Sevoflurane reportedly inhibits adenosine diphosphate-induced platelet aggregation by suppressing thromboxane A2 formation. The increase in intracellular calcium concentration that fosters platelet aggregation, however, is also induced by other cell signaling pathways, such as activation of the production of inositol 1,4,5-triphosphate by thrombin. The current study aimed to clarify the net influence of sevoflurane on thrombin-induced platelet aggregation.
Washed platelets were stimulated by thrombin after incubation with 0.5, 1.0, or 1.5 mM sevoflurane, halothane, or isoflurane. Aggregation curves were measured by an aggregometer. Intracellular calcium concentration was measured fluorometrically using fura-2. Calcium mobilization via plasma membrane calcium channels and the dense tubular system was assessed differentially. Intracellular inositol 1,4,5-triphosphate was measured by radioimmunoassay.
Halothane significantly suppressed aggregation ratios at 5 min compared with those in controls (89 +/- 7%) to 71 +/- 10% (1.0 mM) and 60 +/- 11% (1.5 mM) and the increase in intracellular calcium concentration (controls, 821 +/- 95 nM vs. 440 +/- 124 nM [1.0 mM] or 410 +/- 74 nM [1.5 mM]). Halothane also significantly inhibited release of calcium from the dense tubular system (controls, 220 +/- 48 nM vs. 142 +/- 31 nM [1.0 mM]). Neither sevoflurane nor isoflurane produced a net change in aggregation ratios, intracellular calcium concentration, or calcium mobilization. Halothane (1 mM) significantly suppressed inositol 1,4,5-triphosphate concentrations, whereas neither 1 mM isoflurane nor 1 mM sevoflurane had any effect.
Although sevoflurane has been reported to inhibit human platelet aggregation induced by weak agonists such as adenosine diphosphate, it does not inhibit human platelet aggregation induced by strong agonists such as thrombin.
SEVERAL studies of the effects of inhalation anesthetic agents on platelet function have been reported since Ueda 1demonstrated in 1971 that clinical concentrations of halothane inhibited adenosine diphosphate (ADP)-induced platelet aggregation. Although some reported findings remain controversial, 2–5halothane is considered to inhibit platelet aggregation 6–8whereas isoflurane is not. 8,9Recently, Hirakata et al. 10reported that sevoflurane inhibited platelet aggregation by suppressing the formation of thromboxane A2, concluding that the inhibitory effect of sevoflurane could be mediated by a decrease in cyclooxygenase activity. This study, however, was performed on the platelet aggregation induced by weak agonists such as ADP and epinephrine. Aggregation agonists may be grouped as strong agonists, such as thrombin, and weak agonists, such as ADP. 11Platelet stimulation by strong agonist is mediated by additional or distinct intracellular events compared with stimulation by weak agonists, 11–13suggesting that the inhibition of one of these pathways may not necessarily inhibit platelet aggregation as observed clinically.
To clarify the influence of sevoflurane on thrombin-induced platelet aggregation, halothane was used as a positive control, and isoflurane was used as a negative control. We measured intracellular calcium concentration ([Ca2+]i) simultaneously with platelet aggregation; analyzed calcium mobilization by influx through calcium channels in the plasma membrane or the release from the dense tubular system using the divalent cations, Ni2+and Mn2+; and measured intracellular inositol 1,4,5-triphosphate (IP3) concentrations by radioimmunoassay.
Materials and Methods
This study was approved by the Ethical Committee on Human Research of Kyoto Prefectural University of Medicine, and written informed consent was obtained from all participants.
Preparation of Washed Platelet Suspension Loaded with Fura-2
Venous blood (32 ml) was obtained by antecubital venipuncture from healthy volunteers who had not taken any drug known to affect platelet aggregation for at least 2 weeks. These samples were collected into plastic tubes containing acid–citrate–dextrose solution consisting of 85 mM sodium citrate, 70 mM citric acid, and 110 mM glucose. The ratio of blood to acid–citrate–dextrose solution was 4:1. The samples were centrifuged at 110 g for 15 min. The upper layer was withdrawn as platelet-rich plasma. These platelet-rich plasma samples were incubated immediately with 3 mM fura-2 acetoxymethyl ester for 15 min at 37°C and then centrifuged at 250 g for 15 min in the presence of apyrase (0.3 U/ml). A washed platelet suspension (WPS) loaded with fura-2 was obtained after discarding the supernatant and resuspending the pellet in Tyrode-HEPES buffer at p H 7.4. 14Using a Coulter Counter Model MD II (Coulter Electronics, Hialeah, FL), numbers of platelets in the WPS were adjusted to 105cells/μl for the investigation of aggregation curves and intracellular calcium mobilization, or alternatively to 106cells/μl for the measurement of intracellular IP3concentration.
