Halothane regulates G-protein-dependent phospholipase C activity in turkey erythrocyte membranes.

The ability of halothane to stimulate phospholipase C (PLC) was examined in turkey erythrocyte membranes prepared from [3H]inositol-labeled turkey erythrocytes by measuring [3H]inositol phosphate formation ([3H]InsP) in the presence and absence of G-protein activation. In the presence of guanosine 5'-3-O-(thio)triphosphate) (GTP gamma S), halothane (0.5-10 mM) caused a dose-dependent activation of PLC. The EC50 value for halothane-induced PLC activation was 2.8 +/- 0.3 mM. Halothane (0.1-30 mM) had no effect on PLC activity in the absence of G-protein activation and did not affect Ca(2+)-dependent PLC activity. The activation of PLC by GTP gamma S occurred after an initial lag period of 60 s which was followed by a linear increase in [3H]InsP. Halothane dose-dependently decreased the lag period for GTP gamma S-induced PLC activation (minimal value 15 s) and increased the rate of [3H]InsP formation at all time points following this lag. As a result, halothane shifted the EC50 value for GTP gamma S-induced PLC activation to the left (4-fold) and increased its maximal response. Halothane also caused a dose-dependent activation of PLC in the presence of AlF4-. Half-maximal stimulation of AlF4(-)-activated PLC occurred with an EC50 value of 2.9 +/- 0.4 mM halothane, which is similar to the halothane dose giving half-maximal stimulation of PLC in the presence of GTP gamma S. At low doses (0.1-0.3 mM) halothane inhibited both isoproterenol- and adenosine 5'-O-(2-thiodiphosphate) (ADP beta S)-induced [3H]InsP formation, whereas at higher concentrations it stimulated PLC independent of the presence of these agonists. At concentrations chosen to reflect their different membrane/buffer partition coefficients, both hexanol (5 mM) and benzyl alcohol (20 mM) fluidized turkey erythrocyte membranes to the same degree as halothane (5 mM). However, these agents had no effect on GTP gamma S- or AlF(4-)-induced PLC activity, indicating that halothane-induced PLC activation was not secondary to changes in bulk lipid fluidity properties. Halothane also stimulated [3H]inositol bisphosphate and [3H]inositol trisphosphate formation in intact erythrocytes. These data demonstrate that the anesthetic halothane can stimulate G-protein-dependent PLC activity and modify the responsiveness of this signaling system to activation by receptor-linked agonists.

Halothane is a volatile anesthetic and as such has a pharmacological potency that is a function of its octanol/water partition coefficient, suggesting a hydrophobic site of action within the cell membrane (1,2). Although the molecular mechanism(s) underlying anesthesia are still largely unknown, recent evidence favors specific sites of action at the level of protein-protein or lipid-protein interactions (1)(2)(3).
The transduction of extracellular signals into intracellular second messengers by G-protein'-regulated receptor-effector mechanisms relies on interactions between a number of intrinsic and membrane-associated proteins. As such, these processes are likely to be influenced by the membrane lipid environment and may be susceptible to modification by anesthetics and other agents that perturb membrane structure and protein-lipid interactions.
However, relatively little is known about the effects of halothane on signal transduction and second messenger signaling processes. Halothane has been shown to inhibit muscarinic receptor regulation of adenylate cyclase activity in rat heart membranes (17). Whole cell studies have shown that vasopressin-induced InsPl formation and Ca2+ signaling in vascular smooth muscle cells are inhibited by halothane (18), although halothane had no effect on thyrotropin-releasing hormone-induced InsP formation in GH3 pituitary cells (19).

Halothane Effects on
Phospholipase C 15551 membranes possess a G-protein-regulated, polyphosphoinositide-specific PLC that is functionally coupled to a PZypurinergic receptor (22,23). Recent studies from our group and others have demonstrated that turkey erythrocyte PLC can also be coupled independently to @-adrenergic receptors (24,25). Turkey erythrocyte PLC has been purified as a 150-kDa protein, which is distinct from PLC-/3 or -7, and reconstituted with erythrocyte membranes to confer both G-protein-and receptor-regulated enzyme activity (26,27). The Gprotein involved in the regulation of this 150-kDa PLC has been purified as a 43-kDa protein demonstrating strong reactivity with antiserum against a 12-amino acid sequence from the carboxyl terminus of G, and GI1 (28).
