Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Effects of periadolescent fluoxetine and paroxetine on elevated plus-maze, acoustic startle, and swimming immobility in rats while on and off-drug

Effects of periadolescent fluoxetine and paroxetine on elevated plus-maze, acoustic startle, and... Rationale: Whether selective serotonin reuptake inhibitors (SSRIs) exposure during adolescent brain development causes lasting effects remains unresolved. Objective: Assess the effects of fluoxetine and paroxetine 60 days after adolescent exposure compared with when on-drug. Methods: Male Sprague-Dawley littermates (41 litters) were gavaged on postnatal days 33-53 with fluoxetine (3 or 10 mg/kg/day), paroxetine (3, 10 or, 17 mg/kg/day), or water; half were tested while on-drug (21 litters) and half after 60 days off-drug (20 litters). Results: The highest dose of the drugs reduced body weight gain during treatment that rebounded 1 week post- treatment. On-drug, no significant group differences were found on elevated plus maze time-in-open, zone entries, or latency to first open entry; however, the high dose of paroxetine significantly reduced head-dips (N = 20/ group). No significant effects were found on-drug for acoustic startle response/prepulse inhibition (ASR/PPI) although a trend (p < 0.10) was seen, which after combining dose levels, showed a significant increase in ASR amplitude for both fluoxetine and paroxetine (N = 20-21/group). No differences on immobility time were seen in the Porsolt forced swim test or in plasma corticosterone at the end of forced swim (N-19-21/group). Off-drug, no effects were seen in the elevated plus maze (N = 16/group), ASR/PPI (N = 20/group), forced swim (N = 19-20/ group), or plasma corticosterone (N = 19/group). At the doses tested, fluoxetine and paroxetine induced minor effects with drug on-board but no residual, long-term adverse effects in rats 60 days after drug discontinuation. Conclusions: The data provide no evidence that fluoxetine or paroxetine have long-term adverse effects on the behaviors measured here after adolescent to young adult exposure. Keywords: fluoxetine, paroxetine, acoustic startle response, elevated plus maze, forced swim test, corticosterone, adolescent brain development Background been used with success in adults for treatment of Selective serotonin reuptake inhibitors (SSRIs) are depression, premenstrual syndrome, obsessive compul- widely used antidepressants that inhibit the reuptake of sive behavior, panic disorder, generalized anxiety disor- serotonin (5-HT) into the presynaptic terminal by bind- der, social anxiety disorder, and post-traumatic stress ing to the serotonin transporter (SERT). SSRIs have disorder [1-3]. A meta-analysis of adult trials indicates that suicidality may be increased shortly after antide- pressant treatment is begun regardless of whether the * Correspondence: charles.vorhees@cchmc.org; michael.williams@cchmc.org drug is an SSRI or tricyclic antidepressant [4]. Division of Neurology, Department of Pediatrics, Cincinnati Children’s Research Foundation and University of Cincinnati College of Medicine, 3333 SSRIs have also been used in children and adolescents, Burnet Ave. Cincinnati, OH, 45229, USA primarily for depression [5]. It has been suggested that Full list of author information is available at the end of the article © 2011 Vorhees et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Vorhees et al. Behavioral and Brain Functions 2011, 7:41 Page 2 of 13 http://www.behavioralandbrainfunctions.com/content/7/1/41 children and adolescents may be at higher risk of adolescent exposure to two widely used SSRIs (fluoxe- adverse effects perhaps because SSRIs are contraindi- tine and paroxetine) has immediate or long-term effects. cated for some types of depression (such as bipolar dis- The previous experiments tested animals 21-56 or 30 order) that are often unrecognized in younger patients days after drug exposure. To determine if effects occur [5]. A recent meta-analysis of randomized controlled while on-drug and persist after drug cessation, we trea- trials of SSRI treatment for depression in children and ted two cohorts of rats identically and tested one cohort adolescents found benefit from SSRI treatment, but not while on-drug and the other after drug discontinuation. all SSRIs were equally effective. Fluoxetine, sertraline, We used the Porsolt FST because it is related to depression, the therapeutic target of these drugs. We and citalopram were most effective in this age group compared with other SSRIs [6]. assessed sensorimotor gating using ASR/PPI because Both efficacy and adverse effects of SSRIs in younger this is a widely used marker for abnormalities in neu- patients are ongoing areas of investigation. Modeling the ropsychiatric disorders [13-16]. We used the EPM to effects of juvenile SSRI treatment in rats [7,8] and mice determine whether SSRIs cause changes in anxiety, and [9] demonstrated that between postnatal (P) day 21 and we assessed plasma corticosterone after FST to deter- 35, there are differences in drug response [10]. However, mine if stress reactivity was altered by SSRI exposure. data on SSRI exposure during adolescence in rodents There are many SSRIs. We focused on paroxetine has received less attention. There are two issues con- because it showed residual effects previously [11] and cerning age-related effects of SSRI exposure: (1) efficacy fluoxetine because it is FDA approved for use in juvenile and (2) long-term or adverse effects. and adolescent patients (6-18 years of age). We used the In one experiment, male Wistar rats were treated with same treatment days as de Jong et al. [11], i.e., P33-62, 15 mg/kg of paroxetine or 30 mg/kg of fluvoxamine by an age range from puberty to young adulthood in rats. gavage from postnatal days (P) 33-62 and tested 20-21 One cohort was tested while on-drug after 25 days of days later for sexual behavior, 23-24 days later in the ele- prior treatment (tested on P57-62) and the other cohort vated plus-maze (EPM), 31 days later for acoustic startle off-drug (P122-127; 60 days post-treatment). We tested response (ASR) with prepulse inhibition (PPI), 35-36 doses from the literature showing effects after adoles- days later in the Porsolt forced swim test (FST), and 56 cent exposure [11,12]. A third dose of paroxetine was days later in an elevated T-maze [11]. Both drugs included to match the dose used by de Jong et al. [11]. decreased time in open arms in the EPM, but in an ele- vated T-maze in which 5 trials were given from a closed Methods and materials stem to open arms, only on trial-2 did fluvoxamine sig- Animals and treatments nificantly increase time to enter an open arm. On the Male (251-275 g) and female (151-175 g) Sprague-Daw- converse arrangement (open to closed entry), no change ley CD, IGS rats (Charles River, Raleigh, NC) were bred was seen. No change in ASR/PPI, immobility in the FST, in house after one to four weeks of acclimation to the or in 8-OH-DPAT-induced 5-HT -related behaviors vivarium and offspring were the subjects for these 1A was observed (lower lip retraction or sexual facilitation). experiments. Evidence of pregnancy was designated as A reduction in ejaculatory frequency was found in the E0 (embryonic day zero) and most females delivered on drug groups during the third week of testing, but not E22. Birth was designated P0 and litters were culled to 6 during weeks 1 and 2; no differences were found for eja- males and 2 females on P1. Litters not having a suffi- culatory latency, mount frequency, or intromissions. cient number of males were not included in the experi- More recently the effects of fluoxetine were tested in ment. At total of 41 litters were enrolled in the adolescent mice [12]. C57BL/6J and BALB/cJ male mice protocol. All litters were treated identically, the only dif- were treated with fluoxetine in drinking water at con- ference being that offspring from 21 litters were tested centrations of 80 or 160 mg/L on P21-42 or P56-84 while on drug (Group A testing began on P57) and off- with testing 30 days later on open-field, EPM, cued con- spring from the remaining 20 litters were tested begin- ditioned fear and extinction, theFST,plasmacorticos- ning 60 days after the last day of treatment (Group B terone, and brain 5-HT and 5-HIAA. The lower dose testing began on P122). Only males were tested, there- was calculated to be 9-10 mg/kg/day and the higher fore, 6 animals/litter × 41 litters = 246 animals in the dose 17-18 mg/kg/day. In neither strain, at neither dose experiment. Animals were maintained in polycarbonate did significant effects emerge [12]. No changes in 5-HT, shoebox cages (46 × 24 × 20 cm) on a 14 h light/10 h 5-HIAA, or corticosterone after the FST were seen. dark cycle (lights on at 600 h) with food (NIH-07 diet) Hence, the evidence is divided as to whether SSRI and water available ad lib in a vivarium maintained at exposure during adolescent brain development has 21 ± 1°C with 50 ± 10% humidity and accredited by the immediate and/or untoward lasting effects. The purpose Association for the Accreditation and Assessment of Laboratory Animal Care (AAALAC). The research of the present experiment was to test whether Vorhees et al. Behavioral and Brain Functions 2011, 7:41 Page 3 of 13 http://www.behavioralandbrainfunctions.com/content/7/1/41 protocol was approved by the Cincinnati Children ’s and Group B had 20 litters resulting N/group for each Research Foundation’s Institutional Animal Care and test ranging from 19-21/test for Group A and 16-20/test Use Committee. Offspring were separated from their for Group B. The tests were (in order): EPM, ASR/PPI, dams on P28 and housed 2 per cage for the remainder and the Porsolt FST (immediately followed by decapita- of the experiment. Each cage contained a semicircular tion and blood collection for determination of plasma stainless steel enclosure (17.8 cm long with the open corticosterone). There were 2 days of separation side 18.4 cm across and 10.2 cm high with the open between each test. Group A was tested in the EPM on side down made of 16-gauge stainless steel) as an addi- P57, ASR/PPI on P59, and FST on P61-62. Group B was tional environmental enrichment. tested in the EPM on P122, ASR/PPI on P124, and FST on P126-127. For Group A, all tests while on-drug were Drugs and doses given not less than 1 h and not more than 4 h after Animals were weighed daily during treatment and weekly each daily dose. thereafter. All animals received a daily gavage from P33-62 of 5 ml/kg body weight of USP grade water alone or water Elevated plus maze containing one of the drugs. Drugs were obtained as This test was conducted on a single day. The apparatus hydrochloride salts and all doses are expressed as the salt was constructed of high-density, black polyethylene form. The treatment groups were as follows: Control polymer. Each arm was 10 × 50 cm, with two opposing (water), Flu3 (Fluoxetine 3 mg/kg), Flu10 (fluoxetine 10 open arms and two perpendicular opposing arms with mg/kg), Par 3 (paroxetine 3 mg/kg), Par10 (paroxetine 10 high side walls (closed arms). The walls were 50 cm mg/kg), and Par 17 (paroxetine 17.04; this dose matched high in the closed arms and the open arms had 0.6 cm the dose of paroxetine reported previously [11] as 15 mg/ curbs on the edges to prevent falls. The apparatus was kg expressed as the freebase). Fluoxetine HCl was obtained mounted 50 cm above the floor. During testing the from Eli Lilly and Company (Indianapolis, IN) and paroxe- room was lit by a single dimmed halogen lamp (aimed tine HCl was obtained from Suzhou ChonTech Pharma- upward to reflect off the ceiling). Each animal was tested Chem Technology Company (Suzhou, China). for 5 min and movements were recorded on a DVD recorder using an overhead camera. Behavior was later Drug stability and potency played back and scored by an observer blind to treat- Drugs were mixed weekly in 5 ml of water in the appro- ment group for time-in-open, number of open entries, priate concentrations to deliver the specified doses. In latency to first open entry, and head dips. An arm was order to verify drug concentrations for each solution, a scored as entered if the animal’s head and two front feet crossedaboundarylinebetweenthecenterregionand set of drug solutions at each concentration was made and sent to ABC Laboratories (Columbia, MO) for veri- that arm. fication. In order to ensure stability of these solutions a set of the highest concentration solution of each drug Acoustic startle/PPI was made and analyzed at 0, 3, or 7 days post-mixing by This test was conducted on a single day. Acoustic startle ABC Laboratories. Finally, drug potency was assayed on reactivity with reflex modified inhibition by prepulse sti- both drug stocks at the end of the experiment to ensure mulation (ASR-PPI) was measured in four identical SR that no degradation had occurred by sending