[1] 24w Sexually dimorphic responses of rats to fluoxetine in the forced swimming test are unrelated to the function of the serotonin transporter in the brain

[1] 127w Foundation ("Equipment Grants") and CNPq (Proc. 472446/2012-6) to CLO. All procedures carried out in the present study complied with the current laws in Brazil. Primary data are available in the Open Science Framework (osf.io/3bdea). Highlights:Fluoxetine may favour the activity of rats in the repeated forced swimming (rFST) Fluoxetine favoured activity in the rFST more in male than in female rats SERT captured serotonin similarly in brains of male and female rats Fluoxetine in vivo failed to inhibit SERT functioning in brains of both sexes of rats Sexually dimorphic responses to fluoxetine in rFST seemed unrelated to SERT function Limitations:Sample sizes were chosen based on a qualitative evaluation of previous literature.There was no power analysis for sample size calculation. Group allocation was not randomisedThere was no treatment concealment.

[1] 114w Discovery of new antidepressants requires improvement of research models for successful translation (Belzung, 2014;Berton, Hahn, & Thase, 2012;Kesselheim, Hwang, & Franklin, 2015). Because Major Depression or other affective disorders are complex, multifactorial disorders with debilitating subjective symptoms, their modelling in silico, in vitro, in animals or non-depressed humans present enormous challenges. According to Willner (1997), an "ideal" model for antidepressant research should recreate in laboratory animals the symptoms (face validity), and the neurobiological disturbances (construct validity) observed in depressive patients. Modelled symptoms should be reversible by clinically relevant treatments (predictive validity) (Willner, 1997). "Real" models for depression or antidepressant research have only partial degrees of the different validities (Belzung, 2014;Berton et al., 2012;Willner, 1997).

[2] 140w Exposure of laboratory animals to uncontrollable, unavoidable and inescapable stresses may result in symptoms treatable with antidepressants (Belzung, 2014;Berton et al., 2012;Willner, 1997). In our case, rats forced to swim in an inescapable tank may display swimming, diving, climbing, or immobile behaviours. Immobility, or "behavioural inhibition", may specifically be preventable by antidepressant treatment (Detke & Lucki, 1995;Porsolt, Bertin, & Jalfre, 1978). Although behavioural tests based on models such as "learned helplessness" or "anhedonia" may have a better face value, their predictive validity seems weaker than "behavioural inhibition" (Willner 1997). Besides the good predictive validity, the forced swimming test (FST) or Porsolt test is considered a fast, simple, inexpensive, and a relatively reproducible way of testing antidepressants' effectiveness (Borsini, Lecci, Sessarego, Frassine, & Meli, 1989;Dal-Zotto, Martı , & Armario, 2000;Detke & Lucki, 1995;Kitamura et al., 2004;Mezadri, Batista, Portes, Marino-Neto, & Lino-

[3] 151w This article is protected by copyright. All rights reserved. de-Oliveira, 2011;Porsolt et al., 1978;Ramos-Hryb, Harris, Aighewi, & Lino-de-Oliveira, 2018;Vieira, De Lima, de Pádua Carobrez, & Lino-de-Oliveira, 2008). The variations of Porsolt test such as the repeated FST in rats (Domingues et al., 2018;Mezadri et al., 2011;Possamai et al., 2015) seems to keep the predictive qualities of the original test while simultaneously detecting the onset of antidepressant effects at different points in time. In humans, the effects of antidepressant treatment depend on whether the treatment is acute or chronic. The repeated FST may then offer more predictive validity than traditional Porsolt test once the treatment model is established (Di Simplicio, Norbury, & Harmer, 2012;Lovick, 2013). Furthermore, repeated FST complies with ethical recommendations in animal research, as it requires a smaller sample size and shorter exposure to stress (Domingues et al., 2018) compared to "learned helplessness" models (Dalla, Edgecomb, Whetstone, and Shors, 2008).

[4] 97w Clinical data suggest that women may be more susceptible to depression and anxiety than men and are the primary users of antidepressants (Kendler & Gardner, 2014;Young & Korszun, 2010). Therefore, female rats may be more appropriate heuristic models for antidepressant research than males. In previous works, data suggest that non-treated male and female rats behaved similarly in the repeated FST (Domingues et al., 2018;Mezadri et al., 2011;Possamai et al., 2015). In the present work, the main aim is to compare the responses of male and female rats to the treatment with an antidepressant in the repeated FST.

