Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Cumulative evidence supports the existence of discrete neuroactive steroid binding sites on the -aminobutyric acid (GABA)_A receptor-Cl ionophore complex (Gee et al., 1995; Lambert et al., 1995). Endogenous metabolites of the steroid hormone progesterone appear to interact with a unique recognition site on the GABA_A complex that is distinct from the benzodiazepine and barbiturate binding sites (Gee et al., 1988; Peters et al., 1988; Turner et al., 1989), although specific binding by a radiolabeled steroid has not yet been established. Electrophysiological and ³⁶Cl uptake studies demonstrate that neuroactive steroids exert positive modulatory effects on GABA-evoked activity at nanomolar concentrations (Morrow et al., 1987; Purdy et al., 1990; Woodward et al., 1992). The kinetics of inhibitory postsynaptic currents in single-channel recordings indicate that such effects result from an increase in both the frequency and duration of single GABA_A receptor channel openings (Harrison et al., 1987; Twyman and MacDonald, 1992).

Consistent with their ability to facilitate GABAergic neurotransmission, neuroactive steroids exhibit potent anxiolytic-like effects in a variety of animal models (Gasior et al., 1999). By way of example, the endogenous progesterone metabolites allopregnanolone (3,5-P; 3-hydroxy-5-pregnan-20-one) and pregnanolone (3,5-P; 3-hydroxy-5-pregnan-20-one) produce anxiolytic-like effects in such diverse rodent ethological procedures as light/dark transition (Wieland et al., 1995), elevated plus-maze (Bitran et al., 1991; Rodgers and Johnson, 1998), defensive burying (Picazo and Fern ández-Guasti, 1995), mirrored chamber (Reddy and Kulkarni, 1997), and ultrasonic vocalization (Zimmerberg et al., 1994; Vivian et al., 1997). In addition, neuroactive steroids demonstrate robust anxiolytic-like activity in conflict procedures. For example, endogenous neuroactive steroids increase punished drinking (Crawley et al., 1986; Carboni et al., 1996; Vanover et al., 1999a) and operant punished responding (Wieland et al., 1995; Brot et al., 1997) in rats. Similarly, the synthetic neuroactive steroids Co 3-0593 (Wieland et al., 1997) and alphaxolone (Britton et al., 1991) were shown to increase punished responding in rat operant conflict procedures.

The present experiments were conducted to characterize the in vitro modulatory properties of a novel neuroactive steroid Co 2-6749 (GMA-839; WAY-141839; 3, 21-dihydroxy-3-trifluoromethyl-19-nor-5-pregnan-20-one; Fig. 1) on recombinant human GABA_A receptors as well as to define its anxiolytic-like effects in operant conflict paradigms in rats, pigeons, and squirrel monkeys. In addition, the potential side effect profiles of motor incoordination and interaction with ethanol were evaluated in rats. Furthermore, the behavioral effects of Co 2-6749 were compared with alprazolam in rats and with chlordiazepoxide in pigeons and squirrel monkeys.

View larger version (10K): [in this window] [in a new window]   Fig. 1.   Chemical structure of Co 2-6749.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

GABA_A Receptor Binding

Stable GABA_A Cell Line Preparation. Human 1, 2, 3, and 2L GABA_A receptor subunits were a gift from Peter Seeburg (Max-Planck Institute for Medical Research, Heidelberg, Germany). Human 4, 6, and 2 were cloned as previously described (Yang et al., 1995). Human 5 was cloned from human brain by polymerase chain reaction (PCR) using oligonucleotide primers corresponding to the proposed ends of the coding region based on the human 5 genomic sequence (Knoll et al., 1993). The amino acid sequence derived from this cDNA was identical with the amino acid sequence subsequently reported (Wingrove et al., 1991). Human 3 (Wafford et al., 1994) was cloned from human brain by PCR using oligonucleotide primers derived from the published sequences corresponding to the ends of the coding region. All plasmid DNA for transfection was prepared using two-cycle cesium-chloride-gradient centrifugation. The transfection of the HEK293 cells (CRL 1573; American Type Culture Collection, Manassas, VA) follows the protocol reported previously (Hawkinson et al., 1996).

