UBC Faculty Research and Publications

Voluntary running influences the efficacy of fluoxetine treatment in a model of postpartum depression Gobinath, Aarthi R.; Richardson, Robin J.; Chow, Carmen; Workman, Joanna L.; Lieblich, Stephanie E.; Barr, Alasdair McMillan; Galea, Liisa A. M. Sep 28, 2017

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-Gobinath_A_et_al_Voluntary_running_influences_fluoxetine_2017.pdf [ 1.27MB ]
JSON: 52383-1.0356559.json
JSON-LD: 52383-1.0356559-ld.json
RDF/XML (Pretty): 52383-1.0356559-rdf.xml
RDF/JSON: 52383-1.0356559-rdf.json
Turtle: 52383-1.0356559-turtle.txt
N-Triples: 52383-1.0356559-rdf-ntriples.txt
Original Record: 52383-1.0356559-source.json
Full Text

Full Text

Exercise and fluoxetine for postpartum depression  1  In press September, 2017 Neuropharmacology 1  2 Voluntary running influences the efficacy of fluoxetine treatment in a 3 model of postpartum depression  4 Aarthi R. Gobinath, B.A.1, Robin J. Richardson, B.A. 2, Carmen Chow, M.Sc.2, Joanna L. Workman, 5 Ph.D.2†, Stephanie E. Lieblich, B.Sc.2,4, Alasdair M. Barr, Ph.D.3, Liisa A.M. Galea, Ph.D1,2,4* 6  7 1Program in Neuroscience, 2Department of Psychology, 3Department of Anesthesiology, Pharmacology, 8 and Therapeutics, and 4 Centre for Brain Health 9 University of British Columbia 10  11 *Corresponding Author: 12 Djavad Mowafaghian Centre for Brain Health 13 University of British Columbia 14 2215 Wesbrook Mall 15 Vancouver, BC, CANADA, V6T1Z3  16 liisa.galea@ubc.ca 17  18 †Present Address:  19 University at Albany, State University of New York 20 Department of Psychology 21 1400 Washington Ave. 22 Albany, NY 12222 23  24  25  26  27 Exercise and fluoxetine for postpartum depression  2  HIGHLIGHTS 28  Postpartum CORT (corticosterone) reduced maternal care and hippocampal neurogenesis 29  Postpartum fluoxetine (FLX) prevented CORT-induced reductions in maternal care 30  Exercise, but not FLX, reduced CORT-induced maternal depressive-like behavior 31  FLX increased neurogenesis in dams only in combination with exercise 32   33 Exercise and fluoxetine for postpartum depression  3  ABSTRACT 34 Postpartum depression affects approximately 15% of mothers. Unfortunately, treatment options 35 for postpartum depression are limited. Pharmacological antidepressants such as fluoxetine (FLX) can be 36 controversial due to inconclusive evidence of efficacy during the postpartum and concerns of neonatal 37 exposure to antidepressants. Alternatively, non-pharmacological antidepressants such as exercise may be 38 less controversial but its efficacy in postpartum depression is unclear. To investigate this, we treated rat 39 dams daily with high levels of corticosterone (CORT; 40 mg/kg), to induce a depressive-like phenotype, 40 or oil (vehicle for CORT) during the postpartum period. Within the oil and CORT conditions, four 41 additional antidepressant conditions were created: 1. FLX (10 mg/kg) + exercise (voluntary access to 42 running wheel); 2. FLX + no exercise; 3. Saline (vehicle for FLX) + exercise; 4. Saline + No exercise. We 43 examined maternal care, depressive-like and anxiety-like behavior, stress reactivity, and hippocampal 44 neurogenesis and dams were categorized as “high running” or “low running.” FLX treatment, alone or 45 with high running, prevented CORT-induced disruptions in maternal care. As expected, CORT increased 46 depressive-like behavior but exercise, regardless of running amount, reduced depressive-like behavior. 47 Intriguingly, FLX, but not CORT, increased anxiety-like behavior, which was not mitigated by concurrent 48 exercise. FLX treatment slightly but significantly facilitated serum CORT recovery after forced swim 49 stress. CORT and FLX alone reduced neurogenesis, while exercise coupled with FLX increased density 50 of doublecortin-expressing cells. High running increased density of doublecortin-expressing cells 51 (immature neurons) in comparison to controls. Collectively, these findings indicate that FLX and exercise 52 reverse different endophenotypes of depression in dams, which has translational implications for 53 surveying treatment options of postpartum depression.  54  55 Keywords: antidepressants; doublecortin; glucocorticoids; postpartum depression; hippocampus; exercise 56  57   58 Exercise and fluoxetine for postpartum depression  4  1. Introduction 59 Physical exercise is beneficial for mood (Rethorst and Trivedi, 2013), but its ability to benefit mood 60 during motherhood is unclear. Indeed, exercise increases neurotrophic factors, neuroplasticity, and 61 hippocampal neurogenesis, which may underlie the numerous benefits of exercise on mental well-being 62 (Voss et al., 2013). However, hippocampal volume, trophic factors, and neurogenesis are reduced during 63 pregnancy and lactation (Leuner et al., 2007; Pawluski and Galea, 2007; Hoekzema et al., 2017). 64 Therefore, it is possible that the capacity for exercise to benefit mood and neuroplasticity is compromised 65 during motherhood. It is imperative to evaluate the efficacy of antidepressant interventions, including 66 exercise, in treating depression specifically in mothers. Approximately 15% of mothers suffer from 67 postpartum depression (PPD; Wisner et al., 2013) with 75% of women with maternal depression 68 experiencing their first ever episode of depression (Fisher et al., 2016). Treatment options for PPD are 69 currently limited, and addressing this limitation is necessary for the health of the mother and child. To this 70 end, the following study compares the antidepressant effects of exercise alone or in conjunction with 71 pharmacological antidepressant fluoxetine (FLX) in a rat model of PPD. 72 Systematic reviews evaluating efficacy of postpartum use of selective serotonin reuptake inhibitors 73 (SSRIs) suggest that these drugs are no more effective than placebo in mothers by 3-6 months postpartum 74 (Yonkers et al., 2008; De Crescenzo et al., 2014). Other antidepressants like ketamine (Xu et al., 2017) 75 may have limited preventative benefits in the postpartum. Nonetheless, approximately 6% of mothers use 76 antidepressants after childbirth (Smolina et al., 2015). SSRIs are problematic during pregnancy and the 77 postpartum because they can cross the placenta and remain active in breast milk, potentially affecting 78 infant development (Weissman et al., 2004). Whether developmental SSRI exposure outweighs the 79 adverse effects of maternal depression on child development remains unclear. For this reason, some 80 mothers hesitate to seek antidepressant treatment for PPD, and some women preemptively disengage from 81 prescription antidepressant use during pregnancy (Prady et al., 2016), risking potential relapse in 82 depression. Alternatively, women may be more compliant in using non-pharmacological antidepressants 83 Exercise and fluoxetine for postpartum depression  5  such as exercise due to fewer side effects and concerns for infant outcome. However, systematic reviews 84 of clinical studies investigating the effects of exercise during pregnancy have been inconclusive regarding 85 whether prenatal exercise or postnatal exercise prevents PPD (Dennis, 2004; Ersek and Brunner Huber, 86 2009; Daley et al., 2015; Hinman et al., 2015). Both exercise and FLX are well-validated antidepressants 87 outside of pregnancy and postpartum periods but may have altered efficacy during this time. Thus, this 88 study aims to comprehensively evaluate how the combination of both treatments in comparison to either 89 exercise or FLX alone affects maternal outcome in terms of maternal care, hypothalamic-pituitary-adrenal 90 (HPA) axis activity, affective behavior, and hippocampal neurogenesis in a rat model of PPD.  91 Our laboratory has developed a model of PPD in which a high dose of corticosterone (CORT; 92 primary glucocorticoid in rats) is administered daily during the postpartum period, modelling index 93 depression episode in the postpartum, which accounts for 40% of cases of maternal depression (Fisher et 94 al., 2016). This CORT-induced model of depression capitalizes on the well-established relationship 95 between stress and depression (Angst et al., 2002) including PPD (Seth et al., 2016). Postpartum CORT 96 treatment induces a PPD-like phenotype by reducing time spent with the offspring, increasing depressive-97 like behaviour, and reducing hippocampal neurogenesis and plasticity (Brummelte et al., 2006; 98 Brummelte and Galea, 2010; Workman et al., 2013). We recently reported that maternal postpartum FLX 99 treatment prevented CORT-induced disruptions in maternal care early in the postpartum but was unable to 100 rescue CORT-induced depressive-like behaviour, flattening of HPA axis activity, or reductions in 101 hippocampal neurogenesis in dams (Workman et al., 2016). To expand upon these findings, we sought to 102 determine whether exercise (voluntary running) would prevent CORT-induced depressive-like 103 endophenotypes in dams or potentially boost the efficacy of FLX in the postpartum. We hypothesized that 104 exercise alone, or in conjunction with FLX, would alleviate CORT-induced disruptions in maternal care, 105 depressive-like behavior, HPA axis activity, and hippocampal neurogenesis in dams.  106 2. Materials and methods 107 2.1. Animals 108 Exercise and fluoxetine for postpartum depression  6  Sixty adult female Sprague-Dawley rats (~2 months old) and fifteen adult male Sprague-Dawley 109 rats (2 months old, Charles River, QC) were given rat chow (Jamieson's Pet Food Distributors Ltd, Delta, 110 BC, Canada) and tap water ad libitum. Rats were individually housed in transparent polyurethane bins (27 111 x 25 x 20 cm) with aspen chip bedding and maintained in a 12 h: 12 h light/dark cycle (lights on at 7:00 112 a.m.). All protocols were in accordance with ethical guidelines set by Canadian Council on Animal Care 113 and were approved by the University of British Columbia Animal Care Committee. The timeline of the 114 experiment is depicted in Figure 1. 