[5] 63w Fluoxetine was selected to facilitate comparison with previously published results in male rats (Mezadri et al., 2011;Possamai et al., 2015). Based on data from traditional FST (Gomez, Martinez-Mota, Estrada-Camarena, & Fernandez-Guasti, 2014;Li, Raaby, Sánchez, & Gulinello, 2013;Pic-Taylor et al., 2015), we hypothesise that male and female rats tested in the repeated FST will respond differently to treatment with fluoxetine. The direction of the

[6] 64w This article is protected by copyright. All rights reserved. sexual difference seemed unpredictable since in traditional FST, male rats often present low immobility time after the treatment with fluoxetine (Detke & Lucki, 1995) while female rats may display a decrease (Fernández-Guasti, Olivares-Nazario, Reyes, & Martínez-Mota, 2017), an increase (Ribeiro, Ferigolo, Reis, Barros, & Spritzer, 2000) or no change at all (Barros & Ferigolo, 1998).

[7] 180w Fluoxetine, as well as many different other antidepressants, work to improve mood by increasing the bioavailability of serotonin (5-hydroxytryptamine; 5-HT) (Delgado et al., 1990;Owens & Nemeroff, 1994). The blocking of 5-HT transporters (SERT) is an essential event for the increased availability of 5-HT in the synaptic cleft promoted by fluoxetine and other selective inhibitors of 5-HT reuptake (SSRI). In previous works, the contents of 5-HT and the function of SERT appeared sexually dimorphic in humans (Nishizawa et al., 1997;Sakai et al., 2006) and rats (Dickinson & Curzon, 1986;Linder, Davis, Burnett, & Watts, 2011;Yang, Sampson, Senturk, & Andrews, 2015). These sexually-linked differences in the human serotoninergic system led us to investigate the possibility of sexually dimorphic responses to fluoxetine in rats undergoing repeated FST. Hence, SERT levels were examined in the brains of male and female rats submitted to repeated FST both with and without the treatment with fluoxetine. Due to their known role of helping produce the effects of antidepressants (for review see (Deakin & Graeff, 1991), the frontal cortex and hippocampus were selected as encephalic targets in this study.

[8] 9w This article is protected by copyright. All rights reserved.

[1] 209w All adult female and male rats had healthy appearances throughout the experimental period, and no adverse events were noticed (data not shown). All males and females assigned to the control groups (vehicle-treated rats) performed immobility, swimming, and climbing during the period of the test, as expected (Raw Data are in S1, Means and SEM are in Table S6). Therefore, all rats were kept for further analysis. Except for immobility time, swimming frequency and climbing frequency, all other variables showed the normal distribution for female rats (Kolmogorov-Smirnov one-sample D statistic, p > 0.05). For male rats, behavioural parameters presented normal distribution, except climbing time in retest 2 (Kolmogorov-Smirnov one-sample D statistic, p< 0.05). Over repetition of FST, swimming time increased slightly in male (Friedman ANOVA (N=6, df=2) = 6.3, p=0.04) and female rats (Friedman ANOVA (N=9, df=2) = 2, p=0.4). Climbing time decreased over repeated FST for male (Friedman ANOVA (N=6, df=2)= 6.3, p=0.04) and female rats (Friedman ANOVA (N=9, df=2)= 6.2, p=0.04). Immobility times increased slightly over repetition of FST in control male (Friedman ANOVA (N=6, df=2)=1, p=0.6) or female rats (Friedman ANOVA (N=9, df=2)= 0.4, p=0.8). In summary, the effects of the repetition of FST on swimming, climbing or immobility of male and female naïve rats were small.

[2] 68w In the onset of the experimental period, every female rat was in a random phase of the oestrous cycle (Table S8). Independently of pharmacological treatment, diestrous was the most common oestrous phase found in the afternoon of the test (diestrous= 16, estrous=2, metaestrous=7, proestrus= 9), retest 1 (diestrous= 16, estrous=6, metaestrous=7, proestrus= 5) while in retest 2 other phases were also frequent (diestrous= 11, estrous=6, metaestrous=7, proestrus= 10).

[3] 9w This article is protected by copyright. All rights reserved.

