Beneficial Effect of Propofol on Arterial Adenosine Triphosphate-sensitive K⁺Channel Function Impaired by Thromboxane

The animal care and use committee in Wakayama Medical University (Wakayama, Japan) approved this study. Forty Male Wistar rats (16 to 20 weeks) were obtained from Charles River Japan Inc. (Yokohama, Japan). Rats were anesthetized with inhalation of 3% halothane. Under this anesthetic condition, the rats were killed by exsanguination, and thoracic aortas were harvested. Thoracic aortic rings of 2.5 mm in length were studied in modified Krebs-Ringer’s bicarbonate solution (control solution) of the following composition (in mm): NaCl 119, KCl 4.7, CaCl²2.5, MgSO⁴1.17, KH²PO⁴1.18, NaHCO³25, and glucose 5.5. Endothelial cells were removed mechanically by gentle rubbing of the lumen using a small forceps to avoid the modification mediated by endothelium-derived vasodilator substances as well as endothelial ATP-sensitive K⁺channels.13The removal of endothelial cells was verified by the absence of relaxation in response to acetylcholine (10⁻⁶m).

Eight rings cut from the same artery were studied in parallel. Each ring was connected to an isometric force transducer and suspended in an organ chamber filled with 10 ml of control solution (37°C, pH 7.4) bubbled with 95% O²and 5% CO². The ring was gradually stretched to the optimal point of its length-tension curve, as determined by the contraction to phenylephrine (3 × 10⁻⁷m). In most of studied arteries, optimal resting tension achieved approximately at 1.5 g. Preparations were equilibrated for 90 min. Some rings were treated with an ATP-sensitive K⁺channel antagonist glibenclamide (10⁻⁶m), a superoxide inhibitor Tiron (10 mm), a nonselective NADPH oxidase inhibitor apocynin (1 mm), a hydrogen peroxide scavenger catalase (1200 U/ml), a xanthine oxidase inhibitor allopurinol (10⁻⁴m), propofol (3 × 10⁻⁷or 10⁻⁶m), or dimethyl sulfoxide (2 × 10⁻⁶m). These agents were applied 20 min before the contraction induced by phenylephrine or U46619. During submaximal contraction to phenylephrine (3 × 10⁻⁷m) or a thromboxane A²analogue U46619 (3 × 10⁻⁸m), the concentration-response curves to an ATP-sensitive K⁺channel opener levcromakalim or a voltage-dependent Ca²⁺channel antagonist diltiazem were simultaneously obtained in a cumulative fashion (from 10⁻⁸to 3 × 10⁻⁶m or from 10⁻⁸to 10⁻⁴m, respectively) with 4- to 6-min interval. After the completion of above protocols, all data were calculated. The vasorelaxation was expressed as a percentage of the maximal relaxation in response to papaverine (3 × 10⁻⁴m), which was added at the end of experiments to produce the maximal relaxation (100%) of arteries.

Aortic rings were longitudinally cut and fixed on the bottom of an experimental chamber. The arteries were continuously perfused with control solution (37°C) bubbled with 95% O²–5% CO²gas mixture. A glass microelectrode (tip resistance 40 to 80 MΩ) filled with 3 m KCl and held by a micromanipulator (Narishige, Tokyo, Japan), was inserted into a smooth muscle cell from the intimal side of the vessel.14The electrical signal was amplified using a recording amplifier (Electro 705; World Precision Instruments Inc., Sarasota, FL). The membrane potential was continuously monitored and recorded on a chart recorder (SS-250F-1; SENKONIC Inc., Tokyo, Japan). Membrane potentials in the presence of levcromakalim (10⁻⁷m) in combination with phenylephrine (3 × 10⁻⁷m) or U46619 (3 × 10⁻⁸m) were recorded in arteries treated with or without glibenclamide (10⁻⁶m), Tiron (10 mm), apocynin (1 mm), or propofol (10⁻⁶m). These agents were applied 20 min before membrane potential recordings.

