INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Keywords: antioxidant; ascorbic acid; erythrocytes; cis-parinaric acid; propofol; recycling

We and other researchers have found that propofol (2,6-diisopropylphenol), a novel sedative and anesthetic agent, has potent activity as an antioxidant (1). The characteristics of the hindered phenolic structure of propofol, which are similar to those of tocopherols, butylated hydroxytoluene (BHT), and butylated hydroxyanisole (BHA) (5), might seem to account for the antioxidant activity (3) (Figure 1). These phenolic antioxidants scavenge reactive oxygen species (ROS) to generate less reactive phenoxyl radicals. Kagan and colleagues reported the importance of recycling of phenoxyl radicals to the phenolic form by ascorbic acid, a water-soluble antioxidant (ascorbate-driven recycling of tocopherols) (Figure 2A) (8). This reaction enhances the favorable effects of phenolic antioxidants and reduces their toxic pro-oxidant effects. Thus, the interaction of propofol with ascorbic acid is of importance for its availability as an antioxidant.

View larger version (16K): [in this window] [in a new window]   Figure 1.   Structures of propofol (2,6-diisopropylphenol), BHT (butylated hydroxytoluene), and -tocopherol.

View larger version (19K): [in this window] [in a new window]   Figure 2.   (A) Scheme showing ascorbate-driven recycling of -tocopherol. (B) Scheme showing possible interactions of ascorbic acid with propofol radicals generated by AMVN.

Propofol is usually used for patients undergoing surgery or with critical illnesses that abnormally enhance oxidative stress (3, 9). Under such pathological conditions, erythrocytes become one of the critical targets for free radicals generated in cells and tissues (10). Moreover, these oxidative stresses are sometimes accompanied by deleterious shear mechanical stresses due to instability of osmotic pressure, body fluid imbalance, and circulatory disturbance (13, 14). It has been reported that some phenolic antioxidants function as potent modifiers of biomembranes (15, 16). Considering the structural similarity of propofol to those of phenolic antioxidants, this anesthetic may protect erythrocytes by directly scavenging ROS and/or increasing the resistance of their membranes to hazardous mechanical stresses. Thus, in the present study, we sought to clarify the protective effect of propofol on the oxidative injury of erythrocytes in the presence and absence of ascorbic acid and also to examine its physical effect on erythrocytes as a membrane modifier.

Eight sets of in vitro and clinical studies were designed. First, we investigated the effects of propofol on (1) ROS-induced hemolysis and (2) changes in oxidative stress in erythrocytes by using cis-parinaric acid (a sensitive fluorescent probe for oxidative stress). We then determined the generation of propofol radicals and/or its interaction with ascorbic acid in (3) erythrocyte membranes and in (4) the circulating blood, using electron spin resonance (ESR) to confirm and investigate the results of Studies 1 and 2 in more detail. These four experiments were performed to study the antioxidant effects of propofol. Second, we performed measurements of the effects of propofol on (5) osmotic and (6) mechanical stress-induced hemolysis, and (7  ) membrane fluidity (determined by ESR). All three experiments were performed to examine the physical effects of propofol on erythrocytes. Finally, (8) a clinical study of patients undergoing surgery was performed to confirm the validity of the in vitro effects under in vivo conditions.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Materials and Animals

cis-Parinaric acid was obtained from Sigma (St. Louis, MO), and 2,2'-azobis (2,4-dimethylvaleronitrile) (AMVN) and 2,2,6,6-tetramethyl-1-piperidinyloxy (TEMPO) were purchased from Wako Pure Chemical Industries (Osaka, Japan). Diprivan (the medical product of 2,6-diisopropylphenol dissolved in a lipid emulsion) was obtained from Astra-Zeneca (Osaka, Japan), and 2,6-diisopropylphenol (propofol) and 5-doxylstearic acid (5-DSA) were purchased from Aldrich (Milwaukee, WI). Other chemicals used were of the highest grade commercially available. Oxygen-permeable special Teflon tubing for ESR measurements was obtained from Zeus Industrial Products (Orangeburg, SC). Male scurvy-prone osteogenic disorder Shionogi (ODS) rats were obtained from Clea (Osaka, Japan).

