Abstract

Objectives

Both overproduction of nitric oxide and oxidative injury to the cardiovascular and pulmonary systems contribute to fatal pathophysiology during endotoxemia. We investigated the effect of propofol on oxidative stress-related enzymes in lung (L2), heart (H9C2) and macrophage (NR8383) cells during endotoxemia.

Methods

Experimental endotoxemia was induced by co-culture of Escherichia coli lipopolysaccharide (15 μg/mL) in the abovementioned three types of cells that were under the effect of propofol (15 or 30 μM for 1 or 4 hours). Cellular expression of induced nitric oxide synthase (iNOS), superoxide dismutase (SOD) 1 and 2, and p47phox (representing NADPH oxidase) were determined by immunoblotting. The cellular oxidative burst activity was determined using a dihydroethidium method via flow cytometry to represent the level of reactive oxygen species. The in vivo endotoxemia model was also employed for comparison using a systemic injection of lipopolysaccharide (15 mg/kg) under propofol maintenance (15 or 30 mg/kg/h). The Student t test (two groups) was used for statistical evaluation among the means, and the Newman–Keuls test was used for analysis of variance in the statistical analysis.

Results

In lung L2 cells, propofol significantly reduced the expression of iNOS, SOD1, SOD2, and p47phox under LPS-induced endotoxemia. However, in H9C2 cardiac cells and NR8383 macrophages, only the expression of iNOS was significantly suppressed, but not that of SOD1, SOD2, or p47phox. The level of reactive oxygen species was suppressed in all three kinds of cell. In in vivo animal tissue, except for the suppression of iNOS expression in lung and heart cells, propofol in lung cells produced only SOD1 suppression, but in rat heart the expression of both SOD1 and SOD2 was suppressed.

Conclusion

These results suggest that propofol may have a protective role for lung cells. This effect is associated with its suppression of oxidative-related enzymes, including iNOS, SOD1, SOD2, and p47phox. In cardiac myocytes and macrophages, propofol also provides an antioxidative effect, probably via its inhibition of iNOS. The overall effect of propofol in the organs may be a combination of its effects on various cells. In addition, a reduction in reactive oxygen species plays a major role in the beneficial effect of propofol on experimental endotoxemia.

Keywords

cell line: H9c2(2-1); cell line: L2; cell line: NR8383; epithelial cells; macrophages; propofol;

1. Introduction

Endotoxemia is characterized by massive oxidative stress in the tissues and free radical production. The catastrophic effects of this pathophysiologic change cause much damage to many vital organs. It is always a challenging issue for anesthesiologists when such patients are to undergo surgery. One of the successful ways to terminate the free radical chain reaction is to administer an antioxidant. The commonly used anesthetic propofol is thought to have antioxidant capacity. Furthermore, in our previous studies, propofol was demonstrated to have a differential protection for sepsis-induced lipid peroxidation damage in the heart and lungs.¹ However, the end product of the induced nitric oxide synthase (iNOS) remained unchanged in heart and lung tissue.¹ Therefore the real cellular therapeutic mechanism of propofol in endotoxemia remains unknown.

Propofol is reported to have an effect on enzymes related to reactive oxidative stress. Except for iNOS activity,2, 3 tissue catalase activity⁴ has also been reported to decrease in a testis torsion model. Furthermore, the plasma concentrations of superoxide dismutase (SOD) and myeloperoxidase were also decreased in suprarenal aortic clamping surgery.⁵ SOD and NADPH oxidase are also well known for superoxide production and elimination (Fig. 1). However, few reports have discussed the direct effect of propofol on ROS-related enzymes. In this study, we aimed to delineate the effect of propofol on the free radical-related enzymes involved in experimental endotoxemia via both in vivo and in vitro studies.

Download full-size image

Fig. 1. Representation of the catastrophic pathophysiology of endotoxemia. The overproduction of superoxide and nitric oxide and the activation of those related enzymes was one of the major causes of endotoxemia-induced injury. iNOS = inducible nitric oxide synthase; SOD = superoxide dismutase.

