Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

There is evidence that 5-hydroxytryptamine_1A (5-HT_1A) receptors are involved in anxiety, impulsivity, and depression (De Vry, 1995). Drugs that stimulate the 5-HT_1A receptor are effective anxiolytics and antidepressants (De Vry, 1995; Lucki, 1998). In addition, it has been reported that some people with depressive illness have a blunted hypothermic response to 5-HT_1A receptor stimulation as compared with normal controls (Anderson et al., 1990; Lesch et al., 1990; Cowen et al., 1994; but see Meltzer and Maes, 1995). Animal models that screen for antidepressants have also implicated the 5-HT_1A receptor system (e.g., Porsolt et al., 1978; Marek et al., 1989; Tatarczynska and Chojnacka-Wojcik, 1989; Schreiber and De Vry, 1993).

The notion that the pathogenesis and symptoms of depression involve both environmental and genetic factors, and their interaction, has received much attention (Gershon et al., 1998). Overstreet et al. (1996) have developed selectively bred rats that differ in the magnitude of hypothermia induced by the 5-HT_1A receptor agonist (±)-8-hydroxy-dipropylaminotetralin HBr (8-OH-DPAT). The rats that were more sensitive to the hypothermic effects of 8-OH-DPAT [high 8-OH-DPAT-sensitive (HDS) rats] had a 4°C decrease in body temperature, whereas the resistant rats [low 8-OH-DPAT-sensitive (LDS) rats] had a 1.5°C decrease in body temperature. The HDS and LDS rats also differ in some tasks related to anxiety and depression. HDS rats were found to be more immobile than LDS and randomly bred rats (Overstreet et al., 1996) in the Porsolt swim test, a screen for antidepressants (e.g., Porsolt et al., 1978). The LDS rats were observed to exhibit more social interaction than HDS rats and to have an anxiogenic-like response following intrahippocampal microinjections of 8-OH-DPAT (Gonzalez et al., 1998). However, HDS and LDS rats did not differ in open field activity (Overstreet et al., 1996) or in the propensity to remain in the open arms of an elevated plus maze (Overstreet et al., 1996; Gonzalez et al., 1998). It is not known whether cortical 5-HT_1A receptors mediate these behavioral effects, but HDS rats have a higher density of cortical, but not hippocampal, 5-HT_1A receptors than LDS rats (Knapp et al., 1998). Because HDS and LDS rats exhibit different behavioral profiles on tasks related to anxiety and/or depression, these selectively bred rats may be useful for studying the physiological, biochemical, and genetic aspects of depression (Overstreet et al., 1998).

Because HDS and LDS rats that were selectively bred based on hypothermic sensitivity to 8-OH-DPAT differ in some paradigms related to anxiety and depression (Overstreet et al., 1996; Gonzalez et al., 1998), we hypothesized that the two lines would show different behavioral profiles on the differential reinforcement of low rate (DRL) 72-s operant schedule. The DRL 72-s schedule is a behavioral screen for antidepressant drugs (e.g., McGuire and Seiden, 1980; but see Pollard and Howard, 1986; Marek et al., 1989; Balcells-Olivero et al., 1998). The present hypothesis was based on our previous work that showed outbred Harlan Sprague-Dawley rats compared with Holtzman Sprague-Dawley rats had an exaggerated hypothermic response to 8-OH-DPAT and had a high reinforcement rate on the DRL 72-s task (Balcells-Olivero et al., 1998). In the present study, drug-free DRL 72-s reinforcement rate of LDS rats was lower than that of HDS rats. During drug challenge studies in Holtzman and Harlan Sprague-Dawley rats, several serotonergic-acting drugs (i.e., 8-OH-DPAT, ketanserin, fluoxetine) increased the reinforcement rate of Holtzman but not Harlan Sprague-Dawley rats; desipramine, which is a blocker of the norepinephrine transporter, had similar behavioral effects on both stocks of rats. Because HDS and LDS rats also differed in their hypothermic responsiveness to 5-HT_1A receptor stimulation, we hypothesized that serotonergic-acting drugs would increase the reinforcement rate of LDS but not HDS rats, whereas desipramine would increase the reinforcement rate of both lines of rats. The findings of this study are in agreement with these predictions and support our previous findings with the two stocks of Sprague-Dawley rats (Balcells-Olivero et al., 1998). These results implicate the 5-HT_1A receptor system in behavior maintained by the DRL 72-s schedule.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals

A total of 54 selectively bred male rats obtained from the University of North Carolina's Center for Alcohol Studies were used in these experiments and grouped housed under standard laboratory conditions (constant temperature of 23 ± 1°C and relative humidity of 40-60%); 12-h light/dark cycle with lights on at 7:00 AM). The description of the initial rat stock (maintained by the National Institutes of Health) and the subsequent breeding protocol have been described in detail elsewhere (Overstreet et al., 1994, 1996). The HDS line was established by mating male and female rats from each of 10 litters that had a large hypothermic response to 8-OH-DPAT; conversely, the LDS (low 8-OH-DPAT-sensitive) line was established by mating the rats with the smallest hypothermic response to 8-OH-DPAT. Rats from the 12th generation were used in the temperature studies, and rats from the 14th generation were used for the operant studies. All rats were cared for in accordance with the University of North Carolina, University of Chicago, and National Institutes of Health guidelines.