An oxidative fluorescent dye hydroethidine was used to evaluate levels of superoxide in situ  as described previously.12Cells are permeable to hydroethidine; in the presence of superoxide, hydroethidine is oxidized to fluorescent ethidium bromide, in which form it is trapped by intercalation with DNA. This method provides sensitive detection of superoxide levels in situ . Unfixed rat aortas with 20-μm-thick were placed on glass slides. Hydroethidine (2 × 10⁻⁶m) in phosphate-buffered saline (pH = 7.4) was applied to each slice. Slides were incubated in a light-protected chamber at 37°C for 20 min. Images were obtained with a FLUOVIEW FV300 laser scanning confocal microscope (OLYMPUS Inc., Tokyo, Japan) equipped with a krypton/argon laser. Fluorescence was detected with a 585-nm long-pass filter. Laser settings were identical for acquisition of images from all of arteries. Before the application of hydroethidine, arteries were incubated with the normal solution for 20 min in which phenylephrine (3 × 10⁻⁷m) or U46619 (3 × 10⁻¹⁰, 3 × 10⁻⁹, 3 × 10⁻⁸, or 3 × 10⁻⁷m) was added. In some experiments, Tiron (10 mm), apocynin (1 mm), a thromboxane receptor antagonist SQ29548 (10⁻⁶m), allopurinol (10⁻⁴m), or propofol (3 × 10⁻⁷, 10⁻⁶, or 3 × 10⁻⁶m) was also applied. Aortas exposed to hydroethidine in the sole presence of Tiron (10 mm) served as the control, and the fluorescence in the artery was expressed as a ratio.12 

Cytosolic and membranous fractions were prepared and used for Western immunoblotting analysis.15Arteries were incubated in the modified Krebs-Ringer’s bicarbonate solution (37°C, pH = 7.4, control solution) bubbled with 95% O²–5% CO²gas mixture and thereafter quickly frozen (–80°C). Some aortas were incubated for 20 min in the control solution, the solution with the addition of phenylephrine (3 × 10⁻⁷m), U46619 (3 × 10⁻⁸m), or U46619 (3 × 10⁻⁸m) in combination with propofol (10⁻⁶m). Blood vessels were powdered under liquid nitrogen and solublized in ice-cold sterile water (1 ml) containing 0.1% Triton X-100. The lysate was centrifuged at 600g  for 15 min at 4°C, and the supernatant fluid was used for the measurement of total protein levels.16A portion of the supernatant fluid was centrifuged at 100,000g  for 30 min at 4°C, and the pellet was used as a membrane fraction.

Samples (10 μg for membrane protein) were run on 12.5% sodium dodecyl sulfate (SDS) polyacrylamide gels. Blotted membranes were probed for 120 min at 4°C with anti-Nox1, anti-Nox2, anti-Nox4, anti-p47phox (Upstate Cell Signaling, Lake Placid, NY), anti-p22phox (Santa Cruz Biotechnology, Santa Cruz, CA), and anti-adaptin-α (Thermo Fisher Scientific, Rockford, IL) (1 μg/ml each). After washing with phosphate-buffered saline containing 0.05% Tween 20 for 30 min, the membrane was incubated with horseradish peroxidase-conjugated antiimmunoglobulin G antibody (eBioscience, San Diego, CA) diluted at 1:2000 in phosphate-buffered saline–Tween 20 buffer at room temperature for 60 min. The blots were washed three times for 10 min in phosphate-buffered saline -Tween 20 buffer and subsequently visualized with an enhanced chemiluminescence detection system (Amersham GE Healthcare UK Ltd., Little Chalfont, Buckinghamshire, England), exposed to x-ray film, and analyzed by NIH image software produced by Wayne Rasband (National Institutes of Health, Bethesda, MD). To determine loading/transfer variations of protein, all blots were stained with Ponceau Red (washable, before incubation with antibodies) as well as Coomassie Brilliant Blue (permanent, after the enhanced chemiluminescence detection system). Intensity of total protein bands per lane was evaluated by densitometry. Negligible loading/transfer variation was observed between samples.

The following pharmacological agents were used: allopurinol, apocynin, diltiazem, dimethyl sulfoxide, glibenclamide, levcromakalim, phenylephrine, propofol, SQ29548, Tiron, and U46619 (Sigma Aldrich Inc., St. Louis, MO) and hydroethidine (Polysciences Inc., Warrington, PA). Drugs were dissolved in distilled water such that volumes of less than 60 μl are added to the perfusion system. The stock solutions of apocynin, glibenclamide, levcromakalim, propofol, and Tiron were prepared in dimethyl sulfoxide, and the highest concentration of dimethyl sulfoxide in our experimental condition was 1.74 × 10⁻⁶m. The solution containing hydroethidine (7 mg) was produced using N,N-dimethylacetamide (1 ml). The concentrations of drugs are expressed as final molar concentration.