an aliquot Lab test chambers (San Diego Instruments, San Diego, of each drug to ABC Laboratories for analysis. CA). Each test chamber was calibrated using the manu- Drug concentrations were analyzed via high perfor- facturer’s guidelines and sensitivity was regularly mance liquid chromatography with ultraviolet detection checked using a calibrated oscillation device to ensure (HPLC-UV) at 227 nm. Samples were injected onto a consistent sensitive readings. At the start of each test, Zorbas SB-Phenyl reversed phase column (250 × 4.6 animals were placed in acrylic cylindrical tubes mounted mm, 5 μm). The HPLC system was an Agilent HP 1100. atop piezoelectric force transducers positioned inside The mobile phase was 50% acetonitrile, 50% 0.1% TFA sound attenuating test chambers. Background white aqueous. The flow rate was 1.0 ml/min, attenuation was noise was set at 70 dB. The test paradigm reported pre- 1000 mAU at ambient temperature with a column tem- viously was duplicated as closely as possible [11]. Each perature of 30°C. Sample size was 20 μland runtime test session consisted of a 5 min acclimation period fol- was 10 min. Final drug concentrations were calculated lowed by a 5 × 5 Latin square sequence of trials of 5 from a standard curve in the linear range. different types: no stimulus, startle signal with no pre- pulse, startle signal with prepulse 3 dB above back- Behavioral testing and sample collection ground (73 dB), startle signal with prepulse 5 dB above Group A and B animals were tested in the same beha- background (75 dB), and startle signal with prepulse 10 vioral tasks but at different ages. Group A had 21 litters dB above background (80 dB). Each animal received Vorhees et al. Behavioral and Brain Functions 2011, 7:41 Page 4 of 13 http://www.behavioralandbrainfunctions.com/content/7/1/41 each trial type once in each of the 5 orders, and the using a commercially available EIA kit (IDS, Fountain entire Latin square sequence was repeated; hence, each Hills, AZ). All determinations were made from kits ori- animal received each trial type/order twice. Trials of the ginating from the same lot. same type were averaged together for analysis. The intertrial intervals were 10-20 s with randomized spa- Statistical analyses cing. The interstimulus interval was 100 ms (measured Data for Group A and Group B animals were analyzed from prepulse onset to startle signal onset). The startle separately. Data were analyzed using completely rando- signal was a 120 dB mixed frequency, white noise burst mized block analyses of variance (ANOVA) mixed linear that lasted for 20 ms. The recording window was 100 models (SAS Proc Mixed, SAS Institute, Cary, NC). The ms. Prepulses lasted for 20 ms. Stimulus intensity was Kenward-Roger adjusted degrees of freedom method measured using a Quest sound level meter (SPL scale) was used and can be fractional. In these analyses, litter with the meter placed in the test chamber in the center was the block factor. Data from the EPM and FST had of the test stage with the door closed and the micro- 1-between-subject factor (treatment group). ASR-PPI phone directed upward toward the ceiling-mounted had 1-between-subject factor (treatment group) and 1- speaker. Response amplitude (V = maximum voltage within-subject factor (trial). For body weight, the max change within the recording window) was recorded in ANOVA has 1-between-subject factor (treatment group) units of voltage change (mV). Test chambers were and 1-within-subject factor (age). Interactions were ana- cleaned with 70% ethanol between animals. lyzed using the slice-effect ANOVA method on each level of within-subject factors. A posteriori group com- Porsolt forced swim test parisons were analyzed using the Hochberg step-up Testing was conducted over two successive days using method to control for multiple comparisons. As follow- the original method of Porsolt for rats [17,18]. The up we report effect size (ES) for drug treatment for each apparatus consisted of four clear acrylic cylinders 19 cm behavior. There are no methods for calculating ES for in diameter (i.d.) and 60 cm in height filled to a depth mixed linear ANOVA, therefore, we reanalyzed the data of 30 cm (to avoid water depth concerns [19]) with using general linear model ANOVA (GLM) again with room temperature water (21-23°C). On day-1 rats were litter as a random block factor. SAS GLM provides placed in the cylinders for 15 min and on day-2 for 5 three indices of ES (noncentrality, partial, and semipar- min. Each chamber was separated by black acrylic parti- tial correlation ratios). These reanalyses produced results tions to prevent animals from seeing one another. nearly identical to those obtained from Mixed model Movement was recorded using a video camera and DVD analyses, thereby validating the use of ES values from GLM ANOVA models. We then took the semipartial recorder. Scoring was performed by an experimenter blind with respect to treatment group. Immobility dura- correlation ratios (eta-square) and converted them to tion was scored on day-2 as time spent not swimming Cohen’s f (similar to Cohen’s d, except for ANOVAs or reaching at the walls. Occasional, minor limb move- rather than for t-a test) and followed Cohen’s categori- ments were permitted in order for the animal to main- zation scheme (small ES are around 0.2, medium around tain balance, brace itself against the wall, or tread water 0.5, and large ≥ 0.8 [22]. Significance was taken as p ≤ to keep afloat so long as a coordinated multi-limb swim- 0.05 and data are presented as least square mean ± ming motion was not present. We did not use the SEM. expanded FST method [20,21] because we were not test- ing SSRI efficacy, but rather for toxicity. Results Drug Concentrations and Stability Tissue collection Drug solutions of both compounds were prepared and Upon the completion of the FST, animals were removed sent to ABC Laboratories for analytical chemistry to from the water, carried to another room where they determine drug concentrations by HPLC analysis (Meth- were decapitated (< 30 s from removal from FST), and ods; M-1736-000 and USP 30 Paroxetine HCl). Each blood collected for later plasma corticosterone sample was analyzed in triplicate. The first set of sam- determinations. ples was taken from prepared test solutions for concen- tration verification. Mean concentrations of the test Corticosterone assay solutions were within 3% of target concentrations. Blood was collected in tubes that contained EDTA (2% Stability Assessment: Fluoxetine hydrochloride is in 0.05 ml). Samples were centrifuged for 15 min at 4°C, known to be stable in solution ≥ 7 days but to deter- plasma removed and stored at -80°C until assayed. Cor- mine the stability of paroxetine we tested it at the low ticosterone concentrations were determined by diluting and high concentrations at three time points: immedi- plasma 1:10 in assay buffer and assayed in duplicate ately after preparation and at 3 and 7 days. The Vorhees et al. Behavioral and Brain Functions 2011, 7:41 Page 5 of 13 http://www.behavioralandbrainfunctions.com/content/7/1/41 paroxetine solutions showed no decrease in active drug demonstrating time-dependent growth. The group × content over the 7-day period. week effect was not significant (F(20,960) = 0.50, p > End of Study Drug Purity: After the completion of the 0.96). experiment, crystalline samples of both drugs were sent Body weights during drug treatment (P33-62) were to ABC Laboratories to determine drug purity. Both analyzed with Groups A and B combined. There was a paroxetine HCl and fluoxetine HCl were unchanged significant effect of treatment (F(5,227) = 4.84, p = (mean percent potency of 101%). 0.0003), day (F(29,6922) = 1481.86, p < 0.0001), and treatment × day (F(145,6931) = 1.49, p = 0.0001). Slice- effect ANOVAs on each day showed no treatment General Characteristics There were no deaths; 100% of the animals enrolled in effects on P33-43, however significant effects were the experiment survived and were tested. Group sizes obtained on days P44-62. Since there were no significant varied slightly from the design because of experimenter group differences before P44, only the P44-62 body error or equipment malfunction. The exact N for each weight data are illustrated in Figure 1. Body weights on test is provided in the figure captions. the last day of treatment (P62) were analyzed separately Body weight data were analyzed by age. Once off- for Groups A and B and are shown in Figure 2. As can spring were assigned to treatment groups, their body be seen, for Group A, the Flu10 and Par 17 groups weights from P1-28 were analyzed to ensure that no weighed less than Control. For Group B, only the Flu10 preexisting differences occurred. This analysis showed group weighed less than Control. The Par17 for Group no differences as a function of later assignment to the B was not significantly different from Control (p = 0.12). groups (F(5,200) = 1.09, p > 0.36). The effect of age was Group A animals completed testing by P62, hence significant (F(4,960) = 23327.9, p < 0.0001), there are no body weight data for this group after P62. Figure 1 Body weights during treatment: Data represent least square mean (± SEM) body weights (g) on days on which significant group differences were obtained. Treatment was from P33-62; significant group differences occurred on P44-62. Groups A and B are combined. Group sizes: Group A = 21, Group B = 20. Vorhees et al. Behavioral and Brain Functions 2011, 7:41 Page 6 of 13 http://www.behavioralandbrainfunctions.com/content/7/1/41 remained on P70, P77, or thereafter (i.e., between P84- 126 (not shown)). Elevated plus maze For Group A there were no significant differences in time spent in the open arms, number of open arm entries, or latency to first open arm entry (Figure 4A-C, respectively). A significant treatment group effect was found on number of head dips (F(5,95) = 2.74, p < 0.03). Post hoc group comparisons showed that the Par17 group had fewer head dips than Controls (Figure 4D). No other group comparisons were significant. ES for time in open arms were = 0.15 (small), for arm Figure 2 Body weights: Least square Mean (± SEM) body entries = 0.13 (small), for latency = 0.12 (small), and for weights on the last day of treatment (P62). Group A and B are head dips = 0.27 (small). shown separately. There were 21 litters in Group A and 20 litters in For Group B there were no significant differences in Group B. **p < 0.01 vs. Control. time spent in the open arms, number of open arm entries, latency to first open arm entry, or head dips For Group B, body weights were recorded on P63 and (Figure 4E-H, respectively). ES for time in open arms = weekly thereafter until the end of testing on P126. In 0.18 (small), for arm entries = 0.22 (small), for latency = order to map body weight recovery after the end of 0.16 (small), and for head dips = 0.09 (small). treatment, data for the first two weeks after the end of treatment (P63-77) were analyzed separately. There was Acoustic Startle/PPI no significant treatment main effect (F(5,95) = 1.94, p < ASR-PPI data were analyzed two ways: (a) by prepulse 0.10). The day/week factor in the analysis was significant intensity, and (b) using the 0 prepulse response ampli- (F(2,228) = 4256.37 = p < 0.0001), indicating time- tude as a covariate by ANCOVA in order to assess PPI dependent growth, and there was a significant treatment after controlling for any possible differences in basal × week interaction (F(10,228) = 9.59, p < 0.0001). Slice- startle reactivity. effect ANOVAs on each week showed differences on For the Group A startle response ANOVA, the treat- P63 (p < 0.01) but not on P70 or P77, indicating that ment group main effect was not significant but showed recovery was rapid (< 1 week). This pattern is illustrated a trend (F(5,98.2) = 1.95, p < 0.10). There was no signifi- in Figure 3. As can be seen, only the Flu10 group cant treatment × prepulse interaction. Prepulse intensity weighed significantly less on P63 and no differences was significant (F(3,354) = 100.32, p < 0.0001) showing that prepulse inhibition of ASR was obtained. In the ANCOVA analysis, no significant treatment main effect was seen (F(5,98) = 1.86, p = 0.11). The treatment × PPI interaction was not significant. The prepulse main effect was significant (F(2,236) = 68.01, p < 0.0001). The data are shown in Figure 5(top panel).ESfor Group A = 0.27 (small). An inspection of Figure 5 taken together with the treatment main effect trend (p < 0.10) led us to con- duct two follow-up analyses. In both analyses we com- bined the two Flu groups ’ data and the three Par groups’ data into single Flu and Par groups. In the first follow-up analysis, all prepulse trials were included in a treatment × prepulse ANOVA. The treatment main effect (F(2,39) = 7.13, p < 0.003) and prepulse main effect (F(3,177) = 88.43, p < 0.0001) were both signifi- Figure 3 Body