The influence of intravenous anaesthetics on the activity of enzymes released from polymorphonuclear leucocytes in vitro

Abstract

Background and objective 

Polymorphonuclear leucocytes make a decisive contribution to defence against bacterial infections. In particular, the effects of anaesthetics on non-oxidative bactericidal mechanisms have previously only been superficially examined. Although the influence of anaesthetic agents on oxidative bactericidal activity has been thoroughly examined, our study concentrated on the effect on non-oxidative processes, which appears to have been a neglected field of research.

Methods 

The effects of methohexital, etomidate, ketamine, fentanyl and morphine on the activity of lysozyme and β-glucuronidase released from polymorphonuclear leucocytes have been studied in vitro. The activity of lysozyme was determined by recording the changes in the turbidity of a suspension of micrococcus lysodeicticus caused by the enzymatic action of lysozyme. β-glucuronidase activity was photometrically measured by the enzymatic cleavage of phenolphthalein glucuronic acid.

Results 

High concentrations of methohexital inhibited lysozyme activity; however, etomidate and morphine caused an increase of β-glucuronidase activity in therapeutic plasma concentrations. While there was no effect of etomidate on lysozyme activity, all concentrations tested significantly stimulated β-glucuronidase activity. This result was unexpected because intravenous anaesthetics have previously shown a tendency to suppress polymorphonuclear leucocyte functions. Whereas the inhibition of lysozyme activity by the high concentration of methohexital was no surprise, the increase of β-glucuronidase activity caused by etomidate, ketamine, fentanyl and morphine was quite unexpected.

Conclusions 

At present, the underlying mechanism for the increase of β-glucuronidase activity caused by etomidate, ketamine, fentanyl and morphine is unknown. The fact that there was no influence of these agents on lysozyme activity possibly suggests that the anaesthetic agents have different effects on azurophilic and specific granules. Since in vitro investigations have their limitations, it is too early to draw practical consequences from our study. Moreover, at present it is unclear whether an increase of β-glucuronidase activity in vivo is an advantage or not. In any case, we think it advisable to perform further investigations on the influence of anaesthetic agents on oxygen-independent bactericidal mechanisms.

Introduction

The question whether anaesthetic agents have an effect on perioperative infections has kept researchers occupied for a long time. In 1889, Platania [¹] showed that the administration of curare, alcohol and chloral hydrate considerably promoted the spread of Bacillus anthracis in animals. Erroneously, he tried to explain this phenomenon by a depression of the nervous system. In 1892, Klein and Coxwell [²] described an intensification of bacterial infections caused by the inhalation of chloroform and ether. It was assumed that indistinct ‘chemical alterations…that abolish the normal bacteria-killing properties of the blood and the lymph’ were responsible for this effect. At the turn of the century, Snel [³] confirmed the observations of Platania, Klein and Coxwell. In addition, he discovered that morphine did not intensify infections. Rubin [⁴] was the first to postulate that the impairment of immunity caused by anaesthetic agents was due to a malfunction of leucocytes. In 1911, this assumption was supported by Graham [⁵] who found that ether caused an inhibition of the leucocyte phagocytosis of bacteria. In the human organism, two types of cells are capable of phagocytosis: mononuclear phagocytes and polymorphonuclear leucocytes. Polymorphonuclear leucocytes [^6,7] make an important contribution to the defence against bacterial infections. They are provided with various mechanisms for the destruction of microorganisms. A distinction can be made between oxygen-dependent and oxygen-independent processes. The former is based on the so-called respiratory burst which produces several microbicidal agents (among others superoxide O₂^–, hydrogen peroxide H₂O₂ and the hydroxyl radical OH). However, oxygen-independent killing is the effect of a large number of proteins, which are stored up in polymorphonuclear leucocyte granules. Their release can be primed and caused by a wide variety of stimuli (chemoattractants, cytokines, tumour necrosis factor, growth factors, immunoglobulins, bacterial products and viruses). One of the proteins playing a part in oxygen-independent killing is lysozyme, which destroys bacteria by hydrolysis of cell wall proteoglycans. Lysozyme is contained in azurophilic and specific granules. The enzyme β-glucuronidase is involved in the degradation of bacteria. It is only found in azurophil granules.

In our study, we examined the effects of methohexital, etomidate, ketamine, fentanyl, morphine and their additives on the activity of lysozyme and β-glucuronidase released from polymorphonuclear leucocytes in vitro.