115 2.2. Exercise Treatment 116  Upon arrival in the facility, female rats were randomly assigned to either standard housing with 117 no running wheel (“no exercise;” n=30) or with access to running wheels (“exercise;” n=30; Med 118 Associates Inc., VT, USA) and groups were housed in separate colony rooms. Running wheel activity 119 was recorded daily and in total dams were given access to the running wheel for 8 weeks (± 4 days, due to 120 the variability in length of time to conceive). Dams had voluntary access to running wheel during pre-121 conception, pregnancy, and postpartum periods in order to: 1) prevent additional postpartum stress 122 (besides CORT treatment) of introducing running wheel during the postpartum, 2) prevent any stress 123 associated with removal of the wheel at any point, and 3) allow sufficient acclimation to engage in 124 running wheel activity, which can be sensitive to ovarian hormone changes (Park et al., 2016). 125 2.3. Breeding Procedures 126 Breeding was conducted as previously described (Workman et al., 2016). During pregnancy, 127 dams were left undisturbed besides weekly weighing and cage-changing. One day after birth (birth day = 128 postpartum day 0), all litters were culled to 4 males and 4 females.  129 2.4. Drug Preparation and Treatment  130 Exercise and fluoxetine for postpartum depression  7  Dams were randomly assigned to one of four treatment groups within both exercise and non-131 exercise conditions: 1) CORT/FLX (exercise, n=8; non-exercise, n=9); 2) CORT/saline (exercise, n=8; 132 non-exercise, n=8); 3) oil/FLX (exercise, n=7; non-exercise, n=7); 4) oil/saline (exercise, n=7; non-133 exercise, n=6). Beginning on postpartum day 2, dams received two daily injections of either CORT (40 134 mg/kg, s.c.) or sesame oil (1 ml/kg and 10% EtOH, s.c.) and FLX (10 mg/kg, i.p.) or saline (1 ml/kg and 135 10% DMSO) for 24 consecutive days. Dams received both injections in succession between 8:30 A.M. 136 and 10:30 A.M. CORT (Sigma-Aldrich, St. Louis, MO, USA) and FLX (Sequoia Research Products, 137 Pangbourne, UK) were prepared as previously described (Workman et al., 2016). The CORT dose 138 reliably induces a depressive-like phenotype in dams and impairs maternal care (Brummelte et al., 2006; 139 Brummelte and Galea, 2010; Workman et al., 2013). The FLX dose increased brain derived neurotrophic 140 factor, and hippocampal cell proliferation after 21 days of injections in both sexes (Hodes et al., 2010).  141 2.5. Maternal Behavior 142  Maternal observations were conducted twice per day from PD 2 – 8. Observations occurred at 143 least 2 h after injections or after previous observation. Each observation lasted for 10 min and duration of 144 the following behaviors were scored: nursing (including arched back nursing, blanket nursing, and 145 passive nursing), licking, nursing and licking, self-grooming, and off nest. Time spent running during 146 maternal observation was also quantified in the exercise groups. Data was summated across all 147 observations. 148 2.6. Forced Swim Test (FST) 149 Dams were tested in the FST on postpartum day 22 (FST1) and 23 (FST2) to assess 150 antidepressant efficacy as previously described (Workman et al., 2016). Behavioral testing began 151 approximately at 12:30 p.m.  Briefly, dams were placed into clean water for 15 min on postpartum day 22 152 and for 5 min on postpartum day 23. Sessions were scored blinded for percent time spent immobile 153 during the first and across five minutes using BEST Collection Software (Educational Consulting, Inc., 154 Exercise and fluoxetine for postpartum depression  8  Hobe Sound, FL, USA). We scored the first minute as well as the total of five minutes due to the fact that 155 minute by minute changes are seen in the FST with greater differences in immobility observed early in 156 the test session (des Portes et al., 1998; Mezadri et al., 2011); additionally, because blood was collected 157 after FST1, we reasoned that this may affect immobility in all animals (see below).  158 2.7. Blood Collection and Serum CORT Assay  159  Immediately after FST1, a blood sample was collected via the tail vein to measure stress-induced 160 levels of CORT. During recovery, rats were returned to home cages, and another blood sample was 161 collected 2 h later. Rats in the exercise condition were not allowed access to running wheel during these 162 two hours to prevent exercise-induced elevations in CORT,  163 Blood samples were stored overnight at 4ºC, centrifuged at 10,000 g for 15 min and serum 164 collected and stored at -20 ºC until radioimmunoassay. Total CORT (bound and free) was measured in 165 duplicate using the ImmuChem Double Antibody 125I radioimmunoassay Kit (MP Biomedicals, Solon, 166 OH, USA). The antiserum cross-reacts 0.05% with cortisol, and does not cross-react with dexamethasone 167 (<0.01%). Intra-assay coefficient of variation was 5%.  168 2.8. Novelty Suppressed Feeding 169 Rats were tested for anxiety-like behavior in the novelty suppressed feeding paradigm as 170 previously described (Mahmoud et al., 2016) 24 h after FST2. In this test, rats must resolve an anxiogenic 171 conflict of entering the center of arena to access a Froot Loop after 16 h of food deprivation (Bessa et al., 172 2009). Behavioral testing began approximately at 11:00 a.m.  Latency to feed was recorded as an index of 173 anxiety-like behavior. The trial was terminated either after the rat began to eat or after 10 min which ever 174 came first. Home cage food consumption was measured for one h after the test.  175 2.9. Tissue Collection and Immunohistochemistry 176 Exercise and fluoxetine for postpartum depression  9   Twenty-four h after novelty suppressed feeding test, rats were given an overdose of sodium 177 pentobarbital and perfused with 60 ml cold 0.9% saline followed by 120 ml cold 4% paraformaldehyde. 178 Adrenal glands were collected and weighed. Brains were extracted, postfixed overnight at 4ºC, and 179 transferred to 30% sucrose in phosphate buffer at 4ºC. Brains were rapidly frozen with dry ice and 180 sectioned coronally at 40 µm using a freezing microtome (Leica, Richmond Hill, ON, Canada). Sections 181 were stored in series of 10 in antifreeze (ethylene glycol/glycerol; Sigma) at -20ºC until processing.  182  Immunohistochemistry for doublecortin (DCX), an endogenous marker of immature neurons 183 (Brown et al., 2003), was conducted as described (Gobinath et al., 2016). Briefly, between PBS rinses, 184 sections were treated with 0.3% hydrogen peroxide (30 min) and incubated at 4 ºC for 24 h in primary 185 antibody solution (1:1000, goat anti-DCX; Santa Cruz Biotechnology, Santa Cruz, CA, USA). Sections 186 were rinsed and incubated in a secondary antibody solution (1:500, rabbit anti-goat; Vector Laboratories, 187 Burlington, ON, Canada) at 4ºC for 24 h. Then, rinsed sections were incubated in ABC complex (ABC 188 Elite Kit; 1:1000; Vector) for 4 h. Sections were developed using diaminobenzidine in the presence of 189 nickel (DAB Peroxidase Substrate Kit, Vector), and counterstained with cresyl violet.  190 2.10. Microscopy 191 DCX-expressing cells were quantified in 3 dorsal sections (-2.76 mm to -4.68mm below bregma) 192 and 3 ventral sections (-5.52 mm to -6.60 mm below bregma) using the 40x objective using an Olympus 193 CX22LED brightfield microscope. Areas of these sections were quantified using ImageJ (NIH, Bethesda, 194 MD, USA) and used for density calculations (number of cells per mm2). Additionally, 50 DCX-positive 195 cells in dorsal (n=25) and ventral (n=25) dentate gyrus were randomly selected and classified as either 196 proliferative, intermediate, or post-mitotic as previously described (Gobinath et al., 2016).  197 2.11. Data Analyses  198  Litter characteristics were analyzed using independent t-tests.  