Assessment of Platelet Aggregation Curve
Platelet aggregation curves were investigated using the turbidimetric technique.
Aliquots of WPS (250 μl, including 2.5 × 107platelets per cuvette) were placed in a siliconized glass tube, maintained at 37°C, and stirred at 1,000 rpm throughout the experiments. After adjustment of the extracellular free calcium concentration to 1 mM using CaCl2, platelet aggregation induced by thrombin (final concentration, 0.005–0.020 U/ml) was measured for 5 min using an aggregometer (HEMA TRACER 601; NIKOH Bioscience, Tokyo, Japan).
After incubation for 5 min with halothane, isoflurane, or sevoflurane (final concentration and percentages: for halothane, 0.5 mM [0.5%], 1.0 mM [1.1%], and 1.5 mM [1.6%]; for isoflurane, 0.5 mM [0.9%], 1.0 mM [1.8%], and 1.5 mM [2.7%]; for sevoflurane, 0.5 mM [1.9%], 1.0 mM [3.8%], and 1.5 mM [5.7%]), platelet aggregation was measured for 5 min after addition of thrombin (final concentration, 0.01 U/ml).
Measurement of Intracellular Free Ca2+Concentration and Mobilization
Aliquots (500 μl) of WPS loaded with fura-2, including 5 × 107platelets per cuvette, were added to a fluorometric cuvette stirred at 1,000 rpm within a fluorometer (CAF-110; JASCO, Tokyo, Japan), which repeatedly subjected samples to alternating excitation wavelengths of 340 and 380 nm while measuring emission at 505 nm (sampling frequency, 2 Hz). 15Immediately after adjustment of extracellular free calcium concentration to 1 mM using CaCl2and incubation with each anesthetic agent (final concentrations, 0.0 [control], 0.5, 1.0, or 1.5 mM of halothane, isoflurane, or sevoflurane) for precisely 5 min at 37°C to record basal fluorescence intensity, samples were stimulated for 5 min with thrombin (final concentration, 0.01 U/ml). The ratio of fluorescence intensity at 340 nm relative to that of 380 nm (R340:380) 16,17was recorded at resting and peak phases. At each point, [Ca2+]iwas calculated according to the equation of Tsien et al. , 18using a dissociation constant for fura-2 and Ca2+of 224 nM. After the end of each experiment, cells were treated with 0.3% Triton-X followed by the addition of 3 mM ethylene glycol-bis-(β-amino ethylether)-N,N,N′-tetraacetic acid to obtain maximum and minimum fluorescence, respectively. Data are expressed as the increase in [Ca2+]ifrom its baseline levels.
Ni2+(final concentration, 1 mM) was added to samples instead of CaCl2to block the influx of extracellular calcium. Therefore, the thrombin-induced release of calcium from the dense tubular system was observed. 19,20
Entry of Mn2+was evaluated with a quenching technique to investigate the influx of calcium into the cytosol. 19,20MnCl2(final concentration, 1 mM) was added to the samples, and fluorescence from fura-2 was monitored at the calcium-insensitive excitation wavelength of 360 nm. Mn2+entry, which represents calcium influx, was measured as the rate of decrease of fluorescence after stimulation by thrombin. Data are expressed as percent of initial fluorescence.
Measurement of Intracellular Inositol 1,4,5-Triphosphate Concentration
Intracellular IP3concentration was measured using the technique of Uemura et al. 21After preincubation with 1 mM of Ca2+for 1 min, WPS (120 μl, including 1.2 × 108platelets) first was incubated with halothane, isoflurane, or sevoflurane (final concentration, 1 mM), respectively, for 5 min and then was stimulated with thrombin (final concentration, 0.01 U/ml) for precisely 15, 30, and 60 s. The reactions were terminated by adding ice-cold 20% perchloric acid in an amount equal to one fifth of the original volume. The samples were kept in an ice bath for 20 min and were then centrifuged at 2,000 g for 15 min. The p H of the supernatant was adjusted to 7.5 with 4 N KOH, and insoluble precipitates were removed by centrifugation at 2,000 g for 10 min. The resultant supernatant was lyophilized and stored at −20°C. The lyophilized samples were dissolved in 100 μl distilled water, and the amount of IP3was measured using the Amersham IP3assay system (TRK1,000; Amersham International, Buckinghamshire, United Kingdom). Determinations were made in duplicate, and the results are expressed as pmol/108platelets.
Determination of Anesthetic Concentration
Immediately after an aliquot of each anesthetic agent was added directly to WPS-containing tubes, the tops were tightly sealed with parafilm to minimize evaporation. Anesthetic concentrations in the samples were analyzed using a gas chromatograph equipped with a flame ionization detector (GC-9A; Shimadzu, Kyoto, Japan), a headspace sampler (HSS-2A; Shimadzu), and an integrator (Chromatopac C-R4A; Shimadzu).
