- Email: [email protected]
Anaesthetic sensitivity of fMLP-induced cell signalling in Xenopus oocytes
International Immunopharmacology 6 (2006) 61 – 70 www.elsevier.com/locate/intimp
Anaesthetic sensitivity of fMLP-induced cell signalling in Xenopus oocytes Sigrid Wittmann a,*, Dieter Fro¨hlich a, Andrea Mietens a, Stephen Daniels b a
Department of Anaesthesiology, University of Regensburg, 93042 Regensburg, Germany b Welsh School of Pharmacy, Cardiff University, Cardiff CF10 3 XF, Wales/UK Received 3 April 2005; received in revised form 26 April 2005; accepted 27 July 2005
Abstract FMLP stimulation of Xenopus oocytes expressing fMLP receptors leads to a concentration-dependent biphasic inward current. To identify the evolution of these currents we have examined the effects of blocking various cell signalling pathways. In addition we have analysed the effects of three intravenous anaesthetics on these fMLP-induced currents. Xenopus oocytes were microinjected with cRNA encoding the fMLP receptor and fMLP-stimulated (100 nM) currents measured, using two-electrode voltage-clamp (70 mV), before and after injection of heparin (120 ng ml 1), wortmannin (1 AM), U73122 (5 AM) or buffer. Concentration–response curves were established for the action on fMLP-stimulated currents of thiopentone (5–500 AM), methohexitone (0.2–200 AM) and propofol (0.5–500 AM). Heparin significantly enhanced the fast current ( p b 0.05). Wortmannin had no effect on either current. U73122 inhibited only the slow current ( p b 0.05). All anaesthetics inhibited both currents, with the maximum inhibition for the fast/slow currents 70%/100%, 60%/60% and 100%/100% for thiopentone (IC50 147 / 120 AM), methohexitone (IC50 4.7 / 2.2 AM) and propofol (IC50 33 / 8 AM), respectively. We suggest (a) the slow current arises via the PLC/PKC pathway because it is reduced by the PLC inhibitor U73122, (b) the PI3K- and PLD-mediated pathways are not involved because wortmannin had no effect and (c) activation of the two conductance channels must be different because U73122 reduced the slow but not the fast current. Since both currents are decreased by all three anaesthetics, their inhibition might be mediated through an action at the agonist/receptor, although, since the slow current is consistently more sensitive than the fast, there may be additionally an action on cell signalling. D 2005 Elsevier B.V. All rights reserved. Keywords: Thiopentone; Methohexitone; Propofol; Heparin; U73122; Wortmannin
* Corresponding author. Tel.: +49 941 9447801; fax: +49 941 9447802. E-mail address: [email protected] (S. Wittmann). 1567-5769/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.intimp.2005.07.012
1. Introduction N-formylated peptides, such as fMLP (N-formyl– l-methionyl–l-leucyl–phenylalanine), mediate host
62
S. Wittmann et al. / International Immunopharmacology 6 (2006) 61–70
defence mechanisms such as the oxidative response (production of reactive oxygen derivatives) in human neutrophils [1]. Functional expression of the G-protein coupled fMLP receptor in Xenopus oocytes was reported following microinjection of cRNA [2]. Exposure of Xenopus oocytes expressing the fMLP receptor to fMLP produces a concentration-dependent biphasic inward current with a fast component, carried by Cl, and a slow component, linked to a mixed population of cation channels [2]. Comparative experiments in human neutrophils showed that their activation with fMLP is dependent on store-operated calcium influx that is regulated by chloride channel activation and linked to non-selective cation channels [2]. Chloride and cation channels play a role therefore in both cell systems. Gprotein coupling to the fMLP receptor in human neutrophils has been linked to Gia2 or Gia3 [3,4], whereas in Xenopus oocytes a major role for Gq was reported [2,5]. Intracellular signalling in fMLP-stimulated human neutrophils is connected to three main pathways. The first pathway involves PLC (phospholipase C), PIP2 (PI 4,5-P2, phosphoinositide 4,5diphosphate) and its cleavage products IP3 (Ins 1,4,5-P3, inositol 1,4,5-trisphosphate) and DAG (diacylglycerol). PLC has been shown to be activated by hg subunit complexes of pertussis toxinsensitive or -insensitive G-proteins as well as by a subunits of the Gq family [6,7]. A central element of the second pathway is the PI3K (phosphoinositide 3-kinase). This second pathway is linked to PIP3 (PI 3,4,5-P3, phosphoinositide 3,4,5-trisphosphate) as well as PIP2 [8]. The role of the third PLD (phospholipase D)-associated pathway remains to be resolved. Previous studies suggested that phospholipase D might be activated not only via the hg subunits of the G-protein but also via the phospholipase C-associated pathway [9,10]. The intracellular signalling components involved in fMLP-induced currents in Xenopus oocytes functionally expressing fMLP receptors are currently unknown. However, the existence of many signalling components and their role in responses other than that to fMLP have been shown for Xenopus oocytes. Ginsenoside-induced responses involving PLC have been associated with IP3 and PTX-insensitive G-proteins [11]. M1 muscarinic receptor acti-
vation in Xenopus oocytes has been shown to involve PLC, DAG, IP3 and different PKC isoenzymes [12]. Progesterone treatment of Xenopus oocytes indicates an involvement of PLC and PKC [13]. In addition different PLC isoforms, including PLC h as well as PLD, PI3K and IP3-responsive calcium stores, have been described in Xenopus oocytes [14–17]. This study reports the effects of the signal transduction blockers wortmannin, U73122 and heparin in Xenopus oocytes, functionally expressing fMLP receptors, in order to provide evidence for the fMLP-associated pathway in these cells. In addition we analysed the sensitivity of both the fast and the slow fMLP-induced currents in Xenopus oocytes to the intravenous anaesthetics thiopentone, methohexitone and propofol. Previous studies with human neutrophils have shown a concentration-dependent impairment of the fMLP-induced oxidative response in human neutrophils upon treatment with either thiopentone, methohexitone or propofol [18,19].
2. Materials and methods 2.1. Materials Female extra-large mature Xenopus laevis were obtained from Blades Biological, UK (experiments in Wales) or the University of Bayreuth (experiments in Germany). W. Bautsch (Hannover Medical School) donated the cDNA coding for the human R98 variant of the fMLP receptor in plasmid vector pCDM8. Ultracompetent E. coli MC1061/p3 were obtained from Invitrogen, Netherlands. Other materials were purchased from the following companies: N-formyl–l-methionyl–l-leucyl–phenylalanine (fMLP) and heparin from Sigma, Germany; thiopentone (TrapanalR) from Byk Gulden, Netherlands; propofol (Propofol Abbott 1%) from Abbott, Germany; methohexitone (BrevimytalR) from Lilly, Germany; IntralipidR 10 from Baxter, Germany; and wortmannin and U-73122 from Alexis, Germany. 2.2. Amplification and purification of the plasmid Amplification and purification of the plasmid were performed using ultracompetent E. coli (MC 1061/p3) and the Wizard SV Miniprep DNA Purification System from Promega, UK.
S. Wittmann et al. / International Immunopharmacology 6 (2006) 61–70
63
2.3. Preparation of cRNA
2.6. Experiments with signal transduction blocking drugs
Plasmids containing the cDNA coding for the human fMLP receptor (R98) were linearised with HpaI and cRNA prepared using the mCAP mRNA capping kit from Stratagene, UK. The size of the cRNA was assessed using ethidium bromide-stained agarose gel electrophoresis with RNA reference fragments of defined size.
Each oocyte was stimulated three times with fMLP, in ND96. The oocyte was then de-clamped and heparin, wortmannin or U-73122 was injected intracellularly. The injection volume was 50 nl at a drug concentration designed to achieve intracellular concentrations of 120 ng/ml for heparin, 1 AM for wortmannin and 5 AM for U-73122. Based on previous publications, intracellular blocking drugs were allowed 45 min to achieve their effects [20]. After 45 min of incubation in ND96 without stimulation, the oocyte was re-clamped and stimulated three times with fMLP. In parallel control experiments oocytes were injected with buffer instead of blocking drugs.
2.4. Preparation of oocytes and injection of cRNA X. laevis were killed (experiments in Wales) by decapitation followed by mechanical destruction of the brain and spinal cord or anaesthetized with MS 222 (experiments in Germany) and pieces of ovary tissue were removed by laparotomy. The oocytes were manually stripped from the ovarian lobes using a soft plastic strip. Stage V and VI oocytes were selected, injected with 35–45 nl cRNA using a hydraulic microinjector and placed into sterile pots containing modified Barth’s solution (in mM: NaCl, 100; KCl, 1; NaHCO3, 2; MgSO4, 0.82; Ca(NO3)2, 0.33; CaCl2, 0.41; HEPES, 10; pH 7.4 with NaOH) supplemented with penicillin (100 U cm 3) and streptomycin (100 Ag cm 3). 2.5. Electrophysiological recording Prior to use oocytes were defolliculated by incubation for 15 min in Ca2+-free Barth’s solution containing collagenase (50–100 U cm 3). Electrophysiological recordings were made using two-electrode voltage-clamp (VF 180 amplifiers and CA 100 clamp amplifier, Biologic, France for experiments performed in Wales; Gene Clamp 500, Axon Instruments Inc., USA for experiments performed in Germany) at a holding potential of 70 mV and superfusion with ND96 (in mM: NaCl, 96; KCl, 2; CaCl2, 1.8; MgCl2, 1; HEPES, 5; pH 7.4 with NaOH).
