Synthesis and characterization of a novel gamma-aminobutyric acid type A (GABA_A) receptor ligand that combines outstanding metabolic stability, pharmacokinetics, and anxiolytic efficacy

Abstract

1,4-Benzodiazepines are used in the treatment of anxiety disorders but have limited long term use due to adverse effects. HZ-166 (2) has been shown to have anxiolytic-like effects with reduced sedative/ataxic liabilities. A 1,3-oxazole KRM-II-81 (9) was discovered from a series of six bioisosteres with significantly improved pharmacokinetic and pharmacodynamic properties as compared to 2. Oxazole 9 was further characterized and exhibited improved anxiolytic-like effects in a mouse marble burying assay and a rat Vogel conflict test.

Graphical Abstract

INTRODUCTION

The γ-aminobutyric acid type A receptor (GABA_AR), a heteropentameric chloride ion channel,^{1, 2} is the principle target for the major inhibitory neurotransmitter, γ-aminobutyric acid (GABA), within the central nervous system (CNS). Benzodiazepines (BZDs) are a general class of drugs known to bind to the GABA_AR and are primarily used to treat anxiolytic and obsessive-compulsive disorders.³ The pharmacological action exerted by a BZD is dependent on the discrete subunits of the receptor complex. Classical BZDs, such as diazepam, bind non-selectively to α1-3,5βγ2 GABA_AR at the well-documented BZD binding site located at the α⁺γ⁻ interface.^{4, 5}

A compelling clinical opportunity exists in the development of α2/α3 selective GABA_AR ligands.^{6, 7} Based on the literature, these ligands are expected to result in superior treatments for seizures,^{8, 9} antihyperalgesia^{10, 11} and anxiety^12–14 without causing sedation, amnesia and ataxia, or the propensity for addiction/dependence^{15, 16} or tolerance.¹⁷ The clinical value of classical BDZs is evidenced by the fact that some have been marketed for over 50 years without a suitable replacement. Continued research has seen a number of promising candidates^{8, 18–21} fail due to pharmacokinetic complications²² or a variety of adverse effects.^{19, 23, 24}

Over the past decade a series of α2/α3 subtype selective imidazobenzodiazepines, using a privileged scaffold, have been designed, synthesized and studied. Some of the first compounds, ligands 1 – 3 (XHe-II-053, HZ-166 and JY-XHe-053, respectively; Chart 1) have been characterized previously for their selectivity at GABA_AR containing α2/α3 subunits over α1.²⁵ Previous investigations indicated that ligand 2 can engender anxiolytic-like effects in primates without sedative-like liabilities of non-selective positive allosteric modulators (e.g. diazepam).²⁶ However, ligand 2, as well as ligands 1 and 3, suffered from poor exposure and high clearance in rodents due to hydrolysis of the ester to the carboxylic acid 4, which exhibited low penetration across the blood-brain-barrier. Therefore, the goal of the present study was to discover a lead-like molecule that would retain the anxiolytic-like properties of BZDs 1 – 3 with lower systemic clearance in rats and thus increase plasma and brain exposure. Our design strategies to prevent formation of the carboxylic acid metabolite included blocking a site of oxidation on the ester by replacing the –CH₂ with a –CD₂ (ligand 5, see Scheme S1) and replacing the ester altogether with an ester bioisostere.

Chart 1.

Imidazobenzodiazepines 1 – 5.

The newly synthesized ligands were assessed in vitro and in vivo to determine their metabolic stability, general anxiolytic-like effects, and possible adverse motor effects. As a result, a novel lead compound, KRM-II-81 (9), was identified with exceptional drug stability, pharmacokinetics, anxiolytic-like efficacy, and relatively benign motor-impairing characteristics.