The data were expressed as means ± SD; n refers to the number of rats from which the aorta was taken. Power calculations were done with responses to the reduction of superoxide levels by propofol (3 × 10⁻⁷m) in the aortas treated with U46619 (3 × 10⁻⁸m) as the primary endpoint, and we calculated that a sample size of 5 gave 99% power to detect a change of 13.3% in levels of superoxide at a significance level of 0.05 (SD = 4.3%). Statistical analyses were performed by StatView® version 5.0 (SAS Institute Inc., Cary, NC) and Sample Power 2.0 (SPSS Japan Inc., Tokyo, Japan). Data were evaluated by using repeated-measures of analysis of variance (for comparison among groups in the length-tension study) or a factorial analysis of variance (for comparison among groups in other studies) as appropriate, followed by Student-Newman-Keuls test as a post hoc  analysis. Differences were considered to be statistically significant when P < 0.05.

Levcromakalim concentration-dependently induced relaxation in aortas contracted with phenylephrine (3 × 10⁻⁷m) or U46619 (3 × 10⁻⁸m), whereas the relaxation in those treated with U46619 was attenuated (fig. 1). Glibenclamide (10⁻⁶m) inhibited the relaxation in aortas contracted with either vasoconstrictor agent (fig. 1). Constrictor responses in control aortas (950 ± 100 mg [n = 5] or 933 ± 184 mg [n = 5] for phenylephrine and U46619, respectively) and those in aortas treated with glibenclamide (10⁻⁶m) (920 ± 85 mg [n = 5] and 897 ± 152 mg [n = 5] for phenylephrine and U46619, respectively) did not differ between phenylephrine (3 × 10⁻⁷m) and U46619 (3 × 10⁻⁸m). In addition, glibenclamide did not alter contraction in response to either agent.

Tiron (10 mm) augmented vasodilation in response to levcromakalim in aortas contracted with U46619 (3 × 10⁻⁸m), whereas it did not alter the dilation in those treated with phenylephrine (3 × 10⁻⁷m) (fig. 2). Catalase (1200 U/ml) failed to alter dilation in both aortas contracted with U46619 and those with phenylephrine (fig. 2).

Propofol (3 × 10⁻⁷, 10⁻⁶m) augmented the vasodilator responses to levcromakalim in the aortas contracted with U46619 (3 × 10⁻⁸m) but not those with phenylephrine (3 × 10⁻⁷m) (fig. 3).

In the aortas contracted with U46619 (3 × 10⁻⁸m), apocynin (1 mm), but not allopurinol (10⁻⁴m), augmented the vasodilator responses to levcromakalim, whereas apocynin (1 mm) and Tiron (10 mm) did not alter relaxation in response to diltiazem (fig. 4). Dimethyl sulfoxide (2 × 10⁻⁶m) did not produce any effects on the vasodilation in aortas contracted with U46619 (3 × 10⁻⁸m) (data not shown).

Membrane potential in the presence of levcromakalim (10⁻⁷m) in combination with U46619 (3 × 10⁻⁸m) increased compared with that induced by levcromakalim (10⁻⁷m) in combination with phenylephrine (3 × 10⁻⁷m) (–40.8 ± 1.9 mV and –48.8 ± 2.0 mV, respectively; P < 0.05 compared with phenylephrine). In the presence of levcromakalim (10⁻⁷m) in combination with U46619 (3 × 10⁻⁸m) or phenylephrine (3 × 10⁻⁷m), glibenclamide (10⁻⁶m) significantly produced depolarization, compared with that in aortas without this ATP-sensitive K⁺channel antagonist (–36.2 ± 1.1 mV and –34.8 ± 1.3 mV for condition in the presence of phenylephrine or U46619, respectively; P < 0.05 compared with each control). Tiron (10 mm), apocynin (1 mm), and propofol (10⁻⁶m) similarly induced hyperpolarization in response to levcromakalim (10⁻⁷m) in the presence of U46619 (3 × 10⁻⁸m) (fig. 5), whereas these agents did not affect the resting membrane potential (–44.4 ± 2.7, –42.8 ± 1.9, –44.8 ± 3.6, or –45.2 ± 3.0 mV for control, Tiron, apocynin and propofol, respectively [n = 5 each]; not statistically significant).