2.7. Experiments with intravenous (iv) anaesthetics For all experiments with iv anaesthetics, bath application of the anaesthetic (in ND96) was performed (with the oocyte at a holding potential of 70 mV) for 90 s before the oocyte was stimulated with fMLP (fMLP– anaesthetic mixture). For each anaesthetic a cumulative concentration–response curve was measured (thiopentone 5 to 500 AM, methohexitone 0.2 to 200 AM, and propofol 0.5 to 500 AM). To exclude desensitization phenomena, separate experiments were performed in which only one anaesthetic concentration was tested (thiopentone 50 and 100 AM, propofol 20 and 200 AM, methohexitone 20 and 50 AM). To exclude the possibility that the effects induced by propofol were caused by the solution vehicle, equivalent concentrations of an IntralipidR solution (Intralipid 10% from Pharmacia Upjohn, USA) were tested in parallel to the experiments with propofol. In further control experiments, oocytes were pre-treated with glycerol (Sigma, Germany) to test the effects of this additive.
Table 1 Effects of signal transduction blockade on fMLP-induced currents in Xenopus oocytes expressing the fMLP receptor Blocking drug
Fast current Heparin U73122 Wortmannin Slow current Heparin U73122 Wortmannin
Normalised amplitude of current before injection
Normalised current amplitude post-injection of blocking drug First stimulation
Second stimulation
Third stimulation
1F0
1.33 F 0.17 1.04 F 0.14 1.21 F 0.14
1.66 F 0.22* 0.95 F 0.17 1.14 F 0.08
1.61 F 0.24* 0.94 F 0.25 1.07 F 0.12
1F0
1.36 F 0.29 0.95 F 0.21 1.26 F 0.15
1.79 F 0.35 0.67 F 0.15 0.91 F 0.15
1.14 F 0.19 0.46 F 0.15* 0.78 F 0.13
Data are presented as mean F SEM (n = 6 to n = 8) of the normalised (see text) current amplitude. Significances (*p b 0.05) were calculated for currents obtained after treatment with a blocking drug vs. control currents.
64
S. Wittmann et al. / International Immunopharmacology 6 (2006) 61–70
2.8. Data and statistical analysis Currents recorded from individual oocytes were normalised, with respect to the appropriate control current from that oocyte, so that results could be pooled for analysis. Concentration–response curves were constructed using non-lin-
ear curve fitting routines applied to a standard logistical equation (Origin 6.0, Microcal Software Inc, USA). Homogeneity of variance was analysed using Levene’s test. The statistical difference between treatment groups was tested, where appropriate, using a one-way ANOVA including post hoc testing with Dunnett’s or Bonferroni’s procedure.
Fig. 1. (A and B) The effect of thiopentone on the fMLP-induced fast (A) and slow (B) currents in Xenopus oocytes. The symbols represent the currents obtained from cumulative dose–response curves (n) and single anaesthetic treatment (5). Current amplitudes were normalised to fMLP control without thiopentone treatment and are presented as mean F SEM (n = 8). Significances were determined for anaesthetic concentrations vs. controls *p b 0.05, **p b 0.01). Data (C and D) for methohexitone and (E and F) for propofol are presented analogous to thiopentone data above. For logistic fit of all data see Table 2.
S. Wittmann et al. / International Immunopharmacology 6 (2006) 61–70
3. Results Xenopus oocytes expressing fMLP receptors after injection of cRNA coding for the R98 variant of the human fMLP receptor were voltage-clamped (holding potential 70 mV; two-electrode technique) and stimulated by bath application of fMLP (100 nM) for 30 s. This resulted in a biphasic inward current as described before [2]. 3.1. Blockade of intracellular signal transduction Oocytes were stimulated three times with fMLP following intracellular injection with either heparin, U73122 or wortmannin, with the first stimulation performed 45 min after injection. Heparin enhanced the amplitude of the fast current by up to 66%. The effect of heparin on the slow current was extremely variable, tending to enhance the current amplitude but, on average, this was not a significant effect. Wortmannin did not alter either current, compared to control. U73122 inhibited the slow current, reducing its amplitude by some 54% but was without effect on the fast current (Table 1).
65
Table 2 Effects of anaesthetics on the fMLP-induced currents in Xenopus oocytes: analysis of concentration–response curves IC50 (AM)
Hill coefficient
Final value (AM)
Thiopentone Fast current Slow current
147 F 37 120 F 21
1.1 F 0.2 0.6 F 0.1
0.3 F 0.1 0
Methohexitone Fast current Slow current
4.7 F 5.0 2.2 F 1.3
0.5 F 0.2 0.5 F 0.1
0.4 F 0.1 0.4 F 0.1
Propofol Fast current Slow current
33 F 10 7.9 F 3.3
0.7 F 0.1 0.6 F 0.1
0 0
Data were fitted (Origin V, Microcal Software Inc.) to a logistic function of the following form: y¼
A1 A2 þ A2 1 þ ð x=x0 Þp
where y equals normalised current amplitude, x equals concentration, A 1 equals initial normalised current amplitude, A 2 equals final normalised current amplitude, x 0 equals IC50, and p equals Hill coefficient.
3.2. Treatment with anaesthetics Pre-treatment with thiopentone, methohexitone and propofol led to a concentration-dependent decrease in amplitude of both the fast and the slow currents (Fig. 1). The slow current was more sensitive to anaesthetic treatment as reflected by lower IC50 values. The Hill coefficients suggest that, in all cases, the interaction involves one anaesthetic molecule (Table 2). In parallel experiments, oocytes were treated with single concentrations of each anaesthetic to exclude desensitization phenomena induced by cumulative concentration–response experiments. The currents obtained upon stimulation with a single concentration did not differ significantly from those obtained with the corresponding anaesthetic concentration in the cumulative concentrations–response curve (Fig. 1). Thiopentone pre-treatment using concentrations of 500 AM and higher induced reversible inward currents without fMLP stimulation (results not shown). 3.3. Reversibility of anaesthetic inhibition of fMLP-induced currents After treatment with a single anaesthetic concentration followed by one stimulation with an fMLP/anaesthetic mixture, oocytes were stimulated three times in the absence of the anaesthetic. In general, neither the fast nor the slow current showed any recovery in amplitude, within the 45min period of observation, from the inhibition of the cur-
rents caused by exposure to anaesthetic (Table 3). After experiments to obtain cumulative concentration–response data, the fMLP-induced currents similarly showed no recovery to pre-anaesthetic exposure amplitudes (results not shown). 3.4. Effect of propofol solution adjuvants To analyse the effects of solution adjuvants of commercially available propofol preparations, oocytes were treated in parallel experiments with IntralipidR or glycerol instead of propofol. Effects of IntralipidR concentrations equivalent to propofol 5 and 50 AM were comparable to propofol effects on the fast current, while propofol 500 AM inhibited the fast current significantly more than IntralipidR alone (Table 4). The slow current was inhibited significantly more by propofol than by IntralipidR for all concentrations tested. Treatment of oocytes with glycerol concentrations equivalent to those found in the propofol formulation had no effect on the fMLP-induced currents (data not shown).
4. Discussion We have shown previously that the fast and the slow currents elicited by fMLP stimulation from
66
S. Wittmann et al. / International Immunopharmacology 6 (2006) 61–70
Table 3 Reversibility of anaesthetic inhibition of fMLP-induced currents: anaesthetic concentrations were chosen based on therapeutically effective concentrations Control
Anaesthetic
Recovery, 15 min
Recovery, 30 min
Recovery, 45 min
Fast current Thiopentone, 100 AM Methohexitone, 50 AM Propofol, 20 AM
1F0 1F0 1F0
0.60 F 0.08* 0.62 F 0.07** 0.42 F 0.07**
0.39 F 0.10** 0.17 F 0.07** 0.56 F 0.15**
0.99 F 0.26 0.40 F 0.12** 0.57 F 0.10**
0.36 F 0.07** 0.19 F 0.05** 0.48 F 0.10**
Slow current Thiopentone, 100 AM Methohexitone, 50 AM Propofol, 20 AM
1F0 1F0 1F0
0.46 F 0.08** 0.45 F 0.13** 0.59 F 0.09*
0.47 F 0.10** 0.40 F 0.07** 0.53 F 0.14**
0.63 F 0.10* 0.52 F 0.12** 0.75 F 0.13
0.46 F 0.09** 0.30 F 0.11** 0.58 F 0.14*
Data are presented as mean F SEM (n = 6 to n = 8) of the amplitude of the normalised current. Significances (*p b 0.05, **p b 0.01) were calculated for currents recorded in the presence of anaesthetic or post-anaesthetic recovery vs. control currents.