Male Sprague–Dawley rats under the effect of propofol were subjected to Escherichia coli lipopolysaccharide (LPS) injection to serve as the in vivo septic model, while H9C2, L2, and NR8383 cells (derived from rat cardiac, lung, and macrophage cells, respectively) served as the in vitro model. Different concentrations of propofol were applied to rats or cells with or without LPS. Superoxide production in various cells was analyzed using flow cytometry with dihydroethidium. Expression of SOD1, SOD2, iNOS, and p47phox (a critical component of NADPH oxidase) in various cells and tissues was examined using western blotting. We concluded that propofol had effects on ROS-related enzymes mainly on iNOS in lung cells, with various effects on cardiac cells and macrophages, and also reduced ROS formation in both in vivo and in vitro models of endotoxemia.

2. Materials and methods

2.1. In vitro experimental endotoxemia model

H9C2 (heart-derived rat myocyte), L2 (rat lung cell), and NR8383 (rat macrophage) cells, purchased from the Bioresources Collection and Research Center (BCRC, Hsinchu, Taiwan), were used in the in vitro endotoxemia model. Cells were cultured in 5% CO₂ at 37°C and maintained in Dulbecco's modified Eagle's (H9C2) or F-12K medium (L2, NR8383) with 10% or 15% fetal bovine serum. Cells were stored in liquid nitrogen if not used. Different concentrations of propofol (0, 15, and 30 μM) were applied to the cells with or without LPS (15 μg/mL). Co-culture lasted for 1 or 4 hours in the in vivo model. Cells were then harvested and stored at −80°C for further investigation.

2.2. Animal preparation and in vivo experimental endotoxemia model

The in vivo endotoxemia model was set up as previously described.⁶ All procedures were approved by the Institutional Animal Care and Use Committee of our university. Briefly, male Sprague–Dawley rats (250–380 g) were anesthetized first with inhalation of isoflurane via an endotracheal tube for cannulation of the femoral artery and vein, and tracheal intubation; the anesthetic agent was then changed to propofol according to the animal's body weight (30 or 15 mg/kg/h, n = 5). After stabilization of anesthesia for at least for 30 minutes, the animals were subjected to an intravenous injection of LPS (15 mg/kg) to create the in vivo endotoxemia model. Tissues (heart and lung) were harvested 1 and 4 hours later with 4°C phosphate-buffered saline (PBS) to represent the early and late phases of experimental endotoxemia, respectively.¹

2.3. Protein expression with western blotting

Protein expression was performed as previously described.1, 7 The expression of SOD1 (1:500; Stressgen, Victoria, BC, Canada), SOD2 (1:1000; Stressgen), iNOS (1:1000; BD Biosciences, San Jose, CA, USA) and p47phox (1:1000; BD Biosciences) in various cells and tissues was examined with western blotting. The harvested cells or tissues were minced and homogenized in lysis buffer at 4°C. The concentration of protein was determined using a protein assay kit (Bio-Rad, Hercules, CA, USA). An equal amount of protein (30 μg) was loaded onto polyacrylamide gel (9–12%). Protein was then transferred to nitrocellulose membranes by the wet-transferring method with ice packing. The membranes were incubated overnight with primary antibodies in the above-mentioned dilutions at 4°C, and β-actin (1:5000, Sigma-Aldrich, St Louis, MO) was used as the internal control. After washing with PBS, the membranes were incubated with an appropriate dilution of horseradish peroxidase-linked secondary antibody for 1.5 hours at room temperature. The bands were visualized using enhanced chemiluminescence and quantified by scanning densitometry (ImageJ; NIH, Bethesda, MD, USA).

2.4. Superoxide measurement in cells

Cellular accumulation of superoxide was determined using the dihydroethidium method.¹ To measure levels of ROS, citrated cells were incubated with 1M dihydroethidium for 30 minutes at 37°C, followed by cell washing and resuspension in ice-cold PBS. Cell Quest software was applied for flow cytometric analysis (FACSCalibur; BD). The mean fluorescence intensity of cells, separated by electronic gating in the forward scatter/side scatter dot plot, was measured and quantified in arbitrary units of 10,000 events.

2.5. Statistical analysis

The results are expressed as means ± standard errors of the mean. One-way analysis of variance and the Student t test (two groups) or multiple comparisons were used for statistical evaluation of the differences among means (Newman–Keuls test for post hoc analysis). A value of p < 0.05 was considered to be statistically significant.