Preparation of Human Erythrocytes

Fresh heparinized blood from healthy volunteers was centrifuged at 750 × g for 5 min at 4° C (3). After removing plasma and buffy coat, pellets were washed three times with phosphate-buffered saline (PBS) at pH 7.4, and resuspended in PBS.

Preparation of Human Erythrocyte Ghosts

Fresh human erythrocyte samples, prepared as described above, were lysed in an 8-fold volume of hypotonic phosphate buffer (pH 7.4) containing 100 µM EDTA at 37° C (3). After 45 min, erythrocyte ghosts were washed five times with the same hypotonic buffer by centrifugation at 30,000 × g and 4° C for 20 min. The washed membranes were resuspended in PBS.

Analysis of ROS-induced Hemolysis

Hemolysis of erythrocytes was induced by ROS as described previously (17). Fresh erythrocytes suspended in PBS (hematocrit of 5%) were incubated with 20 mM AMVN at 37° C. After incubation for varying times, each sample was centrifuged at 750 × g and 4° C for 10 min. The absorbance at 415 nm of the supernatant was measured spectrophotometrically in a Hitachi (Tokyo, Japan) U-2000. The extent of hemolysis was expressed as the percentage relative to the complete hemolysis of samples in deionized water.

Analysis of Oxidative Injury of Erythrocytes, Using cis-Parinaric Acid

The assay, which can monitor the peroxidation process in its initial stage in membranes, was first developed by Van den Berg and colleagues (18) and was advanced by Tsuchiya and coworkers (3, 5, 6). It is based on the special structure of the cis-parinaric acid molecule and its conjugated double-bond system, consisting of four sets of double bonds, which allow for fluorescent properties and a high susceptibility to peroxidation.

Erythrocyte ghosts were sonicated under nitrogen gas at 4° C for 10 s in the presence of AMVN to allow the incorporation of the compound in the membranes. cis-Parinaric acid in ethanol was then added to the suspension. Because cis-parinaric acid was quickly and almost thoroughly incorporated into the ghost membranes, the change in fluorescence intensity observed in the suspension reflected the magnitude of oxidative stress incurred. The reaction mixtures contained, in a final volume of 2 ml of PBS, erythrocyte ghosts (0.2 mg of protein), 2.5 µM cis-parinaric acid, and 5 mM AMVN. Fluorescence intensity was measured at 40° C in a Hitachi F-4500 fluorescence spectrophotometer, using a thermostatted cuvette equipped with a magnetic stirrer. The excitation and emission wavelengths were 326 and 416 nm, respectively. Previously, the presence of both 2,6-diisopropylphenol and AMVN in the ghost suspension was found to increase its turbidity greatly, resulting in difficulties in detecting the precise change of fluorescence. However, advances in computerized data processing have enabled the actual signal to be distinguished clearly from the background noise.

Analysis of Propofol Phenoxyl Radical and Ascorbyl Radical, Using ESR

Recycling of propofol phenoxyl radicals by ascorbic acid was analyzed by an ESR technique, according to the method used to demonstrate the recycling of tocopherols described by Scarpa and coworkers (19) and further developed by Packer and coworkers (8, 20). In the present study, AMVN, 2,6-diisopropylphenol (or Diprivan), and ascorbic acid were added sequentially to the erythrocyte ghosts, and generated radical signals were analyzed at each step by ESR. Although the sensitivity of the ESR measurement is not high, it is useful to investigate the species of generated radicals in detail for identification.