Xenopus oocytes expressing the human fMLP receptor are based on activation of different ion currents and channels. The fast current is carried by chloride and the slow current arises from the activation of a mixed population of cation channels [2]. The different time courses of these two currents had suggested that they would arise via different signalling pathways following activation of the G-protein-coupled fMLP receptor. The results obtained in the present study provide support for this conclusion because inhibition of PLC (U73122) decreased the slow current but had no effect on the fast (Cl) current. Heparin acts as a competitive antagonist for IP3 at the IP3 receptor and IP3 is known to play a major role in calcium release after agonist stimulation [21]. In our experiments, heparin increased the amplitude of Table 4 Comparison of the effects of propofol and the solution vehicle IntralipidR: for three concentrations of propofol the equivalent IntralipidR concentrations were tested Control
Fast current
Slow current
1F0 Propofol, 5 AM Intralipid Propofol, 50 AM Intralipid Propofol, 500 AM Intralipid
0.79 F 0.12 0.95 F 0.04 0.56 F 0.08** 0.54 F 0.08** 0.09 F 0.05**# 0.33 F 0.07**
0.56 F 0.10**# 0.99 F 0.08 0.33 F 0.04**# 0.54 F 0.08** 0.05 F 0.02**# 0.53 F 0.08**
Data are presented as mean F SEM (n = 5 to n = 12) of the amplitude of the normalised fMLP-induced currents. Significances were determined for propofol and IntralipidR concentrations vs. controls (**p b 0.01) and for propofol vs. equivalent IntralipidR concentrations (#p b 0.05).
the fast (Cl) current. It is unlikely, therefore, that the Cl current depends on calcium for its activation. The heparin-induced enhanced current might arise through a compensatory activation of other signalling cascades arising from a blockade of the IP3-associated pathway. In particular, if the fast current arises from a direct activation of the Clchannel, by either the a- or hgsubunits of the G-protein, then the enhanced current following IP3 antagonism may arise from a reduction in the cycling of the G-protein subunits due to a reduced availability of intracellular calcium. Results obtained from experiments with human neutrophils are in accordance with the results obtained from our oocyte experiments with U73122. Thus in neutrophils the fMLP-induced oxidative response was inhibited by treatment with U73122, as shown for the slow current recorded from oocytes, indicating a role of either PLC or PLC-associated pathways [22]. Wortmannin acts as a specific and irreversible inhibitor of PI3K in neutrophils and other cells [23]. In addition an inhibitory effect of wortmannin on the fMLP-induced PLD activity has been described [24]. Both PI3K and PLD are supposed to be part of the signalling pathways activated upon fMLP stimulation of neutrophils. Wortmannin’s effects in neutrophils are differential. fMLP-induced production of reactive oxygen derivates in neutropils diminishes upon pretreatment of cells with wortmannin, whereas parameters like the CD11b dependent adhesion and actin polymerisation of neutrophils were not affected. These results suggest the involvement of PI3K/PLD in some but not all neutrophil functions as summarized in Ref. [25]. Data obtained in our experiments
S. Wittmann et al. / International Immunopharmacology 6 (2006) 61–70
suggest no major importance of PI3K and PLD for signalling in fMLP-induced electrophysiological responses in Xenopus oocytes because conductances remained unchanged upon wortmannin treatment of the cells. Thus fMLP-induced signalling, involving the PI3K/PLD pathway, seems to be different for the oxidative response in neutrophils and the electrophysiological responses in oocytes. Based on results obtained with signal transduction blockers, potential signalling pathways for the fMLPinduced currents in Xenopus oocytes are shown in Fig. 2. Although PI3K and PLD exist in oocytes [17,26,27], they do not seem to be of importance for fMLP-induced signalling because wortmannin had no effect on either the fast or slow current. The PLCassociated pathway is therefore highlighted in Fig. 2 compared to the PI3K- and PLD-associated pathways. The ability of U73122 to reduce the amplitude of the slow current suggests that the PLC/PKC pathway is involved in activating the non-selective cation channels that give rise to this current. Since heparin does
67
not impair the slow current, it would appear that only the DAG-associated branch of the pathway is involved. If the PLD-associated pathway in oocytes was similar to that in neutrophils, activation of PLD should lead to increased levels of DAG and activation of PKC. The unaffected slow response upon wortmannin treatment could only be explained by the assumption that the production of DAG upon stimulation of PLD is minor compared to DAG production upon activation of PLC. Thus, wortmannin treatment alone might not be sufficient to significantly impair PKC activity or the slow current. Thiopentone, methohexitone and propofol all reduced, in a concentration-dependent manner, both the fast and slow fMLP-induced currents. The order of potency was found to be methohexitone N propofol N thiopentone (Table 2) whereas in human neutrophils the order of potency against the oxidative response is thiopentone N propofol H methohexitone [18,28]. Although anaesthetics inhibit both fast and
Fig. 2. Suggested signalling pathways in oocytes expressing the seven transmembrane spanning G-protein coupled receptor for fMLP. After agonist (fMLP) binding to the receptor the hg subunits dissociate from the a subunit. The G-protein related to fMLP receptors in Xenopus oocytes is presumably Gq. Via the hg subunits signals are transduced to phosphoinositide 3-kinase (PI3K), phospholipase D (PLD) and phospholipase C (PLC). Chloride channels might be activated by either the a or the hg subunit(s). The mixed cation (X+) channels might be activated by the protein kinase C (PKC). Other components of the intracellular signalling shown in this figure are phosphoinositide 4,5diphosphate (PIP2), phosphoinositide 3,4,5-trisphosphate (PIP3), phosphatidylcholine (PC), phosphatidic acid (PA), diacylglycerol (DAG) and inositol 1,4,5-trisphosphate (IP3). Sites of action of the signal transduction blockers wortmannin, U73122 and heparin are marked with T. For detailed comments see Discussion.
68
S. Wittmann et al. / International Immunopharmacology 6 (2006) 61–70
slow currents in oocytes, the slow current is consistently more sensitive than the fast current (Fig. 1 and Table 2). For all anaesthetics, concentrations within the range of therapeutic plasma levels had significant effects. For thiopentone, methohexitone and propofol, concentrations in the range 4 to 165 AM (1 to 40 Ag/ml), 4 to 63 AM (1 to 18 Ag/ml) and 17 to 28 AM (3 to 5 Ag/ml) are considered to be therapeutically effective plasma concentrations [29–34]. For propofol we could show that the solution adjuvant IntralipidR by itself dose-dependently and differentially affects the fMLP-induced fast and slow current. IntralipidR-induced impairment of currents was in part comparable and in part smaller than the effects of the commercially available propofol solution (Table 4). The effect of these intravenous anaesthetics was found to be long-lasting. This is in accord with neutrophil data, which indicated that nitrous oxide produces a sustained impairment of the oxidative response, of up to 4 h, and that the oxidative response remained impaired for more than 30/60 min after halothane/desflurane treatment. For sevoflurane and isoflurane, however, an immediate reversibility of the effect was shown [35,36]. In conclusion the fast (Cl) current induced by fMLP in oocytes might arise from a direct activation of the chloride channel by the a- or hg-subunits of the activated G-protein. We suggest that the slow current arises following PKC phosphorylation of the non-selective cation channels through activation of PLC. Inhibition by anaesthetics of both the fast and the slow fMLP-induced currents suggests that they might act primarily through an interaction with the agonist/receptor, but the fact that the slow current is consistently more sensitive to anaesthetics provides evidence that the anaesthetics might also affect the intracellular signalling pathways. Whilst it is generally accepted that there are differences between the transduction of receptor binding to physiological effect in the oocyte and the human neutrophil, Xenopus oocytes have been used extensively in experiments aimed at understanding the mechanisms underlying the effects of anaesthetics [20,37–39]. The results described in this paper indicate that anaesthetics inhibit the effect of anaesthetics on fMLP-induced currents in oocytes, as they do the fMLP-induced oxidative response in
human neutrophils. Although there are differences in detail to the anaesthetic-induced inhibition in the two systems, the results described here suggest that an important interaction is a negative allosteric modulation at the fMLP receptor. This not only has implications for the immunological status of patients and may in part explain the increased susceptibility to infection of patients maintained on very long term anaesthesia but also suggests that anaesthetics may affect homeostatic mechanisms, hormone balance and nociception in addition to their primary effect of causing a reversible loss of consciousness.
Acknowledgements We thank Dr. W. Bautsch (Hannover Medical School) for the gift of the cDNA coding for the human fMLP (R98 variant). Grant support: Forschungsfo¨rderprogramm ReForM of the University of Regensburg. Additional financial support: Baxter Germany, Erlangen, Germany, and Abbott Germany, Wiesbaden, Germany.