[4] 174w Effects of fluoxetine treatment on immobility time in the repeated FST were small in males and very small in female rats (Figure 2). Although small, reduction of immobility time by treatment with fluoxetine 2.5 mg/kg/day were significant in comparison with vehicle (Mann-Whitney, Test U (N=6)= 1, p=0.04; Retest 2 U (N=6)= 6, p=0.06; U (N=6)= 2, p=0.009). In female rats, no significant effects of pharmacological treatments (fluoxetine 0, 1, 2.5, 5 mg/kg/day) were observed on immobility time in the repeated FST (Kruskal-Wallis ANOVA, test: H (N=34, df=3) = 1.6, p=0.7; retest 1: H (N=34, df=3)= 2.2, p=0.5; retest 2: H (N=34, df=3)= 1.9, p=0.6). No significant differences were observed between male and female rats treated with vehicle or fluoxetine 2.5 mg/kg/day in the test, retest 1 or retest 2 (Kruskal-Wallis ANOVA, H (N=17, df=1) = 0.5, p=0.5; H (N=17, df=1) = 0.3, p=0.6; H (N=17, df=1) = 0.2, p=0.6). In summary, the effects of chronic treatment with fluoxetine on the behaviour of rats in the repeated FST were higher in males than in females.

[5] 74w In naïve rats, i.e. neither tested nor treated, a slightly higher expression of SERT in the hippocampus of females statistically differed from males (Figure 3, Raw Data are in S2; Mann-Whitney, U (N=5) = 0, p= 0.008). In contrast, a higher expression of SERT in the frontal cortex of females compared to male rats produced non-significant results at alpha= 0.05 (Figure 3, Raw Data are in S2; Mann-Whitney, U (N=5) = 3, p= 0.06).

[6] 53w Contents of 5-HT in the frontal cortex and hippocampus of male naïve rats were similar independent of pre-incubations (vehicle or fluoxetine) or incubations (vehicle or 5-HT) (Figure 3; Kruskal-Wallis ANOVA, frontal cortex: H (N=20, df=3)= 2.18, p=0.53; hippocampus: H (N=20, df=3)= 3.34, p=0.34; Raw Data are in S3; Means and SEM are in

[7] 59w This article is protected by copyright. All rights reserved. Table S4). The same pattern was observed in the frontal cortex of female rats (Kruskal-Wallis ANOVA: H (N=20, df=3)= 7.61, p=0.054). In the hippocampus of female rats, contents of 5-HT were similar independent of pre-incubations (vehicle or fluoxetine) or incubations (vehicle or 5-HT) (Kruskal-Wallis ANOVA: H (N=20, df=3)= 2.93, p=0.40).

[8] 63w Pre-incubation with fluoxetine prevented the increase in contents of 5-HT in the brain of male or female rats (Figure 3). In summary, the expression, the uptake function and the response of SERT to fluoxetine differed minimally between brains of naïve male and female rats. Other relevant data, such as 5-HIAA and the ratio [5-HIAA]/[5-HT] are in the supplementary material (Table S3 and S4).

[9] 46w The basal contents of 5-HT or 5-HIAA in the frontal cortex or hippocampus of male and female rats, i.e. without incubation with 1 uM 5-HT, were similar irrespective of the chronic, in vivo treatment with vehicle or fluoxetine (Figure 4; Raw Data are in Table S5;

[10] 48w Means and SEM are in Table S7). The contents of 5-HT or 5-HIAA in the frontal cortex or hippocampus of male and female rats after incubation with 5-HT 1 uM were higher than basal contents independent of the chronic, in vivo treatment with vehicle or fluoxetine (Figure 4).

[11] 58w In male and female rats treated with vehicle, the contents of 5-HT differed statistically depending on pre-incubation or incubation conditions (Kruskal-Wallis ANOVA; males: frontal cortex (H(N=20, df=3)= 11.09, p=0.001; hippocampus: (H(N=20, df=3)= 14.37, p=0.002; females: frontal cortex (H(N=20, df=3)= 13.26, p=0.004; hippocampus: (H(N=20, df=3)= 12.16, p=0.006). The contents of 5-HT in the frontal cortex and hippocampus of male