CHEMISTRY

The ester bioisosteres 6 – 10 were prepared from compounds 1 – 3, as depicted in Scheme 1. The methyl and ethyl oxadiazoles 6 and 7 were synthesized by treating 2 with an amidoxime in the presence of sodium hydride. The alkyl group was limited to methyl and ethyl to resemble the ethyl ester functionality of 2. The 1,3-oxazoles 8 – 10 were synthesized via the aldehyde derived from 1 – 3 through a two-step reduction-oxidation protocol. The reduction of the ester in the presence of LiAlH₄ afforded the alcohol which was subsequently oxidized with activated manganese dioxide. The oxazoles 8 (KRM-II-82), 9 and 10 (KRM-II-18B) were synthesized individually from the corresponding aldehydes using toluenesufonylmethyl isocyanide (TosMIC) in the presence of potassium carbonate.²⁷

Scheme 1.

Synthesis of heterocyclic bioisosteres 6 – 10.

RESULTS and DISCUSSION

In Vitro Characterization

The GABA_AR ligands depicted in Chart 1, Schemes 1 and 2 were triaged through a combination of in vitro potency, microsomal stability, and in vivo pharmacodynamics (PD) and motor side-effect liabilities to identify optimized compounds for further testing in electrophysiological and in vivo assays (Tables 1 and 2, see Supporting Information for methods).

Table 1.

In vitro data of compounds 1 – 10.

	α3 FLIPRa	Liver microsomal stabilityb (% Remaining, 0.5h, 37 0176C, 4 μM)			Cytotoxicityc (18 h)
	EC50 (μM)	Human	Mouse	Rat	LD50 (μM)
1	0.146	35	6.3	36	n.d.
2	0.844	86	57	58	> 100
3	0.029	6.1	6.1	0.5	n.d.
4	0.47	96	93	99	n.d.
5	1.17	95	63	77	n.d.
6	5.15	81	85	93	> 100
7	3.02	91	92	97	> 100
8	0.321	74	73	72	66.8 ± 9
9	0.937	91	90	90	>100
10	0.011	78	78	68	61.5 ± 10

^aMembrane potential (fluorescence) changes were measured by FLIPR assay using a FLIPR Tetra instrument from Molecular Devices and stably transfected HEK293T cells expressing either α1β3γ2 (n = 2; all compounds EC₅₀ > 20 μM) or α3β3γ2 (n = 1) GABA_ARs; details are listed in methods section.

^bCompounds (n = 1) were incubated with liver microsomes for 30 minutes at 37 °C at a final concentration of 4 μM. The percentage of compound remaining after incubation was determined by LC-MS/MS.

^cCompounds (n = 4) were incubated with HEK293T cells for 18 hours followed by the quantification of viable cells using CellTiter-Glo (Promega); n.d., not determined

Table 2.

Summary of exposure data

Ligand	Total Plasma Concentrations (nM)a				Brain concentrations (T in min) (nM)b
	Time (min)
	15	30	60	120	Conc. of dosed ligand	Conc. of 4 (as a metabolite)
p.o. (10 mg/kg)
2	BQLc	BQLc	BQLc	BQLc	(120) BQLc	(120) 10d
5	BQLc	BQLc	BQLc	BQLc	(120) BQLc	(120) BQLc
4	248	293	161	37.5	(120) BQLc	(120) BQLc
7	n.d.	n.d.	1720	n.d.	(60) 3090	n.d.
9	n.d.	n.d.	3910	n.d.	(60) 3640	n.d.

Sprague-Dawley rats (n = 3 per time point) were dosed p.o. (10 mg/kg).

^aBlood samples were collected at 15, 30, 60 and 120 minutes, extracted, and quantified by LC-MS/MS.

^bBrains were harvested after 60 or 120 minutes followed by extraction and quantification by LC-MS/MS;

^cQuantifiable limit < 12.2 nM;

^dThe three samples averaged 10 nM; n.d. not determined.

