The facilities in which the animals were maintained are fully accredited by the American Association for the Accreditation of Laboratory Animal Care (AAALAC), and the studies described herein were conducted in accordance with the Guide for Care and Use of Laboratory Animals under protocols approved by a local animal care and use committee. All work was done in accordance with U.S. Public Service Policy on Humane Care and Use of Laboratory Animals, amended August 2002. Furthermore, all research protocols were approved by an internal Animal Care and Use Committee.

Male, NIH Swiss mice (Harlan Labs, Indianapolis, IN, USA) weighed between 28 and 32 g and were housed in groups of 10–12 in plastic cages (24 cm × 45 cm × 15 cm high) with sawdust bedding in a temperature-controlled vivarium. A 12-h light/dark cycle was maintained, and all experimental sessions were conducted during the light phase of the cycle. The sessions were conducted between 8:00 and 10:00 a.m. daily.

Rats were used in these experiments as they have been utilized for such studies in the past with great success in generating significant polydipsia (16). Male, Wistar rats were singly housed, and maintained at approximately 85% of their free-feeding weights with free access to water. Twelve adult male, Wistar rats (Harlan Industries, Indianapolis, IN, USA) were used. They were allowed to acclimate to our vivarium for at least 3 days prior to testing. They were 2 months old at the start of the experiments that lasted approximately 6–8 months and were without explicit experimental experience prior to this study. The rats were singly housed, and maintained at approximately 86–91% of their free-feeding weights with post session supplemental feeding of Lab Diet #5001 for rodents (PMI Nutrition International Inc., St. Louis, MO, USA). Body weights were corrected for the normal growth curve of Wistar rats such that actual body weight of each rat increased over the experimental time frame. Water was available continuously in the individual home cages. Rats were inspected daily 7 days/week by experimenters and veterinary staff. Under the conditions of this experiment, there were no health issues observed.

The experiments were conducted using operant behavior test chambers ENV-007 (Med Associates Inc., Georgia, VT, USA), 30.5 cm × 24.1 cm × 29.2 cm. The test chambers were contained within light and sound attenuating shells. A food trough was mounted 2 cm off the grid floor on the centerline. Two response levers were centered 8 cm off the centerline and 7 cm off the grid floor. Three lights were located above each response lever at 15 cm off the grid floor. All events were controlled and lever-press data were recorded by a Compaq computer running MED-PC Version IV (Med Associates Inc., Georgia, VT, USA). The water bottle located within each chamber was weighed before and after experimental sessions.

Rats responded under a fixed-interval 120 s schedule of food delivery in which the first lever press to occur after the elapse of 120 s resulted in the delivery of a 45 mg food pellet. Experimental sessions lasted 90 min. Test compounds or drug vehicle were administered on Tuesdays and Fridays, given the stability of performance (within 15% variation of the individual animal). Non-injection, control performances were on Thursdays. Groups of five to eight rats were randomly selected from the pool of animals that demonstrated stable performance and given one dose of vehicle or test compound. Dose-effect curves and drug interaction data were created by repeated testing in groups of five to eight rats. Test compounds were administered as described in figure legends. The effects of the compound are expressed as a percent of control for each individual animal. Results were analyzed by ANOVA followed by Dunnett’s multiple comparison (p < 0.05).

Although the number of animals/dose were relatively small (5–8/dose group), the effect size was large and the control behavior was tightly controlled and, therefore, sufficient for detection of statistically significant drug effects. Thus, in keeping with the guidelines of animal care and use followed here, we utilized the minimum number of animals required for signal detection. Although order effects from experiment to experiment cannot be ruled out in the present design, the behavior of all animals returned to baseline control values for at least 3 days prior to the animal being used in another drug test. In some cases, systematic replication (e.g., same dose of fluoxetine given in another experiment) was used to verify that drug effects were qualitatively consistent across replication.

Male, Wistar rats (different from those rats used in the schedule-induced polydipsia studies) were used. They were 320 ± 15 g at the initiation of this study. Water was removed from their home cage for 18 h after which water bottles were presented to each rat in an individual housing cage in a quiet room for 60 min. The weight of the water bottle before and after the 60 min period was measured to the nearest 0.1 g. Six to eight rats were studied per dose. It is noted that this deprivation period is considered stressful. After the experiment, the animals were returned to their living cages with continuous access to food and water. No health effects were observed in these animals by direct visual observation and handling.