Experimental Methods

DRL 72-s Schedule. Training. Training rats on the DRL 72-s schedule has been previously described and will only be briefly described here (Richards and Seiden, 1991; Balcells-Olivero et al., 1998). Rats were acclimated to a water-restricted diet in which water was available ad libitum for 20 min after each behavioral session (Monday through Friday) and from Friday afternoon to Sunday afternoon. Behavioral testing was conducted for 1 h/day, Monday through Friday. Rats were trained in chambers to lever press under a DRL 72-s schedule for 0.1 ml of water, and stable behavioral performance was obtained before pharmacological testing began (approximately 16 weeks of training).

DRL 72-s Measures and Analysis. The dependent variables measured were the number of lever presses and reinforcers obtained. The duration of lever presses was also measured to determine whether the drugs produced any gross motor impairments (Fowler and Liou, 1994; Cousins and Salamone, 1996). The inter-response times (IRT) were analyzed by peak deviation analysis to obtain the peak location and peak area (Richards and Seiden, 1991; Richards et al., 1993a). A minimum of 25 lever presses with an IRT >6 s (i.e., pause responses) was required for peak deviation analysis (Richards and Seiden, 1991). The peak location was used to identify the median number of IRTs located above a theoretical random distribution. The peak area is the proportion of obtained pause IRTs above a theoretical random distribution. The peak area is a useful index of schedule control because peak areas that approach zero indicate that the rat is responding randomly (i.e., the obtained and theoretical distributions are identical). Based on previous studies from this laboratory, these measures can be used to define an "antidepressant-like" effect of drugs on the DRL 72-s schedule. The criteria for an antidepressant-like effect include 1) an increase in the number of reinforcers, 2) an increase in the peak location, and 3) little or no effect on the peak area.

Temperature. Core temperature measurements were recorded from miniaturized radio transmitters (resolution of 0.01°C; Minimitters, Sun River, OR) that were implanted into the peritoneal cavity (see Balcells-Olivero et al., 1998 for further details). The ambient air temperature was maintained at 22.5 ± 0.5°C, and the rats' body temperature was averaged once per 3 min. All core body temperatures were reported relative to the 5-min period (i.e., baseline) immediately preceding the initial s.c. injection.

HPLC with Electrochemical Detection (EC) for Tissue Levels of 8-OH-DPAT and Monoamine Levels. HPLC-EC was conducted using a dual-piston pump (Shimadzu, Columbia, MD), a 3-µm C₁₈ column (100 mm × 4.6 mm, Adsorbosphere; Alltech Assoc., Deerfield, IL), and a controller (model 5200; ESA, Chelmsford, MA) with electrodes (model 5014B; ESA). For measuring 8-OH-DPAT, the applied voltages were +500 mV and +800 mV (working versus reference). Mobile phase (0.05 M KH₂PO₄ with 30% methanol, at room temperature, pH 4.5) was pumped at a flow rate of 1.0 ml/min. Frontal cortex, hippocampus, mesencephalon, and hypothalamus were dissected according to the methods of Heffner et al. (1980) and frozen until assayed. On the day of the assay, the tissue was sonicated in 300 µl of 0.1 N perchloric acid and centrifuged at 27,000g for 20 min, 200 µl of supernatant was removed and passed over a previously conditioned Sep-Pak Light cartridge at a rate of ~1 ml/min (T-C₁₈; Waters Corp., Milford, MA). The remaining supernatant was saved for monoamine analysis (described below), and the tissue pellet was stored and frozen in 500 µl of 0.1 N NaOH (20°C). The Sep-Pac Light cartridge was then rinsed with 200 µl of 75% methanol, and the effluent from the second 200 µl of 75% methanol was saved. This 200-µl aliquot was heated until the sample volume was approximately 50 µl (~1 h at 60°C). Twenty microliters was then injected into the HPLC-EC. For each set of samples analyzed, 2 to 4 standards of 2 ng/20 µl 8-OH-DPAT were prepared using the Sep-Pak separation technique; 2 ng of 8-OH-DPAT corresponded roughly to the amount of 8-OH-DPAT on the column. Sample recovery was between 75 and 80%, and brain levels of 8-OH-DPAT were reported directly from levels on the column (pg/mg tissue).

