A newly developed anesthetic based on a unique chemical core

Abstract

Intravenous anesthetic agents are associated with cardiovascular instability and poorly tolerated in patients with cardiovascular disease, trauma, or acute systemic illness. We hypothesized that a new class of intravenous (IV) anesthetic molecules that is highly selective for the slow type of γ-aminobutyric acid type A receptor (GABA_AR) could have potent anesthetic efficacy with limited cardiovascular effects. Through in silico screening using our GABA_AR model, we identified a class of lead compounds that are N-arylpyrrole derivatives. Electrophysiological analyses using both an in vitro expression system and intact rodent hippocampal brain slice recordings demonstrate a GABA_AR-mediated mechanism. In vivo experiments also demonstrate overt anesthetic activity in both tadpoles and rats with a potency slightly greater than that of propofol. Unlike the clinically approved GABAergic anesthetic etomidate, the chemical structure of our N-arylpyrrole derivative is devoid of the chemical moieties producing adrenal suppression. Our class of compounds also shows minimal to no suppression of blood pressure, in marked contrast to the hemodynamic effects of propofol. These compounds are derived from chemical structures not previously associated with anesthesia and demonstrate that selective targeting of GABA_AR-slow subtypes may eliminate the hemodynamic side effects associated with conventional IV anesthetics.

Keywords: anesthesia; drug discovery; gamma amino butyric acid receptor; mechanism.

Fig. 1.

In A, the arrow points to the nitrogen atom of etomidate involved in complexing with the heme iron in the enzyme, 11-β-hydroxylase, thereby inhibiting corticosteroid biosynthesis. A class of IV anesthetic was developed using a chemical core by converting the free nitrogen in the imidazole ring of 1n to a pyrrole ring shown in carbo-1n. The molecular model (B) shows an anesthetic binding site within the transmembrane component of the GABA_AR and the computationally docked structure of carbo-1n. C shows the docking scores of multiple compounds from the series involving 1n, propofol, and etomidate. These served as reference agents to approximate the GABA_AR docking scores of carbo-1n and its derivatives, notably AA and BB (see Fig. 2 for their structures). Their predicted positions noted by arrows predicted a potency making them worthy of subsequent experimental evaluation.

Fig. 2.

Structures of the 11 compounds derived from the in silico structural similarity search for compounds resembling carbo-1n since the latter did not exist.

Fig. 3.

The EC₅₀ concentration for LORR for each compound (A) in bullfrog tadpoles (B) narrowed down the selection of 11 molecules suggested during the high-throughput screening stage to 1 lead compound. The most potent compound appeared to be BB (EC₅₀ = 0.49 µM) (C) which was chosen for the rest of the experiments.

Fig. 4.

(A) Paired-pulse population spike responses were recorded from hippocampal rat brain slices to measure anesthetic effects on tonic currents. Each symbol represents the mean ± SD for n = 5. (B and C) Fast (first pulse) and slow (second pulse) inhibitory pathways in the CA1 area neural circuit. By 100 ms it is too late to have the involvement of fast synapses, but ideal for following the time course of slow synapses. Each symbol represents the mean ± SD for n = 5 in C. Increased stimulation engages inhibitory circuitry, resulting in the paired pulse inhibition (PPI) measure between the first and second population spikes (PSs). Effects on tonic inhibition would contribute equally to both responses. (D) Population spike depression time course produced by propofol, etomidate, and BB showed significant differences in GABA_AR involvement for first (in red) vs. second pulse responses (in blue). Minutes (min) are post-drug application. Each symbol represents the mean ± SD for n = 10 propofol; n = 5 etomidate; n = 7 BB. In D, the following statistical comparisons were made: first pulse propofol (30 min) vs. first pulse etomidate (40 min) P < 0.005; first pulse propofol (30 min) vs. first pulse BB (120 min) P < 0.001. For second pulse responses there were no significant differences between drugs for the depressions produced, since all produced 100% depression.

Fig. 5.

Compound BB dramatically prolongs GABA_AR-mediated IPSCs. (A) Configuration for electrically evoking and recording IPSCs. (A, Top) Photomicrograph (10×) of a transverse hippocampal slice from mouse. Monopolar stainless steel stimulating electrode is placed in the stratum radiatum, and recording pipette in the stratum pyramidale, of hippocampal area CA1. Red box indicates field of magnified image (60×) at Bottom, in which the recording pipette is attached to a pyramidal cell. (B) IPSC duration, measured by the monoexponential decay constant, τ, is significantly prolonged in the presence of BB, but not vehicle (VEH). VEH or 40 µM BB compound was perfused into the recording chamber from time 0 until time 10. Points 1 and 2 indicate times where representative evoked IPSC traces (D) were taken. At Right of graph, the % change of τ versus predrug baseline is shown for both BB and VEH conditions. Individual cell data and group mean ± SD are plotted for the epoch between 10 and 15 min after BB/VEH application. (C) In the same set of experiments shown in B, BB, but not VEH, modestly enhanced IPSC amplitude. Changes of IPSC amplitude versus baseline are again shown at Right of graph for individual cells and group mean ± SD. (D) Representative IPSC traces before and after application of VEH or BB (time points indicated in B). (E and F) BB did not change parameter recording quality, as measured by resistance of the recording electrode in series with the recorded cell (series resistance, E) and the cell’s global resistance to alterations in membrane potential (input resistance, F), indicating little or no effect on tonic GABA_AR currents were produced.

Fig. 6.

Whole-cell oocyte voltage-clamp recordings lead to the above dose–response curves. A illustrates the GABA concentration response curve for α5β3γ2 GABA_ARs that demonstrated an EC₅₀ of 6.6 µM and EC₂₀ of 3.2 µM. B shows the BB concentration-response curve in the presence of GABA EC₂₀-induced currents for alpha5 GABA_AR (α5β3γ2). Error bars refer to SEs from the mean.

Fig. 7.

(A) The dose–response curve for LORR in rats due to IV-administered BB (closed blue circles) showing anesthetic-induced LORR having an ED₅₀ of 4.3 mg/kg. (B) BB had no effect on systolic blood pressure up to 20 mg/kg, whereas IV administration of propofol (closed red squares) significantly reduced systolic blood pressure (n = 5 P < 0.01) at 20 mg/kg; both have minimal effects on (C) heart rate. (D) The diastolic blood pressure was also suppressed compared with propofol (n = 5, P < 0.05 at both 10 mg/kg and 20 mg/kg). Error bars refer to SEs from the mean. The blood pressure and heart rates were recorded at 3 min after initiation of the drug injection.

Fig. 8.

(A) The effect of BB (10 mg/kg IV) given immediately after administering ACTH_1–24 (0.5 mg/kg IV) to cannulated rats previously treated 2 h prior with dexamethasone (0.5 mg/kg, IV). No difference in baseline levels before ACTH and compound administration were detected [F(2, 17) = 1.13, P = 0.35]. ACTH_1–24 along with either vehicle or KSEB caused robust corticosterone synthesis [F(1,17) = 84.9, P = 0.0001]; however there was no difference in corticosterone levels between posttreatment vehicle and BB [F(1, 22) = 0.23, P = 0.64]. (B) The effect of the vehicle or BB (10 mg/kg IV) on corticosterone synthesis expressed as a change in corticosterone levels shows no difference between the vehicle and BB effects [F(2, 16) = 0.22 P = 0.80]. Data are presented as mean ± SEM. ****, statistically significant differences; ns, no statistical difference.