Volatile Anesthetics Selectively Inhibit the Calcium sup 2+ -transporting ATPase in Neuronal and Erythrocyte Plasma Membranes

Thymol-free halothane was obtained from Halcarbon Laboratories (River Edge, NJ), isoflurane was purchased from Anaquest (Liberty Corner, NJ), and sodium pentobarbital was purchased from Anpro Pharmaceutical (Arcadia, CA).

The methods used for preparation of erythrocyte membranes and determination of protein and Calcium²+ concentrations were described previously. [3,9]Briefly, the membranes were prepared in hypotonic solutions from outdated human red blood cells purchased from the local Red Cross. The erythrocytes were lysed and washed several times in the presence of ethylenediamenetetraacetic acid to remove endogenous calmodulin. Ethylenediamenetetraacetic acid was subsequently removed by several washes with Tris buffer. Synaptosomal membranes were prepared from cerebellar lobes purchased from Pel-freez Biologicals (Rogers, AR), which were isolated from Sprague-Dawley rats and frozen in liquid nitrogen within 2 min after the animals were killed. Synaptosomal membranes were prepared by Dr. Zylinska [4,5]by the method of Booth and Clark [10]using Ficoll gradient centrifugation, as described previously. [11,12]Plasma membranes were the dominant component, as determined previously by electron microscopy [11]and marker enzyme activity assays. No endoplasmic reticulum membranes were present as evidenced by lack of any inhibitory effect of thapsigargin at concentrations up to 1 micro Meter. Mitochondrial membranes' contamination was less than 10%, as determined in the presence of specific inhibitors of mitochondrial activities, oligomycin (0.5-2 micro gram/ml) and sodium azide (1 mM). We have established that PMCA of both red blood cells (RBC) and synaptosomal membrane (SM) preparations had similar sensitivity to volatile anesthetics in the presence and absence of calmodulin. Thus, we have decided to routinely use SM preparations that were not totally depleted of calmodulin to avoid a considerable loss of material owing to the extensive ethylenediamenetetraacetic acid washes. The activity assays were normally performed in the absence of exogenous calmodulin except for the assays of barbiturate effects on PMCA in RBC in which 6 nM calmodulin was added to provide stimulation of PMCA activity comparable to that of the enzyme in SM.

Adenosine triphosphatase activities were determined by measurements of inorganic phosphate production, as described previously for the Calcium²+ -ATPase activity of PMCA in RBC. [1]The activity was assayed in the reaction mixture containing 50 mM Tris-maleate, pH 7.4, 120 mM KCl, 8 mM MgCl², 1 mM EGTA, 17.5 micro Meter free Calcium²+, and 3 mM adenosine triphosphate. The activity was calculated as a difference between the activity determined in the above reaction mixture and the reaction mixture in which Calcium²+ was omitted. The PMCA activity was sensitive to orthovanadate.

The ATPase activities not related to PMCA were studied in the presence of the following ions in the reaction mixture just described: (1) Magnesium²+ -ATPase at 8 mM Magnesium²+ (no Calcium²+); (2) Calcium²+ -ATPase at 8 mM Calcium²+ (no Magnesium²+); (3) Sodium sup +, Potassium sup + -ATPase at 15 mM Potassium sup + and 130 mM Sodium sup + (no Calcium²+), in the presence and absence of ouabain. The Sodium sup +, Potassium sup + -ATPase activity was calculated as the activity inhibited by 1 mM ouabain. All other assay conditions were the same as for the PMCA. Activity of the Magnesium²+ -ATPase and Calcium sup 2+ -ATPase was resistant to orthovanadate, the selective inhibitor of ion-transporting ATPases. [13].

The assays were performed in sealed 0.65-ml polypropylene tubes in a total reaction volume of 330 micro liter. The anesthetics were added after the addition of all reagents and membrane preparations, immediately before starting the reaction with adenosine triphosphate. Halothane and isoflurane were delivered to the reaction tube in an airtight Hamilton syringe from saturated solutions in reaction mixture. The assay was performed at 37 degrees C. Aliquots were withdrawn at 15-30 min for colorimetric inorganic phosphate measurement. Steady-state velocities were obtained from plots of inorganic phosphate production that were linear with time.

In parallel to the activity assay, tubes were incubated under identical conditions and used for determination of the effective halothane concentrations during the activity assay. The anesthetic was extracted with heptane, measured by gas chromatography, and the concentrations were calculated as described earlier. [1].

Standard deviations (SD) and standard errors were used to compare groups of data. Data are expressed as the mean+/-SEM of independent experiments performed in triplicates.

