An allosteric coagonist model for propofol effects on α1β2γ2L γ-aminobutyric acid type A receptors

Abstract

Background: Propofol produces its major actions via γ-aminobutyric acid type A (GABA(A)) receptors. At low concentrations, propofol enhances agonist-stimulated GABA(A) receptor activity, and high propofol concentrations directly activate receptors. Etomidate produces similar effects, and there is convincing evidence that a single class of etomidate sites mediate both agonist modulation and direct GABA(A) receptor activation. It is unknown if the propofol binding site(s) on GABA(A) receptors that modulate agonist-induced activity also mediate direct activation.

Methods: GABA(A) α1β2γ2L receptors were heterologously expressed in Xenopus oocytes and activity was quantified using voltage clamp electrophysiology. We tested whether propofol and etomidate display the same linkage between agonist modulation and direct activation of GABA(A) receptors by identifying equiefficacious drug solutions for direct activation. We then determined whether these drug solutions produce equal modulation of GABA-induced receptor activity. We also measured propofol-dependent direct activation and modulation of low GABA responses. Allosteric coagonist models similar to that established for etomidate, but with variable numbers of propofol sites, were fitted to combined data.

Results: Solutions of 19 μM propofol and 10 μM etomidate were found to equally activate GABA(A) receptors. These two drug solutions also produced indistinguishable modulation of GABA-induced receptor activity. Combined electrophysiological data behaved in a manner consistent with allosteric coagonist models with more than one propofol site. The best fit was observed when the model assumed three equivalent propofol sites.

Conclusions: Our results support the hypothesis that propofol, like etomidate, acts at GABA(A) receptor sites mediating both GABA modulation and direct activation.

Figure 1. Direct activation of GABA_A receptors by 10 μM etomidate and 19 μM propofol is equal

Three current traces recorded from a single oocyte expressing α1β2γ2L γ-aminobutyric acid type A (GABA_A) receptors are displayed, illustrating how equi-efficacious solutions of etomidate and propofol were identified. Currents were sequentially activated by superfusion with 10 μM etomidate (ETO) or 19 μM propofol (PRO) as illustrated by the bars above each trace (solid = etomidate; open = propofol), and the oocyte was washed for 15 minutes between traces. Note that the current amplitudes change slightly over time, necessitating control etomidate applications both before and after the experimental (propofol) experiment. The maximum current elicited with 19 μM propofol (arrow) is approximately the average of the etomidate-elicited control currents before and after (dashed lines). Note also that current activation with propofol was significantly slower than with etomidate.

Figure 2. Comparison of GABA modulation by 10 μM etomidate versus 19 μM propofol

Panel A shows three current traces from a single oocyte, illustrating how anesthetic modulation of receptor responses was assessed over a range of γ-aminobutyric acid (GABA) concentrations. After measuring GABA response, oocytes were pre-exposed to anesthetics (etomidate =ETO; propofol = PRO) before co-application of anesthetic and GABA at the same concentration. Oocytes were washed for 15 min between etomidate and propofol experiments. Panel B summarizes mean ± s.d. data from all experiments of this type. Bar color signifies experimental condition: GABA alone (red); GABA + 10 μM ETO (blue); GABA + 19 μM PRO (green). The numbers of oocytes is indicated for each condition within the bar. Significant enhancement was observed at all GABA concentrations, but no significant difference was observed between results for 10 μM etomidate and 19 μM propofol. * indicates p < 0.001; ns indicates p > 0.05.

Figure 3. Effects of 10 μM etomidate versus 19 μM propofol on GABA concentration-responses

Data points are mean ± s.d. measurements of peak currents in at least 5 oocytes per condition, normalized to responses elicited with 10 mM γ-aminobutyric acid (GABA) in the absence of anesthetics. Lines drawn through data represent non-linear least squares fits to logistic functions (equation 1, methods). Fitted parameters are reported as best fit (95% confidence interval). Panel A displays control GABA responses (red squares) and responses in the presence of 10 μM etomidate (ETO; blue squares). Control: Maximum = 101% (98-102); EC₅₀ = 105 μM (95-116 μM); Hill slope = 1.23 (1.10 - 1.35). With etomidate: Maximum = 125% (121 - 129); EC₅₀ = 4.3 μM (3.8-4.9 μM); Hill slope = 1.25 (1.07 - 1.42). The EC₅₀ ratio (eto/cntl) = 0.041 (0.034-0.048). Panel B displays control GABA responses (red circles) and responses in the presence of 19 μM propofol (PRO; blue circles). These data were obtained with different cells from those used for etomidate. Control: Maximum = 102% (99 - 105); EC₅₀ = 97 μM (86-109 μM); Hill slope = 1.3 (1.13-1.48). With propofol: Maximum = 125% (122 - 128); EC₅₀ = 4.2 μM (3.6-4.8 μM); Hill slope = 1.3 (1.11 - 1.51). The EC₅₀ ratio (pro/cntl) = 0.043 (0.036-0.051).

Figure 4. Propofol direct activation and enhancement of low GABA responses

Panel A shows example traces of γ-aminobutyric acid type A (GABA_A) receptor mediated currents stimulated with propofol (PRO). The bars over the traces indicate the period of propofol exposure. Slow activation is observed at 30 μM propofol, and “surge” currents after exposure to 300 μM and higher propofol concentrations indicate a second inhibitory effect. Panel B displays summary data for both propofol direct activation (red triangles) and propofol enhancement of GABA EC2.5 (approx. 5 μM) responses (blue triangles). Data points are mean ± s.d. measurements of peak currents normalized to responses elicited with 10 mM GABA in the absence of anesthetics (n ≥ 4). Lines drawn through data represent non-linear least squares fits to logistic functions (equation 1, methods). Fitted parameters are reported as best fit (95% confidence interval). Propofol direct activation: Maximum = 53 (50-57); EC₅₀ = 106 μM (86-138 μM); Hill slope = 2.5 (1.8 - 3.5). Propofol enhancement of GABA EC2.5: Maximum = 118% (114 - 126); EC₅₀ = 13.6 μM (12.2-16.4 μM); Hill slope = 1.65 (1.4 - 2.10).

Figure 5. A Monod-Wyman-Changeux co-agonist model for propofol and GABA actions

Panel A depicts a Monod-Wyman-Changeux equilibrium co-agonist model, with two functional channel states, R (closed) and O (open). After constraining the number of γ-aminobutyric acid and propofol sites, the model has five free parameters (see equation 2, methods). The basal equilibrium between these states (R/O) in the absence of ligands is L₀. γ-Aminobutyric acid (G) binds to two equivalent sites. K_G is the dissociation constant for γ-aminobutyric acid at closed receptors and K_G* = cK_G is the dissociation constant for γ-aminobutyric acid at open receptors. Propofol (P) binds to three equivalent sites. K_P is the dissociation constant for propofol at closed receptors and K_P* = dK_P is the dissociation constant for propofol at open receptors. Panels B & C show the MWC model together with P_open estimates from Figure 3B and Figure 4B, respectively. Estimated P_open was calculated by correcting data normalized to maximal γ-aminobutyric acid (GABA) responses for the apparent efficacy of GABA. Equation 2 was fitted to average P_open values with L₀ constrained at 50,000. The lines overlying the data represent the best fit (± s.d): K_G = 209 ± 43 μM; c = 0.0017 ± 0.00018; K_P = 97 ± 56 μM; d = 0.019 ± 0.023; n = 2.8 ± 0.79.