Droperidol Inhibits GABA^Aand Neuronal Nicotinic Receptor Activation

The expression vectors for receptor subtype cDNAs were as follows: pSP64 for the human α⁴and β²type nAChR, pMXT for the α⁷type nAChR, pGEMHE for the rat GABA α¹and γ², and pGEMVE for the rat GABA β¹. The restriction enzymes used to linearize the plasmids were XbaI  for the α⁷-type nAChR, AseI  for the α⁴nAChR, PvuII  for the β²nAChR, and nhe1  for all of the GABA^Asubunits. Using a standard protocol, the SP6 RNA polymerase was used to make cRNA from the nACh subunits, and the T7 RNA polymerase was used to make cRNA from the GABA^Asubunits.

Xenopus laevis  oocytes were extracted from anesthetized females and placed in ND-96 medium (96 mm NaCl, 2 mm KCl, 1 mm MgCl², 1.8 mm CaCl²H²O, 5 mm HEPES, 2.5 mm Na-pyruvate, 0.5 mm theophylline, and 10 mg/l gentamicin, adjusted to pH 7.5). The oocyte clusters were incubated in 0.2% collagenase (type IA, Sigma-Aldrich, St. Louis, MO) in ND-96 medium for defolliculation. Oocytes were agitated at 18.5°C for 4 h and afterward were rinsed with Barth medium (88 mm NaCl, 1 mm KCl, 2.4 mm NaHCO³, 15 mm HEPES, pH 7.6). The oocytes were left to recover for 24 h in L-15 oocyte medium (Specialty Media, Phillipsburg, NJ) before injection of cRNA. L-15 oocyte media was obtained from Specialty Media.

Approximately 10 ng of α⁷nAChR cRNA, 10 ng of a 1:1 ratio of α⁴to β²nAChR cRNA, or 10 ng of a 1:1:1 ratio of α¹to β¹to γ²GABA cRNA were injected into individual oocytes in volumes of approximately 100 nl using an automatic injector (Nanoject; Drummond Scientific, Broomall, PA). The oocytes were incubated at 17°C for 2–5 days in ND-96 medium before electrophysiologic recording.

Current recordings were made from whole oocytes at room temperature (19–23°C) using a Gene-Clamp 500 two-microelectrode voltage clamp amplifier with an active ground circuit (Axon Instruments, Inc., Foster City, CA). The recording electrodes were pulled from glass capillary tubing (Drummond) to obtain a resistance between 1 and 5 MΩ and then filled with 3 m KCl. The Ringer solution (115 mm NaCl, 2.5 mm KCl, 1.8 mm BaCl^2,10 mm HEPES, 1 μm atropine, pH 7.4) used for recordings contained atropine to prevent muscarinic receptor stimulation and barium in place of calcium to avoid current amplification by calcium-activated chloride currents. Oocytes were clamped at a holding potential of −60 mV unless otherwise indicated and held in a 125-μl cylindrical channel. Perfusion was applied at a flow rate of 4 ml/min.

γ-Aminobutyric acid, acetylcholine, and other chemicals used were obtained from Sigma-Aldrich (St. Louis, MO). Droperidol was obtained from Abbott Laboratories (North Chicago, IL). Droperidol was made as a stock solution and serially diluted to the appropriate concentration on the day of the experiment. A saturating concentration of 1 mm acetylcholine was used in all experiments with α⁷and α⁴β²nAChRs unless otherwise indicated. A concentration of GABA that was approximately EC²⁰(500 nm GABA) was used in all experiments for the GABA^Aα¹β¹γ²receptor, unless otherwise noted. The oocytes were preequilibrated with droperidol for 2 min before a 2-s coapplication with the agonist. Activation reached its peak during agonist application. To minimize the contribution of nAChR desensitization, 5 min passed between acetylcholine applications. Three minutes passed between GABA applications. Using these time intervals, steady state recordings could be obtained in control experiments. A baseline control response to the agonist was measured before and after each agonist–antagonist coapplication. Responses that did not return to within 80% of baseline values were rejected for analysis. Clampex 7 (Axon Instruments, Inc.) was used for data acquisition, and Microcal Origin 5.0 (Microcal, Northampton, MA) was used for graphics and statistical calculation.

Concentration–response curves for acetylcholine and GABA were fitted to a modified Hill equation:

where IC⁵⁰is the concentration of agonist that elicited 50% of the maximal response, y^maxis the maximal current elicited by the agonist, n is the Hill coefficient, and x is the concentration of agonist. Concentration–response relations for the inhibition were constructed by calculating the current recorded in the presence of antagonist as a percentage of that elicited by the agonist alone. Agonist dose–response curves were normalized to a saturating concentration of agonist, 1 mm acetylcholine or 500 μm GABA, in the absence of droperidol. The data points obtained at each antagonist concentration were averaged, and the calculated mean and SE were fit to a modified Hill equation:

where IC⁵⁰is the concentration of antagonist at which 50% of the response is inhibited, and x and n have the same meanings as above. In studies with multiple agonist applications, the peak current after the first agonist application was compared with that resulting from further applications of agonist with a Student t  test. A Woodhull analysis was performed where the mean current inhibition was plotted on a semilogarithmic scale versus  membrane potential and fit with a linear equation. The resulting equation was compared with the fit equation with a zero slope with an analysis of variance. 17,18 P < 0.05 was considered significant, and data were represented as mean ± SEM.