weights: Least square Mean (± SEM) body cant and the treatment × prepulse interaction showed weights for Group B during the two weeks after the end of treatment. Significant group differences occurred on P63; only the a trend (F(6,177) = 1.89, p < 0.09). Post hoc group Flu10 group weighed significantly less than Controls. From P70 to comparisons (averaged across prepulse intensities) the end of the experiment, there were no significant treatment showed that both the Flu and Par groups differed from group differences. *p < 0.05. Control (mean ± SEM: Control: 149.0 ± 23.5; Flu: Vorhees et al. Behavioral and Brain Functions 2011, 7:41 Page 7 of 13 http://www.behavioralandbrainfunctions.com/content/7/1/41 Figure 4 Elevated plus maze: Data shown are least square means (± SEM) on the EPM during the 5 min test session. A,E: time (s) spent in open arms; B,F: number of open arm entries; C,G: latency to first open arm entry; and D,H: number of open arm head dips. Group A was tested on P57 and Group B on P122. Group sizes for Group A: Number of litters tested = 21; data are based on the following number of subjects/group: Control = 20, Flu3 = 20, Flu10 = 20, Par3 = 20, Par10 = 20, Par17 = 20. One litter’s data were missing because of a DVD playback defect. For Group B: Number of litters tested = 20; data are based on the following number of subjects/group: Control = 16, Flu3 = 16, Flu10 = 16, Par3 = 16, Par10 = 16, Par17 = 16. Data were unrecoverable from 4 litters whose recorded data tracks would not playback from the DVD. Vorhees et al. Behavioral and Brain Functions 2011, 7:41 Page 8 of 13 http://www.behavioralandbrainfunctions.com/content/7/1/41 Figure 5 Acoustic startle/PPI: Least square Mean (± SEM) startle amplitude (V ) measured in units of voltage change (mV). Group A max was tested on P59 and Group B on P124. No treatment group effects were obtained for either Group A or B, nor any treatment × prepulse interactions. Prepulse was significant and showed that the greater the prepulse intensity the greater the inhibition of the startle response. Number of litters in Group A = 21. Progeny group sizes for Group A: Control = 20, Flu3 = 21, Flu10 = 21, Par3 = 20, Par10 = 21, Par17 = 21. For Group A, 1 Control and 1 Par3 animal had missing data because they were tested with the incorrect startle program. Number of litters for Group B = 20. Progeny group sizes for Group B: Control = 20, Flu3 = 20, Flu10 = 20, Par3 = 19, Par10 = 20, Par17 = 20. For Group B, 1 Par3 animal’s data were missing because the test chamber power was not activated. 224.4 ± 23.0; Par: 252.7 ± 23.0). This facilitated startle For Group B the treatment × prepulse ANOVA effect was most pronounced on the no-prepulse trials, showed no significant main effect of treatment group or therefore, the second follow-up ANOVA was with the treatment × prepulse intensity interactions. Prepulse combined dose groups but only on the no-prepulse intensity was significant (F(3,339) = 92.60, p < 0.0001) trials. ES for the latter = 0.44 (medium). With no pre- showing that ASR was significantly modified by the PPI pulses in the analysis, the ANOVA was a simple one- procedure (Figure 5, bottom panel). The ANCOVA way analysis with 3 groups. The treatment effect was analysis showed a similar outcome, no significant treat- significant (F(2,39) = 6.55, p < 0.004). This effect is ment main effect or treatment × prepulse interaction. shown in Figure 6(left panel). Post hoc group compar- The prepulse main effect was significant (F(2,226) = isons revealed that both drug groups showed signifi- 48.20, p < 0.0001). ES = 0.18 (small). cantly increased startle amplitude compared with the As for Group A, follow-up analyses with dose-levels of Control group. ES for the latter = 0.42 (medium). each group combined were conducted for Group B, and Vorhees et al. Behavioral and Brain Functions 2011, 7:41 Page 9 of 13 http://www.behavioralandbrainfunctions.com/content/7/1/41 Figure 6 Acoustic startle for the no-prepulse trials: Data are group least square mean (± SEM) startle amplitude (V ) max measured in units of voltage change (mV). Group A was tested on P59 and Group B on P124 as for Figure 5. Group sizes are as in Figure 5. *p < 0.05, **p < 0.01 compared with Control. there was no significant treatment main effect or treat- ment × prepulse interaction. ES = 0.06 (small). Prepulse was significant (F3,171) = 72.39, p < 0.0001). An addi- tional analysis performed on the no-prepulse trials by Figure 7 Porsolt forced swim test: Least square Mean (± SEM) one-way ANOVA showed no significant treatment effect immobility time on the test trial (day-2; out of 300 s). Animals (Figure 6, right panel). ES = 0.06 (small). received a 15 min training trial 24 h previously. Group A was tested on P61-62 and Group B on P126-127. Group sizes: Number of litters Porsolt FST tested in Group A = 21; data are based on the following number of subjects/group: Control = 19, Flu3 = 21, Flu10 = 20, Par3 = 21, For Group A, there were no significant treatment group Par10 = 21, Par17 = 21. Two Control and one Flu10 animal had effects found on immobility time (Figure 7, top panel) missing data tracks on the DVD because of playback issues. Number or latency to immobility (not shown) in the FST. ES = of litters tested in Group B = 20; data are based on the following 0.20 (small). number of subjects/group: Control = 19, Flu3 = 19, Flu10 = 20, Par3 Similarly, for Group B there were no significant effects = 20, Par10 = 20, Par17 = 20. Tracks for two Control and one Flu3 animal would not playback on the DVD. of treatment group found on immobility time (Figure 7, bottom panel) or latency to immobility (not shown) in the FST. ES = 0.21 (small). disorder, social anxiety disorder, and post-traumatic Corticosterone stress disorder [1-3]. These drugs have been found to be For Group A, all groups showed high corticosterone safer and have fewer side-effects than tricyclic antide- levels at the end of the 5 min FST compared with typi- pressants. SSRIs have also been used in children and cal basal levels (~50 ng/ml) however there were no adolescents, and fluoxetine is approved for use in those treatment differences among the groups (Figure 8, left 6-18 years of age. In children whose nervous system is panel). still developing, there are concerns about the long-term For Group B, again all groups showed increased corti- effects of these drugs. costerone levels in response to the stress at the end of In a recent preclinical experiment examining paroxe- the FST but there were no significant effects of treat- tine and fluvoxamine, it was reported that these drugs ment obtained (Figure 8, right panel). reduced time-in-open in an EPM [11]. Decreased time in open in the EPM is the most widely accepted index Discussion for this test of increased anxiety-like behavior [23]. In Fluoxetine, paroxetine, and many other SSRIs have pro- the experiment by deJong et al. [11] this change ven to be efficacious for the treatment of adults with occurred for both drugs 20-30 days after the end of depression, premenstrual syndrome, obsessive compul- treatment. No change was seen in the number of zone sive behavior, panic disorder, generalized anxiety crossings. These authors also tested the animals in an Vorhees et al. Behavioral and Brain Functions 2011, 7:41 Page 10 of 13 http://www.behavioralandbrainfunctions.com/content/7/1/41 Figure 8 Corticosterone: Mean (± SEM) plasma corticosterone levels obtained immediately after the end of the FST.Nodifferences were observed for either drug compared with Control. Group sizes: Number of litters tested in Group A = 21; data are based on the following number of subjects/group: Control = 21, Flu3 = 21, Flu10 = 21, Par3 = 21, Par10 = 21, Par17 = 21. Number of litters tested in Group B = 20; data are based on the following number of subjects/group: Control = 19, Flu3 = 19, Flu10 = 19, Par3 = 19, Par10 = 19, Par17 = 19. elevated T-maze (the EPM was used with one arm Behaviorally, no treatment-related effects on the EPM blocked). Out of 5 trials for time to enter an open arm were obtained on the principal measure of anxiety, i.e., time-in-open. In addition, no differences were found for after being placed in a closed arm, both SSRI-treated groups entered an open arm faster on trial-2 compared number of zone crossings or latency to first open arm with controls but not on trial-1 or 3-5. When the situa- entry. The Group A Par17 animals showed a significant tion was reversed and they were given an additional trial reduction in head-dips, an effect seen in no other group, by being placed in an open arm and timed for entry however this effect was not seen in Group B when into a closed arm, no differences were found. No differ- tested 60 days after drug cessation. deJong et al. [11] did ences in ASR/PPI were obtained nor were changes not report head-dips, therefore, it is not possible to obtained on immobility time in the FST. make a direct comparison on this variable, however, a We sought to further test the effects of adolescent reductioninhead-dips is generally interpreted as an exposure to SSRIs and followed many of the design fea- increase in anxiety [24], although it is an index of anxi- tures used previously [11], except using more dose levels ety that is secondary to time-in-open and therefore pro- of each drug and including separate groups of animals, vides less persuasive evidence of a significant change in half tested while on-drug (Group A) and half tested off- anxiety. However, this effect would be consistent with drug (Group B; 60 days post-treatment). If long-term the finding of deJong et al. [11]. A recent study showed effects of SSRI treatment were obtained, it would poten- that rats treated from P25-49 with 12 mg/kg of fluoxe- tially represent a concern for the safety of the drugs. tine had reduced time-in-open in the EPM [25] 7 days Both drugs produced body weight reductions at the post-treatment, a finding consistent with reduced head- highest doses tested by P44, i.e., after 12 days of treat- dips that we found while animals were on drug. It is ment, and these reductions remained through the end unclear why adolescent exposed rats and mice show of treatment (P62). But the effects showed recovery to greater inconsistency in EPM responses to SSRIs than non-significant differences one week post-treatment. adult rodents. Some of it may be related to the different Vorhees et al. Behavioral and Brain Functions 2011, 7:41 Page 11 of 13 http://www.behavioralandbrainfunctions.com/content/7/1/41 exposure ages as there is no consistent definition of prepulse trials included, i.e., only the unmodified, basic adolescence in rodents. Doses and routes of drug ASR trials. Both analyses resulted in the same finding: administration as well as duration of treatment and end the pooled fluoxetine and pooled paroxetine groups of treat to test interval vary widely across studies, exhibited increased ASR amplitude compared with Con- increasing the difficulty of discerning patterns. trols. By contrast, when Group B ASR-PPI data were Neither we nor deJong et al. (2006) found any effect analyzed the same way, no residual effect or trend toward an effect was observed. This is in contrast to a on the FST test of swimming despair [17,18] and this study that found that P24-46 fluoxetine reduced ASR at finding is different than that reported by Homberg et al. ahigherdosethan weused(12 mg/kg),but similarto [25] but they treated rats earlier (P25-46); moreover, they found no differences in rats treated at a later age. our data, found no effects at later ages [25]. Other studies have found opposite or paradoxical effects The present data should be viewed within the limita- of adolescent exposure to fluoxetine in mice, but only in tions of the study. As in the deJong et al. [11] and Nor- one of the two strains tested [26]. To the extent that the cross et al. [12] studies, we tested only males. It may be FST is a valid preclinical test of depression, the data do worthwhile to test females as many drugs exhibit sexu- not suggest that adolescent exposure to fluoxetine or ally dimorphic responses. It is also worth considering paroxetine result in long-term changes in immobility. that future experiments might test a broader range of This is consistent with a newer report of chronic fluoxe- doses in order to more thoroughly test for possible tine treatment in mice from P14-42 showing no FST adverse effects. Finally, other behavioral tests might be effects [26]and thedatacited above[12].Wescored worth considering; Norcross et al. [12] also included the FST for immobility time as originally described by open-field and cued fear conditioning with extinction, Porsoltetal.