References [1] Becker EL, Forouhar FA, Grunnet ML, Boulay F, Tardif M, Bormann BJ, et al. Broad immunocytochemical localization of the formylpeptide receptor in human organs, tissues, and cells. Cell Tissue Res 1998;292:129 – 35. [2] Wittmann S, Fro¨hlich D, Daniels S. Characterization of the human fMLP receptor in neutrophils and in Xenopus oocytes. Br J Pharmacol 2002;135:1375 – 82. [3] Boulay F, Naik N, Giannini E, Tardif M, Brouchon L. Phagocyte chemoattractant receptors. Ann N Y Acad Sci 1997;832:69 – 84. [4] Gerard C, Gerard NP. The pro-inflammatory seven-transmembrane segment receptors of the leukocyte. Curr Opin Immunol 1994;6:140 – 5. [5] Burg M, Raffetseder U, Grove M, Klos A, Kohl J, Bautsch W. G[alpha]-16 complements the signal transduction cascade of chemotactic receptors for complement factor C5a (C5a-R) and N-formylated peptides (fMLF-R) in Xenopus laevis oocytes: G[alpha]-16 couples to chemotactic receptors in Xenopus oocytes. FEBS Lett 1995;377:426 – 8. [6] Blank J, Brattain K, Exton J. Activation of cytosolic phosphoinositide phospholipase C by G-protein beta gamma subunits. J Biol Chem 1992;267:23069 – 75. [7] Filtz TM, Paterson A, Harden TK. Purification and G protein subunit regulation of a phospholipase C-beta from Xenopus laevis oocytes. J Biol Chem 1996;271:31121 – 6.
S. Wittmann et al. / International Immunopharmacology 6 (2006) 61–70 [8] Stephens LR, Hughes KT, Irvine RF. Pathway of phosphatidylinositol(3,4,5)-trisphosphate synthesis in activated neutrophils. Nature 1991;351:33 – 9. [9] Thibault N, Burelout C, Harbour D, Borgeat P, Naccache PH, Bourgoin SG. Occupancy of adenosine A2a receptors promotes fMLP-induced cyclic AMP accumulation in human neutrophils: impact on phospholipase D activity and recruitment of small GTPases to membranes. J Leukoc Biol 2002;71:367 – 77. [10] Fensome A, Whatmore J, Morgan C, Jones D, Cockcroft S. ADP-ribosylation factor and rho proteins mediate fMLPdependent activation of phospholipase D in human neutrophils. J Biol Chem 1998;273:13157 – 64. [11] Choi S, Rho SH, Jung SY, Kim SC, Park CS, Nah SY. A novel activation of Ca2+-activated Cl channel in Xenopus oocytes by ginseng saponins: evidence for the involvement of phospholipase C and intracellular Ca2+ mobilization. Br J Pharmacol 2001;132:641 – 8. [12] Kamatchi GL, Tiwari SN, Chan CK, Chen D, Do SH, Durieux ME, et al. Distinct regulation of expressed calcium channels 2.3 in Xenopus oocytes by direct or indirect activation of protein kinase C. Brain Res 2003;968:227 – 37. [13] Johnson J, Capco DG. Progesterone acts through protein kinase C to remodel the cytoplasm as the amphibian oocyte becomes the fertilization-competent egg. Mech Dev 1997;67:215 – 26. [14] Wang YX, Dhulipala PD, Li L, Benovic JL, Kotlikoff MI. Coupling of M2 muscarinic receptors to membrane ion channels via phosphoinositide 3-kinase gamma and atypical protein kinase C. J Biol Chem 1999;274:13859 – 64. [15] Terasaki M, Runft LL, Hand AR. Changes in organization of the endoplasmic reticulum during Xenopus oocyte maturation and activation. Mol Biol Cell 2001;12:1103 – 16. [16] Choi S, Kim HJ, Ko YS, Jeong SW, Kim YI, Simonds WF, et al. Galpha q/11 coupled to mammalian phospholipase C beta 3-like enzyme mediates the ginsenoside effect on Ca2+activated Cl current in the Xenopus oocyte. J Biol Chem 2001; 276:48797 – 802. [17] Carnero A, Dolfi F, Lacal JC. Ras-p21 activates phospholipase D and A2, but not phospholipase C or PKC, in Xenopus laevis oocytes. J Cell Biochem 1994;54:478 – 86. [18] Frohlich D, Rothe G, Schwall B, Schmitz G, Hobbhahn J, Taeger K. Thiopentone and propofol, but not methohexitone nor midazolam, inhibit neutrophil oxidative responses to the bacterial peptide FMLP. Eur J Anaesthesiol 1996;13:582 – 8. [19] Frohlich D, Wittmann S, Rothe G, Schmitz G, Taeger K. Thiopental impairs neutrophil oxidative response by inhibition of intracellular signalling. Eur J Anaesthesiol 2002; 19:474 – 82. [20] Nietgen GW, Honemann CW, Durieux ME. Influence of anesthetics on endogenous and recombinantly expressed G proteincoupled receptors in the Xenopus oocyte. Toxicol Lett 1998;100–101:319 – 27. [21] Ghosh T, Eis P, Mullaney J, Ebert C, Gill D. Competitive, reversible, and potent antagonism of inositol 1,4,5-trisphosphate-activated calcium release by heparin. J Biol Chem 1988;263:11075 – 9.
69
[22] Chen LW, Jan CR. Mechanisms and modulation of formyl– methionyl–leucyl–phenylalanine (fMLP)-induced Ca2+ mobilization in human neutrophils. Int Immunopharmacol 2001;1:1341 – 9. [23] Downey GP, Fukushima T, Fialkow L, Waddell TK. Intracellular signaling in neutrophil priming and activation. Semin Cell Biol 1995;6:345 – 56. [24] Nakamura M, Nakashima S, Katagiri Y, Nozawa Y. Effect of wortmannin and 2-(4-morpholinyl)-8-phenyl-4H-1-benzopyran-4-one (LY294002) on N-formyl–methionyl–leucyl–phenylalanine-induced phospholipase D activation in differentiated HL60 cells: possible involvement of phosphatidylinositol 3kinase in phospholipase deactivation. Biochem Pharmacol 1997;53:1929 – 36. [25] Prossnitz ER, Ye RD. The N-formyl peptide receptor: a model for the study of chemoattractant receptor structure and function. Pharmacol Ther 1997;74:73 – 102. [26] Do SH, Fang HY, Ham BM, Zuo ZMD. The effects of lidocaine on the activity of glutamate transporter EAAT3: the role of protein kinase C and phosphatidylinositol 3-kinase. Anesth Analg 2002;95:1263 – 8. [27] Hehl S, Stoyanov B, Oehrl W, Schonherr R, Wetzker R, Heinemann SH. Phosphoinositide 3-kinase-gamma induces Xenopus oocyte maturation via lipid kinase activity. Biochem J 2001;360:691 – 8. [28] Wittmann S, Daniels S, Ittner KP, Frohlich D. Thiopentone and methohexitone enantiomers do not act stereoselectively on the oxidative response in human neutrophils in vitro. Pharmacology 2004;72:12 – 9. [29] Cordato DJ, Gross AS, Herkes GK, Mather LE. Pharmacokinetics of thiopentone enantiomers following intravenous injection or prolonged infusion of rac-thiopentone. Br J Clin Pharmacol 1997;43:355 – 62. [30] Crankshaw DP, Boyd MD, Bjorksten AR. Plasma drug efflux—a new approach to optimization of drug infusion for constant blood concentration of thiopental and methohexital. Anesthesiology 1987;67:32 – 41. [31] Davidson JA, Boom SJ, Pearsall FJ, Zhang P, Ramsay G. Comparison of the effects of four i.v. anaesthetic agents on polymorphonuclear leucocyte function. Br J Anaesth 1995;74:315 – 8. [32] Nguyen KT, Stephens DP, McLeish MJ, Crankshaw DP, Morgan DJ. Pharmacokinetics of thiopental and pentobarbital enantiomers after intravenous administration of racemic thiopental. Anesth Analg 1996;83:552 – 8. [33] Schickendantz J, Funk W, Ittner KP, Gruber M, Taeger K, Kees F. Elimination of methohexitone after long-term, highdose infusion in patients with critically elevated intracranial pressure. Crit Care Med 1999;27:1570 – 6. [34] Roberts FL, Dixon J, Lewis GT, Tackley RM, PrysRoberts C. Induction and maintenance of propofol anaesthesia. A manual infusion scheme. Anaesthesia 1988;43: 14 – 17. [35] Frohlich D, Rothe G, Schwall B, Schmid P, Schmitz G, Taeger K, et al. Effects of volatile anaesthetics on human neutrophil oxidative response to the bacterial peptide FMLP. Br J Anaesth 1997; 78:718 – 23.
70
S. Wittmann et al. / International Immunopharmacology 6 (2006) 61–70
[36] Frohlich D, Rothe G, Wittmann S, Schmitz G, Schmid P, Taeger K, et al. Nitrous oxide impairs the neutrophil oxidative response. Anesthesiology 1998;88:1281 – 90. [37] Chan CK, Durieux ME. Differential inhibition of lysophosphatidate signaling by volatile anesthetics. Anesthesiology 1997;86:660 – 9. [38] Honemann CW, Nietgen GW, Podranski T, Chan CK, Durieux ME. Influence of volatile anesthetics on thromboxane A2 signaling. Anesthesiology 1998;88:440 – 51.
[39] Hollmann MW, Ritter CH, Henle P, de Klaver M, Kamatchi GL, Durieux ME. Inhibition of m3 muscarinic acetylcholine receptors by local anaesthetics. Br J Pharmacol 2001; 133:207 – 16.