3. Results

3.1. Effects of different concentrations of propofol on ROS-related enzymes and the end product in different rat cells during experimental cellular endotoxemia

3.1.1. The effect of propofol on L2 lung cells

Fig. 2 shows the expression of four different ROS-related enzymes in L2 rat lung cells. Propofol significantly suppressed the expression of the enzymes, not only iNOS [A, control: 1.00 ± 0.02; LPS (4 hours): 4.22 ± 0.14; propofol (30, 4 hours): 1.64 ± 0.06; propofol (15, 4 hours): 2.41 ± 0.07], but also of SOD1 [B, control: 1.00 ± 0.03; LPS (4 hours): 1.53 ± 0.10; propofol (30, 4 hours): 1.20 ± 0.05; propofol (15, 4 hours): 0.79 ± 0.09)], SOD2 [C, control: 1.00 ± 0.04; LPS (4 hours): 1.65 ± 0.01; propofol (30, 4 hours): 1.07 ± 0.02; propofol (15, 4 hour): 0.91 ± 0.04], and p47phox [D, control: 1.00 ± 0.01; LPS (4 hours): 1.38 ± 0.01; propofol (30, 4 hours): 0.73 ± 0.01; propofol (15, 4 hours): 0.64 ± 0.06]. This means that propofol not only had suppressive effects on ROS-produced enzymes in lung cells but also had effects on ROS-eliminating enzymes. However, the reduced ratios of iNOS (68.2%) and p47 (46.8%) were higher than those of the SOD enzymes (SOD1 17.1%, SOD2 32.5%).

Download full-size image

Fig. 2. Representative western blots of inducible nitric oxide synthase (iNOS; A), superoxide dismutase 1 (SOD1; B), superoxide dismutase 2 (SOD2; C), and p47phox (the critical component of NADPH oxide; D) with densitometric analysis of the amount of protein relative to β-actin from L2 rat lung cells at 1 or 4 hours with or without lipopolysaccharide (LPS; 15 μg/mL) in cells that were maintained in propofol (15 or 30 μM). Values are mean ± standard errors, n = 4 dishes per group. *p < 0.05 versus expression in the no-LPS group; #p < 0.05 versus the LPS-only group.

3.1.2. The effect of propofol on H9C2 cardiac cells

Fig. 3 demonstrates the effects of propofol on septic H9C2 cardiac myocyte cells. Propofol also significantly suppressed the expression of iNOS [A, control: 1.00 ± 0.12; LPS (4 hours): 2.04 ± 0.28; propofol (30, 4 hours): 1.43 ± 0.25; propofol (15, 4 hours): 1.63 ± 0.27], but not of SOD1 [B, control: 1.00 ± 0.02; LPS (4 hours): 0.91 ± 0.11; propofol (30, 4 hours): 0.95 ± 0.13; propofol (15, 4 hours): 0.69 ± 0.14], SOD2 [B, control: 1.00 ± 0.01; LPS (4 hours): 0.97 ± 0.00; propofol (30, 4 hours): 0.91 ± 0.10; propofol (15, 4 hours): 0.70 ± 0.13], or p47phox [D, control: 1.00 ± 0.03; LPS (4 hours): 1.14 ± 0.02; propofol (30, 4 hours): 1.00 ± 0.02; propofol (15, 4 hours): 1.12 ± 0.01]. The reduced iNOS ratio (48.1%) was also less than that seen in L2 lung cells. This suggested that propofol might have a greater suppressive effect on the expression of iNOS in lung cells compared with cardiac myocytes.

Download full-size image

Fig. 3. Representative western blots of inducible nitric oxide synthase (iNOS; A), superoxide dismutase 1 (SOD1; B), superoxide dismutase 2 (SOD2; C), and p47phox (D) with densitometric analysis of the amount of protein relative to β-actin from H9C2 rat cardiac cells at 1 or 4 hours with or without lipopolysaccharide (LPS; 15 μg/mL) in cells that were maintained in propofol (15 or 30 μM). Values are mean ± standard errors, n = 4 dishes per group. *p < 0.05 versus expression in the no-LPS group; #p < 0.05 versus the LPS-only group.