Protein concentration was measured using the BioRad protein microassay (Hercules, CA). Total calcium was measured by atomic absorption, and free Calcium²+ concentrations were calculated based on the constants given by Schwartzenbach et al. [14].

(Figure 1) shows specific activities of several ATPases that are integral protein constituents of SM and RBC plasma membranes. Four activities have been determined in the synaptosomal membranes of which the Sodium sup +, Potassium sup + -ATPase was the highest (2), followed by PMCA (1), and Magnesium²+ -ATPase (3). In addition, there was a Calcium²+ -ATPase activity (4) that in contrast to PMCA did not require Magnesium²+ and was not involved in Calcium²+ transport. It was very similar to the Magnesium²+ -ATPase activity in accordance with the possibility that the two activities may represent the same enzyme. [15,16]. This Calcium²+ -ATPase was insensitive to orthovanadate in contrast to the PMCA, which was totally inhibited by 0.1 mM orthovanadate. At 0.1 mM orthovanadate, the Magnesium²+ -ATPase was inhibited by only 20%, indicating that only a small part of the activity pertains to a P-type Magnesium²+ -ATPase.

As compared to SM, in RBCs, all ATPase activities were significantly lower. For example, the activity of PMCA was 1.0+/- 0.25 as compared to 31+/-4.5 micro mol Pⁱ/mg protein/h in SM. Furthermore, in contrast to SM in erythrocytes, the PMCA activity was dominant whereas the Sodium sup +, Potassium sup + -ATPase was very low (0.50+/-0.18), in the same range as the Magnesium²+ -ATPase (0.38+/-0.06 micro mol Pi/mg/h). The orthovanadate-insensitive Calcium²+ -ATPase activity was not detectable in erythrocyte membranes.

As shown previously, [1]increasing concentrations of halothane reduced the PMCA activity in erythrocytes. Figure 2demonstrates that halothane also reduce the PMCA activity in synaptosomal membranes. This inhibition was comparable to that of erythrocyte enzyme: it was dose-dependent and occured at anesthetic concentrations of halothane (Table 1). Also, the effects of isoflurane (Figure 3and Table 1) on the PMCA activity were comparable in the two membranes.

The effects of halothane and isoflurane on PMCA activity were subsequently compared to their effects on the other ATPase activities present in plasma membranes. Figure 4shows the data for halothane in synaptosomal membranes. Twofold to threefold higher anesthetic concentrations were required to half-maximally inhibit the Sodium sup +, Potassium sup + -ATPase as compared to the PMCA. At the highest halothane concentrations studied, the Magnesium²+ -ATPase activity was inhibited by only [nearly equal] 30%. The inhibition pattern of the Calcium²+ -ATPase activity closely resembled that of the Magnesium sup 2+ -ATPase, in line with the possibility that the two activities represented the same enzyme.

(Figure 5) compares activity patterns of the three enzymes in erythrocyte membranes in the presence of isoflurane. Similar to the SM, also in erythrocytes, PMCA activity was distinctly more sensitive to the anesthetic than the activities of Sodium sup +, Potassium sup + -ATPase and Magnesium²+ -ATPase.

(Figure 6) shows effects of sodium pentobarbital on ATPase activities in SM. None of the enzymes were inhibited at anesthetic concentrations of the barbiturate, i.e., 50-100 micro meter [17,18](Figure 6and Table 1). At higher concentrations, dose-dependent effects were observed: the half-maximal inhibition of the PMCA activity occurred at 11 mM pentobarbital, comparable to the Sodium sup +, Potassium sup + -ATPase. The Magnesium²+ -ATPase activity was half maximally inhibited at I sub 50 = 16-17 mM barbiturate concentration.

Similar inhibition patterns also were demonstrated for the three ATPase activities in red cell membranes, with an I⁵⁰of 10-11 mM for PMCA (Table 1), 17 mM for the Sodium sup +, Potassium sup + - ATPase, and I⁵⁰> 20 mM for the Magnesium²+ -ATPase. [19].