Anaesthetic sensitivity of fMLP-induced cell signalling in Xenopus oocytes Sigrid Wittmann a,*, Dieter Fro¨hlich a, Andrea Mietens a, Stephen Daniels b a
Department of Anaesthesiology, University of Regensburg, 93042 Regensburg, Germany b Welsh School of Pharmacy, Cardiff University, Cardiff CF10 3 XF, Wales/UK Received 3 April 2005; received in revised form 26 April 2005; accepted 27 July 2005
Abstract FMLP stimulation of Xenopus oocytes expressing fMLP receptors leads to a concentration-dependent biphasic inward current. To identify the evolution of these currents we have examined the effects of blocking various cell signalling pathways. In addition we have analysed the effects of three intravenous anaesthetics on these fMLP-induced currents. Xenopus oocytes were microinjected with cRNA encoding the fMLP receptor and fMLP-stimulated (100 nM) currents measured, using two-electrode voltage-clamp (70 mV), before and after injection of heparin (120 ng ml 1), wortmannin (1 AM), U73122 (5 AM) or buffer. Concentration–response curves were established for the action on fMLP-stimulated currents of thiopentone (5–500 AM), methohexitone (0.2–200 AM) and propofol (0.5–500 AM). Heparin significantly enhanced the fast current ( p b 0.05). Wortmannin had no effect on either current. U73122 inhibited only the slow current ( p b 0.05). All anaesthetics inhibited both currents, with the maximum inhibition for the fast/slow currents 70%/100%, 60%/60% and 100%/100% for thiopentone (IC50 147 / 120 AM), methohexitone (IC50 4.7 / 2.2 AM) and propofol (IC50 33 / 8 AM), respectively. We suggest (a) the slow current arises via the PLC/PKC pathway because it is reduced by the PLC inhibitor U73122, (b) the PI3K- and PLD-mediated pathways are not involved because wortmannin had no effect and (c) activation of the two conductance channels must be different because U73122 reduced the slow but not the fast current. Since both currents are decreased by all three anaesthetics, their inhibition might be mediated through an action at the agonist/receptor, although, since the slow current is consistently more sensitive than the fast, there may be additionally an action on cell signalling. D 2005 Elsevier B.V. All rights reserved. Keywords: Thiopentone; Methohexitone; Propofol; Heparin; U73122; Wortmannin
* Corresponding author. Tel.: +49 941 9447801; fax: +49 941 9447802. E-mail address: [email protected] (S. Wittmann). 1567-5769/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.intimp.2005.07.012
1. Introduction N-formylated peptides, such as fMLP (N-formyl– l-methionyl–l-leucyl–phenylalanine), mediate host
62
S. Wittmann et al. / International Immunopharmacology 6 (2006) 61–70
defence mechanisms such as the oxidative response (production of reactive oxygen derivatives) in human neutrophils [1]. Functional expression of the G-protein coupled fMLP receptor in Xenopus oocytes was reported following microinjection of cRNA [2]. Exposure of Xenopus oocytes expressing the fMLP receptor to fMLP produces a concentration-dependent biphasic inward current with a fast component, carried by Cl, and a slow component, linked to a mixed population of cation channels [2]. Comparative experiments in human neutrophils showed that their activation with fMLP is dependent on store-operated calcium influx that is regulated by chloride channel activation and linked to non-selective cation channels [2]. Chloride and cation channels play a role therefore in both cell systems. Gprotein coupling to the fMLP receptor in human neutrophils has been linked to Gia2 or Gia3 [3,4], whereas in Xenopus oocytes a major role for Gq was reported [2,5]. Intracellular signalling in fMLP-stimulated human neutrophils is connected to three main pathways. The first pathway involves PLC (phospholipase C), PIP2 (PI 4,5-P2, phosphoinositide 4,5diphosphate) and its cleavage products IP3 (Ins 1,4,5-P3, inositol 1,4,5-trisphosphate) and DAG (diacylglycerol). PLC has been shown to be activated by hg subunit complexes of pertussis toxinsensitive or -insensitive G-proteins as well as by a subunits of the Gq family [6,7]. A central element of the second pathway is the PI3K (phosphoinositide 3-kinase). This second pathway is linked to PIP3 (PI 3,4,5-P3, phosphoinositide 3,4,5-trisphosphate) as well as PIP2 [8]. The role of the third PLD (phospholipase D)-associated pathway remains to be resolved. Previous studies suggested that phospholipase D might be activated not only via the hg subunits of the G-protein but also via the phospholipase C-associated pathway [9,10]. The intracellular signalling components involved in fMLP-induced currents in Xenopus oocytes functionally expressing fMLP receptors are currently unknown. However, the existence of many signalling components and their role in responses other than that to fMLP have been shown for Xenopus oocytes. Ginsenoside-induced responses involving PLC have been associated with IP3 and PTX-insensitive G-proteins [11]. M1 muscarinic receptor acti-
vation in Xenopus oocytes has been shown to involve PLC, DAG, IP3 and different PKC isoenzymes [12]. Progesterone treatment of Xenopus oocytes indicates an involvement of PLC and PKC [13]. In addition different PLC isoforms, including PLC h as well as PLD, PI3K and IP3-responsive calcium stores, have been described in Xenopus oocytes [14–17]. This study reports the effects of the signal transduction blockers wortmannin, U73122 and heparin in Xenopus oocytes, functionally expressing fMLP receptors, in order to provide evidence for the fMLP-associated pathway in these cells. In addition we analysed the sensitivity of both the fast and the slow fMLP-induced currents in Xenopus oocytes to the intravenous anaesthetics thiopentone, methohexitone and propofol. Previous studies with human neutrophils have shown a concentration-dependent impairment of the fMLP-induced oxidative response in human neutrophils upon treatment with either thiopentone, methohexitone or propofol [18,19].
2. Materials and methods 2.1. Materials Female extra-large mature Xenopus laevis were obtained from Blades Biological, UK (experiments in Wales) or the University of Bayreuth (experiments in Germany). W. Bautsch (Hannover Medical School) donated the cDNA coding for the human R98 variant of the fMLP receptor in plasmid vector pCDM8. Ultracompetent E. coli MC1061/p3 were obtained from Invitrogen, Netherlands. Other materials were purchased from the following companies: N-formyl–l-methionyl–l-leucyl–phenylalanine (fMLP) and heparin from Sigma, Germany; thiopentone (TrapanalR) from Byk Gulden, Netherlands; propofol (Propofol Abbott 1%) from Abbott, Germany; methohexitone (BrevimytalR) from Lilly, Germany; IntralipidR 10 from Baxter, Germany; and wortmannin and U-73122 from Alexis, Germany. 2.2. Amplification and purification of the plasmid Amplification and purification of the plasmid were performed using ultracompetent E. coli (MC 1061/p3) and the Wizard SV Miniprep DNA Purification System from Promega, UK.
S. Wittmann et al. / International Immunopharmacology 6 (2006) 61–70
63
2.3. Preparation of cRNA
2.6. Experiments with signal transduction blocking drugs
Plasmids containing the cDNA coding for the human fMLP receptor (R98) were linearised with HpaI and cRNA prepared using the mCAP mRNA capping kit from Stratagene, UK. The size of the cRNA was assessed using ethidium bromide-stained agarose gel electrophoresis with RNA reference fragments of defined size.
Each oocyte was stimulated three times with fMLP, in ND96. The oocyte was then de-clamped and heparin, wortmannin or U-73122 was injected intracellularly. The injection volume was 50 nl at a drug concentration designed to achieve intracellular concentrations of 120 ng/ml for heparin, 1 AM for wortmannin and 5 AM for U-73122. Based on previous publications, intracellular blocking drugs were allowed 45 min to achieve their effects [20]. After 45 min of incubation in ND96 without stimulation, the oocyte was re-clamped and stimulated three times with fMLP. In parallel control experiments oocytes were injected with buffer instead of blocking drugs.
2.4. Preparation of oocytes and injection of cRNA X. laevis were killed (experiments in Wales) by decapitation followed by mechanical destruction of the brain and spinal cord or anaesthetized with MS 222 (experiments in Germany) and pieces of ovary tissue were removed by laparotomy. The oocytes were manually stripped from the ovarian lobes using a soft plastic strip. Stage V and VI oocytes were selected, injected with 35–45 nl cRNA using a hydraulic microinjector and placed into sterile pots containing modified Barth’s solution (in mM: NaCl, 100; KCl, 1; NaHCO3, 2; MgSO4, 0.82; Ca(NO3)2, 0.33; CaCl2, 0.41; HEPES, 10; pH 7.4 with NaOH) supplemented with penicillin (100 U cm 3) and streptomycin (100 Ag cm 3). 2.5. Electrophysiological recording Prior to use oocytes were defolliculated by incubation for 15 min in Ca2+-free Barth’s solution containing collagenase (50–100 U cm 3). Electrophysiological recordings were made using two-electrode voltage-clamp (VF 180 amplifiers and CA 100 clamp amplifier, Biologic, France for experiments performed in Wales; Gene Clamp 500, Axon Instruments Inc., USA for experiments performed in Germany) at a holding potential of 70 mV and superfusion with ND96 (in mM: NaCl, 96; KCl, 2; CaCl2, 1.8; MgCl2, 1; HEPES, 5; pH 7.4 with NaOH).