Membrane Preparation. Membranes from stable HEK293 cell lines expressing human recombinant GABA_A receptor subunit combinations and well washed rat brain cortical homogenates were prepared as described previously (Hawkinson et al., 1996).

[³⁵S]t-Butylbicyclophosphorothionate (TBPS) Assay. Steroid inhibition of 2 nM [³⁵S]TBPS (60-100 Ci/mmol; NEN, Boston, MA) binding was examined in 200 mM NaCl/50 mM sodium phosphate buffer (pH 7.4) as previously described (Hawkinson et al., 1996). The GABA concentrations were chosen to standardize conditions among cell lines and were either the approximate IC₅₀ for inhibition of TBPS binding (rat brain) or the concentration producing the peak TBPS binding from the biphasic GABA concentration-effect curve (recombinant receptors) as indicated in Table 1. The incubation and filtration were done as previously described (Hawkinson et al., 1996) or in 96-well plates (2.0 ml; Beckman, Fullerton, CA) followed by filtration through GF/B 96-well filter plates (Packard, Meriden, CT), and rinsed three times with ~1.5 ml of ice-cold assay buffer. In the latter case, Microscint scintillation cocktail (50 µl; Packard) was added to each well of the dried filter plates, and they were then sealed, shaken vigorously for 5 min, and counted for 5 min/well on a TopCount 6-detector scintillation counter (Packard).

Data Analysis. Nonlinear curve fitting was done using Prism (GraphPad, San Diego, CA). The concentration of test compound producing 50% inhibition (IC₅₀) of specific binding and the maximal extent of inhibition (I_max) were determined for the individual experiments and then the means ± S.E.M. were calculated.

Nontarget Receptor Binding Assays

The effects of Co 2-6749 on binding at cytosolic steroid receptors (PanLabs ProfilingScreen) and neurotransmitter receptors (NovaScreen) were also determined. The activity of Co 2-6749 was compared, for each assay, with that of a reference compound with known affinity. Each experiment was replicated three times.

Geller-Seifter

Animals and Apparatus. Male rats (Sprague-Dawley; Charles River Laboratories, Wilmington, MA; n = 6/group) weighing 250 to 300 g were housed individually in a room with a 12:12-h light/dark cycle. Rats were kept on a restricted diet of Purina Lab Chow (St. Louis, MO) to maintain stable body weights at 85% of their free-feeding young adult levels. Water was freely available in the home cage.

Procedure. During the daily 63-min sessions rats were allowed to press a lever to receive access to a sweetened milk solution. Initially, rats were allowed to respond on a continuous reinforcement schedule and progressed rapidly to 30-s, 1-min, and 2-min variable interval (VI) schedules of reinforcement. On the continuous reinforcement schedule, rats received access to a sweetened milk solution after every lever press. On the VI schedules, milk was available at variable intervals, eventually at an average of once every 2 min. Once responding was stable, four brief periods of punished responding (3 min) were introduced. The punished periods alternated with five periods of unpunished responding (the first unpunished component was 3 min, the others lasted 12 min). The unpunished component consisted of the milk reinforcer being available at infrequent and variable intervals (VI 2 min). The punished component consisted of each response resulting in a brief electric footshock (250-ms duration) in addition to the milk reinforcer and was signaled by a light and a tone. In untreated animals, the electric footshock was sufficient to decrease responding during the punished component. Shock intensity was 0.2 mA initially, and then increased daily in increments of 0.02 mA to gradually suppress lever pressing to five responses or fewer per punished component based on individual performance. An increase in punished responding compared with baseline behavior was interpreted as an anxiolytic-like effect. Unpunished responding was measured to evaluate nonspecific effects on behavior.