Data from maternal observations, 199 FST, and novelty suppressed feeding test were analyzed using ANOVA with exercise, CORT, and FLX 200 Exercise and fluoxetine for postpartum depression  10  as between-subjects factors. CORT concentrations after the FST were analyzed using repeated measures 201 ANOVA with time as the within-subjects factor. The density of DCX-expressing cells was analyzed using 202 repeated measures ANOVA with region (dorsal, ventral) as the within-subjects factor. Post hoc 203 comparisons used Newman-Keuls. A priori comparisons were subjected to Bonferroni corrections. Post-204 hoc analyses of running data categorized dams as either high or low runners using a median split.  All 205 data were analyzed using Statistica software (v. 9, StatSoft, Inc., Tulsa, OK, USA). All effects were 206 considered statistically significant if p ≤ 0.05. 207 3. Results 208 3.1. Running was reduced during the final week of gestation and with postpartum FLX   209  Before drug treatment, dams maintained consistent levels of running from pre-conception through 210 the second week of gestation (p’s > 0.45). Dams ran less during the third week of gestation in comparison 211 to previous time points (p’s < 0.001; main effect of week). Due to the high variability in running we 212 categorized dams as either high or low running using a median split on amount of running before drug 213 treatment began (Figure 2A), and subsequent post-hoc analyses were conducted with high and low 214 running as a variable. However, because this resulted in an unbalanced design due to skewed sample sizes 215 within these groups data are presented as exploratory analyses. 216 Running wheel activity increased from the second to the third week of the postpartum period in 217 saline-treated dams (p < 0.001). However, this increase was not observed in FLX-treated dams [p = 0.47; 218 interaction between postpartum week and FLX; F (2, 52) = 8.24; p < 0.001; Figure 2B; Table 1].  219 3.2. Postpartum FLX under sedentary or high running conditions prevented CORT-induced reductions in 220 time spent nursing and increases in time spent off nest 221  Postpartum CORT decreased time spent nursing in comparison to controls (p = 0.04), and 222 concurrent FLX prevented this reduction in CORT-treated dams [p < 0.01; interaction between CORT 223 Exercise and fluoxetine for postpartum depression  11  and FLX; F (1, 52) = 4.132, p = 0.05; Figure 3A]. No other significant or main effects or interactions 224 were present (all p’s > 0.2). Follow-up analyses indicate that postpartum FLX significantly increased time 225 spent nursing in high running dams only, indicating greater efficacy with FLX in high runners [p < 0.001; 226 interaction between running amount/FLX; F (2, 48) = 5.42; p = 0.008; Figure 3C]. 227  A priori comparisons reveal that postpartum CORT/FLX decreased time off nest in comparison to 228 postpartum CORT alone [p < 0.01; interaction between CORT/FLX; F (1, 52) = 2.83, p = 0.098; Figure 229 3B.] Follow-up analyses indicate that postpartum FLX significantly decreased time spent off nest in 230 comparison to saline in high running dams only [p = 0.003; interaction between running amount/FLX; F 231 (2, 48) = 3.43; p = 0.04, Figure 3D]. 232  Neither postpartum CORT nor FLX treatment interacted with time spent running during maternal 233 observations (all p’s > 0.08). There were no significant main or interacting effects of maternal exercise, 234 CORT, and FLX treatment on time spent licking (all p’s > 0.10). No exercise/CORT-treated dams spent 235 significantly more time nursing + licking in comparison to no exercise/oil-treated dams [p=0.05; 236 interaction between exercise and CORT; F (1, 52) = 4.67; p = 0.04]. There were no other significant main 237 or interacting effects of maternal exercise, CORT, or FLX treatment on time spent nursing + licking. 238 Postpartum FLX treatment reduced time spent self-grooming in comparison to saline (main effect of 239 FLX; p = 0.009). There were no other significant main or interacting effects of maternal exercise, CORT, 240 and FLX treatment on time spent self-grooming (all p’s > 0.06). These data are summarized in Table 2. 241 3.3. Exercise reduced immobility while postpartum CORT or FLX increased immobility in the forced 242 swim test 243  Exercise decreased time spent immobile in comparison to no exercise [main effect of exercise, 244 F(1, 49) = 19.97; p < 0.001; Figure 4A]. Postpartum CORT and postpartum FLX increased time spent 245 immobile in comparison to controls [main effect of CORT, F(1, 49) = 10.06; p = 0.003; main effect of 246 FLX; F(1, 49) = 11.53; p = 0.001; Figure 4A]. There were no significant effects of running amount (p’s > 247 Exercise and fluoxetine for postpartum depression  12  0.27). We used body mass as a covariate to account for its influence on immobility, and it did not 248 significantly alter our results with main effects of exercise, CORT and FLX still evident. 249  Across the 5 min session, the main effect of exercise to reduce time spent immobile persisted (p = 250 0.03; Table 3). However, there were no longer any significant effects of either CORT or FLX or any 251 significant interactions (all p’s > 0.16).  252 3.4. Maternal postpartum CORT impaired, while FLX treatment facilitated, CORT recovery after stress in 253 the CORT-treated dams 254  Immediately after FST1, both oil-treated and CORT-treated dams showed similar levels of serum 255 CORT (p = 0.10). However, 2 h after FST1, CORT-treated dams maintained elevated serum CORT levels 256 in comparison to oil-treated dams [p < 0.001; interaction between time and CORT; F (1, 43) = 6.831, p = 257 0.02; Figure 4B]. A priori comparisons revealed that 2 h after forced swim test, serum CORT levels were 258 significantly lower in oil/saline, oil/FLX, and CORT/FLX dams (p’s < 0.001) but not CORT/saline dams 259 [p = 0.13; interaction between time, CORT, and FLX; F(1, 42) = 1.08; p = 0.07; Figure 4B]. No other 260 significant main or interacting effects were present (all p’s > 0.11). There were no significant effects of 261 running amount (p’s > 0.12). 262 3.5. Postpartum FLX treatment increased anxiety-like behaviour in the novelty suppressed feeding test 263  FLX treatment increased latency to feed in novelty suppressed feeding in comparison to saline 264 (main effect of FLX; p = 0.047; Figure 4C). No other main or interaction effects were present (p’s > 265 0.29). FLX increased latency to feed only among high running dams (p < 0.001) but not low (p = 0.9) or 266 no running dams [p = 0.14; interaction between running amount and FLX; F (2, 48) = 4.4; p = 0.018; 267 Figure 4D].   268  FLX-treated dams and CORT-treated dams consumed significantly less food 1 h after testing than 269 saline-treated dams (main effect of FLX; p = 0.01; main effect of CORT; p = 0.004; Table 4). No other 270 Exercise and fluoxetine for postpartum depression  13  main or interaction effects were present (p’s > 0.521). We used food consumption as a covariate for the 271 novelty suppressed feeding data: there was no significant effect of the covariate (p = 0.13) although it did 272 render the effect of postpartum FLX treatment to be a trend (p = 0.07). 273 3.6. Postpartum CORT reduced density of DCX-expressing cells whereas exercise and FLX together 274 increased DCX in exercising dams; high running increased density of DCX-expressing cells in ventral 275 hippocampus  276  FLX increased density of DCX-expressing cells in exercising dams (compared to all other groups 277 p’s < 0.03) but decreased the density of DCX-expressing cells in non-exercising dams [p = 0.007; 278 exercise/FLX interaction; F (1, 52) = 3.91; p = 0.05; Figure 5B].   279  Exercise increased density of DCX-expressing cells in ventral (p<0.001) but not dorsal dentate 280 gyrus [p = 0.14; region/exercise interaction; F (1, 52) = 10.97; p = 0.002; Figure 5C]. In ventral 281 hippocampus, high running increased density of DCX-expressing cells compared to low (p < 0.001) or no 282 running dams [p < 0.001; region/running amount interaction; F (2, 48) = 6.7; p = 0.003; Figure 5D]. 283 There were no significant differences in dorsal hippocampus (p’s > 0.42).  284 Postpartum CORT decreased density of DCX-expressing cells in both dorsal (p < 0.001; Figure 285 5E) and ventral regions (p < 0.001; Figure 5F). The density of DCX-expressing cells was greater in dorsal 286 compared to the ventral dentate gyrus in CORT-treated [p < 0.001] but not oil-treated dams, [p =0.27; 287 interaction between region/CORT; F (1, 52) = 4.52, p = 0.04]. 288 3.7. Maternal postpartum CORT decreased the proportion of post-mitotic DCX-expressing cells  289  Of saline-treated dams, maternal postpartum CORT tended to decrease proportion of post-mitotic 290 DCX-expressing cells in comparison to oil regardless of hippocampal region [p = 0.06; type, CORT, and 291 FLX interaction; F (2, 108) = 3.80, p = 0.03; Figure 5G] with no other significant main or interaction 292 effects (p’s > 0.05). 