[12] 145w This article is protected by copyright. All rights reserved. and female rats treated with vehicle increased significantly after their incubation with 1 uM of 5-HT (Figure 4). The contents of 5-HT in the frontal cortex of female rats treated with vehicle in vivo and incubated with 5-HT in vitro were statistically different from the control group (treated with vehicle and incubated with the vehicle; Mann-Whitney U (N=5)= 0, p=0.01; Figure 4). The contents of 5-HIAA increased significantly in the frontal cortex of female rats treated with vehicle (Figure 4 Concerning sex differences, no statistically significant differences in the contents of 5-HT or 5-HIAA were observed between males and females, independent of pre-incubation or incubation. However, ratios [5-HIAA]/[5-HT] of frontal cortex from rats treated in vivo with fluoxetine 2.5 and incubated in vitro with vehicle differed significantly between sexes (Figure 4, Mann-Whitney: (U (N=5)= 2, p=0.03).

[1] 79w The main findings of the present study suggest that behaviour in the repeated FST respond to chronic treatment with fluoxetine in a sexually dimorphic fashion. Chronic treatment with fluoxetine reduced immobility time of male rats in the repeated FST while failing to do so in females. The data, therefore, suggest that no apparent relationship is present between the function of SERT and turnover of 5-HT in the brain with the sexually dimorphic response of rats in the repeated FST.

[2] 66w In control conditions, female and male rats had times of swimming and climbing increased significantly over the repeated FST without significant changes in the immobility time. Results of female rats matched the previous studies reporting stable immobility times over the repetition of swimming or fluctuations in the oestrous cycle (Colom-Lapetina et al., 2017;Domingues et al., 2018). Null results of repeated swimming in male rats were unlike

[3] 9w This article is protected by copyright. All rights reserved.

[4] 53w preceding studies reporting a small, but significant, increase in the immobility time in repeated FST (Mezadri et al., 2011;Possamai et al., 2015). This last result may be explained by the daily treatment with sucrose 10% in the present experiment compared to the intraperitoneal injection previously used (Mezadri et al., 2011;Possamai et al., 2015).

[5] 58w However, the immobility time of male rats in the Porsolt test increased with the intake of sucrose (35% for nine weeks), as in other studies (Lemos et al., 2016). Alternatively, the small effect size (Button et al., 2013) may explain the contrasts between present data in male rats and the literature (Mezadri et al., 2011;Possamai et al., 2015).

[6] 184w As expected (Mezadri et al., 2011;Possamai et al., 2015), male rats responded to fluoxetine treatment at the dose of 2.5 mg/Kg with reduced immobility time in the repeated FST. Females either failed to respond or responded, paradoxically, with higher immobility time to chronic treatment with this dosage (2.5mg/Kg) of fluoxetine as also observed in the pilot study. Absent or paradoxical responses to fluoxetine (Li et al., 2013;Pic-Taylor et al., 2015) or other antidepressants (Ribeiro et al., 2000;Thelen, Sens, Mauch, Pandit, & Pitychoutis, 2016) were previously observed in female rats. Sexual hormones may impact the responses of female rats to fluoxetine (Benmansour, Weaver, Barton, Adeniji, & Frazer, 2012;Craft, Kostick, Rogers, White, & Tsutsui, 2010;de Oliveira et al., 2004;De Vry, Maurel, Schreiber, De Beun, & Jentzsch, 1999;Enríquez-Castillo et al., 2008;Erika Estrada-Camarena, López-Rubalcava, Hernández-Aragón, Mejía-Mauries, & Picazo, 2011;E Estrada-Camarena, Rivera, Berlanga, & Fernández-Guasti, 2008;Flores-Serrano et al., 2013;Gomez et al., 2014;Li et al., 2013;Lifschytz, Shalom, Lerer, & Newman, 2006;McNamara et al., 2013;Mitic, Simic, Djordjevic, Radojcic, & Adzic, 2013;Molina-Hernández & Téllez-Alcántara, 2001, 2011;Pic-Taylor et al., 2015;Récamier-Carballo, Estrada-Camarena, Reyes, & Fernández-Guasti, 2012;Sell et al., 2008;Shah & Frazer, 2014). Indeed, ovariectomized

[7] 70w This article is protected by copyright. All rights reserved. rats, with or without estrogen replacement, presented male-like response to fluoxetine in the FST (Benmansour et al., 2012;Erika Estrada-Camarena et al., 2011;E Estrada-Camarena et al., 2008;Gomez et al., 2014;Molina-Hernández & Téllez-Alcántara, 2001, 2011;Récamier-Carballo et al., 2012;Sell et al., 2008;Shah & Frazer, 2014) while intact females continue to display no response (Pic-Taylor et al., 2015) or paradoxical response (Li et al., 2013).