[17,18].Wedidnotscore thetestusing but even more tests might be considered such as fear- the recently suggested indices of active swimming and induced acoustic startle facilitation. There are many wall climbing [20]. While we cannot rule-out the possi- tests of anxiety, conditioned fear, sensorimotor gating, bility that these additional indices might have uncovered and depression, of which only a subset were used here other effects, we did not include them because our or in the other studies cited that tested for residual interest was not differentiating among different classes effects. A more extensive battery of tests might reveal of SSRIs, which was the reason these measures were effects not detected herein. introduced, but rather whether immobility was a long- Overall, the data show that for the doses of fluoxetine and paroxetine tested by an oral route during adolescent term consequence of drug exposure after adolescent to early adult brain development (operationally defined treatment. Others have administered SSRIs by osmotic pump in as P33-62), caused minor and only transient reductions order to maintain plasma concentrations in animals in body weight gain, a small but significant ASR facilita- within the same range as human therapeutic concentra- tion during treatment that did not remain 60 days post- tions [27,28]. The intent of these studies was different treatment, and a high-dose only (Par17) reduction in than ours. These studies were designed to assess the EPM head-dips while on-drug but not off-drug. The molecular targets of SSRIs. For this purpose maintaining findings,whenconsideredinterms of thepower to a constant drug concentration is desirable but this goal detect differences in the present experimental design (≥ was not necessary for our purposes of determining if 20 animals per group), multiple dose levels, use of a these drugs have persistent long-term effects long after within-litter design that improves subject matching, and cessation. inclusion of both on- and off-drug cohorts, suggest that deJong et al. [11] reported no changes in ASR/PPI. there is no signal of adverse effects present in the data We replicated their parameters and in agreement, that might raise concern over the long-term safety of observed no changes in ASR or PPI from either drug 60 the drugs when treatment is during an interval spanning days post-treatment. However, while on-drug, we saw a adolescent brain development in rats, in agreement with trend toward ASR facilitation that did not interact with previous findings in mice [12]. It was recently reported PPI. To further explore the data, we conducted follow- that fluoxetine given by continuous infusion via mini- up analyses by combining the two fluoxetine groups pumps from P14-42 results in anxiogenic effects in mice into one pooled group by averaging the data for the while on-drug on tests of novelty-induced hypophagia in fluoxetine animals in each litter together to create a sin- Swiss-Webster (SW) and C57BL/6 Charles River (B6) mice, however there were no effects in the EPM in the gle, merged group. We did the same for the paroxetine, SW strain and reduced time-in-open in the B6 strain merging the data from all three dose levels together but no effects on open-field center time or FST in either among littermates to create a single paroxetine group. Two follow-up analyses were performed. These were: (1) strain. These effects disappeared off-drug [26] which is with all prepulse trials included and (2) with no- entirely consistent with the present findings. Thus, while Vorhees et al. Behavioral and Brain Functions 2011, 7:41 Page 12 of 13 http://www.behavioralandbrainfunctions.com/content/7/1/41 inspections at regular intervals to ensure protocol compliance. DLG SSRIs induce relatively reliable effects on-drug in adult conducted the corticosterone assays. MRS was in charge of study blinding rodents, their effects from adolescent exposure are less of animals to ensure that technicians had no knowledge of treatment consistent. Our data are consistent in that we saw only groups, made coded drug solutions, and provided drug solutions to ABC Laboratories blinded for potency and stability assays on drug solutions. MTW small effects while on drug and no long-term effects was Co-Principal Investigator, participated in the design, data analyses, and long after drug discontinuation. The latter provides interpretation of the data, and contributed to the manuscript. All authors some evidence that these drugs are not neurotoxic when have read and approved the final manuscript. given during adolescent stages of brain development at Received: 20 April 2011 Accepted: 5 October 2011 least on the behavioral indices used here. Published: 5 October 2011 Financial competing interests References 1. Tang SW, Helmeste D: Paroxetine. Expert Opin Pharmacother 2008, The research reported herein was supported by a 9:787-794. research contract to Cincinnati Children’s Research 2. Goddard AW, Shekhar A, Whiteman AF, McDougle CJ: Serotoninergic Foundation (CCRF) under CCRF required provisions of mechanisms in the treatment of obsessive-compulsive disorder. Drug Discov Today 2008, 13:325-332. academic freedom from sponsor influence, viz., that work 3. Shah NR, Jones JB, Aperi J, Shemtov R, Karne A, Borenstein J: Selective performed under the auspices of the CCRF are under the serotonin reuptake inhibitors for premenstrual syndrome and sole discretion of CCRF and the investigators. Under premenstrual dysphoric disorder: a meta-analysis. Obstet Gynecol 2008, 111:1175-1182. these provisions, the principal investigators are free to 4. Hall WD, Lucke J: How have the selective serotonin reuptake inhibitor publish all results as they deem scientifically appropriate. antidepressants affected suicide mortality? Aust N Z J Psychiatry 2006, In addition, the CCRF authors have received no reimbur- 40:941-950. 5. Henry C, motes-Mainard J: SSRIs, suicide and violent behavior: is there a sements, fees, salary, honoraria, or other benefits from need for a better definition of the depressive state? Curr Drug Saf 2006, the sponsor. Dr. Morford is an employee of Eli Lilly and 1:59-62. Company. The article processing charge was paid by 6. Usala T, Clavenna A, Zuddas A, Bonati M: Randomised controlled trials of selective serotonin reuptake inhibitors in treating depression in children CCRF. The CCRF authors declare that they have no and adolescents: a systematic review and meta-analysis. Eur stock or other financial interest in the sponsoring com- Neuropsychopharmacol 2008, 18:62-73. pany and stand to make no financial gain or loss from 7. Reed AL, Anderson JC, Bylund DB, Petty F, El RH, Happe HK: Treatment with escitalopram but not desipramine decreases escape latency times the publication of these data now or in the future. The in a learned helplessness model using juvenile rats. Psychopharmacology CCRF authors hold no patent now or in the future (Berl) 2009, 205:249-259. related to this manuscript. The authors have received no 8. Reed AL, Happe HK, Petty F, Bylund DB: Juvenile rats in the forced-swim test model the human response to antidepressant treatment for reimbursements, fees, funding, or salary from an organi- pediatric depression. Psychopharmacology (Berl) 2008, 197:433-441. zation that holds or has applied for patents relating to 9. Mason SS, Baker KB, Davis KW, Pogorelov VM, Malbari MM, Ritter R, Wray SP, the content of the manuscript. The CCRF authors declare Gerhardt B, Lanthorn TH, Savelieva KV: Differential sensitivity to SSRI and tricyclic antidepressants in juvenile and adult mice of three strains. Eur J no financial conflicts of interest. Pharmacol 2009, 602:306-315. 10. Bylund DB, Reed AL: Childhood and adolescent depression: why do Non-financial competing interests children and adults respond differently to antidepressant drugs? Neurochem Int 2007, 51:246-253. The authors declare no non-financial competing inter- 11. de Jong TR, Snaphaan LJ, Pattij T, Veening JG, Waldinger MD, Cools AR, ests (political, personal, religious, ideological, academic, Olivier B: Effects of chronic treatment with fluvoxamine and paroxetine intellectual, commercial, or any other) in relation to this during adolescence on serotonin-related behavior in adult male rats. Eur Neuropsychopharmacol 2006, 16:39-48. manuscript. 12. Norcross M, Mathur P, Enoch AJ, Karlsson RM, Brigman JL, Cameron HA, Harvey-White J, Holmes A: Effects of adolescent fluoxetine treatment on fear-, anxiety- or stress-related behaviors in C57BL/6J or BALB/cJ mice. Acknowledgements Psychopharmacology (Berl) 2008, 200:413-424. The authors’ thank Mary S. Moran for assistance with the statistical analyses 13. Swerdlow NR, Light GA, Cadenhead KS, Sprock J, Hsieh MH, Braff DL: Startle and to Holly Johnson and Lindsey Burns for drug treatment and behavioral gating deficits in a large cohort of patients with schizophrenia: testing. This project was supported by a research agreement between relationship to medications, symptoms, neurocognition, and level of Cincinnati Children’s Research Foundation and Eli Lilly and Company, function. Arch Gen Psychiatry 2006, 63:1325-1335. Indianapolis, Indiana. The data, analyses, interpretation, and writing of the 14. Hsieh MH, Swerdlow NR, Braff DL: Effects of background and prepulse manuscript were conducted by the authors, with final content decisions characteristics on prepulse inhibition and facilitation: implications for residing with the Cincinnati authors. neuropsychiatric research. Biol Psychiatry 2006, 59:555-559. 15. Kask K, Gulinello M, Backstrom T, Geyer MA: Sundstrom-Poromaa, I. Author details Patients with premenstrual dysphoric disorder have increased startle Division of Neurology, Department of Pediatrics, Cincinnati Children’s response across both cycle phases and lower levels of prepulse Research Foundation and University of Cincinnati College of Medicine, 3333 inhibition during the late luteal phase of the menstrual cycle. Burnet Ave. Cincinnati, OH, 45229, USA. Eli Lilly and Company, Greenfield, Neuropsychopharmacology 2008, 33:2283-2290. IN, 46140 USA. 16. Geyer MA: The family of sensorimotor gating disorders: comorbidities or diagnostic overlaps? Neurotox Res 2006, 10:211-220. Authors’ contributions 17. Porsolt RD, LePichon M, Jalfre M: Depression: A new animal model CVV was Principal Investigator, participated in the design, data analyses, and sensitive to antidepressant treatments. Nature 1977, 226:730-731. interpretation of the data, oversaw the project, and wrote the manuscript. LRM participated in the design and provided on-sight quality assurance Vorhees et al. Behavioral and Brain Functions 2011, 7:41 Page 13 of 13 http://www.behavioralandbrainfunctions.com/content/7/1/41 18. Porsolt RD, Anton G, Blavet N, Jalfre M: Behavioural despair in rats: A new model sensitive to antidepressant treatments. Eur J Pharmacol 1978, 47:379-391. 19. Detke MJ, Lucki I: Detection of serotonergic and noradrenergic antidepressants in the rat forced swimming test: the effects of water depth. Behav Brain Res 1996, 73:43-46. 20. Cryan JF, Valentino RJ, Lucki I: Assessing substrates underlying the behavioral effects of antidepressants using the modified rat forced swimming test. Neurosci Biobehav Rev 2005, 29:547-569. 21. Cryan JF, Page ME, Lucki I: Differential behavioral effects of the antidepressants reboxetine, fluoxetine, and moclobemide in a modified forced swim test following chronic treatment. Psychopharmacology (Berl) 2005, 182:335-344. 22. Cohen J: Statistical Power Analysis for the Behavioral Sciences. Hillsdale, NJ: Lawrence Erlbaum Associates;, 2 1988. 23. Walf AA, Frye CA: The use of the elevated plus maze as an assay of anxiety-related behavior in rodents. Nat Protoc 2007, 2:322-328. 24. Fernandes C, File SE: The influence of open arm ledges and maze experience in the elevated plus-maze. Pharmacol Biochem Behav 1996, 54:31-40. 25. Homberg JR, Olivier JD, Blom T, Arentsen T, van BC, Schipper P, Korte- Bouws G, van LG, Reneman L: Fluoxetine exerts age-dependent effects on behavior and amygdala neuroplasticity in the rat. PLoS ONE 2011, 6: e16646. 26. Oh JE, Zupan B, Gross S, Toth M: Paradoxical anxiogenic response of juvenile mice to fluoxetine. Neuropsychopharmacology 2009, 34:2197-2207. 27. Owens MJ, Knight DL, Nemeroff CB: Paroxetine binding to the rat norepinephrine transporter in vivo. Biol Psychiatry 2000, 47:842-845. 28. McHugh PC, Rogers GR, Glubb DM, Joyce PR, Kennedy MA: Proteomic analysis of rat hippocampus exposed to the antidepressant paroxetine. J Psychopharmacol 2010, 24:1243-1251. doi:10.1186/1744-9081-7-41 Cite this article as: Vorhees et al.: Effects of periadolescent fluoxetine and paroxetine on elevated plus-maze, acoustic startle, and swimming immobility in rats while on and off-drug. Behavioral and Brain Functions 2011 7:41. Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color figure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Behavioral and Brain Functions Springer Journals

Effects of periadolescent fluoxetine and paroxetine on elevated plus-maze, acoustic startle, and swimming immobility in rats while on and off-drug

Download PDF
13 pages

Loading next page...