The data presented here demonstrate that halothane, at concentrations within the anesthetic range, alters the sensitivity of PLC to G-protein-dependent activation and modifies the responsiveness of this signaling system to activation by receptor-linked agonists. This effect seems to be dependent on the expression of an activated G-protein. This work provides the first demonstration of a volatile anesthetic regulating G-protein-dependent PLC activity.

EXPERIMENTAL PROCEDURES
Labeling of Turkey Erythrocytes with pH]Inositol-Turkey erythrocytes were prepared essentially according to the method of Harden et al. (20), as described previously (24, 29). Briefly, fresh turkey erythrocytes in Alsever's solution (72 mM NaCl, 0.8 M glucose, 27 mM trisodium citrate, 10% citric acid, pH 6.1) were centrifuged at 1,100 X g for 5 min and the resultant supernatant removed by aspiration before resuspending the packed cells in 4-6 volumes of icecold HEPES buffer (1.5 mM HEPES, 150 mM NaCI, pH 7.2). This centrifugation and washing procedure was repeated a further two times before washing the cells in inositol-free Dulbecco's modified Eagle's medium. Finally, 5 ml of packed cells were resuspended in medium containing 9 ml of inositol-free Dulbecco's modified Eagle's medium, 2 ml of chicken serum, 0.4 mg/ml gentamycin, and 200-400 pCi of [3H]inositol. The cell suspension was then incubated at 37 "C in a shaking water bath for 16-18 h under a gas phase of O,/COz (95:5%).
Preparation of ~Hllnositol-labeled Turkey Erythrocyte Ghosts-[3H]Inositol-labeled turkey erythrocytes were lysed for 1 h in ice-cold lysing buffer (5 mM sodium phosphate, 5 mM MgClZ, 1 mM EGTA, pH 7.4). The lysed cells were then centrifuged at 17,700 X g for 5 min at (4 "C) and the supernatant discarded. The remaining unlysed cells were removed by aspiration before resuspending the membranes in lysing buffer. The membranes were further centrifuged at 17,700 X g for 5 min and then three more times at 8,000 X g for 5 min. After each centrifugation step any unlysed cells were removed and the membranes resuspended in the lysing buffer. Microscopic examination of erythrocyte ghosts prepared in this way showed that they were composed of morphologically intact red cell ghosts, all of which were fully permeable to trypan blue and ethidium bromide.
Assay of Phospholipase C Activity-Assays were initiated by adding membrane aliquots (150 pl, 0.7-0.9 mg of protein) to tubes containing 450 p1 of assay buffer consisting of 1 mM MgS04, 115 mM KCl, 5 mM KH2P04, 1 mM EGTA, 1 mM CaEGTA, and 10 mM HEPES, pH 7.0, at 37 "C. Under these conditions the free Ca2+ concentration was 300 nM as measured with fura-2. All incubations were performed in the absence of ATP or an ATP-regenerating system to eliminate the agonist effects of adenine nucleotides acting at purinergic receptors and to prevent ATP from acting as a substrate for CAMP formation (24). Reactions were terminated by the addition of perchloric acid (4% final concentration) and samples stored on ice for 20 min. The perchloric acid precipitates were sedimented by centrifugation and the resulting supernatants neutralized by addition of 2 M Tris base. Total inositol phosphates were then measured by counting 600 pI of the neutralized supernatants in 10 ml of scintillation fluid. We have demonstrated previously that [3H]InsP2 and 13H]InsP3 are the only products of guanine nucleotide-and receptor-stimulated P L c in ATP-depleted turkey erythrocyte membranes (24, 29). For measurements of inositol phosphates in intact cells, turkey erythrocytes (150 pl, 6-9 mg of protein) were incubated in Dulbecco's modified Eagle's medium (450 pl) and the reactions terminated by the addition of perchloric acid. The perchloric acid supernatants from these samples were neutralized by addition of a freshly prepared 1:l (v/v) mixture of Freonltri-n-octylamine, and the inositol phosphates were then separated by anion exchange chromatography on columns of Dowex 1 anion exchange resin (200-400 mesh, formate form) (24, 29).