293 Exercise and fluoxetine for postpartum depression  14  3.8. Amount of running during the postpartum positively correlated with density of DCX-expressing cells 294 in the ventral hippocampus only in control dams 295  Average running wheel activity positively correlated with density of DCX-expressing cells in the 296 ventral hippocampus among oil/saline-treated dams (r = 0.81; p = 0.02; Figure 5H) but not among dams 297 of the other drug treatments (p’s > 0.40). No other dependent variables significantly correlated with time 298 spent running.  299 3.9. Exercise reduced body mass during pregnancy; postpartum CORT reduced body mass by the end of 300 the postpartum 301  During pregnancy, all dams gained body mass (main effect of week; p = 0.001) but exercise dams 302 maintained a lower body mass than non-exercising dams (main effect of exercise; p = 0.001; Table 3). As 303 expected, maternal postpartum CORT reduced body mass of the dams in comparison to oil treatment 304 (main effect of CORT; p < 0.001) but no other significant main or interaction effects were present (p’s > 305 0.08). 306 3.10. Exercise did not affect litter parameters 307   Exercise had no significant effects on litter outcome [number of pups: p = 0.36, litter body mass: 308 p = 0.16; litter sex ratio: p=0.45; Table 5]. There were no significant effects of running amount (p’s > 309 0.35).  310 3.11. Exercise increased relative adrenal mass among oil/saline dams whereas FLX increased relative 311 adrenal mass among non-exercising/oil dams 312  Exercise increased relative adrenal mass in comparison to no exercise among oil/saline treated 313 dams (p < 0.001) but not under any drug/hormone condition (p’s > 0.68). Postpartum CORT reduced 314 relative adrenal mass in comparison to oil regardless of exercise or FLX exposure (p’s < 0.001). Under no 315 Exercise and fluoxetine for postpartum depression  15  exercise/oil conditions, FLX treatment increased relative adrenal mass in comparison to saline [p<0.001; 316 exercise, CORT, and FLX interaction; F (1, 50) = 5.33; p = 0.03].  317 4. Discussion 318   Few studies have investigated antidepressant efficacy in models of maternal depression, raising 319 concerns regarding how to best treat PPD. Postpartum CORT reduced maternal care, increased 320 depressive-like behavior, and reduced hippocampal neurogenesis, which is consistent with our prior 321 results. Intriguingly, voluntary exercise, but not FLX treatment, prevented CORT-induced depressive-like 322 behavior and increased hippocampal neurogenesis. Additionally, FLX alone or the combination of high 323 maternal exercise and postpartum FLX prevented CORT-induced reductions in maternal care. These 324 results suggest that voluntary exercise, particularly with greater running amount, could be used as an 325 effective treatment or adjunct treatment with FLX. One caveat is that FLX and high running also 326 increased anxiety-like behavior which is often an initial side effect of FLX in patients (Messiha, 1993). 327 Collectively, these data suggest that exercise and FLX differentially impact CORT-induced 328 endophenotypes of PPD (summarized in Table 6).   329 4.1. Postpartum FLX prevented CORT-induced disruptions in maternal care under sedentary and high 330 running conditions; exercise did not impact litter parameters.  331  We replicated previous findings that FLX treatment prevented CORT-induced reductions in 332 maternal care in sedentary conditions (Workman et al., 2016). Interestingly, voluntary exercise itself was 333 not sufficient to prevent CORT-induced reductions in maternal care, but rather the combination of high 334 levels of voluntary running and postpartum FLX treatment were sufficient. Possible mechanisms 335 underlying this observation are changes in hippocampal plasticity (discussed below) or alterations in 336 serotonin transmission. High intensity aerobic exercise but not low or moderate levels increased serum 337 serotonin levels in humans (both men and women; Zimmer et al., 2016), and running is associated with 338 increased brain serotonin in rodents (only males studied; Liu, Wu, Liu, Li, & Xie, 2017; Wang, Chen, 339 Exercise and fluoxetine for postpartum depression  16  Zhang, & Ma, 2013). Serotonin transmission can in part regulate maternal behavior (Lerch-Haner et al., 340 2008). Therefore, modification of maternal behavior by FLX under sedentary and high running conditions 341 may be indicative of altered serotonergic tone. Cortical and hippocampal serotonin levels are naturally 342 lower during pregnancy and postpartum (Desan et al., 1988) and exposure to glucocorticoids may 343 exacerbate this reduction (Bambico et al., 2009). Although the exact role of serotonin in maternal 344 behaviour is still under investigation, serotonin turnover is higher in the preoptic area of the 345 hypothalamus, a region crucial for expression of maternal behaviour (Lonstein et al., 2003). Therefore, 346 under high CORT conditions, it is possible that FLX preserves maternal behavior by increasing synaptic 347 serotonin levels and that high running further bolsters serotonin transmission and subsequently the 348 efficacy of FLX. Further research could explore the interaction between stress and serotonin in the 349 postpartum brain, particularly the hippocampus and hypothalamus, to clarify the neural mechanism 350 underlying CORT-induced perturbations to maternal behaviour. Although the experiment did not 351 originally intend to distinguish high and low-running dams, an interesting observation is that neither 352 CORT nor FLX significantly affected maternal care in low-running dams; future studies can attempt to 353 characterize whether low intensity exercise is beneficial for maternal care or is not sufficient as an adjunct 354 therapy to FLX.  355 Strenuous exercise could impair milk production (Treadway and Lederman, 1986) or alter milk 356 nutrition content (Matsuno et al., 1999) in women, thereby impacting infant development. Although our 357 study did not directly measure this, our findings indicate that exercise (regardless of high or low running) 358 during gestation did not significantly affect size or sex ratio of birth litter. Furthermore, continuing to 359 exercise after parturition did not affect pup growth by the time of weaning, suggesting that any effects of 360 exercise on milk content were minimal. These observations are in line with similar findings that maternal 361 exercise has little impact on infant growth in humans (Dewey et al., 1994; Su et al., 2007; Daley et al., 362 2012), lactational competency (Prentice, 1994) or pup growth in rodents (Rosa et al., 2012). Importantly, 363 while exercise itself did not affect postnatal growth, it was not sufficient to prevent the negative effects of 364 Exercise and fluoxetine for postpartum depression  17  maternal postpartum CORT and FLX exposure on pup growth or maternal body mass. Therefore, 365 although exercise itself may not necessitate concern during breastfeeding, its limited efficacy in metabolic 366 measures of dams and offspring may be important to note in cases of maternal depression or maternal use 367 of FLX. 368 4.2. Postpartum CORT treatment impaired serum CORT recovery after stress and FLX treatment 369 improved recovery in the CORT-exposed dams  370  Maternal postpartum FLX facilitated serum CORT recovery after acute stress (FST1) in CORT-371 treated rats. Prior research suggests that antidepressant efficacy may be partly attributed to the ability of 372 antidepressants to normalize HPA axis activity (Ising et al., 2007). The postpartum is characterized by 373 higher basal levels of CORT, attenuated HPA axis activation, and lower CORT-binding-globulin levels 374 (i.e. increased free CORT), and these alterations may contribute to mood disturbances in mothers (Seth et 375 al., 2016).  Although we previously found that FLX had no effect on CORT recovery 90 minutes after 376 stress in CORT-treated dams, the means were in the predicted direction (Workman et al., 2016). In the 377 present study maternal FLX improved CORT recovery 2 h after swim stress, which may be reflective of 378 timing- or condition-specific effects. The present study measured serum CORT recovery after FST1 379 whereas the previous study (Workman et al., 2016) measured CORT recovery after FST2 (the “test” 380 session). Thus, it is possible that FLX facilitated CORT recovery in response to the initial acute stress 381 exposure but not later after the second and shorter stress exposure. Importantly all dams had comparable 382 stress-induced levels of serum CORT regardless of CORT treatment. Together with the stress recovery 383 measure, this indicates that our dose of CORT increased and flattened HPA axis tone to a level relevant to 384 postpartum female rat physiology, thus modelling hypercortsolemia and depression (Stetler and Miller, 385 2011) including PPD (Seth et al., 2016).  