First, AMVN was mixed with an erythrocyte ghost suspension and then briefly sonicated under nitrogen gas at 4° C. Next, 2,6-diisopropylphenol (or Diprivan) was added to the sample. The reaction mixtures contained erythrocyte ghosts (2 mg of protein · ml¹), 130 µM 2,6-diisopropylphenol (or Diprivan), and 200 mM AMVN in PBS. Aliquots of 120 µl were transferred to oxygen-permeable Teflon tubing (0.8-mm i.d. × 12 cm). The tubing was folded into quarters, placed in an ESR quartz tube with a 3-mm open diameter, and then analyzed at 40° C under 80% oxygen with an X-band ESR spectrometer (model JES-RE1X; JEOL, Tokyo, Japan). When ascorbic acid was added to the reaction mixture, the samples were carefully and quickly removed from the tubing, mixed well with ascorbic acid, and returned to the tubing. The ESR spectra were recorded at a modulation amplitude of 0.79 mT, a modulation frequency of 100 kHz, a microwave power of 8 mW, a time constant of 0.1 s, a center field of 336 mT, a receiver gain of 5 × 10³, and a sweep time of 1.25 mT · min¹. For the analysis under alkaline conditions (pH 9), a modulation amplitude of 0.079 mT and a receiver gain of 2 × 10³ were used. A g value of the observed radicals was directly obtained from the scan field using MnO as an external standard. After the reaction, concentrations of ascorbic acid in the mixture were determined by high-performance liquid chromatography (HPLC), as described previously (23).

Direct Detection of Circulating Free Radicals in Vivo, Using ESR

The method employed, which was developed and advanced by Mori and our research team (23, 24), allowed us to detect in vivo free radical formation directly in the circulating blood of living animals, using ESR spectrometry without any labeling or trapping agents. Briefly, male ODS rats (200-250 g), which inherently lack the ability to synthesize ascorbic acid (25), were anesthetized with pentobarbital sodium. Catheters were inserted into the femoral artery and vein, and the blood was extracorporeally circulated through an ESR quartz flow cell (volume, 160 µl; JEOL), placed in an ESR cavity at 38° C. A small amount of heparin (100 units/rat) was used as an anticoagulant. Ascorbic acid (0.15 M) was continuously infused at a rate of 20 µl · min¹. Signals for the ascorbyl radical in the circulating blood were recorded in a JEOL X-band spectrometer (model JES-RE1X) at a modulation amplitude of 0.079 mT, a modulation frequency of 100 kHz, a microwave power of 8 mW, a time constant of 0.3 s, a center field of 334.6 mT, a receiver gain of 1 × 10³, and a sweep time of 3.3 mT · min¹. The concentration of the ascorbyl radical was directly determined by a comparison with the amplitude of the ESR signal from a known concentration of TEMPO.

Analysis of Hemolysis of Erythrocytes Induced by Osmotic Stress

Fresh human erythrocytes (20 µl, hematocrit of 35%) were mixed with 2 ml of hypotonic PBS solution, and incubated in a shaking water bath at 37° C for 30 min (26). After incubation, the samples were centrifuged at 750 × g for 10 min at 4° C, and the extent of hemolysis was determined spectrophotometrically as described above.

Analysis of Hemolysis of Erythrocytes Induced by Mechanical Stress

Fresh human erythrocytes in PBS (hematocrit of 35%) were circulated in 30 cm of Teflon tubing (80 µm in diameter) at a flow rate of 2 ml · min¹ at 37° C by a Perista minipump (model SJ-1211; ATTO, Tokyo, Japan). After 30 min of circulation, the extent of hemolysis was measured spectrophotometrically.