[8] 78w Although the oestrous cycle phases were not controlled in the present study, all females displaying paradoxical responses to fluoxetine in the repeated FST were in the diestrus phase, characterised by low levels of estrogen and high levels of progesterone in the plasma (Donner & Lowry, 2013). Although no hormone measurements were performed in the present study, we speculate that low levels of gonadal hormones may favour the paradoxical responses of female rats to fluoxetine in the repeated FST.

[9] 127w Besides the gonadal axis, multiple mechanisms may participate in sexually dimorphic responses to antidepressants (Sramek & Cutler, 2011). Because fluoxetine acts as an inhibitor of 5-HT reuptake, differences in SERT function in male and female rats could be a plausible explanation for the sexually dimorphic responses to fluoxetine (Linder et al., 2011;Sakai et al., 2006). However, present data indicated similar expression and function of SERT in the hippocampus of male or female naïve rats. In the frontal cortex, slightly higher expression and function of SERT were seen in female naïve rats as compared to males. Also, in vitro pre-incubation with fluoxetine inhibited equally the uptake of 5-HT in the frontal cortex of male or female naïve rats indicating a sexually independent function for SERT in this region.

[10] 60w Additionally, independent of sex, the uptake of exogenous 5-HT by cortex and hippocampus of rats after repeated FST was about ten times that of naïve rats. Upregulation of function of the SERT after FST was observed in male (Racca et al., 2005) and female rats (present study), explaining the lack of SERT inhibition by in vivo treatment with fluoxetine (present

[11] 35w This article is protected by copyright. All rights reserved. study). In other words, the sexual dimorphisms of SERT seem low or absent in the frontal cortex or hippocampus of rats in control or stressed conditions.

[12] 88w In vivo administration of fluoxetine failed to inhibit the uptake of exogenous 5-HT and the catabolism of 5-HT into 5-HIAA in the hippocampus and frontal cortex of females or male rats submitted to the repeated FST. The basal turnover of 5-HT, i.e. ratio between [5-HIAA] and [5-HT] without the loading of 5-HT, was not affected by the chronic treatment with fluoxetine. Indeed, chronic treatment with fluoxetine increased more than ten times the basal turnover of 5-HT in the hippocampus of male or female rats submitted to the FST.

[13] 59w Nevertheless, the high basal turnover of 5-HT in the frontal cortex of female rats submitted to the repeated FST remained high after the treatment with fluoxetine. Overall, basal turnover of 5-HT in the frontal cortex of rats submitted to the FST did not vary in a sexually dimorphic fashion to the chronic treatment with fluoxetine as the behaviour did.

[14] 131w In summary, Wistar rats responded to fluoxetine in a sexually dimorphic fashion in the repeated FST despite the similarities in the function of SERT in their hippocampus and frontal cortex. Therefore, differences in the turnover of 5-HT in the brain of males and females may not explain the sexually dimorphic response of rats to fluoxetine in the repeated FST. The reasons for the sexually dimorphic response of rats to fluoxetine remain unknown, but other mechanisms involved in antidepressant action may play a role. For instance, neurogenesis (Hodes, Yang, Van Kooy, Santollo, & Shors, 2009;Kuipers, Trentani, van der Zee, & den Boer, 2013) and 5-HT 1A expression and function (Goel, Innala, & Viau, 2014;Pitychoutis, Dalla, Sideris, Tsonis, & Papadopoulou-Daifoti, 2012) may display sexual dimorphism and are currently under investigation in our laboratory.