Droperidol inhibited the activation of the GABA^Aα¹β¹γ²receptor in a biphasic manner with concentration-dependent inhibition between 10 nm and 1 μm that was reversed at concentrations higher than 5 μm (fig. 1A). The maximal inhibition of the GABA response by droperidol was 24.7 ± 3.0% and occurred at concentrations between 20 nm and 1 μm droperidol. At high concentrations (100 μm), droperidol can activate the GABA^Aα¹β¹γ²receptor in the absence of GABA (fig. 1B, left ). There was no effect of droperidol 100 μm on uninjected oocytes. The inhibitory response to droperidol is shown in figure 1Cas a percentage of the maximal possible effect. The IC⁵⁰concentration for droperidol is 12.6 ± 0.5 nm.

γ-Aminobutyric acid concentration–response curves were constructed in the presence and absence of two different inhibitory concentrations of droperidol (100 nm and 1 μm) and a higher concentration at which inhibition was no longer observed (5 μm). The GABA dose–response curves in the presence of 100 nm and 1 μm droperidol shifted the curve to the right at low concentrations of GABA, but at high agonist concentrations droperidol had no effect (fig. 2A). Droperidol at 5 μm had no significant effect on activation by GABA at any concentration (fig. 2B). To determine whether the inhibition of the GABA^Aα¹β¹γ²receptor by droperidol was voltage-dependent, the percent inhibition by 100 nm droperidol, when 500 nm GABA activated the receptor, was determined at a range of holding potentials from −100 to −20 mV. Inhibition of the GABA^Aα¹β¹γ²receptor activation increased with membrane hyperpolarization (fig. 2C). To determine whether inhibition by droperidol was dependent on channel activation, we measured peak current responses to repeated applications of GABA, at 3-min intervals, during a continuous application of droperidol at concentrations from 100 nm to 1 μm. There was no increment in the degree of inhibition of the GABA^Aα¹β¹γ²receptor by droperidol with repeated agonist application (fig. 2D;P > 0.05, t  test).

When activated by 1 mm acetylcholine, droperidol inhibited the α⁷nAChR with an IC⁵⁰of 5.8 ± 0.5 μm (fig. 3A). Droperidol only slightly inhibited the activation of the α⁴β²nAChR at concentrations of 10 μm and greater. Droperidol at 1 μm did not inhibit the α⁴β²nAChR, and there was only 10–15% inhibition with 10 μm droperidol (fig. 3B).

Acetylcholine concentration–response curves were constructed in the presence and absence of droperidol at approximately its IC⁵⁰concentration (6 μm). The slope of the acetylcholine dose–response curve was shallower in the presence of droperidol, and the maximal current amplitude was reduced (fig. 4A). To determine whether the inhibition of the α⁷nAChR by droperidol was voltage-dependent, the percent inhibition by droperidol when the receptor was activated by 1 mm acetylcholine was determined at a range of holding potentials from −100 to −30 mV. Inhibition of the α⁷nAChR increased slightly with membrane hyperpolarization; however, the change was not statistically significant (fig. 4B;P > 0.05, analysis of variance). To determine whether inhibition by droperidol was dependent on channel activation, we measured peak current responses to repeated applications of acetylcholine, at 5-min intervals, during a continuous application of droperidol (fig. 4C). There was no increment in the degree of inhibition of the α⁷nAChR by droperidol with repeated agonist application (P > 0.05, t  test).

Droperidol's modulation of the GABA^Areceptor is distinct from other general anesthetics that have been studied. Droperidol has two actions on the GABA^Areceptor that appear to occur at different molecular sites with different affinities. At low droperidol concentrations, from 10 nm to 1 μm, droperidol inhibits the maximal GABA activation by approximately 25% (figs. 1A and C). Inhibition by droperidol is only seen when the receptor is activated by less than saturating concentrations of GABA (fig. 2A). Thus, there may be little effect of droperidol in the setting of classic synaptic activation by the transient application of a high concentration of agonist. However, there is evidence that GABA^Areceptors have important physiologic actions other than the mediation of synaptic transmission. There exists a persistent tonic background current as a result of activation of a pharmacologically and functionally distinct GABA^Areceptor that is preferentially potentiated by propofol. 19It is possible that droperidol might inhibit such a current.