Analysis of Erythrocyte Membrane Fluidity by ESR

Stearic spin label, 5-DSA, which is an analog of stearic acid possessing a nitroxide radical ring at the carbon occupying position 5 of the acyl chain, has been used as a probe to analyze membrane fluidity (15, 27). An ethanol solution of 5-DSA (0.1 mg · ml¹) was dried in a test tube (0.1 ml/tube) under nitrogen gas to form a thin film. The human erythrocyte suspension in PBS (hematocrit of 50%) was then added to the test tube and incubated at 37° C for 20 min with gentle shaking to incorporate the spin label into membranes. The labeled erythrocytes were washed twice with PBS to remove unbound spin label and resuspended in PBS (hematocrit of 50%). The spin-labeled erythrocytes thus obtained were incubated with various concentrations of 2,6-diisopropylphenol (or Diprivan) at 37° C for 10 min, and 50 µl of the suspension was transferred to a glass capillary for ESR analysis. The spectra were recorded at 25° C at a modulation amplitude of 0.16 mT, a modulation frequency of 100 kHz, a microwave power of 8 mW, a time constant of 0.3 s, a center field of 336 mT, a receiver gain of 2 × 10³, and a sweep time of 0.25 mT · min¹.

Clinical Study: Analysis of Red Cell Counts before and after Surgical Operation

Twenty-eight patients who were scheduled for total gastrectomy and classified as ASA (American Society of Anesthesiologist) physical status I or II were studied. They were randomly divided into two groups; one group was anesthetized with propofol and the other was anesthetized with sevoflurane. Both groups were equally treated with continuous epidural infusion of lidocaine during the operation, and other anesthetic parameters were identical between the two groups. Blood cells were counted before the start of anesthesia and 10 h after the end of anesthesia for each patient.

Approval and Data Processing

All protocols were approved by the Research Committee of Osaka City University Medical School. Experiments were repeated at least five times, and data are expressed as means ± SEM (standard error of the mean). Statistical analysis was performed by analysis of variance (ANOVA), and a p value of less than 0.05 was considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Antioxidant Effects of Propofol on Erythrocytes

ROS-induced hemolysis. When incubated in control PBS at 37° C for 12 h, human erythrocytes showed a negligible amount of hemolysis. However, in the presence of lipophilic radical initiator AMVN, they showed marked hemolysis (Figure 3). The time required for 50% hemolysis (HT₅₀) was 95 min. The 2,6-diisopropylphenol effectively suppressed the AMVN-induced hemolysis (HT₅₀ = 230 min), suggesting that it protects erythrocytes by acting as an antioxidant. Although ascorbic acid, a water-soluble antioxidant, scarcely suppressed the AMVN-induced hemolysis, the presence of both ascorbic acid and 2,6-diisopropylphenol significantly delayed hemolysis (HT₅₀ = 340 min) compared with 2,6-diisopropylphenol alone, suggesting that ascorbic acid enhanced the antioxidant activity of 2,6-diisopropylphenol. Diprivan also showed the same effect as 2,6-diisopropylphenol.

View larger version (26K): [in this window] [in a new window]   Figure 3.   Effect of 2,6-diisopropylphenol on AMVN-induced oxidative hemolysis and the synergistic effect with ascorbic acid. Fresh human erythrocytes were suspended in PBS at a hematocrit of 5% and then incubated at 37° C with 20 mM AMVN, in the presence or absence of 50 µM 2,6-diisopropylphenol (or Diprivan) or 100 µM ascorbic acid as indicated. Hemolysis was spectrophotometrically analyzed, as described in METHODS.

Changes in oxidative stress of erythrocytes determined by cis-parinaric acid fluorescence. Figure 4 shows the excitation and emission spectra of cis-parinaric acid during oxidation by AMVN. Because the fluorescence spectra of cis-parinaric acid were clearly separated from the basal levels, the decrease in the fluorescence intensity could be ascribed to the decomposition of cis-parinaric acid.

View larger version (27K): [in this window] [in a new window]   Figure 4.   Excitation and emission spectra of cis-parinaric acid incorporated into erythrocyte membranes at the beginning and after completion of oxidation by AMVN (20 min). The reaction mixtures contained erythrocyte ghosts (0.1 mg of protein · ml1), 2.5 µM cis-parinaric acid, and 5 mM AMVN in a volume of 2 ml of PBS. Fluorescence intensity was measured at an emission wavelength of 426 nm and an excitation wavelength of 326 nm at 40° C.