Drug Administration

The vehicle for 8-OH-DPAT HBr (Research Biochemicals International; Natick, MA) was 0.9% saline; the vehicle for ketanserin tartrate (Research Biochemicals International), desipramine HCl (Research Biochemicals International), WAY-100635 HCl (gift from Wyeth-Ayerst, Princeton, NJ) and fluoxetine HCl (gift from Eli Lilly, Indianapolis, IN) was distilled water. Doses were calculated and expressed in terms of the salts and injected into a volume of 1.0 ml/kg body weight. Doses of 8-OH-DPAT (vehicle, 0.05, 0.1, 0.2, and 0.4 mg/kg s.c.), ketanserin (vehicle, 1.5, 3.0, 6.0, and 12.0 mg/kg i.p.), desipramine (vehicle, 1.25, 2.5, 5.0, and 10.0 mg/kg i.p. in the first experiment; vehicle, 0.15, 0.3, 0.625, 1.25, and 2.5 mg/kg i.p. in the second experiment), and fluoxetine (vehicle, 2.5, 5.0, 7.5, and 10.0 mg/kg i.p.) for the DRL 72-s studies were chosen based on previous work from this laboratory (Marek et al., 1989; Richards and Seiden, 1991; Richards et al., 1993b; Balcells-Olivero et al., 1998). Injections were made on any given day in ascending order of dose, 60 min before the behavioral test session. At least 7 days elapsed between 8-OH-DPAT injections, and a minimum of 4 days elapsed between injections of either ketanserin or desipramine. Rats were exposed first to 8-OH-DPAT, then to ketanserin, desipramine, and fluoxetine; at least 2 weeks of training without drugs was given between the administration of the different drugs, and no effects of this repeated drug procedure on DRL 72-s performance were detected. Rats in the desipramine replication experiment were only injected with desipramine.

Doses of 8-OH-DPAT for the temperature study (vehicle, 0.1, 0.3, and 1.0 mg/kg, s.c.) were based on previous work from this laboratory (Balcells-Olivero et al., 1998). Doses of WAY-100635 that were used to antagonize 8-OH-DPAT's hypothermic effects were based on pilot work and other published reports (0.003, 0.03, and 0.3 mg/kg s.c.; Forster et al., 1995). Because HDS rats are more sensitive to 8-OH-DPAT's hypothermic effects than LDS rats, HDS rats received 0.2 mg/kg 8-OH-DPAT, whereas LDS rats received 1.0 mg/kg 8-OH-DPAT following WAY-100635. Saline or 8-OH-DPAT was injected once a day for 4 consecutive days, in ascending order of dose. The rats were injected with WAY-100635 10 min before injection with 8-OH-DPAT.

The dose of 0.4 mg/kg 8-OH-DPAT was chosen for the pharmacokinetic study because this dose was found to produce robust hypothermia in both lines of rats.

Data Analysis

The data are reported as means ± S.E. The baseline data (no drug administered) from the DRL 72-s schedule, obtained during the week preceding drug injections, were analyzed by between-subjects t test with Bonferroni's correction (Keppel, 1991). DRL data obtained following pharmacological challenges were compared with vehicle within each line of rat by repeated measures t test with Bonferroni's correction (Keppel, 1991). The vehicle data were the averages of the data obtained on the four Tuesdays preceding each drug injection. The control data were the average of the behavioral data obtained on each of the four Wednesdays preceding drug injections (drug-free responding; control data were not used in any statistical analysis, but each of the drugs had independent control and vehicle measures).

For the 8-OH-DPAT challenge temperature experiments, the average body temperature, the maximal hypothermic change from baseline, and the time to attain maximum hypothermia were analyzed by a two-way repeated measures (between factor: lines of rats; within factor: time) mixed ANOVA using the Huynh-Feldt correction (Keppel, 1991). Post hoc comparisons between the vehicle and each dose of 8-OH-DPAT within each line of rat were conducted using Dunnett's test; comparisons between HDS and LDS rats at each dose of 8-OH-DPAT were also conducted with Tukey's test. In the WAY-100635/8-OH-DPAT temperature experiment, HDS and LDS rats were analyzed differently because the homogeneity of variance assumption of ANOVA was not satisfied in HDS rats (Keppel, 1991). Therefore, the temperature and time course data from the HDS rats were analyzed by Wilcoxon's test, whereas the LDS rats were analyzed by a within-subject ANOVA. Dunnett's test was used for comparisons between saline and the 8-OH-DPAT-alone condition and each WAY-100635/8-OH-DPAT condition in LDS rats, with separate error terms for each comparison being used (Keppel, 1991, pp 356-358). Tissue levels of 8-OH-DPAT in HDS and LDS rats were compared using a t test. Alpha was set at <0.05 for all experiments.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