386  Previous research demonstrated that exercise increases basal levels of glucocorticoids and adrenal 387 gland mass, particularly the adrenal cortex (Droste et al., 2007). Although the present study observed that 388 exercise increased adrenal mass, exercise did not affect stress-induced or recovery serum CORT levels. 389 Exercise and fluoxetine for postpartum depression  18  However, it is possible that postpartum-induced glucocorticoid changes masked any exercise-induced 390 changes in CORT and that exercise may have not been able to further stimulate HPA axis activity under 391 the conditions of postpartum physiology. As expected, CORT treatment reduced relative adrenal mass as 392 we have previously shown (Workman et al., 2016). This likely reflects exogenous CORT treatment 393 suppressing HPA axis activity and subsequently endogenous CORT output. 394 4.3. Maternal exercise reduced depressive-like behaviour whereas maternal postpartum FLX increased 395 anxiety-like behaviour and depressive-like behaviour  396   Exercise, regardless of CORT or FLX, reduced depressive-like behavior in FST2, consistent with 397 previous findings in male and non-parturient female rodents (Lapmanee et al., 2013; Yau et al., 2014; Han 398 et al., 2015). Our findings are also consistent with clinical findings that exercise is efficacious for PPD 399 (Armstrong and Edwards, 2003; Daley et al., 2015). However, an element of social support also 400 accompanied those clinical exercise interventions, obscuring the degree to which depression relief was 401 attributable to exercise. Findings from the present study indeed point to the role of exercise itself in 402 reducing maternal depressive-like behavior. Notably, another possible explanation is that exercise 403 increased overall cardiovascular and muscular fitness or reduced adipose tissue, rendering exercise dams 404 more adept than sedentary dams in active coping behavior in FST2. Although these factors have not been 405 explicitly studied in forced swim test, voluntary running in rats is beneficial for cardiovascular and 406 skeletal muscle health (Pósa et al., 2015; Mobley et al., 2017) which may ultimately contribute to forced 407 swim test performance. However, using body mass as a covariate still resulted in the main effect of 408 exercise to reduce immobility in the FLX. Indeed, voluntary running was substantially reduced during the 409 postpartum, limiting the likelihood that robust cardiovascular fitness was involved. Moreover, if more 410 running directly increased fitness, then it would then be reasonable to predict that high running dams 411 would be the most fit. However, there were no significant effects of high or low running in the FST. 412 These data collectively suggest that maternal exercise can reduce depressive-like behavior in this test.   413 Exercise and fluoxetine for postpartum depression  19  Maternal FLX treatment increased both depressive-like behaviour and anxiety-like behavior 414 (which was partially explained by reduced food consumption), partially consistent with other studies 415 (Mahmoud et al., 2016; Pawluski et al., 2012).  A lower dose of FLX also increased both anxiety-like and 416 depressive-like behaviors in non-parous female rats (Mahmoud et al., 2016). Similarly, prenatal stress 417 followed by postpartum FLX treatment of rat dams increased anxiety-like behavior in the elevated zero 418 maze (Pawluski et al., 2012). Although SSRIs are prescribed to improve mood, few studies have tested 419 whether antidepressants are efficacious in the postpartum, and of those conducted, the findings are 420 inconclusive. Studies of antidepressant efficacy are limited by high attrition rates, differences in the 421 baseline depression score, low sample size, and equal efficacy between SSRI and placebo treatment 422 (Yonkers et al., 2008; Sharma and Sommerdyk, 2013; De Crescenzo et al., 2014; Hantsoo et al., 2014). 423 Research from preclinical models, including the present study, supports the limited efficacy of long-term 424 SSRI treatment in dams (Workman et al., 2016), unless dams are continuously exposed to the SSRI via 425 osmotic mini-pump (Haim et al., 2016). Further, preclinical research has observed efficacy of FLX on 426 maternal depressive-like behavior early in the postpartum on postpartum day 7 (Salari et al., 2016). This 427 is an interesting observation as the present study and previous work with this model (Workman et al., 428 2016) found early beneficial effects of FLX on CORT-induced disruptions in maternal care (postpartum 429 days 2-8). Thus, it is possible that SSRIs have a positive effect early in the postpartum but are ineffective 430 with prolonged use, i.e. by the end of the postpartum period. Indeed, 57% of depressed patients who find 431 antidepressant relief with SSRIs also relapse despite continual use of medication (Byrne and Rothschild, 432 1998), and the trajectory of this cyclical relief and relapse with SSRI treatment may be different in 433 mothers. It should also be noted that the effect of postpartum FLX treatment was apparent early in FST2 434 but not over the entire duration of FST2; however, this partially contradicts findings from Workman et al., 435 2016, perhaps due to the additional stress (tail bleed) after FST1. PPD is a heterogenous disease with 436 distinct symptom profiles depending on history of depression, timing of index episode, and presence of 437 concurrent mood disorders (Postpartum Depression: Action Towards Causes and Treatment  Consortium, 438 2015). An interesting avenue of future research would be to utilize rodent models to discern whether there 439 Exercise and fluoxetine for postpartum depression  20  are precise time windows and duration of exposure when SSRI intervention in a maternal depression 440 model can yield positive effects on mood.  441 4.4. The combination of exercise and FLX treatment increased neurogenesis, and high running alone 442 increased neurogenesis in ventral hippocampus 443  Maternal exercise, specifically high running, increased neurogenesis in the ventral 444 dentate gyrus of dams, consistent with studies in male and non-parturient female rodents (van Praag et al., 445 1999; Yau et al., 2014) and confirming this exercise-induced neurogenic effect in the postpartum. 446 Interestingly, this increase occurred specifically in the ventral dentate gyrus, which is functionally 447 relevant for stress regulation (Fanselow and Dong, 2010). Interestingly, others have observed that 448 exercise predominantly increases neurogenesis in dorsal but not ventral hippocampus in male rodents 449 (Tanti & Belzung, 2013; Vivar et al., 2016), highlighting that exercise affects different sub-regions of the 450 hippocampus differently during motherhood. This combinatory effect of exercise and FLX was also 451 observed in maternal behaviour, particularly in high runners, and perhaps contributed to preventing 452 CORT-induced disruptions in maternal care. Because neurogenesis levels are low during the postpartum 453 due to increased circulating glucocorticoids (Leuner et al., 2007; Workman et al., 2016), it is possible that 454 a more potent manipulation such as high running was necessary to bolster neurogenesis. Furthermore, the 455 difference in voluntary running activity between high runners and low runners in the present study was 456 greatest during pre-conception and gestation. After parturition, all the dams engaged in similar levels of 457 running wheel activity. This suggests that for high runners, activity earlier in the experiment was more 458 influential on hippocampal plasticity, resulting in increased density of DCX-expressing cells by the end of 459 the postpartum. It is noteworthy that DCX expression can last for up to three weeks (Brown et al., 2003). 460 Thus, while high running dams engaged in the most amount of running prior to parturition, high running 461 likely had limited direct effects on DCX expression by the end of the postpartum period (approximately 3 462 weeks after parturition). However, running prior to parturition could have influenced DCX expression 463 indirectly by the end of the postpartum period perhaps by increasing trophic factors that promoted 464 Exercise and fluoxetine for postpartum depression  21  generation of DCX-expressing cells. Further research could determine how running influenced the 465 survival of cells generated prior to parturition in the postpartum hippocampus by using BrdU labelling; 466 however, it should be noted that BrdU given during gestation could have cytotoxic effects for the 467 developing offspring which may influence maternal outcomes. Further research could determine how 468 running influenced the survival of cells generated prior to parturition in the postpartum hippocampus by 469 using BrdU labelling; however, it should be noted that BrdU in the doses given to adult rats (200mg/kg) 470 can have cytotoxic effects for the developing offspring which may influence developmental and maternal 471 outcomes and thus was not used in the present study. 