[1] 284w Animals, animal welfare and ethical statements: Male (n= 12) and female (n= 34) Wistar rats (RGD Cat# 13508588, RRID: RGD_13508588) arrived in the laboratory at postnatal day (PND) 21 (post-weaning) from the central vivarium of the Federal University of Santa Catarina. Groups of 4-5 female or male rats were housed in separated rooms (Biosafety level 1) in white boxes (measuring 17 x 32 x 39 cm) with sawdust bedding and a grid lid. Food (Nuvital®, Brazil) and water were available ad libitum. Both rooms were kept in an inverted light/dark cycle (lights on at 19:00 h.) at room temperature (21°C ± 2). Cages were cleaned three times a week by the experimenters. Experimental procedures initiated when rats reached PND 35 weighing between 200 -250g (females) and 250 -300g (males). For euthanasia rats were anaesthetised with urethane (1.75 g/Kg) immediately after retest 2 and beheaded for dissection of cerebral structures. The following measures were taken to minimize the animals' distress: (1) all procedures occurred during the dark phase of light cycle; (2) after every forced swim session, rats were allowed to dry off in a clean cage placed in a warm, dimly lit room before returning to their home cages; (3) each female rat was covered with a cloth to facilitate handling during the collection of vaginal secretion; (4) pharmacological treatment was performed using a method of spontaneous ingestion of a palatable solution (adapted from (Schleimer, Johnston, & Henderson, 2005)); (5) for euthanasia, rats were transferred under anaesthesia to a room distant from the animal house. All procedures were performed according to Brazilian laws, guidelines for animal care and welfare and approved by the Ethics Committee of the Federal University of Santa Catarina (CEUA-UFSC, PP00764).

[2] 9w This article is protected by copyright. All rights reserved.

[3] 13w All behavioural procedures and animal manipulations occurred from 1 pm to 6 pm.

[4] 59w No systematic randomisation was performed for experimental group assignment. The sample sizes chosen for the behavioural experiments were the same used in previous studies with males (Mezadri et al., 2011;Possamai et al., 2015) or females (Domingues et al., 2018;Ribeiro et al., 2000). Experimenters were blind to outcome assessment but not to drug administration. Every experiment was carried out once.

[5] 201w Experiment 1 (Figure 1): Males (n=6/group) and female rats (n=8-9/group, see legends for exact sample size per group) (PND 21) were allowed to adapt to laboratory conditions for two weeks before the seven days of training for drug treatment . After the training period, cages of females and males were assigned to one given treatment (experimental unit, details in section 2.3.1). The oestrous cycle was evaluated daily from PND 43-46 to 57-60 (details section 2.3.2.). All rats were tested in the repeated FST from PND 43-46 to 57-60 (details in section 2.3.3). Due to a large number of rats tested, the experiment was split into three blocks on three different days throughout the week. Every experimental block had at least one subject of each condition allocated in the following order: male vehicle; female vehicle; female fluoxetine 1 mg/Kg/day; male 2.5 mg/Kg/day; female 2.5 mg/Kg/day; female 5 mg/Kg/day. We encountered a pilot study (data not shown) in which male rats responded to Fluoxetine 2.5 mg/Kg/day as expected (Possamai et al., 2015) while females did not, so a dose-response curve was performed only on females in the present study. Immobility time was the primary outcome while swimming and climbing times were secondary outcomes.

[6] 9w This article is protected by copyright. All rights reserved.

[7] 63w Experiment 2: Male (n=5) and female rats (n=5) (PND 21) were allowed to adapt to laboratory conditions for two weeks before the seven days of training for drug treatment (PND 57-60) (details in section 2.3.4). Brain regions were dissected using Western Blot to determine SERT protein activity (details in section 2.3.4). No oestrous cycle evaluation or behavioural testing was performed in these rats.

[8] 200w Experiment 3: Male (n=5) and female rats (n=5) were allowed to adapt to laboratory conditions for two weeks before the seven days of training for drug treatment (PND 57-60 (details in section 2.3.5). Brain regions were dissected to perform an uptake assay (details in section 2.3.5). No oestrous cycle evaluation or behavioural testing was performed in these rats. Contents of 5-HT measured in the HPLC (details in section 2.3.5), contents of 5-HIAA, and the resulting ratios [5-HIAA]/[5-HT] were the experimental outcomes. Experiment 4 (Figure 1): Male (n=5) and female rats (n=5) (PND 21) were allowed to habituate to laboratory conditions for two weeks before the seven days of training for drug treatment . After the training period, cages containing groups of females or males were assigned to one treatment (experimental unit, see details in section 2.3.1). No oestrous cycle evaluation was performed in these rats. All rats were tested in the repeated FST from PND 43-46 and 57-60 (see details in section 2.3.3). Euthanasia occurred immediately after Retest 2 and brain regions were dissected to perform an uptake assay (see details in section 2.3.5). Concentrations of 5-HT, 5-HIAA were primary outcomes while behaviours in the repeated FST were secondary outcomes.