Methods

Permission to conduct the study was obtained from the Ethic Committee of the Medical Faculty of Justus-Liebig-University Giessen.

Blood samples

Heparinized venous blood samples (10 units heparin sodium mL⁻¹; Sigma, Deisenhofen, Germany) were obtained from 10 male volunteers, who had not taken any medication for a period prior to the study and were free of all infective symptoms.

Isolation of polymorphonuclear leucocytes

Polymorphonuclear leucocytes were isolated using a modification of the method described by Hjorth and colleagues [⁸]. Percoll 55% and Percoll 74% were prepared by diluting isotonic Percoll (Sigma, Deisenhofen, Germany) with 0.9% NaCl (normal saline) solution. Next, 4 mL of Percoll 55% were poured into a polystyrene tube, underlayered with 4 mL of Percoll 74% and overlayered with 4 mL of heparinized blood. The tube was centrifuged at 350 g for 20 min at 20°C. The polymorphonuclear leucocytes band was removed with a Pasteur pipette and resuspended in 10 mL of phosphate-buffered saline (PBS; Gibco, Karlsruhe, Germany). After centrifugation at 350 g for 10 min at 20°C, the supernatant fluid was decanted. The few erythrocytes contaminating the pellet were lysed by adding 2 mL of purified water. After 20 s, isotonia was re-established by the addition of 1 mL of 2.7% saline (Merck, Darmstadt, Germany). After 7 mL of PBS had been administered, the tube was centrifuged at 350 g for 10 min at 20°C. The supernatant fluid was then decanted. The purity of the polymorphonuclear leucocyte yield was microscopically examined, and the viability was confirmed as ≥ 95% by the trypan blue exclusion test [⁹]. The polymorphonuclear leucocyte concentration was adjusted to 8.8 × 10⁶ mL⁻¹ by adding PBS containing 0.99 g glucose (Merck, Darmstadt, Germany) per 100 mL.

Stimulation of enzyme release

Stimulation of enzyme release was conducted according to the descriptions of Metcalf and colleagues [¹⁰]. Next, 5 µg cytochalasin B (Sigma, Deisenhofen, Germany) were added to 1 mL of PMNL suspension which was incubated for 10 min at 37°C. Lysosomal enzyme release in response to soluble stimuli, e.g. N-formylmethionylleucylphenylalanine (NFMLP) is significantly enhanced when PMNL are pretreated with the fungal metabolite cytochalasin B. The respective anaesthetic agent was then administered. After another incubation for 30 min at 37°C, giving the anaesthetics time to act on the polymorphonuclear leucocytes, 20 µL NFMLP (10⁻³ M; Sigma, Deisenhofen, Germany) were added, and the suspension was incubated for 20 min at 37°C. After centrifugation at 350 g for 10 min at 20°C, the supernatant fluid was collected.

Measurement of lysozyme activity

The activity of released lysozyme was determined by a standard assay (Testomar-Lysozym; Behring, Marburg, Germany) which detects changes in the turbidity of a suspension of Micrococcus lysodeikticus caused by the enzymatic activity of lysozyme. Next, 25 µL-testomar solution was added to 220 µL supernatant. Extinction was photometrically measured at 546 nm. The activity of released lysozyme was determined using a calibration curve previously established with egg white lysozyme (Behring, Marburg, Germany).

Measurement of β-glucuronidase activity

β-Glucuronidase activity was photometrically measured by the enzymatic cleavage of phenolphthalein glucuronic acid. We used the method described by Metcalf and colleagues [¹⁰]. Briefly, 0.7 mL of acetate buffer (1.158 g of sodium acetate and 0.65 mL of glacial acetic acid were brought up to 200 mL with purified water; pH 4.5; both chemicals by Sigma, Deisenhofen, Germany), 0.2 mL of phenolphthalein glucuronic acid (0.01 M; Sigma, Deisenhofen, Germany) and 0.1 mL of supernatant were mixed and incubated for 10 h at 37°C. Then, 1 mL of glycine buffer [3.26 g of glycine, 2.53 g of sodium chloride, 4.56 mL of sodium hydroxide (0.5 g mL⁻¹ purified water) were brought up to 200 mL with purified water; pH 10.5; chemicals by Sigma, Deisenhofen, Germany] was added and extinction was measured at 540 nm. The concentration of phenolphthalein was determined using a calibration curve previously established with phenolphthalein (Sigma, Deisenhofen, Germany).