2.7. Experiments with intravenous (iv) anaesthetics For all experiments with iv anaesthetics, bath application of the anaesthetic (in ND96) was performed (with the oocyte at a holding potential of 70 mV) for 90 s before the oocyte was stimulated with fMLP (fMLP– anaesthetic mixture). For each anaesthetic a cumulative concentration–response curve was measured (thiopentone 5 to 500 AM, methohexitone 0.2 to 200 AM, and propofol 0.5 to 500 AM). To exclude desensitization phenomena, separate experiments were performed in which only one anaesthetic concentration was tested (thiopentone 50 and 100 AM, propofol 20 and 200 AM, methohexitone 20 and 50 AM). To exclude the possibility that the effects induced by propofol were caused by the solution vehicle, equivalent concentrations of an IntralipidR solution (Intralipid 10% from Pharmacia Upjohn, USA) were tested in parallel to the experiments with propofol. In further control experiments, oocytes were pre-treated with glycerol (Sigma, Germany) to test the effects of this additive.
Table 1 Effects of signal transduction blockade on fMLP-induced currents in Xenopus oocytes expressing the fMLP receptor Blocking drug
Fast current Heparin U73122 Wortmannin Slow current Heparin U73122 Wortmannin
Normalised amplitude of current before injection
Normalised current amplitude post-injection of blocking drug First stimulation
Second stimulation
Third stimulation
1F0
1.33 F 0.17 1.04 F 0.14 1.21 F 0.14
1.66 F 0.22* 0.95 F 0.17 1.14 F 0.08
1.61 F 0.24* 0.94 F 0.25 1.07 F 0.12
1F0
1.36 F 0.29 0.95 F 0.21 1.26 F 0.15
1.79 F 0.35 0.67 F 0.15 0.91 F 0.15
1.14 F 0.19 0.46 F 0.15* 0.78 F 0.13
Data are presented as mean F SEM (n = 6 to n = 8) of the normalised (see text) current amplitude. Significances (*p b 0.05) were calculated for currents obtained after treatment with a blocking drug vs. control currents.
64
S. Wittmann et al. / International Immunopharmacology 6 (2006) 61–70
2.8. Data and statistical analysis Currents recorded from individual oocytes were normalised, with respect to the appropriate control current from that oocyte, so that results could be pooled for analysis. Concentration–response curves were constructed using non-lin-
ear curve fitting routines applied to a standard logistical equation (Origin 6.0, Microcal Software Inc, USA). Homogeneity of variance was analysed using Levene’s test. The statistical difference between treatment groups was tested, where appropriate, using a one-way ANOVA including post hoc testing with Dunnett’s or Bonferroni’s procedure.
Fig. 1. (A and B) The effect of thiopentone on the fMLP-induced fast (A) and slow (B) currents in Xenopus oocytes. The symbols represent the currents obtained from cumulative dose–response curves (n) and single anaesthetic treatment (5). Current amplitudes were normalised to fMLP control without thiopentone treatment and are presented as mean F SEM (n = 8). Significances were determined for anaesthetic concentrations vs. controls *p b 0.05, **p b 0.01). Data (C and D) for methohexitone and (E and F) for propofol are presented analogous to thiopentone data above. For logistic fit of all data see Table 2.
S. Wittmann et al. / International Immunopharmacology 6 (2006) 61–70
3. Results Xenopus oocytes expressing fMLP receptors after injection of cRNA coding for the R98 variant of the human fMLP receptor were voltage-clamped (holding potential 70 mV; two-electrode technique) and stimulated by bath application of fMLP (100 nM) for 30 s. This resulted in a biphasic inward current as described before [2]. 3.1. Blockade of intracellular signal transduction Oocytes were stimulated three times with fMLP following intracellular injection with either heparin, U73122 or wortmannin, with the first stimulation performed 45 min after injection. Heparin enhanced the amplitude of the fast current by up to 66%. The effect of heparin on the slow current was extremely variable, tending to enhance the current amplitude but, on average, this was not a significant effect. Wortmannin did not alter either current, compared to control. U73122 inhibited the slow current, reducing its amplitude by some 54% but was without effect on the fast current (Table 1).
65
Table 2 Effects of anaesthetics on the fMLP-induced currents in Xenopus oocytes: analysis of concentration–response curves IC50 (AM)
Hill coefficient
Final value (AM)
Thiopentone Fast current Slow current
147 F 37 120 F 21
1.1 F 0.2 0.6 F 0.1
0.3 F 0.1 0
Methohexitone Fast current Slow current
4.7 F 5.0 2.2 F 1.3
0.5 F 0.2 0.5 F 0.1
0.4 F 0.1 0.4 F 0.1
Propofol Fast current Slow current
33 F 10 7.9 F 3.3
0.7 F 0.1 0.6 F 0.1
0 0
Data were fitted (Origin V, Microcal Software Inc.) to a logistic function of the following form: y¼
A1 A2 þ A2 1 þ ð x=x0 Þp
where y equals normalised current amplitude, x equals concentration, A 1 equals initial normalised current amplitude, A 2 equals final normalised current amplitude, x 0 equals IC50, and p equals Hill coefficient.
3.2. Treatment with anaesthetics Pre-treatment with thiopentone, methohexitone and propofol led to a concentration-dependent decrease in amplitude of both the fast and the slow currents (Fig. 1). The slow current was more sensitive to anaesthetic treatment as reflected by lower IC50 values. The Hill coefficients suggest that, in all cases, the interaction involves one anaesthetic molecule (Table 2). In parallel experiments, oocytes were treated with single concentrations of each anaesthetic to exclude desensitization phenomena induced by cumulative concentration–response experiments. The currents obtained upon stimulation with a single concentration did not differ significantly from those obtained with the corresponding anaesthetic concentration in the cumulative concentrations–response curve (Fig. 1). Thiopentone pre-treatment using concentrations of 500 AM and higher induced reversible inward currents without fMLP stimulation (results not shown). 3.3. Reversibility of anaesthetic inhibition of fMLP-induced currents After treatment with a single anaesthetic concentration followed by one stimulation with an fMLP/anaesthetic mixture, oocytes were stimulated three times in the absence of the anaesthetic. In general, neither the fast nor the slow current showed any recovery in amplitude, within the 45min period of observation, from the inhibition of the cur-
rents caused by exposure to anaesthetic (Table 3). After experiments to obtain cumulative concentration–response data, the fMLP-induced currents similarly showed no recovery to pre-anaesthetic exposure amplitudes (results not shown). 3.4. Effect of propofol solution adjuvants To analyse the effects of solution adjuvants of commercially available propofol preparations, oocytes were treated in parallel experiments with IntralipidR or glycerol instead of propofol. Effects of IntralipidR concentrations equivalent to propofol 5 and 50 AM were comparable to propofol effects on the fast current, while propofol 500 AM inhibited the fast current significantly more than IntralipidR alone (Table 4). The slow current was inhibited significantly more by propofol than by IntralipidR for all concentrations tested. Treatment of oocytes with glycerol concentrations equivalent to those found in the propofol formulation had no effect on the fMLP-induced currents (data not shown).
4. Discussion We have shown previously that the fast and the slow currents elicited by fMLP stimulation from
66
S. Wittmann et al. / International Immunopharmacology 6 (2006) 61–70
Table 3 Reversibility of anaesthetic inhibition of fMLP-induced currents: anaesthetic concentrations were chosen based on therapeutically effective concentrations Control
Anaesthetic
Recovery, 15 min
Recovery, 30 min
Recovery, 45 min
Fast current Thiopentone, 100 AM Methohexitone, 50 AM Propofol, 20 AM
1F0 1F0 1F0
0.60 F 0.08* 0.62 F 0.07** 0.42 F 0.07**
0.39 F 0.10** 0.17 F 0.07** 0.56 F 0.15**
0.99 F 0.26 0.40 F 0.12** 0.57 F 0.10**
0.36 F 0.07** 0.19 F 0.05** 0.48 F 0.10**
Slow current Thiopentone, 100 AM Methohexitone, 50 AM Propofol, 20 AM
1F0 1F0 1F0
0.46 F 0.08** 0.45 F 0.13** 0.59 F 0.09*
0.47 F 0.10** 0.40 F 0.07** 0.53 F 0.14**
0.63 F 0.10* 0.52 F 0.12** 0.75 F 0.13
0.46 F 0.09** 0.30 F 0.11** 0.58 F 0.14*
Data are presented as mean F SEM (n = 6 to n = 8) of the amplitude of the normalised current. Significances (*p b 0.05, **p b 0.01) were calculated for currents recorded in the presence of anaesthetic or post-anaesthetic recovery vs. control currents.