472 Interestingly, FLX increased neurogenesis only under exercise conditions while FLX reduced 473 neurogenesis under sedentary conditions. This interaction is important for two reasons. First, under 474 sedentary conditions, FLX decreased neurogenesis in addition to increasing depressive-like and anxiety-475 like behavior in dams. Collectively, these observations appear to be contrary to the expected 476 antidepressant effects of FLX, but they underscore the gaps in our understanding how antidepressant 477 interventions specifically in the postpartum affect the maternal brain and behavior.  Indeed, studies 478 examining effects of SSRIs on the maternal hippocampus observe either no significant effect on 479 neurogenesis (Workman et al., 2016) or a neurogenic effect when the SSRI is administered continuously 480 via osmotic mini-pump (Pawluski et al., 2012). As previously mentioned, there is limited evidence for 481 postpartum antidepressant efficacy, and it would be reasonable to expect limited effects of SSRIs in the 482 postpartum hippocampus. Secondly, exposure to both exercise and FLX increased neurogenesis in the 483 postpartum hippocampus despite FLX reducing amount of voluntary running in the postpartum. Thus, it 484 is possible that the high running earlier in the experiment with its commensurate elevated neurogenesis is 485 required for the neurogenic attributes of postpartum FLX. This may be due to increased serotonergic tone 486 because exercise during pregnancy increased density of serotonin-positive cells in the dorsal raphe 487 nucleus by the end of the postpartum (Seo et al., 2013). Additional research is necessary to understand 488 how interventions in pregnancy alter the postpartum brain. It is also plausible that within the normal 489 Exercise and fluoxetine for postpartum depression  22  conditions of postpartum physiology, including reduced neurogenesis and increased basal levels of 490 glucocorticoids, a more robust regimen of antidepressants may be necessary to elicit changes in the 491 postpartum hippocampus. Given the concerns regarding maternal SSRI exposure on child development, it 492 may be of interest for further research in PPD treatments to consider a multimodal approach (i.e. a 493 combination of pharmacological and non-pharmacological) as opposed to higher doses of SSRIs or other 494 pharmacological antidepressants. Indeed, there is evidence that cell survival in the dorsal raphe nucleus (a 495 major source of serotonergic innervation) is reduced in postpartum rat dams, an effect mediated by 496 presence of pups (Holschbach and Lonstein, 2017). Furthermore, (Banasr et al., 2001)  showed that 497 serotonin and estradiol interact to increase in cell proliferation in the dentate gyrus of adult female rats but 498 more research should discern the dynamics between serotonin and hippocampal neurogenesis in the 499 postpartum brain.  500 In conclusion, we show that exercise, but not FLX, can reduce maternal depressive-like behavior 501 and that exercise coupled with FLX increased hippocampal neurogenesis. Further, FLX alone or in 502 combination with high running prevented CORT-induced disruptions in maternal behavior. These 503 independent and coordinated effects of an SSRI and exercise as antidepressants on CORT-induced 504 endophenotypes of PPD highlight that maternal physiology is unique and underscore the importance of 505 reproductive experience as a factor in translation of animal models of mood disorders. This study 506 indicates that the postpartum affects efficacy of different types of antidepressant interventions and 507 contributes a better understanding of maternal mood disorders such as PPD.    508 ACKNOWLEDGMENTS  509  The authors would like to thank Lucille Hoover, Sophie McAlpine, Tamara Bodnar, Wayne Yu, 510 and Christine Ausman for their assistance and contributions throughout the experiment. This work was 511 funded by a CIHR operating grant (MOP 142308) to LAMG. 512 CONFLICTS OF INTEREST 513 Exercise and fluoxetine for postpartum depression  23   The authors have nothing to declare.  514 REFERENCES 515 1. Armstrong K, Edwards H (2003) The effects of exercise and social support on mothers reporting 516 depressive symptoms: a pilot randomized controlled trial. Int J Ment Heal Nurs 12:130–138. 517 2. Bambico FR, Nguyen NT, Gobbi G (2009) Decline in serotonergic firing activity and desensitization 518 of 5-HT1A autoreceptors after chronic unpredictable stress. Eur Neuropsychopharmacol 19:215–228. 519 3. Banasr M, Hery M, Brezun JM, Daszuta A (2001) Serotonin mediates oestrogen stimulation of cell 520 proliferation in the adult dentate gyrus. Eur J Neurosci 14:1417–1424. 521 4. Bessa JM, Ferreira D, Melo I, Marques F, Cerqueira JJ, Palha JA, Almeida OF, Sousa N (2009) The 522 mood-improving actions of antidepressants do not depend on neurogenesis but are associated with 523 neuronal remodeling. Mol Psychiatry 14:739,764-773. 524 5. Brown JP, Couillard-Despres S, Cooper-Kuhn CM, Winkler J, Aigner L, Kuhn HG (2003) Transient 525 expression of doublecortin during adult neurogenesis. J Comp Neurol 467:1–10. 526 6. Brummelte S, Galea LAM (2010) Chronic corticosterone during pregnancy and postpartum affects 527 maternal care, cell proliferation and depressive-like behavior in the dam. Horm Behav 58:769–779. 528 7. Brummelte S, Pawluski JL, Galea LA (2006) High post-partum levels of corticosterone given to dams 529 influence postnatal hippocampal cell proliferation and behavior of offspring: A model of post-partum 530 stress and possible depression. Horm Behav 50:370–382. 531 8. Byrne SE, Rothschild AJ (1998) Loss of antidepressant efficacy during maintenance therapy: possible 532 mechanisms and treatments. J Clin Psychiatry 59:279–288. 533 9. Daley AJ, Blamey R V, Jolly K, Roalfe AK, Turner KM, Coleman S, McGuinness M, Jones I, Sharp 534 DJ, MacArthur C (2015) A pragmatic randomized controlled trial to evaluate the effectiveness of a 535 facilitated exercise intervention as a treatment for postnatal depression: the PAM-PeRS trial. Psychol 536 Med 45:2413–2425. 537 10. Daley AJ, Thomas A, Cooper H, Fitzpatrick H, McDonald C, Moore H, Rooney R, Deeks JJ (2012) 538 Maternal exercise and growth in breastfed infants: a meta-analysis of randomized controlled trials. 539 Pediatrics 130:108–114. 540 11. De Crescenzo F, Perelli F, Armando M, Vicari S (2014) Selective serotonin reuptake inhibitors 541 (SSRIs) for post-partum depression (PPD): a systematic review of randomized clinical trials. J Affect 542 Disord 152–154:39–44. 543 12. Dennis CL (2004) Treatment of postpartum depression, part 2: a critical review of nonbiological 544 interventions. J Clin Psychiatry 65:1252–1265. 545 13. Desan PH, Woodmansee WW, Ryan SM, Smock TK, Maier SF (1988) Monoamine neurotransmitters 546 and metabolites during the estrous cycle, pregnancy, and the postpartum period. Pharmacol Biochem 547 Behav 30:563–568. 548 14. des Portes V, Pinard JM, Billuart P, Vinet MC, Koulakoff A, Carrié A, Gelot A, Dupuis E, Motte J, 549 Berwald-Netter Y, Catala M, Kahn A, Beldjord C, Chelly J (1998) A novel CNS gene required for 550 neuronal migration and involved in X-linked subcortical laminar heterotopia and lissencephaly 551 syndrome. Cell 92:51–61. 552 15. Dewey KG, Lovelady CA, Nommsen-Rivers LA, McCrory MA, Lonnerdal B (1994) A randomized 553 study of the effects of aerobic exercise by lactating women on breast-milk volume and composition. 554 N Engl J Med 330:449–453. 555 16. Droste SK, Chandramohan Y, Hill LE, Linthorst AC, Reul JM (2007) Voluntary exercise impacts on 556 the rat hypothalamic-pituitary-adrenocortical axis mainly at the adrenal level. Neuroendocrinology 557 86:26–37. 558 17. Ersek JL, Brunner Huber LR (2009) Physical activity prior to and during pregnancy and risk of 559 postpartum depressive symptoms. J Obs Gynecol Neonatal Nurs 38:556–566. 560 18. Fanselow MS, Dong H-W (2010) Are the Dorsal and Ventral Hippocampus Functionally Distinct 561 Exercise and fluoxetine for postpartum depression  24  Structures? Neuron 65:7–19. 562 19. Fisher SD, Wisner KL, Clark CT, Sit DK, Luther JF, Wisniewski S (2016) Factors associated with 563 onset timing, symptoms, and severity of depression identified in the postpartum period. J Affect 564 Disord 203:111–120. 565 20. Gobinath AR, Workman JL, Chow C, Lieblich SE, Galea LA (2016) Maternal postpartum 566 corticosterone and fluoxetine differentially affect adult male and female offspring on anxiety-like 567 behavior, stress reactivity, and hippocampal neurogenesis. Neuropharmacology 101:165–178. 568 21. Haim A, Albin-Brooks C, Sherer M, Mills E, Leuner B (2016) The effects of gestational stress and 569 Selective Serotonin reuptake inhibitor antidepressant treatment on structural plasticity in the 570 postpartum brain — A translational model for postpartum depression. Horm Behav 77:124–131. 