Preferred potency for the α3β3γ2 subtype over the α1β3γ2 (all ligands EC₅₀ > 20 μM) GABA_AR in the FLIPR assay was displayed by all compounds; however, the oxadiazoles 6 (EC₅₀ = 5.15 μM) and 7 (EC₅₀ = 3.02 μM) were consistently less potent than their oxazole counterpart 9 (EC₅₀ = 0.94 μM) at α3β3γ2 GABA_AR.

Ligands containing a 2′-F in the pendant phenyl ring, 3 and 10 were especially potent at α3β3γ2 with EC₅₀ values of 29 nM and 11 nM, respectively. The deuterated ethyl ester 5 had a similar activity in comparison to non-deuterated 2, while the acid 4 displayed an EC₅₀ of 0.47 μM for the α3β3γ2 GABA_AR. The lack of interaction with the α1β3γ2 GABA_AR predicted low risk of motor impairment and sedative effects with these ligands.^{8, 9} In addition, these compounds are expected to be devoid of tolerance and dependence.^15–17 Based on physicochemical properties, we anticipated that most molecules would have adequate blood-brain-barrier penetration, with the exception of the hydrophilic carboxylic acid 4. As expected, appreciable quantities of carboxylic acid 4 were not detected in the brain either by dosing 4 itself or monitoring for 4 as the metabolite of esters 2 and 5 (Table 2).

To assess metabolic stability all of the compounds were incubated for 30 minutes with human, mouse, or rat liver microsomes (Table 1). Consistent with our hypothesis of blocking a possible oxidation site in the ester, replacing the –CH₂ of 2 with a –CD₂ in 5 resulted in increased metabolic stability across all species. In addition, replacement of the C(3)-ester in 1 – 3 with a heterocycle, 6 – 10, resulted in improved metabolic stability for all related 2′-X ligands in liver microsomes. Cytotoxicity was evaluated by exposing human embryonic kidney cells (HEK293T) to the compounds for 18 hours. Ligand 2 and bioisosteres 6, 7, and 9 were non-toxic at a dose of 100 μM (Table 1). However, oxazoles 8 and 10 displayed LD₅₀’s of 66.8 μM and 61.5 μM, respectively, while all other ligands had LD₅₀ values higher than 100 μM.

In Vivo Exposure Characterization

Selected compounds were dosed in vivo in rats to determine their plasma and brain exposure profiles (Table 2). Both esters 2 and 5 had poor plasma and exposure when dosed either i.v. (1 mg/kg, see Table S1) or p.o. (Table 2). Unfortunately, the increased metabolic stability of 5 as compared to 2 in rat liver microsomes did not translate into increased in vivo exposure. However, we observed appreciable plasma levels 1 hour after dosing (10 mg/kg p.o.) of oxadiazole 7 and excellent levels of oxazole 9 (1720 nM and 3910 nM, respectively). Significant concentrations in the brain were observed for both compounds 60 minutes after administration.

Pharmacodynamic Assessment

Marble burying is a useful model of anxiety and has been used to evaluate potential anxiolytic compounds.²⁸ Herein, we evaluated the anxiolytic-like effects of compounds 2, 6 – 10 in mice using a marble burying assay (Figure 1).^{28, 29}

Figure 1.

Assessment of anxiolytic-like activity of 2 and 6 – 10 in the marble burying assay. Male, NIH Swiss Webster mice (n = 10) were injected i.p. with either vehicle or a test compound (10 or 30 mg/kg) 30 minutes prior to testing. Data were analyzed using ANOVA (Dunnett’s test: ^* P < 0.05 vs. vehicle). ^a Mild sedation-like behavior observed at 30 mg/kg.

At 10 mg/kg, only oxadiazole 7 produced a significant reduction in marble burying as compared to vehicle. At 30 mg/kg all compounds significantly decreased marble-burying. Mice treated with 30 mg/kg of 7, however, showed weak signs of sedation. To evaluate the potential for anti-anxiety versus motor side effects of these compounds, a motorsensory study using mice on a rotating rod (rotarod) was carried out (Figure 2) with the same mice as studied for marble-burying.