View larger version (29K): [in this window] [in a new window]   Figure 5.   Time course of changes in fluorescence intensity of cis-parinaric acid incorporated into erythrocyte membranes, and the inhibitory effect of 2,6-diisopropylphenol on cis-parinaric acid oxidation. Experimental conditions are described in the legend to Figure 3. Inset: Dose dependency of the inhibition of 2,6-diisopropylphenol on cis-parinaric acid oxidation.

Detection of propofol radical in erythrocyte membranes. The mode of interaction of ascorbic acid and propofol during the oxidation of erythrocyte membranes was analyzed by an ESR technique. When 2,6-diisopropylphenol was added to the AMVN-incorporated erythrocyte membranes, a broad ESR signal with a g value of 2.005 was observed, although high amplitudes of modulation (such as 0.79 mT) were required for detection (Figure 6A). In the absence of AMVN, no such signal appeared (spectrum not shown). Thus, it is likely that the observed radical signal was derived from 2,6-diisopropylphenol, which may have scavenged the peroxyl radicals generated in the AMVN-initiated lipid peroxidation of erythrocyte membranes. Ascorbic acid transiently eliminated this broad signal with the concomitant appearance of another large signal representing the ascorbyl radical. The large signal decreased slowly with concomitant recovery of the initial broad signal. At this time, ascorbic acid in the reaction mixture was not detectable by HPLC, indicating that it was consumed completely. The same sequence of ESR spectra was also observed with Diprivan (data not shown).

View larger version (32K): [in this window] [in a new window]   Figure 6.   ESR spectra of phenoxyl radicals of 2,6-diisopropylphenol generated by AMVN-induced oxidation of erythrocyte membranes, and the effect of ascorbic acid on the 2,6-diisopropylphenol radicals. (A) Spectra at neutral pH. (B) Spectra under alkaline conditions. The reaction mixtures contained erythrocyte ghosts (2 mg of protein · ml1), 130 µM 2,6-diisopropylphenol, and 200 mM AMVN in PBS, and 30 mM ascorbic acid was added during the course of the reaction. The spectra in (A) were recorded at 40° C and pH 7.4 under approximately 80% oxygen with an X-band ESR spectrometer (model JES-RE1X; JEOL, Tokyo, Japan) at a modulation amplitude of 0.79 mT, a modulation frequency of 100 kHz, a microwave power of 8 mW, a time constant of 0.1 s, a center field of 336 mT, a receiver gain of 5 × 103, and a sweep time of 1.25 mT · min1. The spectra in (B) were recorded at nearly pH 9, with a modulation amplitude of 0.079 mT and a receiver gain of 2 × 103.

Propofol-induced ascorbyl radical formation in circulating blood in vivo. To test the possible interaction between ascorbic acid and propofol in vivo, free radicals in the circulating blood of living rats was analyzed, using real-time monitoring by an ESR spectrometer. As reported previously, when ascorbic acid was continuously infused into the circulation, a typical double-peak spectrum for the ascorbyl radical was observed in the circulating blood (Figure 7) (23, 24). After injection of 5 µmol of Diprivan, the signal for the ascorbyl radical increased significantly (0.175 ± 0.007 µM), and gradually decreased thereafter.

View larger version (19K): [in this window] [in a new window]   Figure 7.   ESR spectra of ascorbyl radicals generated in the circulating blood of living ODS rats while receiving 0.15 µM ascorbic acid at a rate of 20 µl · min1. Propofol was given to rats by injection of 10 µl of Diprivan. Spectra were recorded at 38° C with an X-band ESR spectrometer (model JES-RE1X; JEOL) at a modulation amplitude 0.079 mT, a modulation frequency of 100 kHz, a microwave power of 8 mW, a time constant of 0.3 s, a center field of 334.6 mT, a receiver gain of 1 × 103, and a sweep time of 3.3 mT · min1. Insets: Actual ESR spectra of ascorbyl radicals at the indicated time points.