Fluorescence Anisotropy Measurements-Turkey erythrocyte membranes were labeled with DPH and fluorescence anisotropy (steady state) measurements performed as described previously (30, 31) using an SLM 48000 spectrofluorometer (SLM Instruments, Champagne IL) in the T-format mode. Excitation was at 360 nm, and emission was observed at 430 nm. Measurements of DPH fluorescence lifetimes were obtained with a Liconix 4210 NB heliumcadmium laser as the excitation source using the phase modulation technique, as described previously (32, 33).
Analysis of Data-The concentrations of agonist producing halfmaximal stimulation (ECso) were obtained by computer-assisted curve fitting (ALLFIT) (34). Fluorescence lifetime data analysis was performed using GLOBALS UNLIMITED software (Laboratory of Fluorescence Dynamics, Department of Physics, University of Illinois, Urbana) as described by (35) and the data fitted to minimal values of the reduced x' parameter. The experimental error used in these analyses was taken as the standard deviation of averaged values for phase and modulation at each frequency (-0.002 and 0.2" in the modulation and phase, respectively).
Materials-Freshly drawn turkey erythrocytes were obtained from Cocalico Biologicals Inc. my0- [2-~H]Inositol (15 Ci/mmol) was obtained from Du Pont-New England Nuclear. GTPyS was obtained from Boehringer Mannheim and halothane from Aldrich. Chicken serum and Dulbecco's modified Eagle's medium were from Life Technologies, Inc. DPH and fura-2 were obtained from Molecular Probes. All other drugs and chemicals were obtained from Sigma or Fisher.

Effect of Halothane on PLC-
The ability of halothane to stimulate PLC in turkey erythrocyte membranes was determined by examining its effects on [3H]InsP formation in the presence and absence of G-protein activation. Fig. 1 shows that in the presence of the nonhydrolyzable guanine nucleotide GTP+, halothane induced a dose-dependent activation of PLC. Significant increases in [3H]InsP formation were observed within the anesthetic range of between 0.5 and 1.0 mM halothane (2, 12) and maximal responses obtained with 10 mM halothane. The E C~O for halothane-induced [3H]InsP formation under these assay conditions was 2.8 -+ 0.3 mM. At maximal halothane concentrations, the increase in [3H]InsP formation was 110-120% above that elicited by GTP+ alone. This is similar in magnitude to the activation of PLC in Halothane Effects on Phospholipase C response to a maximal concentration of the &adrenergic receptor agonist isoproterenol and about half of the maximal response produced by the P2,.-purinergic agonist ADPps in erythrocyte membranes (see Fig. 7 and Ref. 24).
In the absence of GTPyS, halothane (0.1-30 mM) failed to stimulate [3H]InsP formation (not shown). The effect of halothane on PLC in the absence of G-protein activation was further investigated by examining whether halothane could enhance Ca2+-stimulated PLC activity in erythrocyte membranes. As can be seen from Fig. 2, increasing Ca2+ from 0.07-3 WLM resulted in a %fold increase in [3H]InsP formation. Higher concentrations of Ca2+ had inhibitory effects on PLC activity (not shown). These effects of Ca2+ on turkey erythrocyte PLC activity are very similar to those described previously by Harden et al. (21). At concentrations of halothane which induced substantial increases in GTPyS-dependent PLC activity, halothane did not cause any further increased [3H]InsP formation in the presence of Ca2+ (Table I). These data demonstrate that halothane does not enhance either basal or Ca'+-stimulated PLC activity in turkey erythrocyte membranes and suggest that G-protein activation may be a prerequisite for the expression of the effect of halothane.