Anaesthetics and additives

The following concentrations of anaesthetics and corresponding additives were tested. The medium concentrations correlate to the immediately appearing serum levels after intravenous injections of clinically relevant doses [^11–15].

• •methohexital (Brevimytal Natrium; Eli Lilly, Bad Homburg, Germany): 1.7, 17 and 170 µg mL⁻¹.
• •sodium carbonate (pharmacy of Justus-Liebig-University, Giessen, Germany): 0.1, 1 and 10 µg mL⁻¹.
• •etomidate (Hypnomidate; Janssen-Cilag, Neuss, Germany): 0.032, 0.32 and 3.2 µg mL⁻¹.
• •propylene glycol (Sigma, Deisenhofen, Germany): 0.0056, 0.056, and 0.56 µL mL⁻¹.
• •ketamine (Ketanest 10; Parke-Davis, Berlin, Germany): 0.236, 2.36 and 23.6 µg mL⁻¹.
• •benzethonium chloride (Sigma; Deisenhofen, Germany): 0.00236, 0.0236 and 0.236 µg mL⁻¹.
• •fentanyl (Fentanyl-Janssen; Janssen-Cilag, Neuss, Germany): 0.0035, 0.035 and 0.35 µg mL⁻¹.
• •morphine (Merck, Darmstadt, Germany): 0.127, 1.27 and 12.7 µg mL⁻¹.

Statistics

All tests were performed in duplicate. The results were mean values of two estimations. The Pearson Stephens test was used to check normal distribution. The Bartlett test examined homogeneity of variance (P ≤ 0.1). If the requirements were met, analysis of variance for repeated measures and the Scheffé test were conducted. If the requirements were not fulfilled, the Friedman analysis of variance and the Miller test were performed. A probability of P ≤ 0.05 was regarded as significant.

Results

The high concentration of methohexital significantly (P ≤ 0.05) decreased lysozyme activity (Figure 1). However, there was no statistically relevant influence on β-glucuronidase activity (Figure 2). The solvent sodium carbonate did not alter lysozyme and β-glucuronidase activity.

Figure 1.: The influence of various concentrations of methohexital (µg mL⁻¹) on the activity of lysozyme (egg white lysozyme equivalent mg L⁻¹) released from polymorphonuclear leucocytes in vitro (n = 10; *P ≤ 0.05).

Figure 2.: The influence of various concentrations of methohexital (µg mL⁻¹) on the activity of β-glucuronidase (phenolphthalein, mg L⁻¹) released from polymorphonuclear leucocytes in vitro (n = 10).

While etomidate did not influence lysozyme activity (Figure 3), all concentrations tested significantly (P ≤ 0.01) increased β-glucuronidase activity (Figure 4). The solvent propylene glycol did not produce any effects.

Figure 3.: The influence of various concentrations of etomidate (µg mL⁻¹) on the activity of lysozyme (egg white lysozyme equivalent mg L⁻¹) released from polymorphonuclear leucocytes in vitro (n = 10).

Figure 4.: The influence of various concentrations of etomidate (µg mL⁻¹) on the activity of β-glucuronidase (phenolphthalein, mg L⁻¹) released from polymorphonuclear leucocytes in vitro (n = 10; **P ≤ 0.01).

Ketamine did not change lysozyme activity (Figure 5). However, the high concentration significantly (P ≤ 0.01) increased β-glucuronidase activity (Figure 6). The additive benzethonium chloride did not produce any relevant effects.

Figure 5.: The influence of various concentrations of ketamine (µg mL⁻¹) on the activity of lysozyme (egg white lysozyme equivalent mg L⁻¹) released from polymorphonuclear leucocytes in vitro (n = 10).

Figure 6.: The influence of various concentrations of ketamine (µg mL⁻¹) on the activity of β-glucuronidase (phenolphthalein, mg L⁻¹) released from polymorphonuclear leucocytes in vitro (n = 10; **P ≤ 0.01).

There was no relevant influence of fentanyl on lysozyme activity (Figure 7), but the low and the high concentration significantly (P ≤ 0.01) increased β-glucuronidase activity (Figure 8).

Figure 7.: The influence of various concentrations of fentanyl (µg mL⁻¹) on the activity of lysozyme (egg white lysozyme equivalent mg L⁻¹) released from polymorphonuclear leucocytes in vitro (n = 10).