[9] 9w This article is protected by copyright. All rights reserved.

[10] 225w Vehicle (10% sucrose solution, 1.5 mL) or fluoxetine (Sigma-Aldrich Inc., St. Louis, USA dissolved in the vehicle) oral administration was standardised after a pilot study (data not shown, a method adapted from Schleimer et al., 2005). Oral administration was chosen to reduce the putative stress associated with chronic treatment. Briefly, male and female rats trained for seven days (between 1 pm and 2 pm) to learn to drink the vehicle solution spontaneously. The training session consisted of transferring every rat from the home cage to an individual cage (measuring 13 x 17 x 18 cm), allowing exploration for 2 minutes. After that, a needleless syringe containing vehicle was offered, and rats were allowed to drink the solution while the experimenter gently pressed the plunger. At the end of 1 week of training, all rats consumed the whole volume of sucrose solution (2 mL). Therefore, vehicle or the different doses of fluoxetine were offered daily to male and female rats assigned to control or experimental groups, respectively. Treatment occurred from the day of the first test onwards (Figure 1; details in section 2.2). Males received only fluoxetine 0 or 2.5 mg/kg/day/13 days (Mezadri et al., 2011;Possamai et al., 2015) while a dose-response study was performed for female rats (0,1, 2.5, 5 mg/kg/day/13 days). Solutions were freshly prepared and administered daily from 1 pm to 2 pm.

[11] 72w Female rats' home cages were transferred from the animal room to the adjacent experimental room, which was illuminated by red light. Each rat was removed individually from their home cage, placed on the countertop and covered with a cloth. The experimenter lifted the tail of the animal to expose the genitals, and vaginal secretion was collected with a plastic Pasteur pipette containing a drop of saline solution (NaCl 0.9%). Vaginal smears were

[12] 109w This article is protected by copyright. All rights reserved. transferred to a slide under optical microscopy for oestrous cycle determination, according to Domingues et al. (2018) (adapted from (Caligioni, 2009)). The majority of nucleated epithelial cells were in the proestrus phase, while the majority of enucleated cornified cells were in the oestrous phase. Similar proportions of nucleated epithelial cells, enucleated cornified cells, and leucocytes were in the metaestrous phase, and the majority of leucocytes indicated the diestrous phase. After the procedures, the rats were returned to their home cages. In the days of behavioural testing, oestrous cycle evaluation occurred before the swimming session (from 1 pm to 6 pm).

[13] 9w This article is protected by copyright. All rights reserved.

[14] 186w Male and female naïve rats (see section 2.2.4, Experiment 2) were anaesthetised with urethane (1.75 g/Kg) and beheaded for dissection of the frontal cortex and hippocampus, which were stored at -80°C until analysis. Frozen samples were homogenised with a pestle in lysis buffer (200 ul RIPA, Thermofisher, Cat # 23225 plus 100 uL protease inhibitor, Sigma Cat # P8340). Protein contents were determined in the supernatant after centrifugation (10 minutes, 4°C, 10,000 rpm) using the BCA method according to the manufacturer description (Thermofisher; Cat # 23225). Addition of beta-mercaptoethanol buffer (30uL/sample) yielded a final concentration 5ug/uL of protein per sample. Before electrophoresis (80 min at 90V), samples were boiled for 5 minutes, homogenised, and pipetted in the wells of an acrylamide gel (5uL/well, 25ug total protein per well). After blotting (1h, 300 mA, 100V in a mixture of 20% methanol, 0.1% SDS, tank buffer), nitrocellulose membranes were incubated in blocking solution (10% BSA in TBS, 4°C, 1h) and then further incubated overnight at 4°C with the primary antibody against SERT (Millipore, ab9322, polyclonal Rabbit, molecular weight 71kDa dilution 1:5000) or β-actin (antibody Santa Cruz Biotechnology, sc-81178

[15] 57w This article is protected by copyright. All rights reserved. monoclonal antibody, dilution 1:500). After washings (5 times for 5 minutes with TBS-T, 50mM Tris-Cl, 150mM NaCl, pH 7.6 adjusted with 1M HCl+0,5% Tween 20), membranes were incubated for 2h at room temperature with the secondary antibody (peroxidaseconjugated horse anti-rabbit IgG, Cell signalling #7076, dilution 1:3000) under agitation.