Effects of Halothane on the Kinetics of G-protein-dependent PLC Activation-Previous studies in turkey erythrocyte membranes have demonstrated that the stimulatory effects of receptor agonists on PLC activity reside in their ability to modify the kinetics of G-protein-dependent PLC activation (22,24,29). In erythrocyte membranes the activation of both adenylate cyclase and PLC by guanine nucleotides is preceded by an initial lag period of 1-2-min duration, which can be decreased, but not abolished, in the presence of &adrenergic and P2,-purinergic receptor agonists (22, 24, 36, 37). It has been suggested that this rate-limiting step for G-proteindependent PLC activation reflects the time required to exchange GTP for GDP on the bound a-regulatory subunit of G, (22). Consistent with these findings, we observed an initial lag period of about 60 s in the presence of 30 PM GTPyS before a steady-state rate of [3H]InsP formation was achieved (Fig. 3). The addition of halothane decreased the lag period for GTPyS-induced [3H]InsP formation in a concentrationdependent manner and enhanced the rate of [3H]InsP formation at all time points following this lag period. In common with receptor agonists, halothane did not completely abolish the lag period but decreased it to a minimal value of about 15   s. The basal rate of [3H]InsP formation was unchanged throughout this time course and was unaffected by halothane treatment (data not shown).
Effect of Halothane on GTPyS Dose-Response Curves-To determine whether the effects of halothane on the kinetics of GTPyS-induced PLC activation translate into changes in the sensitivity to GTPyS for PLC activation, the effects of halothane on GTPyS dose-response curves were examined. As shown in Fig. 4, halothane stimulated PLC by increasing the sensitivity and maximal extent of activation by GTP-yS. In the absence of halothane, GTP-yS caused a dose-dependent increase in PLC activity with an ECm of 26 +-7 pM. In the presence of 5 mM halothane the maximal response to GTPyS increased by 77% and the EC, shifted by 4-fold to a value of 6 k 0.9 WM. These results are very similar to those obtained with ADPPS and isoproterenol on PLC activation curves in the presence of GTPyS (22, 24) and are consistent with the ability of halothane and receptor agonists to decrease the lag period and increase the rate of GTPyS-dependent PLC activation.

Effects of Halothane on AlK-induced PLC Actiuation-
A1F: is a nonspecific activator of G-proteins which has been suggested to interact with the GDP-bound form of the aregulatory subunit and mimic the y-phosphate of GTP. Consequently, A1F; stimulates PLC by a mechanism that is independent of guanine nucleotide exchange. We therefore examined the ability of halothane to stimulate PLC in the presence of A1F; as a means of further assessing the role of guanine nucleotide exchange in the action of halothane. In the presence of 30 mM NaF and 10 ph4 AICls (AIR) halothane induced a dose-dependent activation of PLC with an observed EC, value of 2.9 k 0.4 mM (Fig. 5 ) . At maximal concentrations of halothane, PLC activity was more than %fold higher than the activity observed with AlF; alone. Both the EGO value and the magnitude of halothane-induced PLC activation in the presence of AlF; are similar to the values obtained for halothane in the presence of GTPyS. Furthermore, halothane also caused a leftward shift in the dose-response curve for A1F;-induced PLC activation. In the presence of 10 pM A1C13, NaF caused a dose-dependent activation of PLC which was evident over the range 0.1-30 mM F- (Fig. 6). At higher concentrations PLC activity was inhibited (data not shown), probably because of nonspecific effects. Halothane significantly enhanced [3H]InsP formation at every concentration of Ftested and caused a substantial shift in the activation curve to give an EC60 value of 1.5 f 0.2 mM for NaF. This provides the first example in which an activator of the turkey erythrocyte G-protein-dependent PLC is able to enhance PLC activity in the presence of both GTPyS and AlF; and indicates that the ability of halothane to activate PLC is independent of the manner in which the G-protein is activated.