U46619 (3 × 10⁻⁹to 3 × 10⁻⁷m) concentration-dependently increased in situ  levels of superoxide within vascular smooth muscle compared with those incubated with U46619 (3 × 10⁻¹⁰m) or phenylephrine (3 × 10⁻⁷m) (fig. 6). Tiron (10 mm), apocynin (1 mm), and SQ29548 (10⁻⁶m), but not allopurinol (10⁻⁴m), similarly reduced levels of superoxide in the aortas exposed to U46619 (3 × 10⁻⁸m) (fig. 6). Propofol (3 × 10⁻⁷to 3 × 10⁻⁶m) concentration-dependently reduced levels of superoxide in these arteries (fig. 6).

Protein expression of NADPH oxidase subunits (p47phox, but not p22phox, Nox2, Nox1, and Nox4) increased in membrane fractions of aortas exposed to U46619 (3 × 10⁻⁸m) but not in those to phenylephrine (3 × 10⁻⁷m) (fig. 7), whereas total protein fractions did not differ among groups (data not shown). The augmentation of expression in arteries treated with U46619 was abolished by propofol (10⁻⁶m) (fig. 7).

In the current study, phenylephrine (3 × 10⁻⁷m) and U46619 (3 × 10⁻⁸m) similarly contracted aortas, whereas the latter solely inhibited the dilation as well as hyperpolarization in response to levcromakalim. In addition, a selective ATP-sensitive K⁺channel antagonist glibenclamide almost abolished dilation and significantly reduced hyperpolarization irrespective of the agent used for contraction.7,17Therefore, these results suggest that a thromboxane A²analogue, but not an α¹-adrenoceptor agonist, impairs the vascular ATP-sensitive K⁺channel function and that this difference between vasoconstrictor agents is not dependent on the degree of contraction each produced. It is of importance to note that increased oxidative stress seen in some pathophysiological conditions induces malfunction of ATP-sensitive K⁺channels in vascular smooth muscle cells.8,9A superoxide inhibitor Tiron and a nonselective NADPH oxidase inhibitor apocynin (which also acts as a superoxide inhibitor), but not a hydrogen peroxide scavenger catalase, recovered ATP-sensitive K⁺channel function in the current study, indicating that superoxide via  NADPH oxidase is a most likely source of the K⁺channel inhibition in blood vessels induced by thromboxane A².8,18 

Vascular smooth muscle cells contain several superoxide-producing pathways, among which NADPH oxidase is predominant.2Findings in cultured cells from pulmonary arterial endothelium or corpus cavernosum documented that acute exposure to a thromboxane A²mimetic predisposes to superoxide production in these cells via  NADPH oxidase activation, including the membrane translocation of gp91phox (Nox2) or p47phox.4–6Considering these previous results, we have conducted the current study to simultaneously evaluate roles of major NADPH oxidase subunits related to Nox2 activation, including Nox2, p22phox, and p47phox, in the oxidative stress produced by thromboixane A²in the intact artery.1,2As we expected, U46619, but not phenylephrine, increased levels of superoxide within the vascular smooth muscle, and this augmentation was completely inhibited by a superoxide scavenger Tiron, a nonselective NADPH oxidase inhibitor apocynin, or a thromboxane receptor antagonist SQ29548.8,18,19In the current study, protein expression of NADPH oxidase subunits was also examined because recent studies have questioned the selectivity of apocynin as a NADPH oxidase inhibitor.18We have confirmed that the expression of p47phox is solely augmented by the exposure toward U46619. It is well known that interactions between membrane-bound subunits such as Nox2 or p22phox and cytosolic subunits, including p47phox are essential upon activation of Nox2 and that assembly of p47phox is most important among these to support the action of Nox2.1Indeed, our results that membrane translocation of p47phox was predominant upon the application of thromboxane A²are consistent with a previous study on cultured cells form corpus cavernosum.4Therefore, it can be concluded that thromboxane receptor activation in the intact arterial smooth muscle increases levels of superoxide resulting from activation of a Nox2-related NADPH oxidase subunit p47phox. However, the signal transduction pathway to produce p47phox translocation coupling the activation of thromboxane receptor remains to be determined.