Male, Wistar rats (different from those rats used in the schedule-induced polydipsia studies) were used. They were 331 ± 11 g at the initiation of this study. Compound 1 (5.6 mg/kg, s.c.) was evaluated for its ability to reduce drinking. (±)-isoproterenol (30 μg/kg, s.c.) was given to induce drinking (n = 5–20 rat/group). Ten minutes after vehicle or (±)-isoproterenol injection, rats were given either vehicle or Compound 1 and a 60 min session began in which the rats in individual housing cages were given access to water 20 min after the second injection. The weight of the water bottle before and after the 60 min period was measured to the nearest 0.1 g. After the experiment, the animals were returned to their living cages with continuous access to food and water. No health effects were observed in these animals by direct visual observation and handling.

Mice (8/dose group) were used in these studies as the behavior has been well characterized in this species and strain as described here. Studies with mice were conducted as described (17). Mice were acclimated to the experimental room for 60 min under normal overhead fluorescent lighting. Mice were subsequently dosed with vehicle or compound and after a specified pretreatment interval (generally 30 min), were placed in a 17 cm × 28 cm × 12 cm high plastic tub with ~ 5 mm sawdust shavings on the floor along with a preweighed multiply gauze pad (51 mm²; 3 g). Mice were left in the tub for 60 min after which the weight of the gauze pad that had not been torn off by the mouse was obtained. The weight of the gauze used for nest construction was determined by subtraction.

Studies with mice (12/dose group) were conducted as described (17) since this behavior is robust in this species and mouse strain. Separate groups of mice were used in these experiments that were conducted in a dimly lit testing room. The mice were not injected with any drug or vehicle more than once in these studies as the same mouse was evaluated post drug administration sequentially on the rotarod task and then in the marble-burying assay. After 60 min acclimation to the experimental room, mice were dosed with vehicle or compound and after a specified pretreatment interval (noted in the drug section immediately below), were placed on a rotarod (Ugo Basile 7650) operating at a speed of 6 revolutions/min, and observed for falling. Mice that fell off the rotarod on two occasions during 2 min were scored as failing. Mice were not pretrained on this task. After the rotarod task, they were placed in a 17 cm × 28 cm × 12 cm high plastic tub with 5 mm sawdust shavings on the floor that were covered with 20 blue marbles (1.5 cm diameter) placed in the center. Mice were left in the tub for 30 min. The number of marbles buried (2/3 covered with sawdust) was counted.

Locomotor activity of mice (12/dose group) was assessed in order to compare the drug effects thereon to the effects in this species in marble-burying and nestlet shredding. Locomotor activity was measured with a 20 station photobeam activity system (San Diego Instruments, San Diego, CA, USA) with seven photocells per station. Locomotor activity was recorded as the number of ambulations, where ambulation was defined as the breaking of adjacent photobeams. Separate groups of mice were tested. Mice were injected with either vehicle or compound and placed into a plastic tub measuring 41 cm × 20 cm × 15 cm for 60 min. Seven photobeam emittors and detectors were spaced 6 cm apart along the length of the cage. Successive breaks of photobeams (locomotion) were counted in 10 min intervals.

Effects of compounds on schedule-induced polydipsia, marble burying, nestlet shredding, and locomotor activity were analyzed by ANOVA followed by post hoc Dunnett’s tests, and the effects on rotarod performance were assessed by Fisher’s exact test. The smallest dose required to significantly produce an effect was defined as the minimal effective dose (MED). ED₅₀ values (±95% confidence limits) were calculated from log-linear regression analysis of the dose–response curves. In studies with two variables (e.g., day and treatment), two-way ANOVA was used. Statistical probabilities of less than 0.05 were considered to be significant.

Chlordiazepoxide HCl, pentobarbital Na, d-amphetamine SO₄, chlorpromazine HCl, clomipramine HCl, and isoproterenol HCl (Sigma Chemical Co., St. Louis, MO, USA), MDL100907 (Tocris Bioscience, Bristol, UK), and citalopram HBr (synthesized by Eli Lilly and Co., Indianapolis IN, USA). were dissolved in 0.9% NaCl. All compounds were prepared just prior to dosing and administered i.p. in a volume of 1 ml/kg body weight (mice) or 10 ml/kg body weight (rats). Fluoxetine was administered s.c. For marble burying, compounds were administered either 30 min, 15 min (pentobarbital), or 10 min (d-amphetamine) prior to behavioral testing. For nestlet shredding and locomotor acitivity, compounds were administered immediately prior to testing. Drug doses are expressed as the salt.