Xenopus oocytes expressing the human fMLP receptor are based on activation of different ion currents and channels. The fast current is carried by chloride and the slow current arises from the activation of a mixed population of cation channels [2]. The different time courses of these two currents had suggested that they would arise via different signalling pathways following activation of the G-protein-coupled fMLP receptor. The results obtained in the present study provide support for this conclusion because inhibition of PLC (U73122) decreased the slow current but had no effect on the fast (Cl) current. Heparin acts as a competitive antagonist for IP3 at the IP3 receptor and IP3 is known to play a major role in calcium release after agonist stimulation [21]. In our experiments, heparin increased the amplitude of Table 4 Comparison of the effects of propofol and the solution vehicle IntralipidR: for three concentrations of propofol the equivalent IntralipidR concentrations were tested Control
Fast current
Slow current
1F0 Propofol, 5 AM Intralipid Propofol, 50 AM Intralipid Propofol, 500 AM Intralipid
0.79 F 0.12 0.95 F 0.04 0.56 F 0.08** 0.54 F 0.08** 0.09 F 0.05**# 0.33 F 0.07**
0.56 F 0.10**# 0.99 F 0.08 0.33 F 0.04**# 0.54 F 0.08** 0.05 F 0.02**# 0.53 F 0.08**
Data are presented as mean F SEM (n = 5 to n = 12) of the amplitude of the normalised fMLP-induced currents. Significances were determined for propofol and IntralipidR concentrations vs. controls (**p b 0.01) and for propofol vs. equivalent IntralipidR concentrations (#p b 0.05).
the fast (Cl) current. It is unlikely, therefore, that the Cl current depends on calcium for its activation. The heparin-induced enhanced current might arise through a compensatory activation of other signalling cascades arising from a blockade of the IP3-associated pathway. In particular, if the fast current arises from a direct activation of the Clchannel, by either the a- or hgsubunits of the G-protein, then the enhanced current following IP3 antagonism may arise from a reduction in the cycling of the G-protein subunits due to a reduced availability of intracellular calcium. Results obtained from experiments with human neutrophils are in accordance with the results obtained from our oocyte experiments with U73122. Thus in neutrophils the fMLP-induced oxidative response was inhibited by treatment with U73122, as shown for the slow current recorded from oocytes, indicating a role of either PLC or PLC-associated pathways [22]. Wortmannin acts as a specific and irreversible inhibitor of PI3K in neutrophils and other cells [23]. In addition an inhibitory effect of wortmannin on the fMLP-induced PLD activity has been described [24]. Both PI3K and PLD are supposed to be part of the signalling pathways activated upon fMLP stimulation of neutrophils. Wortmannin’s effects in neutrophils are differential. fMLP-induced production of reactive oxygen derivates in neutropils diminishes upon pretreatment of cells with wortmannin, whereas parameters like the CD11b dependent adhesion and actin polymerisation of neutrophils were not affected. These results suggest the involvement of PI3K/PLD in some but not all neutrophil functions as summarized in Ref. [25]. Data obtained in our experiments
S. Wittmann et al. / International Immunopharmacology 6 (2006) 61–70
suggest no major importance of PI3K and PLD for signalling in fMLP-induced electrophysiological responses in Xenopus oocytes because conductances remained unchanged upon wortmannin treatment of the cells. Thus fMLP-induced signalling, involving the PI3K/PLD pathway, seems to be different for the oxidative response in neutrophils and the electrophysiological responses in oocytes. Based on results obtained with signal transduction blockers, potential signalling pathways for the fMLPinduced currents in Xenopus oocytes are shown in Fig. 2. Although PI3K and PLD exist in oocytes [17,26,27], they do not seem to be of importance for fMLP-induced signalling because wortmannin had no effect on either the fast or slow current. The PLCassociated pathway is therefore highlighted in Fig. 2 compared to the PI3K- and PLD-associated pathways. The ability of U73122 to reduce the amplitude of the slow current suggests that the PLC/PKC pathway is involved in activating the non-selective cation channels that give rise to this current. Since heparin does
67
not impair the slow current, it would appear that only the DAG-associated branch of the pathway is involved. If the PLD-associated pathway in oocytes was similar to that in neutrophils, activation of PLD should lead to increased levels of DAG and activation of PKC. The unaffected slow response upon wortmannin treatment could only be explained by the assumption that the production of DAG upon stimulation of PLD is minor compared to DAG production upon activation of PLC. Thus, wortmannin treatment alone might not be sufficient to significantly impair PKC activity or the slow current. Thiopentone, methohexitone and propofol all reduced, in a concentration-dependent manner, both the fast and slow fMLP-induced currents. The order of potency was found to be methohexitone N propofol N thiopentone (Table 2) whereas in human neutrophils the order of potency against the oxidative response is thiopentone N propofol H methohexitone [18,28]. Although anaesthetics inhibit both fast and
Fig. 2. Suggested signalling pathways in oocytes expressing the seven transmembrane spanning G-protein coupled receptor for fMLP. After agonist (fMLP) binding to the receptor the hg subunits dissociate from the a subunit. The G-protein related to fMLP receptors in Xenopus oocytes is presumably Gq. Via the hg subunits signals are transduced to phosphoinositide 3-kinase (PI3K), phospholipase D (PLD) and phospholipase C (PLC). Chloride channels might be activated by either the a or the hg subunit(s). The mixed cation (X+) channels might be activated by the protein kinase C (PKC). Other components of the intracellular signalling shown in this figure are phosphoinositide 4,5diphosphate (PIP2), phosphoinositide 3,4,5-trisphosphate (PIP3), phosphatidylcholine (PC), phosphatidic acid (PA), diacylglycerol (DAG) and inositol 1,4,5-trisphosphate (IP3). Sites of action of the signal transduction blockers wortmannin, U73122 and heparin are marked with T. For detailed comments see Discussion.
68
S. Wittmann et al. / International Immunopharmacology 6 (2006) 61–70
slow currents in oocytes, the slow current is consistently more sensitive than the fast current (Fig. 1 and Table 2). For all anaesthetics, concentrations within the range of therapeutic plasma levels had significant effects. For thiopentone, methohexitone and propofol, concentrations in the range 4 to 165 AM (1 to 40 Ag/ml), 4 to 63 AM (1 to 18 Ag/ml) and 17 to 28 AM (3 to 5 Ag/ml) are considered to be therapeutically effective plasma concentrations [29–34]. For propofol we could show that the solution adjuvant IntralipidR by itself dose-dependently and differentially affects the fMLP-induced fast and slow current. IntralipidR-induced impairment of currents was in part comparable and in part smaller than the effects of the commercially available propofol solution (Table 4). The effect of these intravenous anaesthetics was found to be long-lasting. This is in accord with neutrophil data, which indicated that nitrous oxide produces a sustained impairment of the oxidative response, of up to 4 h, and that the oxidative response remained impaired for more than 30/60 min after halothane/desflurane treatment. For sevoflurane and isoflurane, however, an immediate reversibility of the effect was shown [35,36]. In conclusion the fast (Cl) current induced by fMLP in oocytes might arise from a direct activation of the chloride channel by the a- or hg-subunits of the activated G-protein. We suggest that the slow current arises following PKC phosphorylation of the non-selective cation channels through activation of PLC. Inhibition by anaesthetics of both the fast and the slow fMLP-induced currents suggests that they might act primarily through an interaction with the agonist/receptor, but the fact that the slow current is consistently more sensitive to anaesthetics provides evidence that the anaesthetics might also affect the intracellular signalling pathways. Whilst it is generally accepted that there are differences between the transduction of receptor binding to physiological effect in the oocyte and the human neutrophil, Xenopus oocytes have been used extensively in experiments aimed at understanding the mechanisms underlying the effects of anaesthetics [20,37–39]. The results described in this paper indicate that anaesthetics inhibit the effect of anaesthetics on fMLP-induced currents in oocytes, as they do the fMLP-induced oxidative response in
human neutrophils. Although there are differences in detail to the anaesthetic-induced inhibition in the two systems, the results described here suggest that an important interaction is a negative allosteric modulation at the fMLP receptor. This not only has implications for the immunological status of patients and may in part explain the increased susceptibility to infection of patients maintained on very long term anaesthesia but also suggests that anaesthetics may affect homeostatic mechanisms, hormone balance and nociception in addition to their primary effect of causing a reversible loss of consciousness.
Acknowledgements We thank Dr. W. Bautsch (Hannover Medical School) for the gift of the cDNA coding for the human fMLP (R98 variant). Grant support: Forschungsfo¨rderprogramm ReForM of the University of Regensburg. Additional financial support: Baxter Germany, Erlangen, Germany, and Abbott Germany, Wiesbaden, Germany.