HDS Rats Have Higher Reinforcement Rate under Drug-Free Conditions

Figure 1 shows the means ± S.E. of several important metrics of DRL 72-s responding in the HDS and LDS rats. The HDS rats exhibited an antidepressant-like profile compared with the LDS rats. The response rate was lower and reinforcement rate was higher for HDS rats [t(20) = 2.93, t(20) = 3.75, respectively, P < .05]. Peak deviation analysis indicated that the median peak location was shifted to the right in HDS rats compared with LDS rats [t(20) = 4.17, P < .05]. The burst ratio was higher in HDS rats [t(20) = 2.07, P < .05]. There was no difference in peak area between the two lines of rats, indicating that the proportion of nonrandom responses was similar in both lines. The median duration of each lever press also did not significantly differ between the two lines.

View larger version (28K): [in this window] [in a new window]   Fig. 1.   Comparison of baseline DRL 72-s performance between HDS and LDS rats. Top panel, histogram graphically showing the IRT distribution of HDS and LDS rats. Bottom panel, mean (±S.E.) of various parameters of DRL 72-s responding between the HDS and LDS rats. *P < .05, significantly different from LDS rats.

Effects of Drugs on DRL 72-s Performance

8-OH-DPAT Produced Antidepressant-Like Response in LDS But Not HDS Rats. As shown in Fig. 2, there was a reduction in response rate following administration of 8-OH-DPAT in HDS rats (0.2 and 0.4 mg/kg 8-OH-DPAT; t(10) = 4.96, t(10) = 4.69, respectively, P < 0.05); however, the response rate of LDS rats was not affected by 8-OH-DPAT. All of the LDS rats made >25 pause responses at the doses of 8-OH-DPAT tested, but several HDS rats did not meet this criterion (number of HDS failing to make >25 pause responses: 1 HDS rat at 0.1 mg/kg, 5 HDS rats at 0.2 mg/kg, and 7 HDS rats at 0.4 mg/kg 8-OH-DPAT). Only the LDS rats showed an increased reinforcement rate, and this effect was significant at 0.4 mg/kg 8-OH-DPAT [t(10) = 4.55, P < .05]; no effects were seen in HDS rats at any dose. The peak area, peak location, response duration, and burst ratio metrics were not affected by 8-OH-DPAT in either line.

View larger version (36K): [in this window] [in a new window]   Fig. 2.   Mean (±S.E.) effect of 8-OH-DPAT on several parameters of DRL 72-s responding in HDS and LDS rats (expressed as percentage change from control). The parameters shown are response rate (A; HDS control = 66.1 ± 6.9, LDS control = 92.5 ± 8.4), reinforcements (B; HDS control = 17.5 ± 1.5, LDS control = 10.6 ± 1.1), median peak location (C; HDS control = 64.3 ± 5.0, LDS control = 55.4 ± 4.9), and peak area (D; HDS control = 0.384 ± 0.029, LDS control = 0.310 ± 0.024). *P < .05, different from HDS vehicle.  ^P < .05, different from LDS vehicle.

Ketanserin Produced an Antidepressant-Like Response in LDS But Not HDS Rats. As shown in Fig. 3, there was a reduction in response rate at 0.375, 0.75, 3.0, and 6.0 mg/kg ketanserin in HDS rats [t(9) = 5.16, t(9) = 3.44, t(9) = 5.97, t(9) = 7.26, respectively, P < .05], and LDS rats showed significant reductions in response rate at 0.375, 3.0, and 6.0 mg/kg ketanserin [t(9) = 3.76, t(9) = 3.64, 3.46, respectively, P < .05]. Several HDS rats did not meet the >25 pause responses criterion (1 vehicle HDS rat, 2 HDS rats at 0.375 mg/kg, 1 HDS rat at 0.75 mg/kg, 2 HDS rats at 1.5 mg/kg, 3 HDS rats at 3 mg/kg, and 6 HDS rats at 6.0 mg/kg ketanserin). Ketanserin increased the number of reinforcers obtained by LDS rats. The number of reinforcers received by LDS rats was significantly increased at 0.375, 3.0, and 6.0 mg/kg ketanserin [t(9) = 4.60, t(9) = 3.59, t(9) = 4.49, respectively, P < .05]. LDS rats also showed an increase in median peak location [t(9) = 5.02, P < .05, at 0.375 mg/kg ketanserin]. Ketanserin also increased the median duration of responses at the 0.75 mg/kg dose [t(9) = 3.36, P < .05). The peak area and burst ratio measures were not significantly affected by ketanserin in either line of rat.