Physical Effects of Propofol on Erythrocytes

Hypotonic treatment of erythrocytes. Hemolysis of human eryth-rocytes was induced by hypotonic stress (Figure 8). Fifty percent hemolysis (HC₅₀) was induced by a 41% isotonic solution. Erythrocytes showed high resistance to hypotonic conditions in the presence of 2,6-diisopropylphenol (HC₅₀ = 37%). The same effects were also observed with Diprivan.

View larger version (16K): [in this window] [in a new window]   Figure 8.   Effect of 2,6-diiso- propylphenol on hypotonic hemolysis. Fresh human erythrocytes were suspended in PBS at a hematocrit of 35% and incubated at 37° C for 30 min with the various osmolar solutions, in the presence or absence of 50 µM 2,6-diisopropylphenol (or Diprivan). Hemolysis was spec- trophotometrically analyzed, as described in METHODS.  control;  2,6-diisopropylphenol;  Diprivan.

Hemolysis induced by mechanical stress. Hemolysis of human erythrocytes was also induced by mechanical stress present in the model circulation. After 30 min of circulation at a rate of 2 ml · min¹, 2.0 ± 0.1% of the erythrocytes were hemolyzed. The presence of 2,6-diisopropylphenol significantly decreased this hemolysis by 8.3%, as compared with the control (Figure 9).

View larger version (21K): [in this window] [in a new window]   Figure 9.   Effect of 2,6-diiso- propylphenol on hemolysis induced by mechanical stress. Fresh human erythrocytes in PBS (hematocrit of 30%) were circulated for 30 min at a rate of 2 ml · min1 at 37° C in a model circulatory system equipped with a resistance circuit, in the presence or absence of 30 µM 2,6-diisopropylphenol (or Diprivan). After 30 min, hemolysis was spectrophotometrically analyzed.

Membrane fluidity of erythrocytes. The inset of Figure 10 shows an actual ESR spectrum of 5-DSA embedded in erythrocyte membranes, which represents the rapid anisotropic motion of the spin probes (15). The values of outer (2T_//  ) and inner (2T  ) hyperfine splitting obtained from this spectrum gave the order parameter (S), which indicated the degree of motion of a spin probe intercalated into a phospholipid acyl chain, and reflected the dynamic state of the membrane molecules. Practically, the order parameter was calculated according to the following equation:

View larger version (19K): [in this window] [in a new window]   Figure 10.   Effect of 2,6-diiso- propylphenol on the membrane fluidity of erythrocytes. The stearic spin label 5-DSA was incorporated into erythrocyte membranes, as described in METHODS. ESR spectra of incorporated 5-DSA (shown in the inset), were recorded at 25° C at a modulation amplitude of 0.16 mT, a modulation frequency of 100 kHz, a microwave power of 8 mW, a time constant of 0.3 s, a center field of 336 mT, a receiver gain of 2 × 103, and a sweep time of 0.25 mT · min1 under various concentrations of 2,6-diisopropylphenol (or Diprivan). Order parameters were then calculated from the values of T//   and T, using the formula shown in RESULTS.  Diprivan;  2,6-diisopropylphenol.

Clinical Study

Pre- and postoperative red cell counts in patients receiving propofol and sevoflurane anesthesia: The mean age, height, and weight of the patients were 59 ± 4 yr, 158 ± 3 cm, and 54 ± 1 kg, respectively, in the propofol group, and 58 ± 3 yr, 157 ± 3 cm, and 52 ± 3 kg, respectively, in the sevoflurane group. Duration of anesthesia, urine output, and fluid intake were 264 ± 12 min, 482 ± 55 ml, and 3,183 ± 143 ml, respectively, in the propofol group, and 247 ± 16 min, 470 ± 50 ml, and 2,830 ± 287 ml, respectively, in the sevoflurane group. No significant differences in these parameters were present between the two groups. A blood transfusion was not performed in any patient in either group during the perioperative period. Figure 11 shows pre- and postoperative red blood cell (RBC) counts and intraoperative blood loss for both groups. Preoperative RBC counts and blood loss were not significantly different between the two groups, whereas the postoperative RBC count in the propofol group was significantly higher than in the sevoflurane group. This finding indicated that the difference in postoperative RBC count between the two groups might have been induced by a mechanism based on the differences in the anesthetic agents used.