 
/lp/springer-journals/effects-of-periadolescent-fluoxetine-and-paroxetine-on-elevated-plus-HbelDokO2u

References (62)

Publisher
Springer Journals
Copyright
Copyright © 2011 by Vorhees et al; licensee BioMed Central Ltd.
Subject
Biomedicine; Neurosciences; Neurology; Behavioral Therapy; Psychiatry
eISSN
1744-9081
DOI
10.1186/1744-9081-7-41
pmid
21974752
Publisher site
See Article on Publisher Site

Abstract

Rationale: Whether selective serotonin reuptake inhibitors (SSRIs) exposure during adolescent brain development causes lasting effects remains unresolved. Objective: Assess the effects of fluoxetine and paroxetine 60 days after adolescent exposure compared with when on-drug. Methods: Male Sprague-Dawley littermates (41 litters) were gavaged on postnatal days 33-53 with fluoxetine (3 or 10 mg/kg/day), paroxetine (3, 10 or, 17 mg/kg/day), or water; half were tested while on-drug (21 litters) and half after 60 days off-drug (20 litters). Results: The highest dose of the drugs reduced body weight gain during treatment that rebounded 1 week post- treatment. On-drug, no significant group differences were found on elevated plus maze time-in-open, zone entries, or latency to first open entry; however, the high dose of paroxetine significantly reduced head-dips (N = 20/ group). No significant effects were found on-drug for acoustic startle response/prepulse inhibition (ASR/PPI) although a trend (p < 0.10) was seen, which after combining dose levels, showed a significant increase in ASR amplitude for both fluoxetine and paroxetine (N = 20-21/group). No differences on immobility time were seen in the Porsolt forced swim test or in plasma corticosterone at the end of forced swim (N-19-21/group). Off-drug, no effects were seen in the elevated plus maze (N = 16/group), ASR/PPI (N = 20/group), forced swim (N = 19-20/ group), or plasma corticosterone (N = 19/group). At the doses tested, fluoxetine and paroxetine induced minor effects with drug on-board but no residual, long-term adverse effects in rats 60 days after drug discontinuation. Conclusions: The data provide no evidence that fluoxetine or paroxetine have long-term adverse effects on the behaviors measured here after adolescent to young adult exposure. Keywords: fluoxetine, paroxetine, acoustic startle response, elevated plus maze, forced swim test, corticosterone, adolescent brain development Background been used with success in adults for treatment of Selective serotonin reuptake inhibitors (SSRIs) are depression, premenstrual syndrome, obsessive compul- widely used antidepressants that inhibit the reuptake of sive behavior, panic disorder, generalized anxiety disor- serotonin (5-HT) into the presynaptic terminal by bind- der, social anxiety disorder, and post-traumatic stress ing to the serotonin transporter (SERT). SSRIs have disorder [1-3]. A meta-analysis of adult trials indicates that suicidality may be increased shortly after antide- pressant treatment is begun regardless of whether the * Correspondence: charles.vorhees@cchmc.org; michael.williams@cchmc.org drug is an SSRI or tricyclic antidepressant [4]. Division of Neurology, Department of Pediatrics, Cincinnati Children’s Research Foundation and University of Cincinnati College of Medicine, 3333 SSRIs have also been used in children and adolescents, Burnet Ave. Cincinnati, OH, 45229, USA primarily for depression [5]. It has been suggested that Full list of author information is available at the end of the article © 2011 Vorhees et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Vorhees et al. Behavioral and Brain Functions 2011, 7:41 Page 2 of 13 http://www.behavioralandbrainfunctions.com/content/7/1/41 children and adolescents may be at higher risk of adolescent exposure to two widely used SSRIs (fluoxe- adverse effects perhaps because SSRIs are contraindi- tine and paroxetine) has immediate or long-term effects. cated for some types of depression (such as bipolar dis- The previous experiments tested animals 21-56 or 30 order) that are often unrecognized in younger patients days after drug exposure. To determine if effects occur [5]. A recent meta-analysis of randomized controlled while on-drug and persist after drug cessation, we trea- trials of SSRI treatment for depression in children and ted two cohorts of rats identically and tested one cohort adolescents found benefit from SSRI treatment, but not while on-drug and the other after drug discontinuation. all SSRIs were equally effective. Fluoxetine, sertraline, We used the Porsolt FST because it is related to depression, the therapeutic target of these drugs. We and citalopram were most effective in this age group compared with other SSRIs [6]. assessed sensorimotor gating using ASR/PPI because Both efficacy and adverse effects of SSRIs in younger this is a widely used marker for abnormalities in neu- patients are ongoing areas of investigation. Modeling the ropsychiatric disorders [13-16]. We used the EPM to effects of juvenile SSRI treatment in rats [7,8] and mice determine whether SSRIs cause changes in anxiety, and [9] demonstrated that between postnatal (P) day 21 and we assessed plasma corticosterone after FST to deter- 35, there are differences in drug response [10]. However, mine if stress reactivity was altered by SSRI exposure. data on SSRI exposure during adolescence in rodents There are many SSRIs. We focused on paroxetine has received less attention. There are two issues con- because it showed residual effects previously [11] and cerning age-related effects of SSRI exposure: (1) efficacy fluoxetine because it is FDA approved for use in juvenile and (2) long-term or adverse effects. and adolescent patients (6-18 years of age). We used the In one experiment, male Wistar rats were treated with same treatment days as de Jong et al. [11], i.e., P33-62, 15 mg/kg of paroxetine or 30 mg/kg of fluvoxamine by an age range from puberty to young adulthood in rats. gavage from postnatal days (P) 33-62 and tested 20-21 One cohort was tested while on-drug after 25 days of days later for sexual behavior, 23-24 days later in the ele- prior treatment (tested on P57-62) and the other cohort vated plus-maze (EPM), 31 days later for acoustic startle off-drug (P122-127; 60 days post-treatment). We tested response (ASR) with prepulse inhibition (PPI), 35-36 doses from the literature showing effects after adoles- days later in the Porsolt forced swim test (FST), and 56 cent exposure [11,12]. A third dose of paroxetine was days later in an elevated T-maze [11]. Both drugs included to match the dose used by de Jong et al. [11]. decreased time in open arms in the EPM, but in an ele- vated T-maze in which 5 trials were given from a closed Methods and materials stem to open arms, only on trial-2 did fluvoxamine sig- Animals and treatments nificantly increase time to enter an open arm. On the Male (251-275 g) and female (151-175 g) Sprague-Daw- converse arrangement (open to closed entry), no change ley CD, IGS rats (Charles River, Raleigh, NC) were bred was seen. No change in ASR/PPI, immobility in the FST, in house after one to four weeks of acclimation to the or in 8-OH-DPAT-induced 5-HT -related behaviors vivarium and offspring were the subjects for these 1A was observed (lower lip retraction or sexual facilitation). experiments. Evidence of pregnancy was designated as A reduction in ejaculatory frequency was found in the E0 (embryonic day zero) and most females delivered on drug groups during the third week of testing, but not E22. Birth was designated P0 and litters were culled to 6 during weeks 1 and 2; no differences were found for eja- males and 2 females on P1. Litters not having a suffi- culatory latency, mount frequency, or intromissions. cient number of males were not included in the experi- More recently the effects of fluoxetine were tested in ment. At total of 41 litters were enrolled in the adolescent mice [12]. C57BL/6J and BALB/cJ male mice protocol. All litters were treated identically, the only dif- were treated with fluoxetine in drinking water at con- ference being that offspring from 21 litters were tested centrations of 80 or 160 mg/L on P21-42 or P56-84 while on drug (Group A testing began on P57) and off- with testing 30 days later on open-field, EPM, cued con- spring from the remaining 20 litters were tested begin- ditioned fear and extinction, theFST,plasmacorticos- ning 60 days after the last day of treatment (Group B terone, and brain 5-HT and 5-HIAA. The lower dose testing began on P122). Only males were tested, there- was calculated to be 9-10 mg/kg/day and the higher fore, 6 animals/litter × 41 litters = 246 animals in the dose 17-18 mg/kg/day. In neither strain, at neither dose experiment. Animals were maintained in polycarbonate did significant effects emerge [12]. No changes in 5-HT, shoebox cages (46 × 24 × 20 cm) on a 14 h light/10 h 5-HIAA, or corticosterone after the FST were seen. dark cycle (lights on at 600 h) with food (NIH-07 diet) Hence, the evidence is divided as to whether SSRI and water available ad lib in a vivarium maintained at exposure during adolescent brain development has 21 ± 1°C with 50 ± 10% humidity and accredited by the immediate and/or untoward lasting effects. The purpose Association for the Accreditation and Assessment of Laboratory Animal Care (AAALAC). The research of the present experiment was to test whether Vorhees et al. Behavioral and Brain Functions 2011, 7:41 Page 3 of 13 http://www.behavioralandbrainfunctions.com/content/7/1/41 protocol was approved by the Cincinnati Children ’s and Group B had 20 litters resulting N/group for each Research Foundation’s Institutional Animal Care and test ranging from 19-21/test for Group A and 16-20/test Use Committee. Offspring were separated from their for Group B. The tests were (in order): EPM, ASR/PPI, dams on P28 and housed 2 per cage for the remainder and the Porsolt FST (immediately followed by decapita- of the experiment. Each cage contained a semicircular tion and blood collection for determination of plasma stainless steel enclosure (17.8 cm long with the open corticosterone). There were 2 days of separation side 18.4 cm across and 10.2 cm high with the open between each test. Group A was tested in the EPM on side down made of 16-gauge stainless steel) as an addi- P57, ASR/PPI on P59, and FST on P61-62. Group B was tional environmental enrichment. tested in the EPM on P122, ASR/PPI on P124, and FST on P126-127. For Group A, all tests while on-drug were Drugs and doses given not less than 1 h and not more than 4 h after Animals were weighed daily during treatment and weekly each daily dose. thereafter. All animals received a daily gavage from P33-62 of 5 ml/kg body weight of USP grade water alone or water Elevated plus maze containing one of the drugs. Drugs were obtained as This test was conducted on a single day. The apparatus hydrochloride salts and all doses are expressed as the salt was constructed of high-density, black polyethylene form. The treatment groups were as follows: Control polymer. Each arm was 10 × 50 cm, with two opposing (water), Flu3 (Fluoxetine 3 mg/kg), Flu10 (fluoxetine 10 open arms and two perpendicular opposing arms with mg/kg), Par 3 (paroxetine 3 mg/kg), Par10 (paroxetine 10 high side walls (closed arms). The walls were 50 cm mg/kg), and Par 17 (paroxetine 17.04; this dose matched high in the closed arms and the open arms had 0.6 cm the dose of paroxetine reported previously [11] as 15 mg/ curbs on the edges to prevent falls. The apparatus was kg expressed as the freebase). Fluoxetine HCl was obtained mounted 50 cm above the floor. During testing the from Eli Lilly and Company (Indianapolis, IN) and paroxe- room was lit by a single dimmed halogen lamp (aimed tine HCl was obtained from Suzhou ChonTech Pharma- upward to reflect off the ceiling). Each animal was tested Chem Technology Company (Suzhou, China). for 5 min and movements were recorded on a DVD recorder using an overhead camera. Behavior was later Drug stability and potency played back and scored by an observer blind to treat- Drugs were mixed weekly in 5 ml of water in the appro- ment group for time-in-open, number of open entries, priate concentrations to deliver the specified doses. In latency to first open entry, and head dips. An arm was order to verify drug concentrations for each solution, a scored as entered if the animal’s head and two front feet crossedaboundarylinebetweenthecenterregionand set of drug solutions at each concentration was made and sent to ABC Laboratories (Columbia, MO) for veri- that arm. fication. In order to ensure stability of these solutions a set of the highest concentration solution of each drug Acoustic startle/PPI was made and analyzed at 0, 3, or 7 days post-mixing by This test was conducted on a single day. Acoustic startle ABC Laboratories. Finally, drug potency was assayed on reactivity with reflex modified inhibition by prepulse sti- both drug stocks at the end of the