Figure 2.

Assessment of ataxic effects of 2 and 6 – 10 in the rotarod assay. Mice, as treated in the marble burying assay, were placed on a rotarod set at 4 r.p.m. and testing time was 2 min. Mice not falling off during the test were given a “Success” designation, while mice that fell once were assigned a “Partial” designation. Mice falling twice during the 2 minute time period failed the test

At a concentration of 10 mg/kg little change in motorsensory behavior was observed for all the compounds tested with the exception of compounds 6 and 10, which showed very minor impairment. At 30 mg/kg, the effective anxiolytic dose for most compounds, we observed a very weak increase in ataxia for all compounds except 9, which importantly did not significantly alter rotarod performance compared to vehicle and displayed zero failures.

From an in vivo perspective, the mouse marble burying assay allowed us to screen bioisosteres 6 – 10 for PD as well as potential anxiolytic-like effects while the rotarod assay permitted detection of motor side-effect liabilities. By triaging the molecules through a combination of in vitro potency, microsomal stability, and in vivo PD and motor side-effect liabilities we selected oxazole 9 for further characterization in electrophysiology, pharmacokinetic and in vivo rodent anxiolytic activity.

Electrophysiology

Oxazole 9 and for comparison ester 2 were assessed for electrophysiological efficacy profiles in recombinant α1β3γ2, α2β3γ2, α3β3γ2 and α5β3γ2 GABA_A receptors expressed in Xenopus laevis oocytes as described previously.³⁰ To minimize cell-to-cell variability, oxazole 9 was compared directly with ethyl ester 2 in the same cells at 100 nM, 1 μM, and 10 μM (Table 3).

Table 3.

Efficacy of 9 and 2 at αxβ3γ2 GABA_AR subtypes as % of control current.

Change of GABA EC3-6 current (100%), n=3–4
	9			2
	100 nM	1 μM	10 μM	100 nM	1 μM	10 μM
α1	n.d.	155.2 ± 9.9	231.7 ± 13.75	n.d.	147.4 ± 11.43	178.4 ± 19.55
α2	100 ± 11.85	200.6 ± 15.67	258.3 ± 18.07	111 ± 6.48	207.4 ± 19.72	244 ± 18
α3	109.3 ± 6.33	187.5 ± 25.25	332 ± 48.14	119.1 ± 8.62	195.2 ± 32.82	305 ± 31
α5	n.d.	157.5 ± 11	246.4 ± 6.15	n.d.	134 ± 9.4	158.5 ± 2.6

Dose-dependent modulation of GABA (EC_3-6 concentration) by ligands 9 and 2 on Xenopus laevis oocytes expressing GABA_AR subtypes α_1-3,5β3γ2 was determined by two-electrode voltage clamp current recordings. Data represents mean ± SEM.

While oxazole 9 is slightly more efficacious than ethyl ester 2 in modulating α1β3γ2 receptors at 10 μM, it is of comparable efficacy at α2β3γ2 and α3β3γ2 receptors at all concentrations. Most importantly and indicated by the electrophysiological results above, oxazole 9 is less active at α1β3γ2 receptors than at α2β3γ2 and α3β3γ2 receptors at all concentrations rendering it an α1 sparing GABA_AR ligand. Interestingly, it is significantly more efficacious than 2 at α5β3γ2 receptors (but only at the 10 μM dose), which very recently has been mechanistically-connected to possible anxiolytic activity.³¹

The oxazole 9, along with 2 and 5, were also assessed with the high-throughput Ion Works Barracuda (IWB) electrophysiology platform and compared to zolpidem and diazepam in HEK293 cells expressing human α1β3γ2, α2β3γ2, and α3β3γ2 GABA_ARs (see Supporting Information, Figure S1). The results of these concentration-dependent studies indicate that oxazole 9 modulates α2 (EC₅₀ = 118 nM) and α3 (EC₅₀ = 205 nM) containing GABA_ARs to a greater extent than α1-containing receptors (EC₅₀ = 489 nM).