Figure 8.: The influence of various concentrations of fentanyl (µg mL⁻¹) on the activity of β-glucuronidase (phenolphthalein mg L⁻¹) released from polymorphonuclear leucocytes in vitro (n = 10; **P ≤ 0.01).

While morphine did not cause any change in lysozyme activity (Figure 9), the medium and the high concentration significantly (P ≤ 0.05) increased β-glucuronidase activity (Figure 10).

Figure 9.: The influence of various concentrations of morphine (µg mL⁻¹) on the activity of lysozyme (egg white lysozyme equivalent mg L⁻¹) released from polymorphonuclear leucocytes in vitro (n = 10).

Figure 10.: The influence of various concentrations of morphine (µg mL⁻¹) on the activity of β-glucuronidase (phenolphthalein mg L⁻¹) released from polymorphonuclear leucocytes in vitro (n = 10; *P ≤ 0.05).

Discussion

In our study, there was no significant influence of methohexital on the activity of β-glucuronidase released from polymorphonuclear leucocytes in vitro. However, the high concentration, which is certainly of no relevance in clinical practice, caused an inhibition of lysozyme activity. The solvent sodium carbonate did not have any effect. In previous studies, a depression of polymorphonuclear leucocyte adherence [¹⁶] and phagocytosis [¹⁷] caused by methohexital had been found. While high doses cause an inhibition, practically relevant concentrations do not alter the respiratory burst [^18–20]. Polymorphonuclear leucocyte chemotaxis is impaired by methohexital [^21,22].

While there was no effect of etomidate on lysozyme activity, all concentrations tested significantly stimulated β-glucuronidase activity. This result was unexpected as intravenous anaesthetics in former studies showed a tendency to suppress polymorphonuclear leucocyte functions. The solvent propylene glycol did not influence enzyme liberation. It is known that etomidate inhibits polymorphonuclear leucocyte adherence [¹⁶], phagocytosis and bactericidal activity [²³]. However, it does not affect chemotaxis [^22,24]. It is not clear whether etomidate inhibits the respiratory burst [²⁵] or not [²⁶].

In our study, ketamine did not have any influence on lysozyme activity. The high concentration, however, significantly increased β-glucuronidase activity. The additive benzethonium chloride did not have any effect. Ketamine is known for its inhibition of the adherence [²⁷], phagocytosis [¹⁷] and the killing of bacteria [²³]. With the possible exception of extremely high concentrations [²⁸], it does not alter the respiratory burst [^26,29]. Furthermore, there is no influence of ketamine on chemotaxis [²²] and polymorphonuclear leucocyte aggregation [²⁹].

In an abstract published in 1986, Mathieu described inhibitory effects of fentanyl and morphine on lysozyme and β-glucuronidase activities [³⁰]. We are not able to confirm these results. In our investigation, fentanyl did not change lysozyme activity. However, relevant concentrations significantly stimulated β-glucuronidase activity. In previous studies, it appeared that there was no influence of fentanyl on polymorphonuclear leucocyte adherence [³¹], respiratory burst [³²], phagocytosis and bactericidal activity [³³].

Morphine did not alter lysozyme activity. However, there was a significant increase in β-glucuronidase activity. It is known that morphine does not influence polymorphonuclear leucocyte adherence [³¹] and aggregation [²⁹]. Furthermore, there probably is no effect of clinically relevant concentrations on chemotaxis [³⁴] and the respiratory burst [^18,29]. Morphine inhibits phagocytosis [¹⁷] but increases bactericidal activity [³⁵].

While the influence of anaesthetic agents on oxidative bactericidal activity has been examined thoroughly, our study concentrated on the effect on non-oxidative processes, which up to date seems to have been a neglected field of research. Whereas the inhibition of lysozyme activity by the high concentration of methohexital was no surprise, the increase in β-glucuronidase activity caused by etomidate, ketamine, fentanyl and morphine was absolutely unexpected. At present, the underlying mechanism is unknown. The fact that there was no influence of these agents on lysozyme activity possibly suggests that the anaesthetic agents have different effects on azurophilic and specific granules. Because in vitro investigations have their limitations, it is too early to draw practical consequences from our study. Moreover, at present it is unclear whether an increase in β-glucuronidase activity in vivo is an advantage or not. In any case, we think it advisable to perform further investigations on the influence of anaesthetic agents on oxygen-independent bactericidal mechanisms.