]IP³mobilizes Calcium²+ from intracellular organella DTS. [14,15]Because, in platelets, IP³is the primary regulator for Calcium²+ release from intracellular stores and because the time course of the increase in [IP³]ⁱinduced by thrombin was very similar to that of the change in [Calcium²+]ⁱ(Figure 2and Figure 4), we suggest that IP³is an important determinant of [Calcium²+]ⁱduring agonist stimulation. Furthermore, there is evidence that IP³can directly enhance Calcium²+ influx from the extracellular space. [26]Therefore, inhibition of increases in [IP³]ⁱby halothane might in itself account for the observed effects on both [Calcium²+]ⁱand aggregation.

Our results are in general agreement with studies in a variety of cell types, in which halothane treatment has been associated with inhibition of the increase in [Calcium²+]ⁱmediated by IP³. [16,42,43]These studies have demonstrated that halothane alters Calcium²+ homeostasis, an action that underlies the in vivo effect of the anesthetic. However, Smart et al. [17]and Rooney et al. [44]showed that halothane induced IP³formation in SH-SY5Y human neuroblastoma cells and turkey erythrocytes, respectively. Rooney and colleagues also showed activation of phospholipase C activity by halothane. [44]These apparent discrepancies may result from the differences of cell types and species and/or the selective effects of halothane on certain receptors, G-proteins, or phospholipase C isozymes.

Cyclic 3',5'-adenosine monophosphate is another mechanism known to regulate [Calcium²+]ⁱ. In this study, halothane moderately but significantly increased the [cAMP]ⁱof human platelets. This result is consistent with the previous suggestion by Walter et al. [4]that the impairment of platelet aggregation observed with halothane incubation might result from halothane-induced activation of platelet adenylate cyclase, resulting in a higher cAMP concentration in the cytosol. Several effects of cAMP on platelet Calcium sup + metabolism have been established, [45,46]including stimulation of Calcium²+ efflux [6]and stimulation of Calcium²+ uptake into DTS, [7]both of which result in a decrease in [Calcium²+]ⁱ. In addition, cAMP-dependent protein phosphorylation has been reported to attenuate IP sub 3 -mediated Calcium²+ release from DTS. [47]However, more recent data obtained with a pure preparation of the catalytic subunit of protein kinase seem to contradict this evidence. [48]Thus, the molecular mechanism by which cAMP regulates IP³-mediated Calcium²+ responses in intact platelets is far from fully established. Primarily as a consequence of its effect on [Calcium²+]ⁱ, cAMP influences certain other platelet responses, including phosphorylation of myosin light chain kinase [10]: cAMP-dependent protein kinase phosphorylates myosin light chain kinase and inhibits its activity in vitro [49]via a mechanism independent of Calcium²+ (Figure 1).

In this study, forskolin was used to investigate the role of cAMP in the inhibition by halothane of platelet aggregation. When forskolin was used at concentrations that inhibited platelet aggregation to the same extent as halothane (1 and 2 mM), the [cAMP]ⁱincreased to a much greater extent than with halothane (Figure 7). Thus, the inhibitory effect of halothane on platelet aggregation appears to involve additional mechanisms other than an increase of [cAMP]ⁱ.

In summary, halothane in clinical concentrations inhibits thrombin-induced human platelet aggregation mainly by decreasing [Calcium sup 2+]ⁱwithout inhibiting agonist-receptor binding. The inhibitory effect of halothane on [Calcium²+]ⁱmay be mediated both by a decrease in [IP³]ⁱand by an increase in [cAMP]ⁱ. We suggest that, while cAMP both suppresses [Calcium²+]ⁱand acts via a Calcium²+ -independent pathway to inhibit platelet aggregation, IP³is a crucial regulator of aggregation in platelets and that the inhibitory action of halothane on aggregation is explicable by its effects on IP³.

The authors thank A. Namiki, M.D., Ph.D., Professor and Chairman of Anesthesiology, Sapporo Medical University School of Medicine, for directing the research and reviewing the manuscript; H. Kanoh, M.D., Ph.D., Sapporo Medical University, K. Matsuno, M.D., Ph.D., Hokkaido University, T. L. Croxton, Ph.D., M.D., T. S. Kickler, M.D., and P. A. Stephens, Ph.D., The Johns Hopkins Medical Institutions, for comments and suggestions concerning this manuscript; and T. Mitani, Ph.D., Hokkaido Red Cross Blood Center, for technical assistance.