Data Analysis. The number of lever presses during punished and unpunished components was determined for each rat. Drug injections were administered on Tuesdays and Fridays if baseline levels of responding remained stable. Saline injections on Mondays and Thursdays served as control. For test sessions, differences between vehicle control and drug treatment were computed for punished (drug  vehicle) and unpunished (% control) lever presses for each rat. Thus, each rat served as its own control. The data were then averaged across rats and the standard error of the means were calculated. The minimum effective dose (MED) was determined to be the dose at which punished responses (drug  vehicle) were increased to a value of 20. The minimum suppressive dose (MSD) was determined to be the dose at which unpunished responses (% control) decreased to 75%. Additionally, statistical comparisons were made on the individual difference scores after each drug dose and its preceding vehicle baseline score, using one-way ANOVA and individual post hoc comparisons with Bonferroni correction for multiple comparisons. For the experiments of chronic (14-day) dosing, values were averaged across rats and the standard error of the means were calculated. The data are presented graphically.

Rotorod

Animals and Apparatus. Naive male rats (Sprague-Dawley; Harlan Sprague-Dawley, Inc., San Diego, CA; n = 8/group) weighing 200 to 225 g were housed (two per cage) in polycarbonate cages containing sterilized bedding material (Sani-Chips; P.J. Murray, Montville, NJ) in a room maintained at 23.0°C (± 2.5°C) and on a 12:12-h light/dark cycle. Food (Teklad LM 485; Harlan Teklad, Placentia, CA) and water were freely available in the home cage.

Procedure. Before administration of test substance, rats were trained to walk continuously on the drum for a period of 90 s. During testing, rats were given three opportunities to remain on the apparatus continuously for 1 min. Remaining on the apparatus was scored as a pass. Results were treated quantally.

Data Analysis. Each dose-response function (n = 8-24/dose) was based on separate experiments (n = 8/dose) conducted on different days and the results summed. A dose that caused behavioral toxicity (i.e., falling from the rotorod apparatus) in half the animals (toxic dose; TD₅₀) was calculated based on each dose response function using PHARM/PCS version 4.2 software (Springer Verlag, New York). In addition, the 95% confidence intervals were calculated around each TD₅₀. A shift in the dose-response function in the presence of ethanol, such that the 95% confidence intervals of the two TD₅₀ values did not overlap, was considered significant.

Pigeon Conflict

Animals and Apparatus. Seven white Carneaux male pigeons, approximately 1 year old, were obtained from the Palmetto Pigeon Plant (Sumter, SC). All pigeons were experimentally naïve and were maintained individually in cages provided with continuously available water and grit. Lighting in the temperature- and humidity-controlled vivarium was on a 12:12-h light/dark cycle. All pigeons were reduced to approximately 85% of their free-feeding body weights before key peck training and were maintained at this weight for the duration of the study.

Procedure. Pigeons were initially trained to key peck at the center white or red key. Initially, each response resulted in 3-s access to mixed grain. Gradually, the response requirement was raised to 30 (fixed ratio or FR30 schedule), such that every 30th response resulted in 3-s access to mixed grain. Once responding stabilized, shock was introduced during the red keylight stimulus. Shock intensity was varied in individual pigeons to establish a level that suppressed response rate to <10% of unpunished responding. In the presence of a white key, completion of an FR30 resulted in 3-s access to mixed grain. In the presence of the red key, completion of an FR30 resulted in 3-s access to mixed grain and a brief electric shock (1-3 mA, 200 ms). Experimental sessions consisted of 10 (five of each) alternating 3-min components of the unpunished (white key) and punished (red key) schedule conditions. The alternating components were separated by a 30-s timeout period during which all illumination in the chamber was extinguished.