Sevoflurane was obtained from Hoechst Marion Roussel (Tokyo, Japan), halothane from Hoechst Japan (Tokyo, Japan), isoflurane from Dainabott (Tokyo, Japan), fura-2 acetoxymethyl ester from Dojindo Laboratories (Kumamoto, Japan), and apyrase from Sigma Chemical (St. Louis, MO). Thrombin was donated by Mochida Pharmaceutical (Tokyo, Japan). All other compounds were purchased from commercial sources.
Values in this study are expressed as mean ± SD. Unpaired t tests were adopted for analysis of aggregation ratios. One-way analysis of variance followed by Bonferroni correction was adopted for analysis of the increase in [Ca2+]iand release of calcium from the dense tubular system. Two-way analysis of variance followed by Bonferroni correction with repeated measures was used to analyze differences in the calcium influx and the IP3concentration. A P value < 0.05 was regarded as significant for all comparisons.
Thrombin-induced Aggregation Curves
Among four aggregation curves induced by thrombin at concentrations of 0.005, 0.010, 0.015, and 0.020 U/ml (n = 8 each), stable aggregation could not be obtained at a thrombin concentration of 0.005 U/ml (fig. 1A). Among the three stable aggregation curves induced by thrombin at 0.010, 0.015, and 0.020 U/ml, the aggregation ratio at 5 min (89 ± 7%, 86 ± 7%, and 89 ± 6%, respectively) did not differ significantly; typical curves are shown in figure 1A. Accordingly, 0.01 U/ml thrombin was the minimum concentration that induced stable platelet aggregation in this study.
We compared aggregation curves induced by 0.01 U/ml thrombin among 10 preparations (three concentrations of three anesthetic agents plus a control group) from each of the eight specimens.
Aggregation ratios at 5 min (table 1) with 1.0 and 1.5 mM halothane were significantly lower than in controls. Isoflurane and sevoflurane had no influence on the aggregation curves. Typical patterns of aggregation are shown in figure 1B.
Changes in Intracellular Calcium Concentration
We compared the increase in [Ca2+]iinduced by 0.01 U/ml thrombin among the 10 preparations described earlier for aggregation curves (n = 8 for each). Baseline [Ca2+]iwas 97 ± 11 nM in controls. With halothane, baseline [Ca2+]iwas 101 ± 9 nM at 0.5 mM; 96 ± 14 nM at 1.0 mM; and 97 ± 14 nM at 1.5 mM. With isoflurane, baseline [Ca2+]iwas 96 ± 9 nM at 0.5 mM; 92 ± 14 nM at 1.0 mM; and 96 ± 10 nM at 1.5 mM. With sevoflurane, baseline [Ca2+]iwas 88 ± 17 nM at 0.5 mM; 93 ± 11 nM at 1.0 mM; and 104 ± 9 nM at 1.5 mM. None of the three anesthetic agents significantly changed baseline intracellular calcium concentrations.
The thrombin-induced increases in [Ca2+]iare shown in table 2. Values for halothane at 1.0 and 1.5 mM were significantly lower than in controls (P < 0.01). Typical patterns for the R340:380 value, indicating [Ca2+]i, are shown in figure 2.
Analysis of Intracellular Calcium Mobilization
When Ni2+was added to the sample instead of CaCl2, the increase in [Ca2+]iwas 220 ± 48 nM in controls; 142 ± 31 nM with halothane at 1.0 mM; 200 ± 58 nM with isoflurane at 1.0 mM; and 206 ± 64 nM with sevoflurane at 1.0 mM. The increase in [Ca2+]iwas significantly lower (P < 0.01) with halothane than with other agents. Typical patterns for aggregation and for changes in the R340:380 value, indicating [Ca2+]i, are shown in figure 3.
Figure 4shows Mn2+entry, expressed as percent decrease of fluorescence at 360 nm. After stimulation by thrombin, fluorescence decreased rapidly for more than 20–30 s and then gradually decreased. The early phase represented Mn2+entry (i.e., Ca2+influx) caused by thrombin stimulation, and the delayed phase was attributable to passive leakage through the plasma membrane. 19,20Decreases in fluorescence showed no significant difference between control conditions and any of the anesthetic agents for either phase.
Intracellular Inositol 1,4,5-Triphosphate Concentration
Time courses of intracellular IP3concentrations in thrombin-stimulated WPS in the absence (control) and in the presence of 1 mM halothane, isoflurane, or sevoflurane are shown in figure 5. After stimulation by thrombin, intracellular IP3concentrations increased instantly within 15 s and then returned rapidly to baseline levels. The IP3concentrations at rest (0 s) were (in pmol/108platelets) 1.6 ± 1.2 in controls, 1.8 ± 1.4 with halothane, 1.3 ± 1.0 with isoflurane, and 1.7 ± 1.1 with sevoflurane. Halothane significantly suppressed the peak value (2.0 ± 1.5 pmol/108platelets at 15 s) to lower than that seen in controls (3.9 ± 1.9 pmol/108platelets). No significant differences in peak concentration of IP3were observed among control, isoflurane, and sevoflurane groups.