The PMCA of neuronal cells shows the same sensitivity to halogenated volatile anesthetics as the PMCA of erythrocytes. Both are inhibited in a dose-dependent manner at clinical anesthetic concentrations, with the half-maximal effect observed at [nearly equal] 1 minimum alveolar concentration of either halothane or isoflurane. Furthermore, also the Calcium²+ -transport ATPase in endothelial cells of cerebral microvessels has a comparable sensitivity to these anesthetics. [5]The data suggest that independent of the cell type, its specific PMCA activity, and expected differences in its lipid composition and membrane asymmetry, the PMCA is similarly affected by volatile anesthetics in vitro. In line with these observations also is the similarity of the sensitivity to the anesthetics of the Calcium²+ -transporting ATPase purified from either RBC or SM, which is twofold higher than that of the enzyme in the membranes. [1,20]This is an interesting finding because the PMCA of the three plasma membranes may significantly differ in its isoform composition, as indicated by tissue distribution of four PMCA isoforms to which antibodies were raised. [21,22]Isoforms 1 and 4 are expressed in all analyzed human and rat tissues. Whereas RBC contain only PMCA 1 and 4, neuronal tissues have in addition isoforms 2 and 3, with PMCA 2 present specifically in high concentration in cerebellum. Thus, it appears that all PMCA isoforms have similar sensitivity to volatile anesthetics. This would be in agreement with an impairment introduced by anesthetic binding in PMCA domains shared by all isoforms.

Activities of other ATPases that have been studied in the same membrane preparations and under comparable conditions are significantly less sensitive to volatile anesthetics. The relatively low sensitivity of the Sodium sup +, Potassium sup + -ATPase, which together with the PMCA belongs to the ion-transporting P-type ATPases, is in agreement with previous reports. [23,7]The other two ATPase activities are inhibited by volatile anesthetics by only 30% at the maximal studied anesthetic concentrations that totally inhibit the PMCA. Of these two activities in synaptosomal membranes, only 20% of the Magnesium²+ -ATPase activity pertains to a P-type ATPase, as indicated by its response to orthovanadate, a specific inhibitor of this class of ATPases. The Calcium sup 2+ -ATPase activity (unrelated to PMCA) in SM, which is totally resistant to orthovanadate, represents most probably an ecto-ATPase, described by several laboratories, which also may be stimulated by Magnesium²+. [15,24](This would be in agreement with the finding that 80% of Magnesium²+ -ATPase activity in our membrane preparation is not inhibited by orthovanadate). Similar inhibition of the vanadate-insensitive ecto-ATPase activity ([nearly equal] 40%) was observed in a different preparation of rat brain synaptic plasma membranes at 1.5 vol% halothane. [6]In summary, the PMCA is significantly more sensitive to volatile anesthetics than other ATPases studied in the same membrane. In addition, it is more sensitive than an intracellular membrane Calcium²+ -transporting ATPase of SR of both skeletal and cardiac muscle. [25-28].

In contrast to volatile anesthetics, the inhibition caused by barbiturates cannot be regarded as either selective or relevant to their pharmacologic action. Sodium pentobarbital at its clinical concentrations does not inhibit the PMCA studied in either of the two plasma membranes (present study) or in a purified form. [19]Furthermore, the differences in the sensitivity of various ATPases to the barbiturate at supraclinical concentrations are small as compared to volatile anesthetics.

Comparison of the effects of the representatives of the two groups of general anesthetics on the PMCA of different cell types (current report) and on the purified enzyme [19,25]suggests different sites for their action on the PMCA molecule. We have previously [25]proposed that volatile anesthetics permeate enzyme molecules and interact with nonpolar sites (cavities) in the interior, as has been demonstrated earlier for hemoglobin and myoglobin by nuclear magnetic resonance spectroscopy and X-ray crystallography studies. [28,29]Such an interaction is expected to modify conformational substrates of the enzyme and impair its function as shown by anesthetic-induced conformational changes that could result in impaired Calcium²+ binding to the enzyme and loss of its activity. [25]In contrast, it is postulated that barbiturates inhibit the PMCA by binding to the exposed nonpolar patch on enzyme surface that is involved in enzyme activation. [19]This less selective binding would result in a similar overall effect (inhibition of enzyme activity) as with volatile anesthetics, however, at distinctly higher drug concentrations. The molecular events leading from the binding to inhibition of enzyme activity are expected to be different for barbiturates and volatile anesthetics, as suggested for example by twofold to threefold stronger effects of barbiturates on calmodulin-activated PMCA monomers than dimers in contrast to equal concentration dependence observed with volatile anesthetics. [1,19].

The comparative approach allowed us to establish that the inhibition of PMCA by volatile anesthetics in vitro is selective by the following criteria. (1) None of the other ATPases studied are significantly inhibited by volatile anesthetics at 1 or 2 minimum alveolar concentrations, as is the PMCA. (2) Barbiturates do not inhibit the PMCA at their clinical concentrations. (3) The PMCA in the two plasma membranes responds very similarly: it shows the same difference in sensitivity to volatile anesthetics versus to pentobarbital and is significantly more sensitive to the volatile anesthetics than the other enzymes studied.