3.1.3. The effect of propofol on NR8383 macrophage cells

Meanwhile, the effects of propofol on septic NR8383 microphages are showed in Fig. 4. The results were similar to those in H9C2 cardiac cells. Only the expression of iNOS [control: 1.00 ± 0.04; LPS (4 hours): 2.01 ± 0.25; propofol (30, 4 hours): 1.16 ± 0.13; propofol (15, 4 hours): 1.28 ± 0.25] was suppressed by propofol, and not that of SOD1 [control: 1.00 ± 0.03; LPS (4 hours): 1.02 ± 0.05; propofol (30, 4 hours): 1.12 ± 0.04; propofol (15, 4 hours): 1.05 ± 0.04], SOD2 [control: 1.00 ± 0.19; LPS (4 hours): 1.26 ± 0.17; propofol (30, 4 hours): 1.48 ± 0.03; propofol (15, 4 hours): 1.02 ± 0.23], or p47phox [control: 1.00 ± 0.22; LPS (4 hours): 1.15 ± 0.19; propofol (30, 4 hours): 1.47 ± 0.17; propofol (15, 4 hours): 1.26 ± 0.15]. The reduction ratio (26.5%) was the least of all the three cell types.

Download full-size image

Fig. 4. Representative western blots of inducible nitric oxide synthase (iNOS; A), superoxide dismutase 1 (SOD1; B), superoxide dismutase 2 (SOD2; C), and p47phox (D) with densitometric analysis of the amount of protein relative to β-actin from NR8383 rat macrophage cells at 1 or 4 hours with or without lipopolysaccharide (LPS; 15 μg/mL) in cells that were maintained in propofol (15 or 30 μM). Values are mean ± standard errors, n = 4 dishes per group. *p < 0.05 versus expression in the no-LPS group; #p < 0.05 versus the LPS-only group.

3.1.4. The effect of propofol on ROS production in septic cellular condition

Even the effect of propofol exerted on the three cell types was different. The total production of ROS in these three kinds of cell was significantly decreased in LPS-induced septic culture conditions when propofol was applied for 4 hours (Fig. 5) as with L2 [control: 1.00 ± 0.01; LPS (4 hours): 1.49 ± 0.05; propofol (30, 4 hours): 0.34 ± 0.02; propofol (15, 4 hours): 0.39 ± 0.01], H9C2 [control: 1.13 ± 0.13; LPS (4 hours): 1.39 ± 0.08; propofol (30, 4 hours): 0.48 ± 0.03; propofol (15, 4 hours): 0.63 ± 0.19], and NR8383 [control: 1.12 ± 0.12; LPS (4 hours): 1.59 ± 0.10; propofol (30, 4 hours): 0.64 ± 0.05; propofol (15, 4 hours): 0.63 ± 0.19].

Download full-size image

Fig. 5. Representative cellular level of reactive oxygen species (ROS) in L2 lung cells, H9C2 cardiac cells, and NR8383 macrophages at 1 or 4 h after co-culture with lipopolysaccharide (LPS; 15 μg/mL) or without LPS (C means control in this figure). Cells were also maintained under different concentrations of propofol [P(H): 30 μM, P(L):15 μM]. Values are mean ± standard errors, n = 4 dishes per group. *p < 0.05 versus expression in the no-LPS group; #p < 0.05 versus the LPS-only group.

3.2. Effects of different concentrations of propofol on ROS-related enzymes in cardiac and pulmonary tissues during different periods of experimental endotoxemia