View larger version (38K): [in this window] [in a new window]   Fig. 3.   Mean (±S.E.) effect of ketanserin on several parameters of DRL 72-s responding in HDS and LDS rats (expressed as percentage change from control). The parameters shown are response rate (A; HDS control = 56.2 ± 9.4, LDS control = 91.1 ± 9.2), reinforcements (B; HDS control = 15.6 ± 1.6, LDS control = 11.1 ± 1.3), median peak location (C; HDS control = 64.5 ± 5.8, LDS control = 56.6 ± 3.6), and peak area (D; HDS control = 0.368 ± 0.021, LDS control = 0.327 ± 0.024). *P < .05, different from HDS vehicle. ^P < .05, different from LDS vehicle.

Fluoxetine Produced an Antidepressant-Like Response in LDS But Not HDS Rats. As shown in Fig. 4, fluoxetine did not significantly reduce response rate in HDS rats at any dose, but 7.5 mg/kg fluoxetine did reduce the response rate of LDS rats [all t(9) < 2.60, all N.S.; t(9) = 3.12, P < .05, respectively]. Several HDS rats did not meet the >25 pause responses criterion (2 vehicle HDS rats, 3 HDS rats at 2.5 mg/kg and 3 at 5.0 mg/kg, 6 HDS rats at 7.5, and 4 HDS rats at 10.0 mg/kg fluoxetine). The number of reinforcers obtained by LDS rats was increased by 7.5 and 10.0 mg/kg fluoxetine [t(9) = 3.75, t(9) = 3.89, P < .05, respectively]. The median peak location was significantly increased in LDS rats at 7.5 mg/kg fluoxetine [t(10) = 3.92, P < .05]. There were no significant effects of fluoxetine on lever press duration, peak area, or burst ratio in either line of rat.

View larger version (35K): [in this window] [in a new window]   Fig. 4.   Mean (±S.E.) effect of fluoxetine on several parameters of DRL 72-s responding in HDS and LDS rats (expressed as percentage change from control). The parameters shown are response rate (A; HDS control = 59.1 ± 10.3, LDS control = 76.7 ± 6.1), reinforcements (B; HDS control = 17.5 ± 1.8, LDS control = 13.7 ± 1.3), median peak location (C; HDS control = 65.7 ± 3.6, LDS control = 55.6 ± 2.3), and peak area (D; HDS control = 0.397 ± 0.031, LDS control = 0.398 ± 0.019). *P < .05, different from HDS vehicle. ^P < .05, different from LDS vehicle.

Desipramine Produced an Antidepressant-Like Response in LDS and HDS Rats. As shown in Fig. 5, 2.5, 5.0, and 10.0 mg/kg desipramine reduced the response rate of HDS rats [t(9) = 3.77, t(9) = 4.40, t(9) = 5.49, P < .05, respectively]; 0.625, 2.5, 5.0, and 10.0 mg/kg desipramine decreased the response rate of LDS rats [t(10) = 3.18, t(10) = 3.42, t(10) = 3.78, t(10) = 3.80, P < .05, respectively]. Several HDS rats did not meet the >25 pause responses criterion (1 HDS rat at 0.625 mg/kg, 3 HDS rats at 2.5 mg/kg, 3 HDS rats at 5.0 mg/kg, and 6 HDS rats at 10.0 mg/kg desipramine). Desipramine increased the reinforcement rate of HDS and LDS rats; a significant increase in reinforcement rate was observed at 1.25 mg/kg desipramine in HDS rats [t(9) = 3.95, P < .05], and reinforcement rate was increased at 0.625, 2.5, 5.0, and 10.0 mg/kg desipramine in LDS rats [t(10) = 3.51, t(10) = 4.06, t(10) = 4.25, t(10) = 4.10, P < .05, respectively]. The peak location was shifted to the right in HDS and LDS rats following desipramine; the median peak location was shifted to the right in LDS rats at 2.5, 5.0, and 10.0 mg/kg desipramine [t(10) = 3.86, t(9) = 4.35, t(10) = 4.17, P < .05, respectively] and in HDS rats at 1.25, 2.5, and 10.0 mg/kg desipramine [t(8) = 4.96, t(5) = 6.48, t(3) = 9.39, P < .05, respectively]. The peak area was increased by 5.0 mg/kg desipramine in LDS rats [t(9) = 5.09, P < .05]. There were no significant effects of desipramine on burst ratio or response duration in either line of rat.

View larger version (38K): [in this window] [in a new window]   Fig. 5.   Mean (±S.E.) effect of desipramine on several parameters of DRL 72-s responding in HDS and LDS rats (expressed as percentage change from control). The parameters shown are response rate (A; HDS control = 62.4 ± 9.0, LDS control = 81.7 ± 7.7), reinforcements (B; HDS control = 17.2 ± 1.3, LDS control = 13.4 ± 1.3), median peak location (C; HDS control = 64.5 ± 3.5, LDS control = 53.6 ± 2.9), and peak area (D; HDS control = 0.363 ± 0.033), LDS control = 0.376 ± 0.024). *P < .05, different from HDS vehicle.  ^P < .05, different from LDS vehicle.