Effect of Halothane on Agonist-induced PLC Actiuation-
Previous studies from this laboratory have shown that the effects of two agonists (ADPBS and isoproterenol) are nonadditive at maximal doses (24). Presumably these agonists share a common mechanism of PLC activation. Therefore, it  Reactions were allowed to proceed for 5 min, and then total inositol phosphates were extracted and measured as described under "Experimental Procedures." In these experiments All% caused a stimulation that was 651 f 152% of the basal value (1,150   was of interest to examine the additivity of halothane action with ADPPS and isoproterenol in the presence of GTPyS. Fig, 7 shows the effect of a range of halothane concentrations on the PLC response to maximal levels of ADPPS and isoproterenol. At low doses (0.1-0.3 mM), where it had negligible effects on PLC activity in the absence of agonist, halothane inhibited both isoproterenol-and ADPPS-induced [3H]InsP formation by 61 and 28%, respectively (Fig. 7). This somewhat unexpected result is in marked contrast to the lack of any inhibitory effects of halothane in the absence of agonist. At higher concentrations (0.5-30 mM), halothane caused an activation of PLC above the partially inhibited agonist response, which was of a similar magnitude and dose response to the activation caused by halothane in the absence of agonist (see Relationship between Membrane Lipid-disordering and Gprotein-dependent PLC Activity-Since halothane fluidizes the bulk lipid component of membranes, it is possible that PLC activation by halothane is a secondary consequence of a nonspecific effect on membrane properties. To investigate the relationship between lipid disordering and PLC activation, the effects of halothane (5 mM) on fluidity and PLC activity were compared with those of hexanol (5 mM) and benzyl alcohol (20 mM), two other well known membrane fluidizing agents (2, 38). The membranebuffer partition coefficients of these agents differ, so that concentrations were chosen to achieve approximately equal membrane lipid disordering effects (2, 38). All three agents disordered (or fluidized) turkey erythrocyte membrane lipids as reflected by decreased fluorescence anisotropy values of DPH (Table 11). Unlike hexanol and benzyl alcohol, halothane is a fluorescence quenching agent, as shown by a reduction in the major lifetime component of DPH from 9.1 to 8.5 ns (from a biexponential analysis) for 5 mM halothane. This causes an apparent increase in the steady-state anisotropy value, since this parameter is sensitive to the lifetime of the fluorophore (39). As a result, the magnitude of the decrease in anisotropy caused by the halothaneinduced increase in fluidity reflects a slight underestimation of the true fluidity change. Based on time-resolved anisotropy measurements with lipid vesicles and using the equation relating steady-state anisotropy to its time-resolved components (39), it was calculated that the quenching effect of halothane caused the anisotropy change resulting from membrane fluidization to be underestimated by 0.003. Thus, correction of the Ar parameter for halothane in Table I1 in this way would yield a value of -0.012, essentially the same as that for hexanol. Despite the fact that all three agents fluidized turkey erythrocyte membranes, neither hexanol nor benzyl alcohol mimicked the effects of halothane to activate G-protein-dependent PLC activity. As shown in Table I, under conditions in which halothane stimulated PLC activity in the presence of GTPyS or AlF;, hexanol and benzyl alcohol had no significant effect on PLC activity. These agents also had no effect on PLC activity in the absence of G-protein activation (not shown) or on Ca2+-stimulated PLC activity (Table I). These data indicate that the ability of halothane to stimulate G-proteindependent PLC activity cannot be explained simply by its ability to increase membrane fluidity and argue in favor of a more specific locus of halothane action.
Effect of Halothane on PLC Activity in Intact Turkey Erythrocytes-The data of Table I11 provide important confirmation of the ability of halothane to stimulate PLC in intact turkey erythrocytes. In intact erythrocytes prelabeled with [3H]inositol, halothane (5 mM) stimulated the formation of both [3H]InsPz and [3H]InsP3. The response to halothane was comparable with that of isoproterenol and about 30% of the maximal response produced by ADPPS (Table 111). Thus, these data demonstrate that the ability of halothane to stimulate PLC is not solely a property of the membrane preparation. As described previously for receptor agonists in intact erythrocytes (24), no significant increases in [3H]InsP1 were observed. Although halothane did appear to enhance [3H] InsPr formation, these changes were difficult to resolve accurately because of the high levels of [3H]InsPs in turkey erythrocytes.

DISCUSSION
In the present study we have demonstrated that halothane regulates G-protein-dependent PLC activity in turkey erythrocyte membranes. Halothane has dual effects on PLC activity in that it stimulates GTPyS-and A1F;-induced [3H]InsP formation but can also inhibit receptor-mediated PLC activation with even greater potency. Both of these effects are observed within the anesthetic range of between 0.3 and 1 mM halothane, although the EC50 value for halothane action in our studies with erythrocyte ghosts was somewhat higher than its EDso for anesthesia (1,2). This does not preclude a possible role for PLC in halothane anesthesia, since anesthesia is a complex end point that is unlikely to bear a simple relationship to the magnitude of alterations in the activity of the molecular targets of halothane action. It should also be noted that the effects of halothane on PLC could contribute to side effects not associated with anesthesia.