experiment to ensure mulation (ASR-PPI) was measured in four identical SR that no degradation had occurred by sending an aliquot Lab test chambers (San Diego Instruments, San Diego, of each drug to ABC Laboratories for analysis. CA). Each test chamber was calibrated using the manu- Drug concentrations were analyzed via high perfor- facturer’s guidelines and sensitivity was regularly mance liquid chromatography with ultraviolet detection checked using a calibrated oscillation device to ensure (HPLC-UV) at 227 nm. Samples were injected onto a consistent sensitive readings. At the start of each test, Zorbas SB-Phenyl reversed phase column (250 × 4.6 animals were placed in acrylic cylindrical tubes mounted mm, 5 μm). The HPLC system was an Agilent HP 1100. atop piezoelectric force transducers positioned inside The mobile phase was 50% acetonitrile, 50% 0.1% TFA sound attenuating test chambers. Background white aqueous. The flow rate was 1.0 ml/min, attenuation was noise was set at 70 dB. The test paradigm reported pre- 1000 mAU at ambient temperature with a column tem- viously was duplicated as closely as possible [11]. Each perature of 30°C. Sample size was 20 μland runtime test session consisted of a 5 min acclimation period fol- was 10 min. Final drug concentrations were calculated lowed by a 5 × 5 Latin square sequence of trials of 5 from a standard curve in the linear range. different types: no stimulus, startle signal with no pre- pulse, startle signal with prepulse 3 dB above back- Behavioral testing and sample collection ground (73 dB), startle signal with prepulse 5 dB above Group A and B animals were tested in the same beha- background (75 dB), and startle signal with prepulse 10 vioral tasks but at different ages. Group A had 21 litters dB above background (80 dB). Each animal received Vorhees et al. Behavioral and Brain Functions 2011, 7:41 Page 4 of 13 http://www.behavioralandbrainfunctions.com/content/7/1/41 each trial type once in each of the 5 orders, and the using a commercially available EIA kit (IDS, Fountain entire Latin square sequence was repeated; hence, each Hills, AZ). All determinations were made from kits ori- animal received each trial type/order twice. Trials of the ginating from the same lot. same type were averaged together for analysis. The intertrial intervals were 10-20 s with randomized spa- Statistical analyses cing. The interstimulus interval was 100 ms (measured Data for Group A and Group B animals were analyzed from prepulse onset to startle signal onset). The startle separately. Data were analyzed using completely rando- signal was a 120 dB mixed frequency, white noise burst mized block analyses of variance (ANOVA) mixed linear that lasted for 20 ms. The recording window was 100 models (SAS Proc Mixed, SAS Institute, Cary, NC). The ms. Prepulses lasted for 20 ms. Stimulus intensity was Kenward-Roger adjusted degrees of freedom method measured using a Quest sound level meter (SPL scale) was used and can be fractional. In these analyses, litter with the meter placed in the test chamber in the center was the block factor. Data from the EPM and FST had of the test stage with the door closed and the micro- 1-between-subject factor (treatment group). ASR-PPI phone directed upward toward the ceiling-mounted had 1-between-subject factor (treatment group) and 1- speaker. Response amplitude (V = maximum voltage within-subject factor (trial). For body weight, the max change within the recording window) was recorded in ANOVA has 1-between-subject factor (treatment group) units of voltage change (mV). Test chambers were and 1-within-subject factor (age). Interactions were ana- cleaned with 70% ethanol between animals. lyzed using the slice-effect ANOVA method on each level of within-subject factors. A posteriori group com- Porsolt forced swim test parisons were analyzed using the Hochberg step-up Testing was conducted over two successive days using method to control for multiple comparisons. As follow- the original method of Porsolt for rats [17,18]. The up we report effect size (ES) for drug treatment for each apparatus consisted of four clear acrylic cylinders 19 cm behavior. There are no methods for calculating ES for in diameter (i.d.) and 60 cm in height filled to a depth mixed linear ANOVA, therefore, we reanalyzed the data of 30 cm (to avoid water depth concerns [19]) with using general linear model ANOVA (GLM) again with room temperature water (21-23°C). On day-1 rats were litter as a random block factor. SAS GLM provides placed in the cylinders for 15 min and on day-2 for 5 three indices of ES (noncentrality, partial, and semipar- min. Each chamber was separated by black acrylic parti- tial correlation ratios). These reanalyses produced results tions to prevent animals from seeing one another. nearly identical to those obtained from Mixed model Movement was recorded using a video camera and DVD analyses, thereby validating the use of ES values from GLM ANOVA models. We then took the semipartial recorder. Scoring was performed by an experimenter blind with respect to treatment group. Immobility dura- correlation ratios (eta-square) and converted them to tion was scored on day-2 as time spent not swimming Cohen’s f (similar to Cohen’s d, except for ANOVAs or reaching at the walls. Occasional, minor limb move- rather than for t-a test) and followed Cohen’s categori- ments were permitted in order for the animal to main- zation scheme (small ES are around 0.2, medium around tain balance, brace itself against the wall, or tread water 0.5, and large ≥ 0.8 [22]. Significance was taken as p ≤ to keep afloat so long as a coordinated multi-limb swim- 0.05 and data are presented as least square mean ± ming motion was not present. We did not use the SEM. expanded FST method [20,21] because we were not test- ing SSRI efficacy, but rather for toxicity. Results Drug Concentrations and Stability Tissue collection Drug solutions of both compounds were prepared and Upon the completion of the FST, animals were removed sent to ABC Laboratories for analytical chemistry to from the water, carried to another room where they determine drug concentrations by HPLC analysis (Meth- were decapitated (< 30 s from removal from FST), and ods; M-1736-000 and USP 30 Paroxetine HCl). Each blood collected for later plasma corticosterone sample was analyzed in triplicate. The first set of sam- determinations. ples was taken from prepared test solutions for concen- tration verification. Mean concentrations of the test Corticosterone assay solutions were within 3% of target concentrations. Blood was collected in tubes that contained EDTA (2% Stability Assessment: Fluoxetine hydrochloride is in 0.05 ml). Samples were centrifuged for 15 min at 4°C, known to be stable in solution ≥ 7 days but to deter- plasma removed and stored at -80°C until assayed. Cor- mine the stability of paroxetine we tested it at the low ticosterone concentrations were determined by diluting and high concentrations at three time points: immedi- plasma 1:10 in assay buffer and assayed in duplicate ately after preparation and at 3 and 7 days. The Vorhees et al. Behavioral and Brain Functions 2011, 7:41 Page 5 of 13 http://www.behavioralandbrainfunctions.com/content/7/1/41 paroxetine solutions showed no decrease in active drug demonstrating time-dependent growth. The group × content over the 7-day period. week effect was not significant (F(20,960) = 0.50, p > End of Study Drug Purity: After the completion of the 0.96). experiment, crystalline samples of both drugs were sent Body weights during drug treatment (P33-62) were to ABC Laboratories to determine drug purity. Both analyzed with Groups A and B combined. There was a paroxetine HCl and fluoxetine HCl were unchanged significant effect of treatment (F(5,227) = 4.84, p = (mean percent potency of 101%). 0.0003), day (F(29,6922) = 1481.86, p < 0.0001), and treatment × day (F(145,6931) = 1.49, p = 0.0001). Slice- effect ANOVAs on each day showed no treatment General Characteristics There were no deaths; 100% of the animals enrolled in effects on P33-43, however significant effects were the experiment survived and were tested. Group sizes obtained on days P44-62. Since there were no significant varied slightly from the design because of experimenter group differences before P44, only the P44-62 body error or equipment malfunction. The exact N for each weight data are illustrated in Figure 1. Body weights on test is provided in the figure captions. the last day of treatment (P62) were analyzed separately Body weight data were analyzed by age. Once off- for Groups A and B and are shown in Figure 2. As can spring were assigned to treatment groups, their body be seen, for Group A, the Flu10 and Par 17 groups weights from P1-28 were analyzed to ensure that no weighed less than Control. For Group B, only the Flu10 preexisting differences occurred. This analysis showed group weighed less than Control. The Par17 for Group no differences as a function of later assignment to the B was not significantly different from Control (p = 0.12). groups (F(5,200) = 1.09, p > 0.36). The effect of age was Group A animals completed testing by P62, hence significant (F(4,960) = 23327.9, p < 0.0001), there are no body weight data for this group after P62. Figure 1 Body weights during treatment: Data represent least square mean (± SEM) body weights (g) on days on which significant group differences were obtained. Treatment was from P33-62; significant group differences occurred on P44-62. Groups A and B are combined. Group sizes: Group A = 21, Group B = 20. Vorhees et al. Behavioral and Brain Functions 2011, 7:41 Page 6 of 13 http://www.behavioralandbrainfunctions.com/content/7/1/41 remained on P70, P77, or thereafter (i.e., between P84- 126 (not shown)). Elevated plus maze For Group A there were no significant differences in time spent in the open arms, number of open arm entries, or latency to first open arm entry (Figure 4A-C, respectively). A significant treatment group effect was found on number of head dips (F(5,95) = 2.74, p < 0.03). Post hoc group comparisons showed that the Par17 group had fewer head dips than Controls (Figure 4D). No other group comparisons were significant. ES for time in open arms were = 0.15 (small), for arm Figure 2 Body weights: Least square Mean (± SEM) body entries = 0.13 (small), for latency = 0.12 (small), and for weights on the last day of treatment (P62). Group A and B are head dips = 0.27 (small). shown separately. There were 21 litters in Group A and 20 litters in For Group B there were no significant differences in Group B. **p < 0.01 vs. Control. time spent in the open arms, number of open arm entries, latency to first open arm entry, or head dips For Group B, body weights were recorded on P63 and (Figure 4E-H, respectively). ES for time in open arms = weekly thereafter until the end of testing on P126. In 0.18 (small), for arm entries = 0.22 (small), for latency = order to map body weight recovery after the end of 0.16 (small), and for head dips = 0.09 (small). treatment, data for the first two weeks after the end of treatment (P63-77) were analyzed separately. There was Acoustic Startle/PPI no significant treatment main effect (F(5,95) = 1.94, p < ASR-PPI data were analyzed two ways: (a) by prepulse 0.10). The day/week factor in the analysis was significant intensity, and (b) using the 0 prepulse response ampli- (F(2,228) = 4256.37 = p < 0.0001), indicating time- tude as a covariate by ANCOVA in order to assess PPI dependent growth, and there was a significant treatment after controlling for any possible differences in basal × week interaction (F(10,228) = 9.59, p < 0.0001). Slice- startle reactivity. effect ANOVAs on each week showed differences on For the Group A startle response ANOVA, the treat- P63 (p < 0.01) but not on P70 or P77, indicating that ment group main effect was not significant but showed recovery was rapid (< 1 week). This pattern is illustrated a trend (F(5,98.2) = 1.95, p < 0.10). There was no signifi- in Figure 3. As can be seen, only the Flu10 group cant treatment × prepulse interaction. Prepulse intensity weighed significantly less on P63 and no differences was significant (F(3,354) = 100.32, p < 0.0001) showing that prepulse inhibition of ASR was obtained. In the ANCOVA analysis, no significant treatment main effect was seen (F(5,98) = 1.86, p = 0.11). The treatment × PPI interaction was not significant. The prepulse main effect was significant (F(2,236) = 68.01, p < 0.0001). The data are shown in Figure 5(top panel).ESfor Group A = 0.27 (small). An inspection of Figure 5 taken together with the treatment main effect trend (p < 0.10) led us to con- duct two follow-up analyses. In both analyses we com- bined the two Flu groups ’ data and the three Par groups’ data into single Flu and Par groups. In the first follow-up analysis, all prepulse trials were included in a treatment × prepulse ANOVA. The treatment main effect (F(2,39) = 7.13, p < 0.003) and prepulse main effect (F(3,177) = 88.43, p < 0.0001) were both signifi- Figure 3 Body weights: Least square Mean (± SEM) body cant and the treatment × prepulse interaction showed weights for Group B during the two weeks after the end of treatment. Significant group differences occurred on P63; only the a trend (F(6,177) = 1.89, p < 0.09). Post hoc group Flu10 group weighed significantly less than Controls. From P70 to comparisons (averaged across prepulse intensities) the end of the experiment, there were no significant treatment showed that both the Flu and Par groups differed from group differences. *p < 0.05. Control (mean ± SEM: Control: 149.0 ± 23.5; Flu: Vorhees et al. Behavioral and Brain Functions 2011, 7:41 Page 7 of 13 http://www.behavioralandbrainfunctions.com/content/7/1/41 Figure 4 Elevated plus maze: Data shown are least square means (± SEM) on the EPM during the 5 min test session. A,E: time (s) spent in open arms; B,F: number of open arm entries; C,G: latency to first open arm entry; and D,H: number of open arm head dips. Group A was tested on P57 and Group B on P122. Group sizes for Group A: Number of litters tested = 21; data are based on the following number of subjects/group: Control = 20, Flu3 = 20, Flu10 = 20, Par3 = 20, Par10 = 20, Par17 = 20. One litter’s data were missing because of a DVD playback defect. For Group B: Number of litters tested = 20; data are based on the following number of subjects/group: Control = 16, Flu3 = 16, Flu10 = 16, Par3 = 16, Par10 = 16, Par17 = 16. Data were unrecoverable from 4 litters whose recorded data tracks would not playback from the DVD. Vorhees et al. Behavioral and Brain Functions 2011, 7:41 Page 8 of 13 http://www.behavioralandbrainfunctions.com/content/7/1/41 Figure 5 Acoustic startle/PPI: Least square Mean (± SEM) startle amplitude (V ) measured in units of voltage change (mV). Group A max was tested on P59 and Group B on P124. No treatment group effects were obtained for either Group A or B, nor any treatment × prepulse interactions. Prepulse was significant and showed that the greater the prepulse intensity the greater the inhibition of the startle response. Number of litters in Group A = 21. Progeny group sizes for Group A: Control = 20, Flu3 = 21, Flu10 = 21, Par3 = 20, Par10 = 21, Par17 = 21. For Group A, 1 Control and 1 Par3 animal had missing data because they were tested with the incorrect startle program. Number of litters for Group B = 20. Progeny group sizes for Group B: Control = 20, Flu3 = 20, Flu10 = 20, Par3 = 19, Par10 = 20, Par17 = 20. For Group B, 1 Par3 animal’s data were missing because the test chamber power was not activated. 224.4 ± 23.0; Par: 252.7 ± 23.0). This facilitated startle For Group B the treatment × prepulse ANOVA effect was most pronounced on the no-prepulse trials, showed no significant main effect of treatment group or therefore, the second follow-up ANOVA was with the treatment × prepulse intensity interactions. Prepulse combined dose groups but only on the no-prepulse intensity was significant (F(3,339) = 92.60, p < 0.0001) trials. ES for the latter = 0.44 (medium). With no pre- showing that ASR was significantly modified by the PPI pulses in the analysis, the ANOVA was a simple one- procedure (Figure 5, bottom panel). The ANCOVA way analysis with 3 groups. The treatment effect was analysis showed a similar outcome, no significant treat- significant (F(2,39) = 6.55, p < 0.004). This effect is ment main effect or treatment × prepulse interaction. shown in Figure 6(left panel). Post hoc group compar- The prepulse main effect was significant (F(2,226) = isons revealed that both drug groups showed signifi- 48.20, p < 0.0001). ES = 0.18 (small). cantly increased startle amplitude compared with the As for Group A, follow-up analyses with dose-levels of Control group. ES for the latter = 0.42 (medium). each group combined were conducted for Group B, and Vorhees et al. Behavioral and Brain Functions 2011, 7:41 Page 9 of 13 http://www.behavioralandbrainfunctions.com/content/7/1/41 Figure 6 Acoustic startle for the no-prepulse trials: Data are group least square mean (± SEM) startle amplitude (V ) max measured in units of voltage change (mV). Group A was tested on P59 and Group B on P124 as for Figure 5. Group sizes are as in Figure 5. *p < 0.05, **p < 0.01 compared with Control. there was no significant treatment main effect or treat- ment × prepulse interaction. ES = 0.06 (small). Prepulse was significant (F3,171) = 72.39, p < 0.0001). An addi- tional analysis performed on the no-prepulse trials by Figure 7 Porsolt forced swim test: Least square Mean (± SEM) one-way ANOVA showed no significant treatment effect immobility time on the test trial (day-2; out of 300 s). Animals (Figure 6, right panel). ES = 0.06 (small). received a 15 min training trial 24 h previously. Group A was tested on P61-62 and Group B on P126-127. Group sizes: Number of litters Porsolt FST tested in Group A = 21; data are based on the following number of subjects/group: Control = 19, Flu3 = 21, Flu10 = 20, Par3 = 21, For Group A, there were no significant treatment group Par10 = 21, Par17 = 21. Two Control and one Flu10 animal had effects found on immobility time (Figure 7, top panel) missing data tracks on the DVD because of playback issues. Number or latency to immobility (not shown) in the FST. ES = of litters tested in Group B = 20; data are based on the following 0.20 (small). number of subjects/group: Control = 19, Flu3 = 19, Flu10 = 20, Par3 Similarly, for Group B there were no significant effects = 20, Par10 = 20, Par17 = 20. Tracks for two Control and one Flu3 animal would not playback on the DVD. of treatment group found on immobility time (Figure 7, bottom panel) or latency to immobility (not shown) in the FST. ES = 0.21 (small). disorder, social anxiety disorder, and post-traumatic Corticosterone stress disorder [1-3]. These drugs have been found to be For Group A, all groups showed high corticosterone safer and have fewer side-effects than tricyclic antide- levels at the end of the 5 min FST compared with typi- pressants. SSRIs have also been used in children and cal basal levels (~50 ng/ml) however there were no adolescents, and fluoxetine is approved for use in those treatment differences among the groups (Figure 8, left 6-18 years of age. In children whose nervous system is panel). still developing, there are concerns about the long-term For Group B, again all groups showed increased corti- effects of these drugs. costerone levels in response to the stress at the end of In a recent preclinical experiment examining paroxe- the FST but there were no significant effects of treat- tine and fluvoxamine, it was reported that these drugs ment obtained (Figure 8, right panel). reduced time-in-open in an EPM [11]. Decreased time in open in the EPM is the most widely accepted index Discussion for this test of increased anxiety-like behavior [23]. In Fluoxetine, paroxetine, and many other SSRIs have pro- the experiment by deJong et al. [11] this change ven to be efficacious for the treatment of adults with occurred for both drugs 20-30 days after the end of depression, premenstrual syndrome, obsessive compul- treatment. No change was seen in the number of zone sive behavior, panic disorder, generalized anxiety crossings. These authors also tested the animals in an Vorhees et al. Behavioral and Brain Functions 2011, 7:41 Page 10 of 13 http://www.behavioralandbrainfunctions.com/content/7/1/41 Figure 8 Corticosterone: Mean (± SEM) plasma corticosterone levels obtained immediately after the end of the FST.Nodifferences were observed for either drug compared with Control. Group sizes: Number of litters tested in Group A = 21; data are based on the following number of subjects/group: Control = 21, Flu3 = 21, Flu10 = 21, Par3 = 21, Par10 = 21, Par17 = 21. Number of litters tested in Group B = 20; data are based on the following number of subjects/group: Control = 19, Flu3 = 19, Flu10 = 19, Par3 = 19, Par10 = 19, Par17 = 19. elevated T-maze (the EPM was used with one arm Behaviorally, no treatment-related effects on the EPM blocked). Out of 5 trials for time to enter an open arm were obtained on the principal measure of anxiety, i.e., time-in-open. In addition, no differences were found for after being placed in a closed arm, both SSRI-treated groups entered an open arm faster on trial-2 compared number of zone crossings or latency to first open arm with controls but not on trial-1 or 3-5. When the situa- entry. The Group A Par17 animals showed a significant tion was reversed and they were given an additional trial reduction in head-dips, an effect seen in no other group, by being placed in an open arm and timed for entry however this effect was not seen in Group B when into a closed arm, no differences were found. No differ- tested 60 days after drug cessation. deJong et al. [11] did ences in ASR/PPI were obtained nor were changes not report head-dips, therefore, it is not possible to obtained on immobility time in the FST. make a direct comparison on this variable, however, a We sought to further test the effects of adolescent reductioninhead-dips is generally interpreted as an exposure to SSRIs and followed many of the design fea- increase in anxiety [24], although it is an index of anxi- tures used previously [11], except using more dose levels ety that is secondary to time-in-open and therefore pro- of each drug and including separate groups of animals, vides less persuasive evidence of a significant change in half tested while on-drug (Group A) and half tested off- anxiety. However, this effect would be consistent with drug (Group B; 60 days post-treatment). If long-term the finding of deJong et al. [11]. A recent study showed effects of SSRI treatment were obtained, it would poten- that rats treated from P25-49 with 12 mg/kg of fluoxe- tially represent a concern for the safety of the drugs. tine had reduced time-in-open in the EPM [25] 7 days Both drugs produced body weight reductions at the post-treatment, a finding consistent with reduced head- highest doses tested by P44, i.e., after 12 days of treat- dips that we found while animals were on drug. It is ment, and these reductions remained through the end unclear why adolescent exposed rats and mice show of treatment (P62). But the effects showed recovery to greater inconsistency in EPM responses to SSRIs than non-significant differences one week post-treatment. adult rodents. Some of it may be related to the different Vorhees et al. Behavioral and Brain Functions 2011, 7:41 Page 11 of 13 http://www.behavioralandbrainfunctions.com/content/7/1/41 exposure ages as there is no consistent definition of prepulse trials included, i.e., only the unmodified, basic adolescence in rodents. Doses and routes of drug ASR trials. Both analyses resulted in the same finding: administration as well as duration of treatment and end the pooled fluoxetine and pooled paroxetine groups of treat to test interval vary widely across studies, exhibited increased ASR amplitude compared with Con- increasing the difficulty of discerning patterns. trols. By contrast, when Group B ASR-PPI data were Neither we nor deJong et al. (2006) found any effect analyzed the same way, no residual effect or trend toward an effect was observed. This is in contrast to a on the FST test of swimming despair [17,18] and this study that found that P24-46 fluoxetine reduced ASR at finding is different than that reported by Homberg et al. ahigherdosethan weused(12 mg/kg),but similarto [25] but they treated rats earlier (P25-46); moreover, they found no differences in rats treated at a later age. our data, found no effects at later ages [25]. Other studies have found opposite or paradoxical effects The present data should be viewed within the limita- of adolescent exposure to fluoxetine in mice, but only in tions of the study. As in the deJong et al. [11] and Nor- one of the two strains tested [26]. To the extent that the cross et al. [12] studies, we tested only males. It may be FST is a valid preclinical test of depression, the data do worthwhile to test females as many drugs exhibit sexu- not suggest that adolescent exposure to fluoxetine or ally dimorphic responses. It is also worth considering paroxetine result in long-term changes in immobility. that future experiments might test a broader range of This is consistent with a newer report of chronic fluoxe- doses in order to more thoroughly test for possible tine treatment in mice from P14-42 showing no FST adverse effects. Finally, other behavioral tests might be effects [26]and thedatacited above[12].Wescored worth considering; Norcross et al. [12] also included the FST for immobility time as originally described by open-field and cued fear conditioning with extinction, Porsoltetal.