Inhibition is moderately voltage-dependent and was larger at negative potentials (fig. 2C). According to Woodhull, 17blockade by positively charged particles can be considered as a Boltzmann distribution under a potential difference. Accordingly, the voltage dependence of blockade can be estimated with the following equation:

where I^maxis the current amplitude in the absence of droperidol, B is the concentration of droperidol, K^Bis the apparent dissociation constant at the reference potential of 0 mV, δ is the electrical distance from the outer mouth of the channel, z is the charge, and E, F, R, and T have their general meanings. When the mean current inhibition at various potentials is plotted as a semilog distribution (fig. 5A), the slope is −0.019. Using the Henderson Hasselbach equation to calculate the concentration of charged particles, droperidol has a charge of +0.635 at pH 7.4. According to equation 1, δ (z) is calculated to be 0.76 or 76% of the electrical distance. It is therefore likely that droperidol does not reduce GABA affinity by binding to the agonist site; rather, its inhibitory site may be within the membrane electric field. In contrast, figure the slope for the voltage dependence in figure 5Bis not significantly different from 0, suggesting little effect of the membrane electric field in the α⁷nAChR.

At high concentrations (100 μm), droperidol can activate the GABA^Areceptor. Because of limitations in solubility, we were unable to determine if droperidol is capable of full activation of the GABA^Areceptor. Droperidol is not acting as a partial agonist at the GABA^Areceptor because, if this were the case, at low GABA concentrations larger currents would be expected in the combined presence of droperidol and agonist.

It has recently become clear that, unlike agonist binding at the nAChR, GABA binding is not diffusion-dependent. 20Instead, agonist affinity could be predicted by binding rates that were much slower than would be predicted by diffusion. In the GABA^Areceptor, a conformational change in the binding site either precedes or accompanies binding. Droperidol could affect the free energy required for such a conformational change by acting anywhere within the channel. It might either reduce binding or increase the unbinding rates of GABA. Our experiments are not equipped to measure the kinetics. However, another dopamine antagonist, chlorpromanzine, inhibits GABA currents by causing both a reduction in binding and an increase in unbinding rates for GABA in the GABA^Areceptor. 21 

Volatile anesthetics, barbiturates, propofol, and neurosteroids potentiate the GABA response, whereas ketamine, nitrous oxide, and xenon have little effect. 11,16,22–26Droperidol is the only anesthetic drug that has been shown to inhibit the GABA response. Submaximal GABA inhibition may be responsible for a key side effect that limits the utility of droperidol in neuroleptanesthesia. At high antiemetic concentrations and concentrations that are present on emergence from neuroleptanesthesia, droperidol causes anxiety and dysphoria. Because GABA agonists are anxiolytic, this side effect might be caused by inhibition of inhibitory transmission by droperidol.

Fully efficacious GABA antagonists such as bicuculine and picrotoxin cause seizures. One might predict that antagonism of GABA^Areceptor activation by droperidol would cause it to be epileptogenic in clinical use, but it is not. Unlike many other anesthetics, however, droperidol does not reduce seizure activity. 27Neuroleptanesthesia can be used for anesthesia in humans and animals when epileptic activity is desirable, as in seizure mapping and electroconvulsive therapy. 28It is possible that either a 25% maximal inhibition of GABA activity is not sufficient to cause seizures in most situations or the combination with inhibition of excitatory nicotinic activity is protective.

The activation of the α⁷nAChR was inhibited by droperidol (fig. 3). The IC⁵⁰for this inhibition is 5.8 ± 0.53 μm. The inhibition is noncompetitive and minimally voltage-dependent. In consideration of whether the inhibitory concentration is clinically relevant, the shoulder of the concentration–response relation is perhaps more important than the concentration that results in a 50% maximal effect. If at clinically relevant concentrations there is no significant effect on a putative target, it is unlikely that the target is mediating the clinical drug action. In the case of the α⁷nAChR, clinically relevant droperidol concentrations are within the shoulder of the concentration–response relation for inhibition. Droperidol has prolonged central nervous system actions (hours) despite a relatively short terminal half-life (approximately 10 min), and it is thought that this is a result of preferential central nervous system uptake. 3Therefore, central nervous system concentrations may be higher than measured plasma concentrations. It is therefore possible that inhibition of the activation of the α⁷nAChR mediates the anesthetic actions of droperidol. In contrast, there was no significant effect of clinically relevant concentrations of droperidol on the α⁴β²nAChR. Droperidol is unique among anesthetics that have nicotinic effects in that it preferentially acts on α⁷nAChRs. Volatile anesthetics inhibit heteromeric nAChRs more potently than α⁷nAChRs, whereas thiopental and ketamine inhibit both with approximately equal potency. 7–11 

Droperidol is also known to inhibit currents from human neuronal potassium channels in a similar concentration range to its effects on α⁷nAChRs, whereas human neuronal sodium channels are modulated by droperidol at more than 10 times droperidol concentrations. 29,30However, inhibition of voltage-gated potassium currents by droperidol would be predicted to be excitatory in nature.

In conclusion, the α⁷nAChR can be considered as a putative target for mediating neuroleptanesthesia. Droperidol is unusual among anesthetics in that it causes submaximal inhibition of a GABA response. Inhibition of GABA activation may be responsible for anxiety and dysphoria, a common side effect of droperidol when used at anesthetic concentrations.