TABLE III [3H]Inositol phosphate formation in intact turkey erythrocytes Intact [3H]inositol-labeled turkey erythrocytes were incubated in
The stimulatory effect of halothane on PLC appears to be dependent on prior G-protein activation, as shown by the fact that halothane has no effect on either basal or Ca2+-stimulated PLC activity. It has been shown that the sensitivity of endogenous and reconstituted turkey erythrocyte PLC to activation by Ca2+ is not modified by G-protein activation (21, 28). Similar results have also been obtained for G, activation of PLC-Pl in bovine liver (40), suggesting that Ca2+ activates PLC by increasing the intrinsic activity of the enzyme. The inability of halothane to enhance Ca2+-activated PLC demonstrates that the effect of halothane is dependent on the mechanism by which PLC activation is achieved. This argues against a direct effect of halothane on PLC itself.
Further evidence that halothane interacts with G-proteindependent PLC is provided by its effects on the kinetics of GTPyS-induced PLC activation. Thus, halothane dose-dependently decreases the lag period for activation of PLC and increases the steady-state rate of [3H]InsP formation. The fact that halothane changes the time course of guanine nucleotide activation of PLC is not consistent with a mechanism whereby halothane affects the catalytic activity of PLC. These results are similar to those obtained with 8-adrenergic and P2,-purinergic receptor agonists in turkey erythrocyte membranes. In addition, as with receptor agonists, halothane increases the sensitivity and maximal responsiveness of PLC to activation by GTPyS.
The effects of halothane on GTPyS-dependent PLC activity are compatible with a mechanism in which halothane stimulates PLC by enhancing the rate of guanine nucleotide exchange on the a-regulatory subunit of Gdllr as described previously for receptor agonists (22,24). In our previous studies with turkey erythrocyte membranes, an important indicator of the dependence on guanine nucleotide exchange as a site for agonist-induced stimulation of PLC was the finding that agonists do not enhance PLC activity in the presence of AlF;. However, in the present study we observed that halothane induced a dose-dependent activation of A1F;stimulated PLC with an EC50 value similar to that obtained for halothane stimulation of PLC in the presence of GTPyS. There are two possible explanations for the finding that halothane activates PLC similarly in the presence of either GTPyS or AlF;. It could act at two distinct sites, one similar to the site affected by agonists (which apparently enhance the rate of guanine nucleotide exchange), and a second site at a step beyond guanine nucleotide binding which would increase the efficacy of the A1F;-activated G-protein to stimulate PLC. Alternatively, halothane could interact at a single step in the PLC activation cascade which is common to both GTPySand A1F;-stimulated PLC. In the latter case, potential sites of halothane action would be to stimulate the dissociation of the heterotrimeric G-protein or to enhance the interaction between the activated G-protein and PLC. Some evidence to suggest that halothane does not act by exactly the same mechanism as agonist-activated receptors comes from our finding that PLC activation by halothane is entirely additive with the activation by maximal doses of either isoproterenol or ADPpS (after allowing for the inhibition occurring at low levels of halothane'). By contrast, these agonists do not stimulate PLC in an additive manner when they are added in combination (24).
Recent studies have described an alternative mechanism to G-protein a-subunit regulation of PLC activity, in which free G-protein py-subunits can directly activate the enzyme (41, 42) in a manner that appears to be selective for the PLC-pZ isoform (43,44). However, the observation that py-subunits are much less potent than Gadll for PLC activation in both turkey erythrocytes and bovine liver (41, 42) has led to the conclusion that receptor-mediated PLC activation is mediated by a-subunits in these preparations. The effects of halothane on the kinetics of PLC activation make it unlikely that it acts by enhancing the interaction between free Py and PLC in turkey erythrocytes. Modulation of By-subunit association with PLC would not be expected to cause a GTPyS-dependent decrease in the lag period. A role for py-subunits in PLC activation is also unlikely in view of the fact that adenosine, which stimulates the dissociation of G, into as and By-subunits in turkey erythrocyte membranes, has no effect on PLC activity in this preparation (25). It seems more likely that Py stimulation of PLC may provide a mechanism whereby pertussis toxin-sensitive G-proteins can stimulate PLC (41-44).