Data Analysis. Rates of responding (responses/s) were calculated for unpunished and punished components. Mean control rates for both unpunished and punished responding were determined by averaging data from all noninjection and vehicle control sessions that preceded drug test sessions. The effects of each dose were calculated as a percentage of the mean control rates for individual subjects. Drug effects were considered significant in an individual animal when the rate of responding differed by more than 2 standard deviations from the mean control rate of responding. Group means and S.E.M. are presented graphically.

Squirrel Monkey Conflict

Animals and Apparatus. Three adult male squirrel monkeys (Saimiri sciureus) weighing 0.87 to 0.95 kg lived in individual home cages except during experimental sessions. Squirrel monkeys were maintained at 85% of their free-feeding body weights by regulating their daily portion of Purina Monkey Chow. Monkeys were supplemented with fresh fruits and vegetables and had unlimited access to water in their home cages.

Procedure. Monkeys were initially trained to lever press on the right lever in the presence of white or red stimulus lights. Initially, each response (FR1) resulted in a food pellet reinforcer. Gradually, the response requirement was increased to 30 (FR30) such that every 30th response resulted in a food pellet reinforcer. Once responding stabilized, shock was introduced on an FR50 schedule when the red stimulus lights were illuminated. Under this schedule (FR30, food; FR50, shock), completion of 30 responses resulted in a food pellet reinforcer and completion of 50 responses resulted in a mild electric shock. Shock intensity was varied in individual monkeys to establish a level that suppressed response rate to <10% of unpunished responding. Experimental sessions consisted of four response components each preceded by a 10-min timeout period. Each response component consisted of a 3-min period of unpunished responding (white stimulus lights) followed by a short timeout (30 s) followed by a 3-min period of punished responding (red stimulus lights).

Data Analysis. Rates of responding (responses/s) were calculated for each of the unpunished and punished response periods. Mean control rates for both unpunished and punished responding were determined by averaging data from all noninjection and vehicle control sessions that preceded drug test sessions. The effects of each dose were calculated as a percentage of the mean control rates for individual subjects. For both compounds, only response rates in the first component were included in the analysis because maximal drug effects were observed for both drugs in this component. Drug effects were considered significant in any individual animal when the rate of responding differed by more than 2 standard deviations from the mean control rate of responding. Group means and S.E.M. are presented graphically.

Drugs

Co 2-6749 (CoCensys, Inc., Irvine, CA) was dissolved in a vehicle of 50:50 (w/v) hydroxypropyl--cyclodextrin (HPCD):0.9% saline (designated as 50% HPCD) and sonicated overnight for the experiments in rats. Co 2-6749 was diluted with saline to a final concentration of 10% HPCD. Co 2-6749 was dissolved in polyethylene glycol (PEG) 200 for the pigeon and squirrel monkey experiments. Alprazolam (Research Biochemicals International, Natick, MA; Sigma, St. Louis, MO; Wyeth-Ayerst, Princeton, NJ) was dissolved in a vehicle of 90% PEG 400:10% Tween 80 for the experiments in rats. Chlordiazepoxide (Wyeth-Ayerst) was dissolved in either 0.9% NaCl or sterile H₂O for the pigeon and squirrel monkey experiments. Co 2-6749 and/or alprazolam were administered orally in a volume of 5 ml/kg in rats, and 1 to 2 ml/kg in pigeons and monkeys. Chlordiazepoxide was administered i.m. in a volume of 1 ml/kg in pigeons and <0.4 total ml volume in monkeys.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

GABA_A Receptor Binding. Co 2-6749 had moderate potency for inhibition of [³⁵S]TBPS binding in rat brain cortical membranes with an IC₅₀ value of 230 ± 20 nM (Fig. 2; Table 1). Co 2-6749 fully inhibited the binding of [³⁵S]TBPS with an I_max value of 97 ± 2%. The slope of the inhibition curve was close to unity (1.16). Co 2-6749 was 4.5-fold less potent than 3,5-P in this assay.