3.2.1. The effects of propofol on lung tissue during different periods of endotoxemia

The effects of propofol on SOD1, SOD2, and p47phox are shown in Fig. 6. (iNOS expression has not been presented because it had already been shown in our previous study,¹ as the result of culture, that iNOS was all suppressed in both septic lung and heart tissue.) It was shown that the lung expression of SOD1 [control: 1.00 ± 0.06; L1 (propofol 15 mg/kg/h for 1 hour): 0.83 ± 0.09; L4 (propofol 15 mg/kg/h for 4 hours): 0.82 ± 0.13; H1 (propofol 30 mg/kg/h for 1 hour): 0.49 ± 0.04; H4 (propofol 30 mg/kg/h for 4 hours): 0.34 ± 0.05] was significantly suppressed when a high dose of propofol was administrated. SOD2 expression was significantly increased after systemic administration of LPS but not obviously different with various other dosages of propofol [control: 1.00 ± 0.10; L1 (propofol 15 mg/kg/h for 1 hour): 2.32 ± 0.13, L4 (propofol 15 mg/kg/h for 4 hours): 2.18 ± 0.13; H1 (propofol 30 mg/kg/h for 1 hour): 2.11 ± 0.07; H4 (propofol 30 mg/kg/h for 4 hours): 1.76 ± 0.07]. There was no difference in p47phox expression after exposure of lung tissue to propofol. This indicated the complexity of propofol's effect on ROS-related enzymes in the endotoxemic condition.

Download full-size image

Fig. 6. Representative western blots of superoxide dismutase 1 (SOD1), superoxide dismutase 2 (SOD2), and p47phox with densitometric analysis of the amount of protein relative to β-actin from rat lung at 1 or 4 hour after systemic injection of lipopolysaccharide (LPS; 15 mg/kg) in rats that were maintained under propofol (15 or 30 mg/kg/h). C = control, without LPS injection. H1 and H4: propofol 30 mg/kg/h over 1 or 4 hours after LPS injection, respectively. L1 and L4: propofol 15 mg/kg/h over 1 or 4 hours after LPS injection, respectively. Values are means ± standard errors of the mean, n = 5 animals per experimental group. *p < 0.05 versus control.

3.2.2. The effects of propofol on heart tissue during different durations of endotoxemia

The effects of propofol on SOD1, SOD2, and p47phox production in heart tissue are shown in Fig. 7. The expression of SOD1 and SOD2 was similar, and was shown to be increased at low doses but suppressed when high doses of propofol were supplied [SOD1: control: 0.61 ± 0.05; L1 (propofol 15 mg/kg/h for 1 hour): 0.98 ± 0.01; L4 (propofol 15 mg/kg/h for 4 hours): 0.93 ± 0.08; H1 (propofol 30 mg/kg/h for 1 hour): 0.78 ± 0.11; H4 (propofol 30 mg/kg/h for 4 hours): 0.63 ± 0.01; SOD2: control: 0.67 ± 0.09; L1 (propofol 15 mg/kg/h for 1 hour): 0.76 ± 0.06; L4 (propofol 15 mg/kg/h for 4 hour propofol): 1.04 ± 0.05; H1 (propofol 30 mg/kg/h for 1 hour): 0.77 ± 0.10; H4 (propofol 30 mg/kg/h for 4 hours): 0.46 ± 0.02]. However, the expression of p47 was also increased at low dose but was unchanged when high doses of propofol were administrated [control: 0.70 ± 0.08; L1 (propofol 15 mg/kg/h for 1 hour): 0.98 ± 0.11; L4 (propofol 15 mg/kg/h for 4 hours): 0.98 ± 0.06; H1 (propofol 30 mg/kg/h for 1 hour): 1.02 ± 0.02; H4 (propofol 30 mg/kg/h for 4 hours): 0.97 ± 0.00].

Download full-size image

Fig. 7. Representative western blots of superoxide dismutase 1 (SOD1), superoxide dismutase 2 (SOD2), and p47phox with densitometric analysis of the amount of protein relative to β-actin from rat heart at 1 or 4 hours after systemic injection of lipopolysaccharide (LPS; 15 mg/kg) in rats that were maintained under propofol (15 or 30 mg/kg/h). C = control, without LPS injection. H1 and H4: propofol 30 mg/kg/h over 1 or 4 hours after LPS injection, respectively. L1 and L4: propofol 15 mg/kg/h over 1 or 4 hours after LPS injection, respectively. Values are means ± standard errors of the mean, n = 5 animals per experimental group. *p < 0.05 versus control; #p < 0.05 versus the group receiving a low dose of propofol (15 mg/kg/h).