Effects of anesthetic agents on platelet function, such as the aggregation curve, the increase in [Ca2+]i, and IP3production, are measured using stimulated platelets by aggregation agonists. Aggregation agonists can be divided into strong agonists, such as thrombin, collagen, and thromboxane A2, and weak agonists, such as ADP, epinephrine, and serotonin, which depend on secretion to obtain a full response. 11Intracellular signals elicited by strong agonists are different from those elicited by weak agonists. 11–13These differences complicate comparison of platelet function induced by different agonists, which could account for several conflicting reports about the effects of anesthetic agents on platelet function. Further, determination of the appropriate concentration of aggregation agonists is critical, because agents do not inhibit platelet aggregation if stimulation is too intense.
First, we established an optimal concentration of thrombin for studies; 0.01 U/ml thrombin was the minimum stimulation needed to induce stable platelet aggregation (fig. 1A). At this concentration of thrombin, we demonstrated dose-dependent inhibition by halothane of the induced increase of [Ca2+]iand of platelet aggregation (a positive control for investigation of sevoflurane) and no effect of isoflurane (a negative control for sevoflurane). These findings for halothane and isoflurane are consistent with those of previous studies. 4,8–10
Hirakata et al. 10have reported that sevoflurane inhibited ADP-induced platelet aggregation and formation of thromboxane A2by suppression of cyclooxygenase activity, concluding that sevoflurane inhibited human platelet aggregation. However, their study was performed on the platelet aggregation induced by weak agonists, such as ADP or epinephrine. We demonstrated here that sevoflurane does not affect the platelet aggregation induced by the strong agonist, thrombin, as there were no changes in the aggregation curve, increase in [Ca2+]i, calcium influx, release from the dense tubular system, or intracellular IP3concentration. Platelet stimulation by strong agonists is mediated by different intracellular events than is stimulation by weak agonists. 11–13For example, the phospholipase C/protein kinase C pathway is involved in human platelet activation induced by thrombin, 22,23although it has little involvement in ADP-induced aggregation. 24,25Clinical platelet aggregation is thought to be induced by combined activation of multiple pathways mediated by thromboxane A2and IP3formation, accompanied by the increase in the [Ca2+]i. 26The inhibition of a single pathway may not result in the net inhibition of platelet aggregation. Platelet aggregation and calcium release induced by strong agonists such as thrombin reportedly are not inhibited by cyclooxygenase inhibitors 27or thromboxane A2receptor blockers. 28Further, amounts of thromboxane A2formed by thrombin-induced platelet aggregation are approximately 10 times greater than those formed in weak agonist–induced platelet aggregation. 29These data strongly suggest that any decrease in calcium mobilization caused by sevoflurane-induced suppression of the formation of thromboxane A2could be compensated for by the activation of the thrombin-mediated pathway, maintaining normal platelet aggregation.
Halothane has been reported to prolong bleeding time 30and to inhibit thrombin-induced platelet aggregation by decreasing [Ca2+]ivia a decrease in IP3. Halothane also inhibits ADP-induced platelet aggregation by suppression of formation of thromboxane A2and binding to its receptors. 8,9We also demonstrated that halothane suppressed production of IP3and inhibited the increase in [Ca2+]iin a dose-dependent manner by suppressing the release of calcium from the dense tubular system in thrombin-induced platelet aggregation, a finding that is in agreement with previous reports. 8,9These data suggest that an inhibitory effect on platelet aggregation can be observed clinically when the thrombin-induced and the ADP-induced platelet aggregation are suppressed, a situation not present with sevoflurane. In our study, isoflurane also did not inhibit thrombin-induced platelet aggregation, which is consistent with previous observations. 4,8–10Isoflurane suppressed neither production of IP3nor the increase in [Ca2+]iin thrombin-induced platelet aggregation, producing no change in influx of calcium or release of calcium from the dense tubular system.
We demonstrated that a high clinical concentration of sevoflurane (1.5 mM corresponds to approximately 5.7%) does not affect thrombin-induced platelet aggregation, although it is reported to inhibit ADP-induced platelet aggregation. Clinically, platelet aggregation is thought to be induced by combined activation of multiple pathways involved in strong agonists and weak agonists separately. Therefore, any inhibition of platelet aggregation by sevoflurane through the ADP-involved pathway does not mean the net inhibition of clinical platelet aggregation.