View larger version (28K): [in this window] [in a new window]   Fig. 2.   Concentration-effect curves for inhibition of [35S]TBPS binding by Co 2-6749 in rat brain cortical membranes and HEK293 cell membranes expressing recombinant human GABAA receptors. Percentage of control specific [35S]TBPS binding is shown as a function of the log of the Co 2-6749 concentration. Each point represents the mean ± S.E.M. of at least three independent experiments.

Nontarget Receptor Binding. Co 2-6749 exhibited negligible inhibition (IC₅₀ >10 µM) in radioligand binding assays for a large number of nontarget receptors, including cytosolic steroid, inhibitory amino acid, excitatory amino acid, monoamine, and peptide receptors (Table 2).

Geller-Seifter. Vehicle control values of unpunished responding ranged from 379 to 5704 responses and punished responding ranged from 0 to 24 responses, but values were more stable within any given animal (generally ±20%). Note 24 responses after vehicle administration occurred one time in one rat; responding after vehicle administration was usually fewer than 20 responses, the criteria for a minimum effective dose.

View larger version (20K): [in this window] [in a new window]   Fig. 3.   Dose-effect curves for Co 2-6749 (top) and alprazolam (bottom) in the Geller-Seifter conflict procedure in rats. The effects on punished responses (; left axis) and on unpunished responses (; right axis) are shown as a function of dose. The horizontal dashed line represents baseline responses. Each point represents the mean ± S.E.M. of six rats.

View larger version (20K): [in this window] [in a new window]   Fig. 4.   Effects of Co 2-6749 during 14-day repeated administration in the Geller-Seifter procedure in rats. Effects of Co 2-6749 on punished responses (top) and unpunished responses (bottom) are shown as a function of day. The horizontal dashed line represents baseline responses. Each point represents the mean ± S.E.M. of six rats.

Rotorod. Co 2-6749 (10-40 mg/kg, p.o., 60 min) caused a dose-related increase in the number of rats failing the rotorod assay (i.e., falling off the apparatus; Fig. 5). Loss of righting reflex was observed at higher doses (data not shown). The peak effect, including the number exhibiting loss of righting reflex, appeared to be 60 min after administration. The duration of action was dose-related with higher doses lasting longer. The TD₅₀ of Co 2-6749 was 25.3 mg/kg (95% confidence interval: 19.5-32.8) administered alone at the time of peak effect and was 15.5 mg/kg (12.5-19.2) in combination with a subataxic dose (1.0 g/kg, i.p., 30 min) of ethanol (Fig. 5; Table 3).

View larger version (15K): [in this window] [in a new window]   Fig. 5.   Dose-effect curves for Co 2-6749 (circles) and alprazolam (inverted triangles) in the rotorod procedure in rats. The percentage of rats remaining on the rotorod apparatus is shown as a function of dose. The effects of each drug are shown alone (filled symbols) and in the presence of 1.0 g/kg ethanol i.p. (EtOH; open symbols).

Pigeon Conflict. Administration of saline or sterile H₂O engendered rates of responding of 2.84 ± 0.26 responses/s (mean ± S.E.M.) and 0.07 ± 0.05 responses/s in unpunished and punished components, respectively. Compared with control sessions, unpunished responding was 101% of control and punished responding was 80% of control. Administration of PEG 200 (2× volume, p.o.) engendered rates of responding of 3.20 ± 0.12 and 0.27 ± 0.12 responses/s in unpunished and punished components, respectively. Compared with control sessions, unpunished responding was 116% of control and punished responding was 611% of control. The apparent increase in punished responding was reflected by increases in punished responding in three of seven pigeons.

View larger version (11K): [in this window] [in a new window]   Fig. 6.   Dose-effect curves for Co 2-6749 (top) and chlordiazepoxide (bottom) in the pigeon conflict procedure. Effects on punished responses () and on unpunished responses () are shown as a function of dose. The horizontal dashed line represents baseline responses. Note the log scale on the ordinate axis due to the magnitude of effect. Each point represents the mean ± S.E.M. of seven pigeons.