Propofol reportedly possesses potential antioxidant properties caused by its chemical structure similar to that of phenol-based free-radical scavengers such as vitamin E.10Previous in vivo  or in vitro  studies documented that this intravenous anesthetic reduces oxidative stress toward blood vessels.11,12These results suggest that this anesthetic may be protective against the vascular dysfunction caused by increased oxidative stress. Indeed, propofol (3 × 10⁻⁷to 10⁻⁶m) recovered vascular ATP-sensitive K⁺channel function via  reduction of superoxide levels within arterial walls. In addition, 10⁻⁶m of this agent completely inhibited protein expression of a Nox2-related NADPH oxidase subunit p47phox. The plasma concentration of propofol during induction of anesthesia in humans has been reported as up to 3 × 10⁻⁵m, and burst suppression doses of propofol for cerebral protection are up to 6 × 10⁻⁵m.20–22Effective concentrations of propofol (3 × 10⁻⁷to 10⁻⁶m) to inhibit NADPH oxidase activity, resulting in restoration of vascular ATP-sensitive K⁺channel function are probably within clinical range, even if one considers plasma-free concentrations calculated from both above clinical plasma concentrations and the substantial binding of this compound to plasma proteins (from 97 to 98%).21Therefore, our results strongly indicate that clinically relevant concentrations of propofol can play a role as a NADPH oxidase inhibitor. It is crucial to note the difference between results in the current study and our previous study showing the inhibition of propofol on vasorelaxation in response to ATP-sensitive K⁺channel openers.23Upon the comparison of these studies, it has become evident that higher plasma-free concentration of propofol (above 10⁻⁵m) is probably needed to inhibit the K⁺channel function, indicating the benefit of this intravenous anesthetic in the clinical practice considering its protective effect toward oxidative stress in blood vessels. A recent study showing that propofol enhances plasma γ–tocopherol levels in surgical patients, also suggests a multi-mode of action of this agent regarding the organ protection24; further studies are required to clarify the whole scenario why this anesthetic is protective toward oxidative stress.

The current study was conducted in aortic rings as well as the vascular smooth muscle. It is still unclear whether the beneficial effects of propofol seen in our study have any relevance to pathogenesis in such a conduit artery, although recent studies have demonstrated that increased oxidative stress induced by NADPH oxidase plays a crucial role in the aortic atherosclerosis.25It is also true that smaller resistance arteries greatly contribute to the regulation of hemodynamics in vivo ; therefore, further studies are warranted to clarify that propofol similarly reduces oxidative stress in these peripheral vasculatures. In addition, it should be noted that throughout the current study, higher oxygen (hyperoxia) was used to incubate aortic rings because we intended to make a comparison of the current data with our previous study.23Such a higher concentration of oxygen may modify the propofol effects toward oxidative stress. Indeed, a previous study demonstrated that lower oxygen tension is capable of inducing augmentation of a superoxide-inactivating system in the blood vessels.26However, it remains unclear whether changes of oxygen tension influence the results obtained in the current study.

The current study has demonstrated the following new findings. First, thromboxane A²induces overproduction of superoxide via  a Nox2-related NADPH oxidase subunit p47phox in vascular smooth muscle cells of the intact artery. Second, increased levels of superoxide produced by thromboxane A²impair ATP-sensitive K⁺channel function in the vascular smooth muscle. Third, clinically relevant concentrations of propofol restore the channel function acting as a NADPH oxidase inhibitor. Thromboxane A²reportedly plays a role in a variety of cardiovascular diseases, including myocardial infarction, cerebral vasospasm, hypertension, preeclampsia, and thrombotic disorders.3,27,28Accumulating findings demonstrated that ATP-sensitive K⁺channels contribute to pathophysiological vasodilation during hypoxia, acidosis, and inflammation.7,29–31These findings together with our results suggest possibilities that thromboxane receptor activation modifies ATP-sensitive K⁺channel function in above pathophysiological conditions by modulating levels of superoxide via  NADPH oxidase, and that clinically used propofol may be protective against vascular malfunction related to such pathogenesis.