Summary of Drugs on DRL 72-s Performance. One of the most reliable and readily identifiable features of antidepressants in rats responding on a DRL 72-s schedule is an increase in the reinforcement rate, which is usually accompanied by a shift to the right in the IRT distribution. The reinforcement rate of LDS rats was increased by 8-OH-DPAT, ketanserin, and fluoxetine; these drugs did not significantly affect the reinforcement rate of HDS rats. Desipramine was the only drug to significantly increase the reinforcement rate of LDS and HDS rats. Also, desipramine was the only drug to produce a significant shift to the right in IRT distribution (measured by peak location) of both lines of rats.

8-OH-DPAT-Induced Hypothermia and Antagonism with WAY-100635

8-OH-DPAT Produced Greater Hypothermia in HDS Rats. LDS rats had a blunted hypothermic response relative to HDS rats in response to 0.1 to 1.0 mg/kg 8-OH-DPAT (Fig. 6, top panels). There was a significant effect of rat line (F1,9 = 22.1, P < .001) and 8-OH-DPAT treatment (F3,27 = 83.0, P < .001) on average core body temperature (Fig. 6, bottom panel). As indicated by a significant interaction (F3,27 = 12.6, P < .001), the hypothermic differences between HDS and LDS rats increased with the dose of 8-OH-DPAT. There were no significant differences in body temperature between HDS and LDS during the baseline period (HDS = 37.5 ± 0.2°C; LDS = 37.4 ± 0.2°C) or after administration of saline. A similar pattern of results was observed using the maximal hypothermic response as the dependent variable (there was a significant rat line effect (F1,9 = 28.7, P < .001), 8-OH-DPAT effect (F3,27 = 76.8, P < .001), and a significant interaction (F3,27 = 14.9, P < .001)). In contrast, the time to reach maximal hypothermia was not influenced by either rat line or 8-OH-DPAT treatment, and there was no interaction.

View larger version (41K): [in this window] [in a new window]   Fig. 6.   Effects of 8-OH-DPAT on core body temperature of high 8-OH-DPAT-sensitive and low 8-OH-DPAT-sensitive rats. There were no differences in body temperature between the two lines during the baseline period (HDS rats = 37.1 ± 0.2oC; LDS = 37.3 ± 0.1oC). Effect of 8-OH-DPAT on body temperature (top panel) in HDS and LDS rats across the 20-min preinjection period and 120-min postinjection period (injection at arrows). The average and maximum hypothermia, relative to baseline, during the 120-min postinjection period and the time at which this is attained is shown for HDS and LDS rats for each dose of 8-OH-DPAT (bottom panel).

WAY-100635 Antagonized 8-OH-DPAT-Induced Hypothermia in LDS and HDS Rats. As shown in Table 1, the lowest dose of WAY-100635 (0.003 mg/kg WAY-100635) had no significant effect on 8-OH-DPAT's hypothermic effects in HDS rats. However, 0.03 and 0.3 mg/kg WAY-100635 significantly antagonized the hypothermia produced by 8-OH-DPAT, as determined by the average temperature [T(5) = 2.0, T(5) = 2.0, P < .05, respectively] and the maximum hypothermia [T(5) = 2.0, T(5) = 2.0, P < .05, respectively]. WAY-100635 at 0.03 mg/kg significantly increased the time to attain a hypothermic response in HDS rats [T(5) = 2.0, P < .05], but the low and high doses of WAY-100635 had no significant effects on this metric. In LDS rats, there was an overall effect of WAY-100635 on the average temperature and maximum hypothermia induced by 8-OH-DPAT (F3,15= 30.3 and F3,15 = 27.3, respectively, both P < .001). There was no significant effect of WAY-100635 on the time to attain maximum hypothermia in LDS rats.

Tissue Levels of 8-OH-DPAT and Monoamines

Tissue Levels of 8-OH-DPAT Do Not Differ between HDS and LDS Rats. Levels of 8-OH-DPAT in tissue obtained from the frontal cortex, hippocampus, mesencephalon, and hypothalamus of HDS and LDS rats are shown in Table 2. There were no significant differences in levels of 8-OH-DPAT in HDS and LDS rats in any brain region at either 15 or 30 min postinjection [all t(7, 8, 9, or 10) < 1.55, all N.S.].