View larger version (34K): [in this window] [in a new window]   Figure 11.   Red blood cell (RBC) counts before the start of anesthesia and 10 h after the end of anesthesia, and intraoperative blood loss in the propofol anesthesia group (n = 14) and sevoflurane anesthesia group (n = 14). All patients were scheduled for a total gastrectomy, and the operations were uneventful, completed without a blood transfusion. Pre-operative RBC (cells/µl); Post-operative RBC (cells/µl); Blood loss during surgery (ml).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

AMVN, a lipophilic diazo radical initiator, generates peroxyl radicals at a constant rate for several hours during its thermal decomposition in the presence of oxygen, as shown in reactions (1)-(3):

(1)

(2)

(3)

where R represents (CH₃)₂CHCH₂C(CH₃)CN for AMVN (3, 28). Because of the lipophilic nature of AMVN, this compound binds to membranes, forms free radicals, and catalyzes radical chain reactions of unsaturated membranous lipid (3, 5, 28), thereby causing hemolysis (17).

The observation that propofol suppressed AMVN-induced hemolysis (Figure 3) indicates that propofol inhibited the chain reaction initiated by the free radical generation. Because ascorbic acid exhibited its protective effect only in the presence of propofol, ascorbic acid might interact with propofol rather than scavenging free radicals derived from AMVN. This hypothesis is confirmed by the experiments using cis-parinaric acid (Figure 5), which could reveal the change in oxidative stress in membranes with high sensitivity and continuity. Although an artificial effect induced by incorporation of the probe into membranes should be considered when interpreting the results of the cis-parinaric acid study, the agreement of results between the hemolysis and fluorescence studies strengthen the validity of the observed paradoxical antioxidant actions of propofol and ascorbic acid in erythrocytes.

The reason why ascorbic acid alone failed to protect erythrocytes may be explained by its pro-oxidant activity (29, 30). However, the pro-oxidant activity of ascorbic acid often depends on the presence and kind of free forms of transition metals that function as redox cyclers (29). Because free metals were not required in the present AMVN experiments, the pro-oxidant activity might not underlie the present results. Moreover, the concept of pro-oxidant activity does not seem to explain the fact that ascorbic acid enhances the antioxidant effect of propofol in both oxidative hemolysis and cis-parinaric acid fluorescence studies.

It should be noted that the efficacy of antioxidants to scavenge free radicals highly depends on physicochemical properties and subcellular localization of the ligands near the sites of ROS generation (3, 5, 6). Ascorbic acid is water soluble, unlike propofol, which is amphipathic. Thus, it is not surprising that ascorbic acid alone could not inhibit membrane oxidation occurring within lipid membranes of erythrocytes (28).

Enhancement of the protective effect of propofol by ascorbic acid might be explained by another mechanism demonstrated between ascorbic acid and tocopherols (ascorbate-driven recycling of tocopherols) (8, 20). To verify this hypothesis, radical generation during the oxidation of erythrocyte membranes was studied in the presence of propofol, using an ESR technique (Figure 6). The observed spectra suggested the occurrence of propofol radicals. However, they are different from those reported by Murphy and colleagues in an acetone solution using an ultrarapid mixing ESR system (2). It should be noted that phenoxyl radicals can easily form a diradical compound, as shown in reaction (4),

(4)

where P-O^· represents the phenoxyl radical (31, 32). This diradical compound is stable and theoretically gives a five-line ESR spectrum. Thus, the ESR signals in our experiment might reflect the occurrence of diradicals of propofol. Similar ESR spectra derived from propofol oxidation were also observed by Eriksson and colleagues (1).