Squirrel Monkey Conflict. Administration of saline engendered rates of responding of 2.30 ± 0.15 responses/s (mean ± S.E.M.) and 0.06 ± 0.06 responses/s in unpunished and punished components, respectively. Compared with control sessions, unpunished responding was 96% of control and punished responding was 74% of control. Administration of PEG 400/Tween (2× volume, p.o.) engendered rates of responding of 2.83 ± 0.32 and 0.12 ± 0.12 responses/s in unpunished and punished components, respectively. Compared with control sessions, unpunished responding was 118% of control and punished responding was 85% of control.

View larger version (12K): [in this window] [in a new window]   Fig. 7.   Dose-effect curves for Co 2-6749 (top) and chlordiazepoxide (bottom) in the squirrel monkey conflict procedure. Effects on punished responses () and on unpunished responses () are shown as a function of dose. The horizontal dashed line represents baseline responses. Note the log scale on the ordinate axis due to the magnitude of effect. Each point represents the mean ± S.E.M. of three squirrel monkeys.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The present study shows that Co 2-6749 is a selective steroid modulator of the GABA_A receptor complex in vitro. Co 2-6749 was generally less potent than allopregnanolone at inhibiting [³⁵S]TBPS binding to native and recombinant human receptor combinations. Whereas the effects of benzodiazepines at GABA_A receptor isoforms are clearly influenced by subunit composition, neuroactive steroids do not require a strict subunit composition for activity (Lambert et al., 1995). In the present study, the -subtype (1, 2, 3, 5) had little or no influence on the interaction, although Co 2-6749 displayed low potency at 4/632L receptor complexes. The latter result may indicate that Co 2-6749 displays selectivity for 1, 2, 3, and 5-containing complexes or alternatively that it has low activity at 3 relative to 2-containing complexes. Unfortunately, direct comparisons could not be made because membranes from 422L-expressing cells did not bind [³⁵S]TBPS and cells expressing 6b22L were not viable.

Co 2-6749 produced dose-dependent increases in punished responding in rats, pigeons, and monkeys, consistent with anxiolytic-like effects in multiple species. Although anxiolytic-like pharmacology is similar across species, not all clinically useful anxiolytics are robust in all three species; buspirone is a noted exception as reviewed by Pollard and Howard (1990). That the current data are orderly and consistent across species argues for a strong predictive validity to humans.

In rats maintained under the Geller-Seifter procedure, Co 2-6749 exhibited a minimum effective dose of 1.6 mg/kg, p.o. Co 2-6749 was more potent than the prototypical anxiolytic drug alprazolam, which exhibited a minimum effective dose of 8.1 mg/kg, p.o. Co 2-6749 increased punished responding in pigeons and squirrel monkeys at doses that did not affect unpunished responding, an effect and potency similar to that observed with chlordiazepoxide. More importantly, Co 2-6749 increased punished responding in all species tested with an effect size comparable to that of the benzodiazepines. In the Geller-Seifter procedure, the maximum mean number of punished responses was 78.8 for Co 2-6749 and occurred at a dose of 8 mg/kg. Although it is possible that a higher dose of alprazolam might show a further increase of punished responses in rats, the maximum mean number of punished responses of the doses tested was 83.7 and occurred at the highest dose tested (64 mg/kg). Alprazolam engendered a robust increase in unpunished behavior in rats, as well, whereas Co 2-6749 had a lesser effect on unpunished behavior relative to alprazolam. Neither Co 2-6749 nor chlordiazepoxide exhibited an increase of unpunished responding in pigeons or squirrel monkeys.