Tissue Levels of Monoamines Do Not Differ between HDS and LDS Rats. Tissue levels of 5-HT, 5-HIAA, dopamine, norepinephrine, DOPAC, and HVA were measured in several brain regions of HDS and LDS rats (data not shown). HDS rats had significantly higher levels of HVA in the hypothalamus [t(19) = 2.1, P < .05] and hippocampus [t(19) = 2.5, P < .05] than LDS rats. A trend toward higher DOPAC levels was identified in the hippocampus of HDS rats relative to the LDS rats [t(19) = 1.9, P = 0.08]. Although there were modest differences in metabolite levels in several brain regions between the HDS and LDS rats, there were no differences in the levels of 5-HT, 5-HIAA, dopamine, and norepinephrine between the two lines of rats.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

We confirmed that HDS rats and LDS rats differed from each other in their sensitivity to the hypothermic effects of the 5-HT_1A receptor agonist 8-OH-DPAT; the hypothermic effects of 8-OH-DPAT were blocked by the 5-HT_1A antagonist WAY-100635. HDS and LDS rats were found to differ from each other on DRL 72-s performance under drug-free (baseline) conditions. The reinforcement rate under the DRL 72-s schedule was significantly increased in LDS rats but not in HDS rats after administration of three drugs that affect the 5-HT system: 8-OH-DPAT, ketanserin, and fluoxetine. In contrast, the noradrenergic uptake blocker desipramine increased the reinforcement rate in LDS and HDS rats. The relationship between the hypothermic response to 8-OH-DPAT and DRL 72-s performance (during both baseline and drugged conditions) in HDS and LDS rats is consistent with differences previously reported between out-bred Harlan Sprague-Dawley and Holtzman Sprague-Dawley rats (Balcells-Olivero et al., 1998).

HDS and LDS rats differed in drug-free DRL 72-s responding. On the DRL 72-s schedule, LDS rats emitted more responses and obtained fewer reinforcements than HDS rats. Along with obtaining fewer reinforcers than HDS rats, LDS rats had a peak location that was shifted to the left relative to HDS rats, as demonstrated by peak deviation analysis. However, it is important to note that differences in DRL 72-s responding between HDS and LDS rats were not accompanied by differences in peak area; the similar peak area between HDS and LDS rats suggests that both lines of rats were similarly controlled by the contingencies of the operant schedule. The modal duration of each lever press did not differ significantly between HDS and LDS rats, which indicates that there were not any obvious differences in coordinated motor ability (Fowler and Liou, 1994; Cousins and Salamone, 1996). Thus, relative to the LDS rats, the HDS rats perform as if they had been treated with an antidepressant. However, it must be emphasized that these behavioral effects have only been observed in one generation of HDS and LDS rats.

We measured tissue monoamine levels between HDS and LDS rats because previous research indicated that large, central depletions of 5-HT affect DRL responding. There were no differences in tissue levels of 5-HT, dopamine, or norepinephrine between the two lines. This is somewhat surprising because, relative to HDS rats, LDS rats behave as if they have a 5-HT depletion. Intraventricular microinjections of the 5-HT neurotoxin 5,7-dihydroxytryptamine disrupts DRL responding by increasing response rate and decreasing reinforcement rate (Marek et al., 1989; Wogar et al., 1993; Jolly et al., 1999). However, very large 5-HT depletions (approximately >90%) were required for behavioral effects in those studies, and it would seem unlikely a priori that the selective-breeding procedure for HDS and LDS rats would produce such large differences in tissue 5-HT levels. Injection of 5,7-dihydroxytryptamine or 8-OH-DPAT directly into the median raphe also disrupts DRL responding (Azmitia, 1981; Fletcher, 1994, 1995). Considering the baseline behavioral differences between HDS and LDS rats, these data suggest that the two selectively bred lines of rats may have differences that involve the median raphe 5-HT system; however, if differences exist they were not detected by measuring mesencephalic 5-HT tissue levels. A recent microdialysis study indicated that extracellular levels of 5-HT in the hippocampus and frontal cortex are similar under basal conditions (L. E. Gonzalez, M. Parada, L. Hernandez, L. Teneud, D. H. Overstreet., submitted).

In addition to characterizing drug-free DRL 72-s responding in HDS and LDS rats, we also examined the responses of these rats to several classes of drugs: a 5-HT_1A receptor agonist (8-OH-DPAT), a 5-HT₂ antagonist (ketanserin), a selective serotonin reuptake inhibitor (fluoxetine), and a tricyclic antidepressant (desipramine). However, only the LDS rats showed an increase in the number of reinforcements following either 8-OH-DPAT, ketanserin, or fluoxetine. In contrast to fluoxetine, 8-OH-DPAT and ketanserin did not produce a concomitant increase in peak location and therefore were only partially antidepressant-like in LDS rats. Another study in our laboratory using Holtzman Sprague-Dawley rats has also found only a modest antidepressant-like effect of 8-OH-DPAT on the DRL 72-s schedule. Indeed, there are several conflicting reports regarding the antidepressant-like profile of 8-OH-DPAT on this operant schedule (Marek et al., 1989; Kostowski et al., 1992; Jolly et al., 1999). In summary, 8-OH-DPAT, ketanserin, and fluoxetine increased the reinforcement rate of LDS but not HDS rats.