571 22. Han T-K, Lee J-K, Leem Y-H (2015) Chronic exercise prevents repeated restraint stress-provoked 572 enhancement of immobility in forced swimming test in ovariectomized mice. Metab Brain Dis 573 30:711–718. 574 23. Hantsoo L, Ward-O’Brien D, Czarkowski KA, Gueorguieva R, Price LH, Epperson CN (2014) A 575 randomized, placebo-controlled, double-blind trial of sertraline for postpartum depression. 576 Psychopharmacol 231:939–948. 577 24. Hinman SK, Smith KB, Quillen DM, Smith MS (2015) Exercise in Pregnancy: A Clinical Review. 578 Sports Health 7:527–531. 579 25. Hodes GE, Hill-Smith TE, Suckow RF, Cooper TB, Lucki I (2010) Sex-specific effects of chronic 580 fluoxetine treatment on neuroplasticity and pharmacokinetics in mice. J Pharmacol Exp Ther 581 332:266–273. 582 26. Hoekzema E, Barba-Muller E, Pozzobon C, Picado M, Lucco F, Garcia-Garcia D, Soliva JC, Tobena 583 A, Desco M, Crone EA, Ballesteros A, Carmona S, Vilarroya O (2017) Pregnancy leads to long-584 lasting changes in human brain structure. Nat Neurosci 20:287–296. 585 27. Holschbach MA, Lonstein JS (2017) Motherhood and infant contact regulate neuroplasticity in the 586 serotonergic midbrain dorsal raphe. Psychoneuroendocrinology 76:97–106. 587 28. Ising M, Horstmann S, Kloiber S, Lucae S, Binder EB, Kern N, Kunzel HE, Pfennig A, Uhr M, 588 Holsboer F (2007) Combined dexamethasone/corticotropin releasing hormone test predicts treatment 589 response in major depression - a potential biomarker? Biol Psychiatry 62:47–54. 590 29. Lapmanee S, Charoenphandhu J, Charoenphandhu N (2013) Beneficial effects of fluoxetine, 591 reboxetine, venlafaxine, and voluntary running exercise in stressed male rats with anxiety- and 592 depression-like behaviors. Behav Brain Res 250:316–325. 593 30. Lerch-Haner JK, Frierson D, Crawford LK, Beck SG, Deneris ES (2008) Serotonergic transcriptional 594 programming determines maternal behavior and offspring survival. Nat Neurosci 11:1001–1003. 595 31. Leuner B, Mirescu C, Noiman L, Gould E (2007) Maternal experience inhibits the production of 596 immature neurons in the hippocampus during the postpartum period through elevations in adrenal 597 steroids. Hippocampus 17:434–442. 598 32. Liu Z, Wu Y, Liu T, Li R, Xie M (2017) Serotonin regulation in a rat model of exercise-induced 599 chronic fatigue. Neuroscience 349:27–34. 600 33. Lonstein JS, Dominguez JM, Putnam SK, De Vries GJ, Hull EM (2003) Intracellular preoptic and 601 striatal monoamines in pregnant and lactating rats: possible role in maternal behavior. Brain Res 602 970:149–158. 603 34. Mahmoud R, Wainwright SR, Chaiton JA, Lieblich SE, Galea LAM (2016) Ovarian hormones, but 604 not fluoxetine, impart resilience within a chronic unpredictable stress model in middle-aged female 605 rats. Neuropharmacology 107:278–293. 606 35. Matsuno AY, Esrey KL, Perrault H, Koski KG (1999) Low intensity exercise and varying proportions 607 of dietary glucose and fat modify milk and mammary gland compositions and pup growth. J Nutr 608 129:1167–1175. 609 36. Messiha FS (1993) Fluoxetine: adverse effects and drug-drug interactions. J Toxicol Clin Toxicol 610 31:603–630. 611 37. Mezadri TJ, Batista GM, Portes AC, Marino-Neto J, Lino-de-Oliveira C (2011) Repeated rat-forced 612 Exercise and fluoxetine for postpartum depression  25  swim test: reducing the number of animals to evaluate gradual effects of antidepressants. J Neurosci 613 Methods 195:200–205. 614 38. Mobley CB, Holland AM, Kephart WC, Mumford PW, Lowery RP, Kavazis AN, Wilson JM, 615 Roberts MD (2017) Progressive resistance-loaded voluntary wheel running increases hypertrophy and 616 differentially affects muscle protein synthesis, ribosome biogenesis, and proteolytic markers in rat 617 muscle. J Anim Physiol Anim Nutr (Berl). 618 39. Park Y-M, Kanaley JA, Padilla J, Zidon T, Welly RJ, Will MJ, Britton SL, Koch LG, Ruegsegger 619 GN, Booth FW, Thyfault JP, Vieira-Potter VJ (2016) Effects of intrinsic aerobic capacity and 620 ovariectomy on voluntary wheel running and nucleus accumbens dopamine receptor gene expression. 621 Physiol Behav 164:383–389. 622 40. Pawluski JL, Charlier TD, Fillet M, Houbart V, Crispin HT, Steinbusch HW, van den Hove DL 623 (2012) Chronic fluoxetine treatment and maternal adversity differentially alter neurobehavioral 624 outcomes in the rat dam. Behav Brain Res 228:159–168. 625 41. Pawluski JL, Galea LA (2007) Reproductive experience alters hippocampal neurogenesis during the 626 postpartum period in the dam. Neuroscience 149:53–67. 627 42. Pósa A, Szabó R, Kupai K, Baráth Z, Szalai Z, Csonka A, Veszelka M, Gyöngyösi M, Radák Z, 628 Ménesi R, Pávó I, Berkó AM, Varga C (2015) Cardioprotective Effects of Voluntary Exercise in a 629 Rat Model: Role of Matrix Metalloproteinase-2. Oxid Med Cell Longev 2015:1–9. 630 43. Postpartum Depression: Action Towards Causes and Treatment (PACT) Consortium (2015) 631 Heterogeneity of postpartum depression: a latent class analysis. The lancet Psychiatry 2:59–67. 632 44. Prady SL, Pickett KE, Gilbody S, Petherick ES, Mason D, Sheldon TA, Wright J (2016) Variation 633 and ethnic inequalities in treatment of common mental disorders before, during and after pregnancy: 634 combined analysis of routine and research data in the Born in Bradford cohort. BMC Psychiatry 635 16:99. 636 45. Prentice A (1994) Should lactating women exercise? Nutr Rev 52:358–360. 637 46. Rethorst CD, Trivedi MH (2013) Evidence-based recommendations for the prescription of exercise 638 for major depressive disorder. J Psychiatr Pr 19:204–212. 639 47. Rosa B V, Blair HT, Vickers MH, Morel PC, Cockrem JF, Firth EC (2012) Voluntary exercise in 640 pregnant rats improves post-lactation maternal bone parameters but does not affect offspring 641 outcomes in early life. J Musculoskelet Neuronal Interact 12:199–208. 642 48. Salari A-A, Fatehi-Gharehlar L, Motayagheni N, Homberg JR (2016) Fluoxetine normalizes the 643 effects of prenatal maternal stress on depression- and anxiety-like behaviors in mouse dams and male 644 offspring. Behav Brain Res 311:354–367. 645 49. Seo J-H, Kim T-W, Kim C-J, Sung Y-H, Lee S-J (2013) Treadmill exercise during pregnancy 646 ameliorates post‑ traumatic stress disorder‑ induced anxiety‑ like responses in maternal rats. Mol 647 Med Rep 7:389–395. 648 50. Seth S, Lewis AJ, Galbally M (2016) Perinatal maternal depression and cortisol function in 649 pregnancy and the postpartum period: a systematic literature review. BMC Pregnancy Childbirth 650 16:124. 651 51. Sharma V, Sommerdyk C (2013) Are antidepressants effective in the treatment of postpartum 652 depression? A systematic review. Prim care companion CNS Disord 15. 653 52. Smolina K, Hanley GE, Mintzes B, Oberlander TF, Morgan S (2015) Trends and Determinants of 654 Prescription Drug Use during Pregnancy and Postpartum in British Columbia, 2002–2011: A 655 Population-Based Cohort Study Fischer G, ed. PLoS One 10:e0128312. 656 53. Stetler C, Miller GE (2011) Depression and hypothalamic-pituitary-adrenal activation: a quantitative 657 summary of four decades of research. Psychosom Med 73:114–126. 658 54. Su D, Zhao Y, Binns C, Scott J, Oddy W (2007) Breast-feeding mothers can exercise: results of a 659 cohort study. Public Heal Nutr 10:1089–1093. 660 55. Treadway JL, Lederman SA (1986) The effects of exercise on milk yield, milk composition, and 661 offspring growth in rats. Am J Clin Nutr 44:481–488. 662 56. van Praag H, Christie BR, Sejnowski TJ, Gage FH (1999) Running enhances neurogenesis, learning, 663 Exercise and fluoxetine for postpartum depression  26  and long-term potentiation in mice. Proc Natl Acad Sci U S A 96:13427–13431. 664 57. Voss MW, Vivar C, Kramer AF, van Praag H (2013) Bridging animal and human models of exercise-665 induced brain plasticity. Trends Cogn Sci 17:525–544. 666 58. Wang J, Chen X, Zhang N, Ma Q (2013) Effects of exercise on stress-induced changes of 667 norepinephrine and serotonin in rat hippocampus. Chin J Physiol 56:245–252. 668 59. Weissman AM, Levy BT, Hartz AJ, Bentler S, Donohue M, Ellingrod VL, Wisner KL (2004) Pooled 669 analysis of antidepressant levels in lactating mothers, breast milk, and nursing infants. Am J 670 Psychiatry 161:1066–1078. 671 60. Wisner KL, Sit DKY, McShea MC, Rizzo DM, Zoretich RA, Hughes CL, Eng HF, Luther JF, 672 Wisniewski SR, Costantino ML, Confer AL, Moses-Kolko EL, Famy CS, Hanusa BH (2013) Onset 673 timing, thoughts of self-harm, and diagnoses in postpartum women with screen-positive depression 674 findings. JAMA psychiatry 70:490–498. 675 61. Workman JL, Brummelte S, Galea LAM (2013) Postpartum Corticosterone Administration Reduces 676 Dendritic Complexity and Increases the Density of Mushroom Spines of Hippocampal CA3 Arbours 677 in Dams. J Neuroendocrinol 25:119–130. 678 62. Workman JL, Gobinath AR, Kitay NF, Chow C, Brummelte S, Galea LA (2016) Parity modifies the 679 effects of fluoxetine and corticosterone on behavior, stress reactivity, and hippocampal neurogenesis. 680 Neuropharmacology 105:443–453. 