ROS-induced injury was not present with the end product despite its presence in our previous study.¹ That study certainly revealed that the ROS-induced lipid peroxidation in the lungs and heart was reduced in a dose-dependent manner. The overall effect of propofol on various enzymes is summarized in Table 1.

Table 1. Representative summary of results for the various changes in reactive oxygen species-related enzymes within cells and tissues under lipopolysaccharide/endotoxemic condition with or without 4-hour propofol administration.

Download full-size image

4. Discussion

Propofol acts by many mechanisms that may influence the expression of endotoxemia. It is reported to protect the cell membrane from lipid peroxidation by inhibiting the production of malondialdehyde.3, 8 Its antioxidant effect resembles that of vitamin E.⁹ It also inhibits the production of inflammatory cytokines during sepsis, and reduces macrophage infiltration during endotoxemia.¹⁰ Our previous results demonstrated that propofol influenced lipid peroxidation in the heart and lungs and free radical formation in polymorphonuclear cells.¹ We concluded that propofol may prolong endotoxemia in rats mainly through its antioxidant effect on the vital organs.

An in vivo study concurrently focused specifically on the heart, lungs, and polymorphonuclear cells to investigate the protective effect of propofol on experimental endotoxemia. However, that study lacked the protective mechanism of propofol at a cellular level. In the lungs, propofol can suppress the infiltration of neutrophils, and there are reports indicating that propofol reduces NO synthesis in macrophages.¹¹ However, some reports have demonstrated that propofol also has effects on endothelial cells.¹² The major target of propofol in experimental endotoxemia remains unclear. The protective ratio of propofol for various cells or tissues also needs further evaluation.

The beneficial effect of propofol on in vivo experimental endotoxemia was demonstrated to be related to the suppression of ROS-induced injury to the organs.¹ In this study, we clearly showed that the expression of iNOS was significantly suppressed by propofol in all three types of cell. This may indicate that propofol has a rather stronger effect on iNOS than on other ROS-related enzymes. This suppressive effect of propofol has also been reported in many other manuscripts.3, 13 However, a lack of specific targets of propofol on cells was also noted in these papers. Our culture data on iNOS showed unique findings for propofol in lung, heart and macrophage cells. In addition, we revealed various changes in three other ROS-related enzymes.

Very few studies have focused on the effect of propofol on cellular SOD or p47phox activity. It has been reported that propofol had no effect on circulating SOD after general anesthesia in swine¹⁴ and rats with traumatic brain injuriess.¹⁵ It also had no effect on SOD in red blood cells when an oxidative challenge was carried out ex vivo.¹⁶ Our results indicated that propofol could suppress the expression of SOD1, SOD2, and p47phox only in lung epithelial cells (L2), but not in macrophages and cardiac myocytes. As for the expression of SOD1, SOD2, and p47phox in an in vivo endotoxemia model under the effect of propofol, there were inconsistent results compared with culture for SOD1 and SOD2 in the lungs, and SOD1, SOD2, and p47phox in the heart. The real reasons remain unclear. Macrophage infiltration may not be the major reason because our result with NR8383 cells did not support this idea. The other explanation for these inconsistencies may be the local pH change or hypoxic condition induced by the endotoxemia.

There were still some limitations to our study. The ROS cascade pathway induced by endotoxemia involves many enzymes, and our three or four enzymes certainly did not cover the whole picture. However, we still revealed consistent data for iNOS expression. Furthermore, we did not rule out the possibility of a direct effect of propofol itself on a chelation of the free radicals by its phenol ring. Another contribution this study makes lies in the consistency of free radical reduction in all three kinds of cell and the in vivo results.

In conclusion, this study has demonstrated that propofol plays a protective role in in vitro endotoxemic lung cells. This effect is associated with a suppression of oxidative-related enzymes including iNOS, SOD1, SOD2, and p47phox. In cardiac myocytes and macrophages, propofol also provides its antioxidative effect on the activity of iNOS probably via its inhibition. The total effect of propofol in the organs may be a combination of those effects on various cells. In addition, ROS reduction plays a major role in the beneficial effect during endotoxemia.

Acknowledgments

The authors would like to thank Alice Y.W. Chang and Julie Y.H. Chan for their helpful and important assistance in constructing this article.