[16] 61w After washing (5 times for 5 minutes with TBS-T, and then two times with distilled water), membranes were incubated with ECL-Kit (procedures as provided by the manufacturer GE Lifesciences) and read on Chemidoc® MP Imaging System (BioRad). SERT levels in the samples (arbitrary units) were assessed using the chemiluminescence readings of SERT signal normalised to the readings of the β-actin signal.

[17] 63w Male and female rats of Experiment 3 (see section 2.2.3) and 4 (see section 2.2.4) were anaesthetised with urethane (1.75 g/Kg) and beheaded for dissection of the frontal cortex and hippocampus from both hemispheres. Brain tissues were assayed for 5-HT uptake and detection, adapted from Linder et al. (2011). Immediately after dissection, brain structures from both hemispheres were incubated for 30 minutes at

[18] 93w This article is protected by copyright. All rights reserved. liquid chromatography with an electrochemical detector (HPLC-EC). Thawed samples were macerated and incubated in an ultrasound bath for 10 minutes. The supernatant was collected for HPLC analysis after centrifugation at 14,500 r.p.m. at 4°C for 20 minutes. Calibration curves used freshly prepared solutions containing [5-HT] and [5-HIAA] ranging from 10 to 1,000 ng/mL. All samples and calibration curves received dihydroxybenzylamine (DHBA) as an internal standard for the control of electrode sensibility. For more details of the chromatographic conditions, see Dos Santos et al., 2015.

[19] 19w Some dependent variables lacked normality (Kolmogorov-Smirnov one-sample D statistic) and homoscedasticity (Levene's test) indicating the use of non-parametric statistics.

[20] 170w Friedman ANOVA followed by Wilcoxon Matched Pairs Test was performed to assess differences within the repetition of FST. Kruskal-Wallis ANOVA by Ranks followed by a two-tailed Mann-Whitney U test or Chi-squared test was performed to assess differences among pharmacological treatments on behaviour or serotonin measures or oestrous phases, respectively. Two-tailed Mann-Whitney U tests were performed to assess differences between sexes. P-values under alpha 0.05 indicate statistical significance. All statistical analyses were performed with the aid of the software Statistica. Unless otherwise stated, data are expressed as mean + SEM. The mode was used to express the frequency of oestrous phases. The Hedge's g test, unbiased with Confidence Intervals, was used to express the effect size (behavioural inhibition or uptake inhibition) (Effect size calculator created by Jared DeFife, Ph.D., Emory University, 2009; downloaded in June/21/2018 from https://web.cs.dal.ca/~anwar/ds/Excel4.xlsx). No exclusions occurred after data collection, as all rats performed immobility during the test session (Mezadri et al., 2011). The exact sample sizes per group may be found in legends for figures or tables.

[21] 9w This article is protected by copyright. All rights reserved.

[1] 82w In every swimming session, rats were placed in a plastic cylinder (measures: 23 x 52 cm) filled with water (40 cm high, 24°C) in such a way that they were unable to reach the bottom of the tank. In the pretest, rats were allowed to swim for 15 minutes. Every rat submitted to the pretest underwent the 5-minute test swimming session 24 hours later. The test session was then repeated at day seven (retest 1) and fourteen (retest 2) after the pretest.

[2] 146w After every swimming session (pretest, test, retest 1, retest 2), rats were allowed to dry out in a clean cage, placed in a dimly lit, warm room, as described earlier in section 2.1. The assigned oral treatment (fluoxetine or vehicle) was administered once a day, as described earlier in section 2.3.1. The first administration occurred 1h before the test and then repeated daily for the next twelve days. In the days of retest 1 or retest 2, the treatment was performed one hour before the swimming sessions. The following behavioural categories were scored (Domingues et al. 2018): 1-Immobility: rat was floating, with small and occasional movements of forepaws, keeping the head above the water; 2-Swimming: rat was moving the forepaws displacing the body horizontally; 3-Climbing: rat was moving the forepaws vertically, either against the wall or in the centre of the recipient, with or without body