Characterization of Oxazole 9

Following a 1 mg/kg i.v. dose in rats, oxazole 9 exhibited a clearance of 21.7 ml/min/kg with a low Vdss of 1.4 L/kg and a good T_½ of 1.4 hours. Following a 10 mg/kg i.p. dose in rats, the AUC was 16500 nM*hrs with a C_max of 3090 nM occurring at 2.0 hours. Oxazole 9 had an i.p. T_½ of 3.1 hours and very good bioavailability of 69%.

Understanding the relationship between in vivo CNS effects and the unbound concentration of ligand in the brain can be important when assessing dose-response relationships. Thus in a second experiment, the total plasma and brain concentrations were determined 4 hrs following a 10 mg/kg i.p. dose (Table 4). The fraction unbound in plasma and brain was then measured as 25.9% and 17.7%, respectively using in vitro equilibrium dialysis. Subsequently, the unbound concentration in each compartment was calculated by multiplying free fraction by total concentration to provide a C_u,plasma = 552 nM and a C_u,brain = 363 nM at 4 hours, which leads to a K_p,uu = 0.66 since K_p,uu = C_u,brain/C_u,plasma. A full unbound concentration-time profile in plasma and brain of unbound 9 is illustrated in the supporting information (Figure S2).

Table 4.

Total and unbound concentrations at 4 hours of 9 following a 10 mg/kg i.p. dose (n = 3).

	Total Concentration (nM)	Unbound Concentration (Cu) (nM)a	Free Fractionb Fu
Plasma	2130	552	0.259
Brain	2050	363 (Kp,uu = 0.66)	0.177

^aCalculated from total concentration (nM) and fraction coefficient (F_u);

^bDetermined with rat plasma protein and rat brain in vitro using equilibrium dialysis.

Next, oxazole 9 was evaluated further in the “Vogel conflict test” to detect anxiolytic-like behavioral effects. In this assay, water deprived rats were allowed to drink water during an unpunished and punished period in which licking was suppressed by a mild shock during punished periods. Anxiolytics cause significantly increased drinking during the punished period.³² Included in this assay was ester 2 for comparison and chlordiazepoxide (CDAP) as positive control³² (Figure 4).

Figure 4.

Vogel Conflict assessment of anxiolytic-like activity with male, Sprague-Dawley rats (n = 6–8). A. Evaluation of 2 and the anxiolytic CDAP given i.p. at 30 mg/kg and 20 mg/kg respectively; B. Dose response effects of 9 in comparison to CDAP. All compounds were administered 30 min prior to testing. Results were analyzed using ANOVA (Dunnett’s test: ^* P < 0.05; ^** P < 0.01).

Ligand 2 was tested at 30 mg/kg, where it had been active in the marble burying assay (Figure 1). However, no discernable difference was observed from the vehicle in the Vogel assay. In contrast, CDAP induced a large and significant increase in drinking behavior during the punished period without significantly affecting drinking during the non-punished period. The 1,3-oxazole 9 produced no increase in response in comparison to the control when given a 3 mg/kg i.p. dose with a 30 min pretreatment. However, when dosed at 10 mg/kg, 9 increased the rate of response by 240% over the control (P < 0.05). Doses at 30 mg/kg and 60 mg/kg showed similar increases of about 220% and 200% over the control, respectively. In comparison, the marketed anxiolytic, CDAP, produced a 330% response over the control (P < 0.05) when dosed at 20 mg/kg. Neither 9 (up to 60 mg/kg) nor CDAP (20 mg/kg) affected the response rate in the unpunished session. Thus oxazole 9 possesses a significant anxiolytic signature at 10 mg/kg i.p. in the punished session without decreasing the response rate in the unpunished session. The lack of anxiolytic-like effects of 2 in rats using the Vogel conflict test is likely due to the instability of the ester function in rodents and further underscores the unique and improved drug-like properties of 9.