References [1] Becker EL, Forouhar FA, Grunnet ML, Boulay F, Tardif M, Bormann BJ, et al. Broad immunocytochemical localization of the formylpeptide receptor in human organs, tissues, and cells. Cell Tissue Res 1998;292:129 – 35. [2] Wittmann S, Fro¨hlich D, Daniels S. Characterization of the human fMLP receptor in neutrophils and in Xenopus oocytes. Br J Pharmacol 2002;135:1375 – 82. [3] Boulay F, Naik N, Giannini E, Tardif M, Brouchon L. Phagocyte chemoattractant receptors. Ann N Y Acad Sci 1997;832:69 – 84. [4] Gerard C, Gerard NP. The pro-inflammatory seven-transmembrane segment receptors of the leukocyte. Curr Opin Immunol 1994;6:140 – 5. [5] Burg M, Raffetseder U, Grove M, Klos A, Kohl J, Bautsch W. G[alpha]-16 complements the signal transduction cascade of chemotactic receptors for complement factor C5a (C5a-R) and N-formylated peptides (fMLF-R) in Xenopus laevis oocytes: G[alpha]-16 couples to chemotactic receptors in Xenopus oocytes. FEBS Lett 1995;377:426 – 8. [6] Blank J, Brattain K, Exton J. Activation of cytosolic phosphoinositide phospholipase C by G-protein beta gamma subunits. J Biol Chem 1992;267:23069 – 75. [7] Filtz TM, Paterson A, Harden TK. Purification and G protein subunit regulation of a phospholipase C-beta from Xenopus laevis oocytes. J Biol Chem 1996;271:31121 – 6.
S. Wittmann et al. / International Immunopharmacology 6 (2006) 61–70 [8] Stephens LR, Hughes KT, Irvine RF. Pathway of phosphatidylinositol(3,4,5)-trisphosphate synthesis in activated neutrophils. Nature 1991;351:33 – 9. [9] Thibault N, Burelout C, Harbour D, Borgeat P, Naccache PH, Bourgoin SG. Occupancy of adenosine A2a receptors promotes fMLP-induced cyclic AMP accumulation in human neutrophils: impact on phospholipase D activity and recruitment of small GTPases to membranes. J Leukoc Biol 2002;71:367 – 77. [10] Fensome A, Whatmore J, Morgan C, Jones D, Cockcroft S. ADP-ribosylation factor and rho proteins mediate fMLPdependent activation of phospholipase D in human neutrophils. J Biol Chem 1998;273:13157 – 64. [11] Choi S, Rho SH, Jung SY, Kim SC, Park CS, Nah SY. A novel activation of Ca2+-activated Cl channel in Xenopus oocytes by ginseng saponins: evidence for the involvement of phospholipase C and intracellular Ca2+ mobilization. Br J Pharmacol 2001;132:641 – 8. [12] Kamatchi GL, Tiwari SN, Chan CK, Chen D, Do SH, Durieux ME, et al. Distinct regulation of expressed calcium channels 2.3 in Xenopus oocytes by direct or indirect activation of protein kinase C. Brain Res 2003;968:227 – 37. [13] Johnson J, Capco DG. Progesterone acts through protein kinase C to remodel the cytoplasm as the amphibian oocyte becomes the fertilization-competent egg. Mech Dev 1997;67:215 – 26. [14] Wang YX, Dhulipala PD, Li L, Benovic JL, Kotlikoff MI. Coupling of M2 muscarinic receptors to membrane ion channels via phosphoinositide 3-kinase gamma and atypical protein kinase C. J Biol Chem 1999;274:13859 – 64. [15] Terasaki M, Runft LL, Hand AR. Changes in organization of the endoplasmic reticulum during Xenopus oocyte maturation and activation. Mol Biol Cell 2001;12:1103 – 16. [16] Choi S, Kim HJ, Ko YS, Jeong SW, Kim YI, Simonds WF, et al. Galpha q/11 coupled to mammalian phospholipase C beta 3-like enzyme mediates the ginsenoside effect on Ca2+activated Cl current in the Xenopus oocyte. J Biol Chem 2001; 276:48797 – 802. [17] Carnero A, Dolfi F, Lacal JC. Ras-p21 activates phospholipase D and A2, but not phospholipase C or PKC, in Xenopus laevis oocytes. J Cell Biochem 1994;54:478 – 86. [18] Frohlich D, Rothe G, Schwall B, Schmitz G, Hobbhahn J, Taeger K. Thiopentone and propofol, but not methohexitone nor midazolam, inhibit neutrophil oxidative responses to the bacterial peptide FMLP. Eur J Anaesthesiol 1996;13:582 – 8. [19] Frohlich D, Wittmann S, Rothe G, Schmitz G, Taeger K. Thiopental impairs neutrophil oxidative response by inhibition of intracellular signalling. Eur J Anaesthesiol 2002; 19:474 – 82. [20] Nietgen GW, Honemann CW, Durieux ME. Influence of anesthetics on endogenous and recombinantly expressed G proteincoupled receptors in the Xenopus oocyte. Toxicol Lett 1998;100–101:319 – 27. [21] Ghosh T, Eis P, Mullaney J, Ebert C, Gill D. Competitive, reversible, and potent antagonism of inositol 1,4,5-trisphosphate-activated calcium release by heparin. J Biol Chem 1988;263:11075 – 9.
69
[22] Chen LW, Jan CR. Mechanisms and modulation of formyl– methionyl–leucyl–phenylalanine (fMLP)-induced Ca2+ mobilization in human neutrophils. Int Immunopharmacol 2001;1:1341 – 9. [23] Downey GP, Fukushima T, Fialkow L, Waddell TK. Intracellular signaling in neutrophil priming and activation. Semin Cell Biol 1995;6:345 – 56. [24] Nakamura M, Nakashima S, Katagiri Y, Nozawa Y. Effect of wortmannin and 2-(4-morpholinyl)-8-phenyl-4H-1-benzopyran-4-one (LY294002) on N-formyl–methionyl–leucyl–phenylalanine-induced phospholipase D activation in differentiated HL60 cells: possible involvement of phosphatidylinositol 3kinase in phospholipase deactivation. Biochem Pharmacol 1997;53:1929 – 36. [25] Prossnitz ER, Ye RD. The N-formyl peptide receptor: a model for the study of chemoattractant receptor structure and function. Pharmacol Ther 1997;74:73 – 102. [26] Do SH, Fang HY, Ham BM, Zuo ZMD. The effects of lidocaine on the activity of glutamate transporter EAAT3: the role of protein kinase C and phosphatidylinositol 3-kinase. Anesth Analg 2002;95:1263 – 8. [27] Hehl S, Stoyanov B, Oehrl W, Schonherr R, Wetzker R, Heinemann SH. Phosphoinositide 3-kinase-gamma induces Xenopus oocyte maturation via lipid kinase activity. Biochem J 2001;360:691 – 8. [28] Wittmann S, Daniels S, Ittner KP, Frohlich D. Thiopentone and methohexitone enantiomers do not act stereoselectively on the oxidative response in human neutrophils in vitro. Pharmacology 2004;72:12 – 9. [29] Cordato DJ, Gross AS, Herkes GK, Mather LE. Pharmacokinetics of thiopentone enantiomers following intravenous injection or prolonged infusion of rac-thiopentone. Br J Clin Pharmacol 1997;43:355 – 62. [30] Crankshaw DP, Boyd MD, Bjorksten AR. Plasma drug efflux—a new approach to optimization of drug infusion for constant blood concentration of thiopental and methohexital. Anesthesiology 1987;67:32 – 41. [31] Davidson JA, Boom SJ, Pearsall FJ, Zhang P, Ramsay G. Comparison of the effects of four i.v. anaesthetic agents on polymorphonuclear leucocyte function. Br J Anaesth 1995;74:315 – 8. [32] Nguyen KT, Stephens DP, McLeish MJ, Crankshaw DP, Morgan DJ. Pharmacokinetics of thiopental and pentobarbital enantiomers after intravenous administration of racemic thiopental. Anesth Analg 1996;83:552 – 8. [33] Schickendantz J, Funk W, Ittner KP, Gruber M, Taeger K, Kees F. Elimination of methohexitone after long-term, highdose infusion in patients with critically elevated intracranial pressure. Crit Care Med 1999;27:1570 – 6. [34] Roberts FL, Dixon J, Lewis GT, Tackley RM, PrysRoberts C. Induction and maintenance of propofol anaesthesia. A manual infusion scheme. Anaesthesia 1988;43: 14 – 17. [35] Frohlich D, Rothe G, Schwall B, Schmid P, Schmitz G, Taeger K, et al. Effects of volatile anaesthetics on human neutrophil oxidative response to the bacterial peptide FMLP. Br J Anaesth 1997; 78:718 – 23.
70
S. Wittmann et al. / International Immunopharmacology 6 (2006) 61–70
[36] Frohlich D, Rothe G, Wittmann S, Schmitz G, Schmid P, Taeger K, et al. Nitrous oxide impairs the neutrophil oxidative response. Anesthesiology 1998;88:1281 – 90. [37] Chan CK, Durieux ME. Differential inhibition of lysophosphatidate signaling by volatile anesthetics. Anesthesiology 1997;86:660 – 9. [38] Honemann CW, Nietgen GW, Podranski T, Chan CK, Durieux ME. Influence of volatile anesthetics on thromboxane A2 signaling. Anesthesiology 1998;88:440 – 51.
[39] Hollmann MW, Ritter CH, Henle P, de Klaver M, Kamatchi GL, Durieux ME. Inhibition of m3 muscarinic acetylcholine receptors by local anaesthetics. Br J Pharmacol 2001; 133:207 – 16.