Activity of Co 2-6749 after repeated administration also was evaluated. Co 2-6749 continued to produce a robust increase in punished responding after daily dosing for 14 consecutive days. Although further evaluation using different doses and dosing intervals would be useful, the present lack of tolerance contrasts with tolerance to anxiolytic-like and sedative effects after daily dosing of benzodiazepine receptor ligands of both short and long duration (Soderpalm et al., 1989). Interestingly, unpunished responding gradually increased with repeated administration of Co 2-6749. This drift upwards in unpunished responding with repeated administration has been previously observed in multiple species and with neuroactive steroids, benzodiazepines, and barbiturates (Margules and Stein, 1968; Cook and Sepinwall, 1975; Witkin and Barrett, 1981; Wieland et al., 1997). The phenomenon has been attributed to "disinhibition" of a generalized suppression of baseline unpunished responding by the punished component (Margules and Stein, 1968). The interaction between the components of a multiple schedule after the introduction of punishment into one component is known (behavioral contrast), although the direction of the change is generally the opposite of that postulated in the aforementioned case (Brethower and Reynolds, 1962; Terrace, 1968; Cook and Sepinwall, 1975). The variables determining a result of behavioral contrast compared with a result of a change of the behavior in the same direction remain unclear.

In rats, Co 2-6749 exhibited a large therapeutic index (TI) between its minimum effective dose and its behavioral suppressive dose (Geller-Seifter TI = 8.3). In addition, Co 2-6749 showed a large separation between its minimum effective dose and its ataxic dose (rotorod TI = 15.8). This compares favorably to alprazolam that only showed a 3-fold separation between its minimum effective dose in Geller-Seifer and its ataxic dose in the rotorod procedure. It is curious that alprazolam was apparently more potent in producing ataxia in the rotorod procedure (TD₅₀ = 26.8 mg/kg, p.o.) than it was in suppressing unpunished responding in the Geller-Seifter procedure (minimum suppressive dose >64 mg/kg, p.o.). This apparent discrepancy may be due to the fact that the rotorod is a brief test of less than 2-min duration, whereas the Geller-Seifter procedure samples behavior over a period of an hour. It is probable that the ataxic effects of alprazolam are short in duration and diminish during the course of the Geller-Seifter session. Alternatively, the rotorod procedure, which used naïve animals, may be more sensitive to the effects of alprazolam than the Geller-Seifter procedure, which used animals with a history of drug administration.

In addition to its robust anxiolytic-like activity and its minimal sedative/ataxic effects, Co 2-6749 shows another distinct advantage over existing benzodiazepines. Benzodiazepines demonstrate interactions with ethanol that result in increased sedation and motor incapacitation. Previous studies suggest that neuroactive steroids exhibit a lesser propensity for adverse interaction with ethanol than do benzodiazepines (Edgar et al., 1997; Vanover et al., 1999b). The present studies demonstrate that Co 2-6749 shows minimal interaction with ethanol insomuch as the ataxic effects of high doses of Co 2-6749 are shifted less than 2-fold to the left in the presence of a subataxic does of ethanol. In contrast, there was greater than a 2-fold leftward shift in the ataxic effects of alprazolam. Although the differences between the effects of ethanol on Co 2-6749 and alprazolam ataxia are small, it is important to note that even in the presence of ethanol, there is an approximate 10-fold separation between the anxiolytic and ataxic dose of Co 2-6749. This is in contrast to alprazolam, which has a separation of less than 2-fold.

With robust anxiolytic-like activity, a large separation between anxiolytic-like effects and sedation/ataxia, and apparent oral bioavailability, Co 2-6749 makes an ideal candidate for development as a novel anxiolytic drug. In addition, the lack of tolerance in the Geller-Seifter procedure after 14-day dosing demonstrates the potential of this compound as an effective treatment for chronic anxiety disorders. Furthermore, recent results suggest that neuroactive steroids show less abuse potential than do benzodiazepines (Rowlett et al., 1999). Decreased abuse liability together with the lack of interaction with ethanol would present a potential advantage over currently available benzodiazepine anxiolytics.