In contrast to the effects of 8-OH-DPAT, ketanserin, and fluoxetine, desipramine showed an antidepressant-like effect on DRL 72-s behavior in LDS and HDS rats. Both lines of rats showed significant increases in the number of reinforcements received, although this latter effect was much more robust in LDS rats. Indeed, although four doses of desipramine increased reinforcement rate in the LDS rats, only one dose (1.25 mg/kg in both experiments) increased this measure in the HDS rats. These behavioral effects are probably not mediated by desipramine's anticholinergic properties (Gupta et al., 1967), because muscarinic antagonists such as scopolamine produced a decrease in DRL reinforcement rate (McGuire and Seiden, 1980). The peak location metric was also shifted to the right in both HDS and LDS rats, as evidenced by the increase in reinforcement rate. These data show that desipramine produces an "antidepressant-like" effect in both lines of rats. The similar behavioral response to desipramine in the two lines of rats suggest that the differences in responding and reinforcement rate we observed with the 5-HT-acting drugs are not simply a function of baseline differences. Considering the effects of 8-OH-DPAT, ketanserin, and fluoxetine, it appears that drugs that act primarily through the serotonergic system affect DRL responding of the LDS rats, whereas drugs that act primarily through the noradrenergic system affect the LDS and HDS rats. Clearly, because only one dose of desipramine increased the reinforcement rate of HDS rats, it will be important to replicate this effect and to also determine whether other noradrenergic drugs produce an antidepressant-like effect in HDS rats.

The finding that 8-OH-DPAT, ketanserin, and fluoxetine affected the behavior of only the LDS rats, whereas desipramine affected both lines of rats, suggest that there may be similar anatomical or neurochemical substrates in HDS and LDS rats related to the norepinephrine system. However, an important consideration is that these drugs are not entirely selective for the target receptor or monoamine transporter in vivo (e.g., Chen and Reith, 1994; Jordan et al., 1994). Therefore, more selective serotonergic and noradrenergic drugs need to be evaluated in HDS and LDS rats to support the claim of selective drug responsivity.

In agreement with previous reports, 8-OH-DPAT produced dose-dependent decreases in core body temperature in HDS and LDS rats, but this effect was much larger in HDS rats (Overstreet et al., 1994, 1996). LDS rats also have a blunted hypothermic response relative to HDS rats following intraventricular microinjection of 8-OH-DPAT (Knapp et al., 1998). Because 8-OH-DPAT has similar affinity for both the 5-HT_1A and 5-HT₇ receptor subtypes, we used the selective 5-HT_1A antagonist WAY-100635 in an attempt to block the hypothermic effects of 8-OH-DPAT (Forster et al., 1995). In both lines of rats, WAY-100635 antagonized 8-OH-DPAT's hypothermic effects without affecting core body temperature when administered alone. Therefore, it appears that 5-HT_1A receptors mediate the differential thermic response to 8-OH-DPAT in HDS and LDS rats. This conclusion is consistent with that proposed earlier (Overstreet et al., 1996).

5-HT_1A agonist-induced hypothermia may have clinical implications. It is well established that 5-HT_1A receptor stimulation induces hypothermia in rats (Lin et al., 1983; Gudelsky et al., 1986; Koenig et al., 1988) and humans (Anderson et al., 1990; Lesch et al., 1990; Anderson and Cowen, 1992; Meltzer and Maes, 1995). Evidence indicates that some depressed patients differ in the magnitude of 5-HT_1A agonist-induced hypothermia, and some unmedicated, depressed patients have a blunted hypothermic response as compared with normal humans (Anderson et al., 1990; Lesch et al., 1990; Cowen et al., 1994; but see Meltzer and Maes, 1995). Although the physiological or anatomical substrates for these hypothermic differences in humans and rats are presently unclear, there may be important similarities between the blunted hypothermic response to 5-HT_1A receptor agonists in selectively bred LDS rats and in some people with depression.

The baseline DRL 72-s reinforcement rate was lower in LDS rats as compared with HDS rats. In addition, we observed that only LDS rats showed an increase in reinforcement rate following 8-OH-DPAT, ketanserin, or fluoxetine, but desipramine increased the reinforcement rate of both lines of rats. Precisely how these data from DRL 72-s performance and 8-OH-DPAT-induced hypothermia in these lines relate to each other is difficult to ascertain at this time, although we have previously observed a similar pattern in Harlan Sprague-Dawley and Holtzman Sprague-Dawley out-bred rats (Balcells-Olivero et al., 1998). Using lines of rats that differ in the magnitude of 8-OH-DPAT-induced hypothermia and that have other behavioral, pharmacological, and molecular differences may be useful for elucidating the action of antidepressants.