The 5-HT_2C receptor agonists, mCPP (PDSP Data Base; Table 2) and Ro 60-0175 [(19, 20); Table 2], like fluoxetine, significantly decreased both the number of responses emitted and the amount of water consumed (mCPP: responses—F_2,23 = 17.1, p < 0.0001; water—F_2,23 = 13.5, p < 0.001) (Ro-60-0175: responses—F_{4, 39} = 8.6, p < 0.0001; water—F_2,23 = 13.5, p < 0.001) (Figure 5). In contrast, CPD1 dose-dependently suppressed excessive water drinking (F_3,31 = 9.5, p < 0.001) without affecting response output (F_3,31 = 0.37, p = 0.78). Statistically significant reductions of excessive drinking were detected at doses of 3 mg/kg and greater of CPD 1 (Figure 5).

SB-242084 prevented the decreases in excessive water consumption produced by CPD 1 (5.6 mg/kg) (F_4,25 = 6.0, p < 0.001) and significantly enhanced the number of responses emitted (F_4,25 = 3.1, p < 0.05) (Figure 6).

The 5-HT_2A receptor antagonist, MDL100907 (21) was used to access the potential contributions of 5-HT_2A receptors on the actions of CPD 1. Excessive water consumption was significantly reduced (64% decrease) by CPD 1 (5.6 mg/kg, sc, 20 min prior). MDL100907 (0.1 mg/kg, sc, 30 min prior) did not significantly alter the decrease produced by CPD 1 (65%, p > 0.05).

Water deprivation for 18 h engendered increases in water consumption. Deprivation-induced drinking was significantly reduced by 50% by 5.6 mg/kg of CPD 1 (Figure 7).

Isoproterenol increased water consumption by 432% relative to vehicle-treated rats. Isoproterenol-induced drinking was reduced by 77% by 5.6 mg/kg of CPD 1 (F_3,48 = 24.9, p < 0.0001) (Figure 8). In contrast, CPD 1 did not significantly affect normal drinking (Figure 8).

Ro 60-0175 decreased marble-burying at a MED of 0.1 mg/kg and nestlet-shredding at a MED of less than 1 mg/kg. CPD 1 decreased marble-burying at a MED of 10 mg/kg and nestlet-shredding at a MED of 3 mg/kg (Table 4).

Ro 60-0175 decreased locomotor activity at a MED of greater than 10 mg/kg and rotarod performances at a MED of greater than 1 mg/kg. CPD 1 decreased locomotor activity and rotarod performances at a MED of greater than 10 mg/kg (Table 4).

Marble-burying was significantly decreased by CPD 1 (10 mg/kg) compared to vehicle-treated mice (F_1,110 = 59.4, p < 0.0001). Repeated dosing with CPD 1 did not produce tolerance to its suppressant effects. Over 5 days of consecutive dosing effects of CPD 1 (10 mg/kg) on marble-burying did not significantly tolerate over days (Figure 9). There was also a significant impact of dosing day (F_4,110 = 5.2, p < 0.001) but no significant day × treatment interaction (F_4,110 = 0.31, p = 0.87).

Schedule-induced polydipsia was likewise studied over a 5-day dosing period. Although there was a tendency for the effects of CPD 1 (10 mg/kg) to abate over days, these effects were not significant (Figure 10). CPD 1 (10 mg/kg) significantly decreased excessive water consumption (F_1,40 = 128.5, p < 0.0001) and to a lesser extent response output over 5 days (F_1,40 = 11.7, p < 0.05). The factor of day of dosing did not significantly impact effects of CPD1 (responses: F_4,40 = 0.11, p = 0.98; water: F_4,40 = 0.49, p = 0.74) nor was there any significant day × treatment interaction (responses: F_4,40 = 0.06, p = 0.99; water: F_4,40 = 1.37, p = 0.26). There was not a significant attenuation of the effect of CPD 1 over 5 days (F_4,20 = 1.46, p = 0.25).

Rotarod performances of mice were also studied. On day 1, vehicle produced falling in 3/12 mice and failures in 1/12; CPD 1 (10 mg/kg) did not significantly differ compared to vehicle (2/12 falls and 0/12 failures). On the fifth day of dosing, vehicle produced falls in 1/12 mice and failures in 0/12. CPD 1 after 5 days of dosing produced neither falls nor failures in performance in the 12 mice tested.