CONCLUSION

It has long been known that targeting specific GABA_AR α subunits can result in desired pharmacological effects; however, due to the similar topology of the BzR binding site of the various α subunits,^{5, 33} designing selective ligands devoid of adverse effects, such as sedation, ataxia, amnesia, tolerance, and dependence, has been difficult. Results presented here indicate that oxazole 9 has potential as an anxiolytic devoid of these adverse effects. Consequently, oxazole 9 represents a unique and promising lead compound with attributes that include a desirable α2/α3-GABA_AR selective profile, low molecular weight (MW = 351), superb lipophilicity (clogP = 2.3; calculated used the ChemAxon program), good pH=7 solubility (by light scattering), excellent total exposure in plasma and brain and favorable measured fraction unbound in plasma and brain (26% and 18% respectively), leading to in vivo activity in pre-clinical rodent models of anxiety. Preliminary studies have begun in models of epilepsy and neuropathic pain (data to be published) to further characterize 9.

EXPERIMENTAL SECTION

5-(8-Ethynyl-6-(pyridin-2-yl)-4H-benzo[f]-imidazo[1,5-a][1,4]diazepin-3-yl)oxazole (9)

Step 1

8-Ethynyl-6-(pyridin-2-yl)-4H-benzo[f]imidazo[1,5-a][1,4]diazepine-3-carbaldehyde. Ethyl ester 2 (3 g, 8.4 mmol) was placed in an oven dried two neck round bottom flask and dry THF (300 mL) was added. The reaction mixture was cooled to 0 °C and the mixture allowed to stir after which LiAlH₄ (320 mg, 8.42 mmol) was added to the reaction mixture at 0 °C. After 10 min the reaction mixture was stirred at rt with continued stirring for 45 min under an argon atmosphere. After 45 min at rt, analysis of the mixture by TLC (silica gel 1:9 MeOH/EtOAc) indicated the absence of starting ester 2. The reaction mixture was slowly quenched with a cold saturated aq solution of Na₂SO₄ (20 mL) at 0 °C and then the reaction mixture diluted with EtOAc (50 mL). After this, the mixture was filtered through a small pad of celite, followed by an EtOAc wash (30 mL) of the celite, and the filtrate mixture washed and extracted with EtOAc (3 × 30 mL). The combined organic layers were washed with water, brine, and the solvent removed under reduced pressure to furnish the mixture of alcohols (imine alcohol 40% and reduced imine alcohol 60%, via analysis by ¹H NMR spectroscopy) as a yellow solid. This mixture of alcohols was used directly in the next step. The mixture of 2′-pyridylalcohols was dissolved in dry DCM (300 mL) under an argon atmosphere, and activated MnO₂ (10.9 g, 126 mmol) and Na₂CO₃ (2.7 g, 25.3 mmol) were added to the reaction mixture at 0 °C. The mixture was allowed to stir at rt overnight. After complete conversion of alcohol to aldehyde in the reaction mixture as indicated by TLC (silica gel), the reaction mixture was diluted with DCM (50 mL) and filtered through a small pad of celite. This pad of celite was washed with DCM (50 mL). The solvent was removed under reduced pressure and the residue that resulted purified by flash column chromatography (silica gel, EtOAc/DCM 2:1 with 1% each Et₃N and CH₃OH) to afford the pure aldehyde as a white solid (1.03 g, 39% over two steps). ¹H NMR (300 MHz, CDCl₃) δ 10.05 (s, 1H), 8.56 (d, J = 5.0 Hz, 1H), 8.08 (d, J = 7.5 Hz, 1H), 7.97 (s, 1H), 7.78 (ddd, J = 1.5, 6.0 Hz, 1H), 7.77 (dd, J = 1.5, 7.0 Hz, 1H), 7.55–7.57 (m, 2H), 7.38 (ddd, J = 1.5, 5.0 Hz, 1H), 6.00 (br s, 1H), 4.17 (br s, 1H), 3.16(s, 1H): ¹³C NMR (75 MHz, CDCl₃) δ 186.9, 167.7, 156.2, 148.6, 137.7, 137.1, 136.7, 136.3, 135.4, 135.3, 135.0, 127.1, 124.9, 124.0, 122.8, 121.5, 81.5, 79.7, 44.4; HRMS (LCMS-IT-TOF) Calc. for C₁₉H₁₃N₄O (M + H)⁺ 313.1080, found 313.1084.