Further evidence for a specific site of halothane action comes from the finding that other agents with similar effects on bulk lipid fluidity properties did not mimic the effects of halothane on PLC activity. Thus, concentrations of hexanol and benzyl alcohol which fluidized turkey erythrocyte membranes to a degree similar to halothane had no effect on GTPyS-or A1F;-induced PLC activation. This suggests that the effects of halothane on G-protein-dependent PLC are not caused by effects on membrane fluidity (i.e. changes in bulk lipid order or accessibility of substrate). However, this does not exclude the possibility that halothane may interact more potently with a limited membrane domain or the lipid interfacial region of a specific signal transduction protein, with this site of halothane action being relatively insensitive to the other membrane-fluidizing agents. Regardless of the physicochemical nature of the site at which halothane modulates PLC activity, it is clear that the result is a specific modification in the signal transduction pathway either at the level of G-protein activation or the interaction of the G-protein with PLC. In view of the fact that halothane is most likely to act at a locus within the hydrophobic region of the membrane, it is tempting to suggest that it increases the probability or efficacy of G-protein PLC interaction. The catalytic function of PLC itself does not appear to be affected by halothane.
An additional important finding of the present study was that halothane, at concentrations that are well within the anesthetic range (0.1-0.3 mM), significantly inhibited both isoproterenol-and ADPPS-induced [3H]InsP formation. The action of low concentrations of halothane to inhibit agoniststimulated PLC activity most probably occurs at a distinct site from the stimulatory effect of halothane for a number of reasons. First, halothane appears to be more potent in inhib- The inhibition of receptor-stimulated PLC observed at low halothane concentrations may not be maximal since the direct stimulatory effects occurring at halothane concentrations of 0.5 mM and above would obscure any further depression of the receptor-mediated effect. If this were the case, then the PLC stimulation induced by halothane in the presence of agonists would actually represent a synergistic activation, greater than that observed with halothane in the absence of agonist.
iting the effects of agonists, since this occurs at concentrations at which halothane does not stimulate GTPyS-or A1F;activated PLC. Second, since halothane at these concentrations does not inhibit PLC activity in the presence of GTPyS or AlF; (in the absence of agonist), the inhibitory site is presumably specific to the receptor-mediated pathway of PLC activation. One possibility would be that halothane interferes with agonist binding. This seems relatively unlikely in view of the fact that the inhibition by halothane occurs at supramaximal agonist doses. Furthermore, both the P-adrenergic and Pz,-purinergic receptor responses were affected over a similar halothane concentration range. The most obvious site of halothane action would be to modify the interaction between the activated receptor and the G-protein. This could result in a reduced ability of the receptor to stimulate GDP dissociation. Alternatively, halothane might inhibit an additional component involved in the receptor stimulation of PLC activity, which is not shared by GTPyS or A I S . Evidence for such an additional mode of receptor action has come from time course studies of the kinetics of PLC activation, which have shown that maximal PLC activity in the presence of GTPyS never reaches that attained in the presence of GTPyS plus receptor agonists, even after stable linear rates have been achieved under both conditions (22, 24, 29). At present the identity of this additional site of receptor action is unknown.
In conclusion, the data presented here demonstrate that halothane can both inhibit and stimulate PLC activity in turkey erythrocyte membranes in a manner that seems to be dependent on the expression of an activated G-protein. Previous studies have described inhibitory and stimulatory effects of halothane on ion channel activity, synaptic transmission, and [Ca2+Ii homeostasis. However, the underlying mechanism(s) of halothane action in these systems has remained ill defined. Our results demonstrate that halothane stimulates G-protein-dependent PLC activity and activates InsPB formation, and by inference diacylglycerol production, in intact erythrocytes. Both InsP3-induced [Ca2+Ii release and diacylglycerol activation of PKC have been shown to be key regulators of cell function, including ion channel activity and [Ca*+], homeostasis (45, 46). Alterations in the formation of these second messengers could provide at least one mechanism whereby halothane could perturb membrane conductance. Thus, the ability of anesthetics to stimulate second messenger formation may provide a mechanism whereby these drugs can lead to a direct or indirect alteration of ion channel activity and cell excitability.