[17,18].Wedidnotscore thetestusing but even more tests might be considered such as fear- the recently suggested indices of active swimming and induced acoustic startle facilitation. There are many wall climbing [20]. While we cannot rule-out the possi- tests of anxiety, conditioned fear, sensorimotor gating, bility that these additional indices might have uncovered and depression, of which only a subset were used here other effects, we did not include them because our or in the other studies cited that tested for residual interest was not differentiating among different classes effects. A more extensive battery of tests might reveal of SSRIs, which was the reason these measures were effects not detected herein. introduced, but rather whether immobility was a long- Overall, the data show that for the doses of fluoxetine and paroxetine tested by an oral route during adolescent term consequence of drug exposure after adolescent to early adult brain development (operationally defined treatment. Others have administered SSRIs by osmotic pump in as P33-62), caused minor and only transient reductions order to maintain plasma concentrations in animals in body weight gain, a small but significant ASR facilita- within the same range as human therapeutic concentra- tion during treatment that did not remain 60 days post- tions [27,28]. The intent of these studies was different treatment, and a high-dose only (Par17) reduction in than ours. These studies were designed to assess the EPM head-dips while on-drug but not off-drug. The molecular targets of SSRIs. For this purpose maintaining findings,whenconsideredinterms of thepower to a constant drug concentration is desirable but this goal detect differences in the present experimental design (≥ was not necessary for our purposes of determining if 20 animals per group), multiple dose levels, use of a these drugs have persistent long-term effects long after within-litter design that improves subject matching, and cessation. inclusion of both on- and off-drug cohorts, suggest that deJong et al. [11] reported no changes in ASR/PPI. there is no signal of adverse effects present in the data We replicated their parameters and in agreement, that might raise concern over the long-term safety of observed no changes in ASR or PPI from either drug 60 the drugs when treatment is during an interval spanning days post-treatment. However, while on-drug, we saw a adolescent brain development in rats, in agreement with trend toward ASR facilitation that did not interact with previous findings in mice [12]. It was recently reported PPI. To further explore the data, we conducted follow- that fluoxetine given by continuous infusion via mini- up analyses by combining the two fluoxetine groups pumps from P14-42 results in anxiogenic effects in mice into one pooled group by averaging the data for the while on-drug on tests of novelty-induced hypophagia in fluoxetine animals in each litter together to create a sin- Swiss-Webster (SW) and C57BL/6 Charles River (B6) mice, however there were no effects in the EPM in the gle, merged group. We did the same for the paroxetine, SW strain and reduced time-in-open in the B6 strain merging the data from all three dose levels together but no effects on open-field center time or FST in either among littermates to create a single paroxetine group. Two follow-up analyses were performed. These were: (1) strain. These effects disappeared off-drug [26] which is with all prepulse trials included and (2) with no- entirely consistent with the present findings. Thus, while Vorhees et al. Behavioral and Brain Functions 2011, 7:41 Page 12 of 13 http://www.behavioralandbrainfunctions.com/content/7/1/41 inspections at regular intervals to ensure protocol compliance. DLG SSRIs induce relatively reliable effects on-drug in adult conducted the corticosterone assays. MRS was in charge of study blinding rodents, their effects from adolescent exposure are less of animals to ensure that technicians had no knowledge of treatment consistent. Our data are consistent in that we saw only groups, made coded drug solutions, and provided drug solutions to ABC Laboratories blinded for potency and stability assays on drug solutions. MTW small effects while on drug and no long-term effects was Co-Principal Investigator, participated in the design, data analyses, and long after drug discontinuation. The latter provides interpretation of the data, and contributed to the manuscript. All authors some evidence that these drugs are not neurotoxic when have read and approved the final manuscript. given during adolescent stages of brain development at Received: 20 April 2011 Accepted: 5 October 2011 least on the behavioral indices used here. Published: 5 October 2011 Financial competing interests References 1. Tang SW, Helmeste D: Paroxetine. Expert Opin Pharmacother 2008, The research reported herein was supported by a 9:787-794. research contract to Cincinnati Children’s Research 2. Goddard AW, Shekhar A, Whiteman AF, McDougle CJ: Serotoninergic Foundation (CCRF) under CCRF required provisions of mechanisms in the treatment of obsessive-compulsive disorder. Drug Discov Today 2008, 13:325-332. academic freedom from sponsor influence, viz., that work 3. Shah NR, Jones JB, Aperi J, Shemtov R, Karne A, Borenstein J: Selective performed under the auspices of the CCRF are under the serotonin reuptake inhibitors for premenstrual syndrome and sole discretion of CCRF and the investigators. Under premenstrual dysphoric disorder: a meta-analysis. Obstet Gynecol 2008, 111:1175-1182. these provisions, the principal investigators are free to 4. Hall WD, Lucke J: How have the selective serotonin reuptake inhibitor publish all results as they deem scientifically appropriate. antidepressants affected suicide mortality? Aust N Z J Psychiatry 2006, In addition, the CCRF authors have received no reimbur- 40:941-950. 5. Henry C, motes-Mainard J: SSRIs, suicide and violent behavior: is there a sements, fees, salary, honoraria, or other benefits from need for a better definition of the depressive state? Curr Drug Saf 2006, the sponsor. Dr. Morford is an employee of Eli Lilly and 1:59-62. Company. The article processing charge was paid by 6. Usala T, Clavenna A, Zuddas A, Bonati M: Randomised controlled trials of selective serotonin reuptake inhibitors in treating depression in children CCRF. The CCRF authors declare that they have no and adolescents: a systematic review and meta-analysis. Eur stock or other financial interest in the sponsoring com- Neuropsychopharmacol 2008, 18:62-73. pany and stand to make no financial gain or loss from 7. Reed AL, Anderson JC, Bylund DB, Petty F, El RH, Happe HK: Treatment with escitalopram but not desipramine decreases escape latency times the publication of these data now or in the future. The in a learned helplessness model using juvenile rats. Psychopharmacology CCRF authors hold no patent now or in the future (Berl) 2009, 205:249-259. related to this manuscript. The authors have received no 8. Reed AL, Happe HK, Petty F, Bylund DB: Juvenile rats in the forced-swim test model the human response to antidepressant treatment for reimbursements, fees, funding, or salary from an organi- pediatric depression. Psychopharmacology (Berl) 2008, 197:433-441. zation that holds or has applied for patents relating to 9. Mason SS, Baker KB, Davis KW, Pogorelov VM, Malbari MM, Ritter R, Wray SP, the content of the manuscript. The CCRF authors declare Gerhardt B, Lanthorn TH, Savelieva KV: Differential sensitivity to SSRI and tricyclic antidepressants in juvenile and adult mice of three strains. Eur J no financial conflicts of interest. Pharmacol 2009, 602:306-315. 10. Bylund DB, Reed AL: Childhood and adolescent depression: why do Non-financial competing interests children and adults respond differently to antidepressant drugs? Neurochem Int 2007, 51:246-253. The authors declare no non-financial competing inter- 11. de Jong TR, Snaphaan LJ, Pattij T, Veening JG, Waldinger MD, Cools AR, ests (political, personal, religious, ideological, academic, Olivier B: Effects of chronic treatment with fluvoxamine and paroxetine intellectual, commercial, or any other) in relation to this during adolescence on serotonin-related behavior in adult male rats. Eur Neuropsychopharmacol 2006, 16:39-48. manuscript. 12. Norcross M, Mathur P, Enoch AJ, Karlsson RM, Brigman JL, Cameron HA, Harvey-White J, Holmes A: Effects of adolescent fluoxetine treatment on fear-, anxiety- or stress-related behaviors in C57BL/6J or BALB/cJ mice. Acknowledgements Psychopharmacology (Berl) 2008, 200:413-424. The authors’ thank Mary S. Moran for assistance with the statistical analyses 13. Swerdlow NR, Light GA, Cadenhead KS, Sprock J, Hsieh MH, Braff DL: Startle and to Holly Johnson and Lindsey Burns for drug treatment and behavioral gating deficits in a large cohort of patients with schizophrenia: testing. This project was supported by a research agreement between relationship to medications, symptoms, neurocognition, and level of Cincinnati Children’s Research Foundation and Eli Lilly and Company, function. Arch Gen Psychiatry 2006, 63:1325-1335. Indianapolis, Indiana. The data, analyses, interpretation, and writing of the 14. Hsieh MH, Swerdlow NR, Braff DL: Effects of background and prepulse manuscript were conducted by the authors, with final content decisions characteristics on prepulse inhibition and facilitation: implications for residing with the Cincinnati authors. neuropsychiatric research. Biol Psychiatry 2006, 59:555-559. 15. Kask K, Gulinello M, Backstrom T, Geyer MA: Sundstrom-Poromaa, I. Author details Patients with premenstrual dysphoric disorder have increased startle Division of Neurology, Department of Pediatrics, Cincinnati Children’s response across both cycle phases and lower levels of prepulse Research Foundation and University of Cincinnati College of Medicine, 3333 inhibition during the late luteal phase of the menstrual cycle. Burnet Ave. Cincinnati, OH, 45229, USA. Eli Lilly and Company, Greenfield, Neuropsychopharmacology 2008, 33:2283-2290. IN, 46140 USA. 16. Geyer MA: The family of sensorimotor gating disorders: comorbidities or diagnostic overlaps? Neurotox Res 2006, 10:211-220. Authors’ contributions 17. Porsolt RD, LePichon M, Jalfre M: Depression: A new animal model CVV was Principal Investigator, participated in the design, data analyses, and sensitive to antidepressant treatments. Nature 1977, 226:730-731. interpretation of the data, oversaw the project, and wrote the manuscript. LRM participated in the design and provided on-sight quality assurance Vorhees et al. Behavioral and Brain Functions 2011, 7:41 Page 13 of 13 http://www.behavioralandbrainfunctions.com/content/7/1/41 18. Porsolt RD, Anton G, Blavet N, Jalfre M: Behavioural despair in rats: A new model sensitive to antidepressant treatments. Eur J Pharmacol 1978, 47:379-391. 19. Detke MJ, Lucki I: Detection of serotonergic and noradrenergic antidepressants in the rat forced swimming test: the effects of water depth. Behav Brain Res 1996, 73:43-46. 20. Cryan JF, Valentino RJ, Lucki I: Assessing substrates underlying the behavioral effects of antidepressants using the modified rat forced swimming test. Neurosci Biobehav Rev 2005, 29:547-569. 21. Cryan JF, Page ME, Lucki I: Differential behavioral effects of the antidepressants reboxetine, fluoxetine, and moclobemide in a modified forced swim test following chronic treatment. Psychopharmacology (Berl) 2005, 182:335-344. 22. Cohen J: Statistical Power Analysis for the Behavioral Sciences. Hillsdale, NJ: Lawrence Erlbaum Associates;, 2 1988. 23. Walf AA, Frye CA: The use of the elevated plus maze as an assay of anxiety-related behavior in rodents. Nat Protoc 2007, 2:322-328. 24. Fernandes C, File SE: The influence of open arm ledges and maze experience in the elevated plus-maze. Pharmacol Biochem Behav 1996, 54:31-40. 25. Homberg JR, Olivier JD, Blom T, Arentsen T, van BC, Schipper P, Korte- Bouws G, van LG, Reneman L: Fluoxetine exerts age-dependent effects on behavior and amygdala neuroplasticity in the rat. PLoS ONE 2011, 6: e16646. 26. Oh JE, Zupan B, Gross S, Toth M: Paradoxical anxiogenic response of juvenile mice to fluoxetine. Neuropsychopharmacology 2009, 34:2197-2207. 27. Owens MJ, Knight DL, Nemeroff CB: Paroxetine binding to the rat norepinephrine transporter in vivo. Biol Psychiatry 2000, 47:842-845. 28. McHugh PC, Rogers GR, Glubb DM, Joyce PR, Kennedy MA: Proteomic analysis of rat hippocampus exposed to the antidepressant paroxetine. J Psychopharmacol 2010, 24:1243-1251. doi:10.1186/1744-9081-7-41 Cite this article as: Vorhees et al.: Effects of periadolescent fluoxetine and paroxetine on elevated plus-maze, acoustic startle, and swimming immobility in rats while on and off-drug. Behavioral and Brain Functions 2011 7:41. Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color figure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit

Journal

Behavioral and Brain Functions – Springer Journals

Published: Oct 5, 2011

There are no references for this article.