Step 2

5-(8-Ethynyl-6-(pyridin-2-yl)-4H-benzo[f]imidazo[1,5-a][1,4]diazepin-3-yl)oxazole (KRM-II-81, 9). The toluenesulfonylmethyl isocyanide (TosMIC, 750 mg, 3.84 mmol) was placed in a dry two neck round bottom flask and dry MeOH (100 mL) added under an argon atmosphere. At rt, solid K₂CO₃ (1.33 g, 9.6 mmol) and the aldehyde from the preceding step (1.0 g, 3.2 mmol) were added to the reaction mixture and the mixture heated to reflux for 3 to 4 h. After completion of the reaction on analysis by TLC (silica gel, 1:10 MeOH and EtOAc), which was indicated by the absence of aldehyde starting material and complete conversion to the 1,3-oxazole (lower R_f,), the reaction mixture was quenched with cold water. After the mixture was quenched, 1/3 of the solvent was removed under reduced pressure and the product extracted with ethyl acetate (3 × 30 mL). The combined organic layers were washed with water, brine and dried (Na₂SO₄). The solvent was then removed under reduced pressure and the residue purified by flash chromatography (silica gel) to give the pure oxazole 9 as white solid (821 mg, 73%). ¹H NMR (300 MHz, CDCl₃) δ 8.62 (d, J = 4.2 Hz, 1H), 8.12 (s, 1H), 8.06 (d, J = 7.8 Hz, 1H), 7.96 (s, 1H), 7.85 (ddd, J = 1.8, 6.0 Hz, 1H), 7.79 (dd, J = 1.8, 6.6 Hz, 1H), 7.62 (d, J = 8.4 Hz, 1H), 7.55 (d, J = 1.5 Hz, 1H), 7.53 (s, 1H), 7.41 (ddd, J = 1.5, 4.8 Hz, 1H), 5.78 (d, J = 12.9 Hz, 1H), 4.31 (d, J = 12.9 Hz, 1H), 3.71 (s, 1H). ¹³C NMR (75 MHz, CDCl₃) δ 167.9, 156.7, 149.9, 149.0, 146.6, 137.0, 136.4, 135.8, 135.5, 135.3, 129.8, 127.5, 127.0, 124.9, 124.0, 122.8, 122.7, 121.0, 81.8.7, 79.5, 45.3; HRMS (LCMS-IT-TOF) Calc. for C₂₁H₁₄N₅O (M + H)⁺ 352.1188, found 352.1193.

Figure 3.

Total plasma concentration-time profile of 9. Male Sprague-Dawley rats (n = 3 per time point) were dosed i.v. (1 mg/kg) or i.p. (10 mg/kg) and plasma concentrations measured at time points noted.

ABBREVIATIONS

BQL

below quantification limit

BZD(s)

benzodiazepine(s)

CDAP

chlordiazepoxide

CNS

central nervous system

EC₅₀

half maximal effective concentration

Et₃N

trimethylamine

FLIPR

Fluorescence Imaging Plate Reader

GABA

gamma-aminobutyric acid

GABA_AR

gamma-aminobutyric acid type A receptor

HRMS

high-resolution mass spectrometry

PD

pharmacodynamics

TosMIC

toluenesulfonylmethyl isocyanide