On the basis of these ESR experiments with erythrocyte membranes in the presence of ascorbic acid, propofol might seem to be regenerated from its radical at the expense of ascorbic acid, as follows,

(5)

where P-OH and P-O^· represent propofol and its radical, respectively, and AH and A^· represent ascorbic acid and its radical, respectively. The gradual decrease in the signal for ascorbyl radical with concomitant appearance of the signal for propofol radical after the consumption of ascorbic acid also indicate the interaction of propofol and ascorbic acid.

We also analyzed free radical generation in vivo, using ODS rats, which lack the ability to synthesize ascorbic acid (25). This unusual feature for rats is suitable for the investigation of ascorbate metabolism, as humans also lack this ability. Under physiological conditions, ascorbic acid is continuously auto-oxidized in vivo to maintain a steady state concentration of ascorbyl radical in the circulation (23, 24). Because of the fairly low concentrations of ascorbic acid in plasma, ascorbic acid was infused to analyze ESR signals in the circulation. Although it was practically difficult to detect propofol radical as shown in Figure 6, treatment with propofol increased the signal for ascorbyl radical in the circulation (Figure 7), indicating the presence of the interaction of propofol (radical) and ascorbic acid even in vivo, as given by reaction (5).

The putative reaction between propofol and ascorbic acid is summarized in Figure 2B. This regeneration of propofol from its radical is important for the antioxidant activity of this anesthetic to protect erythrocyte membranes, as shown by the interaction between ascorbic acid and tocopherols (8).

The present work also showed that erythrocytes become more resistant to physical stress (Figures 8 and 9) in conjunction with an increase in membrane fluidity (Figure 10) caused by treatment with propofol. Araki and Rifkind reported the importance of membrane fluidity in the stability of erythrocytes (33). It is known that erythrocytes become resistant to hypotonic condition and detergent treatment when membrane fluidity is favorably increased by treatment with tocopherols (15, 34, 35). This finding suggests that a slight increase in membrane fluidity might have a beneficial effect on the stability of erythrocytes. Propofol seems to protect erythrocyte membranes as a membrane stabilizer by the same mechanism.

It should be noted that some experiments in the present study were performed at temperatures different from 37° C. However, temperature differences were the result of the use of the optimum temperature for each experiment, and all temperatures but one were between 37 and 40° C; not extremely far from physiologic range. Only the study for membrane fluidity was performed at 25° C. In that case, the value of the order parameter at 37° C is probably smaller than that at 25° C, which means the fluidity of erythrocyte membranes is higher at 37° C than at 25° C. However, the relation to propofol concentration at 37° C is theoretically analogous to that at 25° C; thus, it seems certain that propofol could increase the fluidity of erythrocyte membranes, even at physiologic temperatures.

The protective effect of propofol on erythrocyte membranes led us to consider verification of its clinical effectiveness. Because a surgical operation generally enhances the generation of ROS and increases mechanical shear stress of vascular endothelial cells and erythrocytes, propofol may have therapeutic potential in patients who undergo a surgical operation. Although many factors must be taken into consideration in this clinical setting, the fact that the decrease in RBC count after surgical operation was smaller in patients who received propofol anesthesia as compared with those given sevoflurane (Figure 11) is consistent with the notion described above. In addition, volatile anesthetics, such as halothane, are reported to decrease the stability of erythrocytes (36, 37). Such a negative effect of volatile anesthetics, which might also be true for sevoflurane, might further increase the difference in RBC counts between the propofol and sevoflurane groups.

In conclusion, the present findings demonstrate that propofol protects erythrocytes against oxidative stress by the aid of ascorbate-driven recycling, as well as against physical stress by increasing membrane fluidity, indicating that propofol may have a high potential as an efficient antioxidant.