681 63. Xu Y, Li Y, Huang X, Chen D, She B, Ma D (2017) Single bolus low-dose of ketamine does not 682 prevent postpartum depression: a randomized, double-blind, placebo-controlled, prospective clinical 683 trial. Arch Gynecol Obs 295:1167–1174. 684 64. Yau SY, Li A, Zhang ED, Christie BR, Xu A, Lee TM, So KF (2014) Sustained running in rats 685 administered corticosterone prevents the development of depressive behaviors and enhances 686 hippocampal neurogenesis and synaptic plasticity without increasing neurotrophic factor levels. Cell 687 Transpl 23:481–492. 688 65. Yonkers KA, Lin H, Howell HB, Heath AC, Cohen LS (2008) Pharmacologic treatment of 689 postpartum women with new-onset major depressive disorder: a randomized controlled trial with 690 paroxetine. J Clin Psychiatry 69:659–665. 691 66. Zimmer P, Stritt C, Bloch W, Schmidt F-P, Hübner ST, Binnebößel S, Schenk A, Oberste M (2016) 692 The effects of different aerobic exercise intensities on serum serotonin concentrations and their 693 association with Stroop task performance: a randomized controlled trial. Eur J Appl Physiol 694 116:2025–2034. 695 67.   696 Exercise and fluoxetine for postpartum depression  27  FIGURE LEGENDS 697 Figure 1. Timeline of Experiment. 698 Figure 2. A) Individual running data based on average revolutions per day before drug treatment began. 699 Individual dams were categorized as “high running” or “low running” based on a median split 700 (n=15/group). B) During pre-conception and gestation periods, high running dams consistently ran 701 significantly more than low running dams. Regardless of high or low running, dams voluntarily ran at 702 similar levels during pre-conception and first two weeks of gestation. Dams ran less during the third and 703 final week of gestation in comparison to earlier periods. After parturition and regardless of drug 704 treatment, all dams ran at comparable levels during the first two postpartum weeks. During the third week 705 of the postpartum, FLX reduced running in comparison to saline in high runners only. * p < 0.05; FLX: 706 fluoxetine; HR: high running; LR: low running.  707 Figure 3. A) Regardless of exercise, CORT treatment reduced time spent nursing in comparison to 708 vehicle. FLX increased time spent nursing in comparison to saline in CORT-treated dams. B) Regardless 709 of exercise, CORT/FLX-treated dams spent less time off nest in comparison to CORT/saline-treated 710 dams. C) FLX increased time spent nursing in comparison to saline in high running dams but not low 711 running dams. D) FLX reduced time spent off nest in comparison to saline in high running dams but not 712 low running dams. In figures 3C and 3D, data from the non-exercising dams are presented again for 713 reference. * p < 0.05; CORT: corticosterone; FLX: fluoxetine; HR: high running; LR: low running. 714 Figure 4. A) Postpartum CORT and postpartum FLX independently increased time spent immobile in 715 FST2 in comparison to oil and saline, respectively. Exercise reduced time spent immobile in comparison 716 to no exercise. B) Immediately after FST1, all dams reached similar levels of stress-induced serum CORT 717 regardless of CORT, FLX, or exercise. After 2 h, CORT-treated dams maintained elevated serum CORT 718 levels in comparison to oil-treated dams. There was a slight but significant effect of FLX to reduce serum 719 CORT levels in comparison to saline among CORT-treated dams. C) FLX increased latency to feed in 720 comparison to saline in novelty suppressed test. D) FLX increased latency to feed in comparison to saline 721 in high running dams but not low or no running dams. * p < 0.05; CORT: corticosterone; FLX: 722 fluoxetine; HR: high running; LR: low running. It should be noted that the main effects of FLX are 723 denoted with * by FLX in the legend.  724 Figure 5. A) Photomicrograph of a post-mitotic DCX-expressing cell. B) FLX reduced density of DCX-725 expressing cells in comparison to saline vehicle in non-exercising dams. However, the combination of 726 exercise and FLX increased density of DCX-expressing cells in comparison to either exercise or FLX 727 alone. C) Exercise increased density of DCX-expressing cells in comparison to no exercise in ventral 728 hippocampus only. D) The neurogenic effect of exercise was significant only in high runners; high 729 running increased density of DCX-expressing cells in comparison to no or low running in ventral 730 hippocampus only. E) CORT reduced density of DCX-expressing cells in comparison to oil vehicle in 731 dorsal hippocampus and F) ventral hippocampus. Exercise increased density of DCX-expressing cells in 732 comparison to no exercise in ventral hippocampus. G) CORT reduced proportion of type 3 (post-mitotic) 733 DCX-expressing cells in comparison to vehicle-treated dams. H) Among vehicle-treated dams, there was 734 a positive correlation in amount of average postpartum running wheel activity and density of DCX-735 expressing cells in ventral hippocampus. * p < 0.05; CORT: corticosterone; DCX+: DCX-expressing; 736 FLX: fluoxetine; HR: high running; LR: low running. For post-mortem analyses:  1) CORT/FLX 737 (exercise, n=8; non-exercise, n=9); 2) CORT/saline (exercise, n=8; non-exercise, n=8); 3) oil/FLX 738 (exercise, n=7; non-exercise, n=7); 4) oil/saline (exercise, n=7; non-exercise, n=6). 739  740 Exercise and fluoxetine for postpartum depression  28   741   742   Pre-Conception Gestation – Week 1 Gestation – Week 2 Gestation – Week 3 Postpartum – Week 1 Postpartum – Week 2 Postpartum – Week 3 Oil/Saline 4571 ± 464 3214 ± 359 7135 ± 1083* Oil/FLX 3010 ± 619 2361 ± 1039 2405 ± 109 CORT/Saline 4849 ± 1946 3091 ± 1102 7896 ± 2163* CORT/FLX 2429 ± 653 1892 ± 403 2203 ± 372 Oil/Saline 2364 2126 2365 Oil/FLX 2452 ± 414 1165 ± 312 2728 ± 506 CORT/Saline 3010 ± 1117 3166 ± 673 4095 ± 529 CORT/FLX 3198 ± 895 2595 ± 732 2314 ± 425 Table 1. Mean running wheel revolutions ± SEM. Voluntary running wheel activity increased from the postpartum week 2 to postpartum week 3 in saline-treated dams but not FLX-treated dams. * p < 0.05 CORT: Corticosterone; FLX: Fluoxetine. Exercise and fluoxetine for postpartum depression  29   743  744   Mean % time licking ± SEM Mean % time nursing + licking ± SEM Mean % time self-grooming ± SEM Oil/Saline 0.4 ± 0.1 10.2 ± 2.1 9.9 ± 1.1 Oil/FLX 0.3 ± 0.2 10.5 ± 1.9 7.9 ± 1.3 * CORT/Saline 0.6 ± 0.2 8.7 ± 2.1 10.7 ± 1.5 CORT/FLX 0.2 ± 0.1 7.7 ± 1.4 6.8 ± 1.2* Oil/Saline 0.3 ± 0.1 7.1 ± 1.0 12.8 ± 1.8 Oil/FLX 0.2 ± 0.1 8.8 ± 1.3 7.5 ± 1.8* CORT/Saline 0.8 ± 0.3 13.4 ± 3.0* 12.8 ± 2.0 CORT/FLX 0.5 ± 0.2 11.2 ± 2.9* 11.2 ± 2.0*  745   746 Table 2. Mean percent time spent licking, nursing + licking, and self-grooming ± SEM. Maternal exercise, CORT, and FLX treatment did not significantly affect time spent licking. Non-exercising/CORT-treated dams spent more time nursing + licking in comparison to non-exercising/oil-treated dams. Postpartum FLX treatment reduced time spent self-grooming in comparison to saline. * p < 0.05; CORT: Corticosterone; FLX: Fluoxetine. Exercise and fluoxetine for postpartum depression  30   747   748  No Exercise Exercise* Oil/Saline 24.4% 13.0% Oil/FLX 20.5% 20.0% CORT/Saline 22.4% 16.7% CORT/FLX 31.0% 15.6% Table 3. Mean percent time immobile over the entire 5 min session of FST2 ± SEM. Maternal exercise maintained its significant effect to reduce time spent immobile over the course of FST2. However, the effects of postpartum CORT and FLX were not apparent in analyzing the entire 5 min session. * p < 0.05; CORT: Corticosterone; FLX: Fluoxetine. Exercise and fluoxetine for postpartum depression  31   749   750  SAL FLX* No Exercise – OIL 32.7 ± 2.2 29.6 ± 2.9 Exercise – OIL 32.0 ± 2.9 28.3 ± 1.9 No Exercise – CORT 28.9 ± 1.3 23.9 ± 1.4 Exercise - CORT 27.6 ± 2.8 23.5 ± 2.3 Table 4. Mean food consumption (g) 1 h after novelty suppressed feeding test ± SEM. FLX reduced food consumption in comparison to saline. * p < 0.05 CORT: Corticosterone; FLX: Fluoxetine. Exercise and fluoxetine for postpartum depression  32   751  752  753   754  Dams Offspring Body Mass G0 (g) Body Mass G21 (g) Body Mass P25 (g) Relative Adrenal Mass (g/100 g bw) # of pups Ratio of male: female pups in birth litter Litter mass at birth (g) Litter mass on P22 (g) Oil/Saline 323 ± 16 0.31 ± 0.01* 559 ± 17 Oil/FLX 336 ± 19 0.29 ± 0.02 523 ± 28 CORT/Saline 276 ± 14* 0.11 ± 0.01 525 ± 32 CORT/FLX 273 ± 7* 0.11 ± 0.01 423 ± 21 Oil/Saline 344 ± 15 0.22 ± 0.02 568 ± 13 Oil/FLX 329 ± 8 0.28 ± 0.02 501 ± 21 CORT/Saline 293 ± 10* 0.10 ± 0.01 504 ± 12 CORT/FLX 274 ± 8* 0.12 ± 0.01 423 ± 12 Table 5. Exercising dams maintained lower body mass from the beginning to end of pregnancy in comparison to no exercise. At the time of perfusion, CORT treatment reduced body mass in comparison to oil and reduced relative adrenal mass in dams. Under vehicle conditions, exercise increased adrenal mass in comparison to no exercise. Under no exercise/oil conditions, FLX increased adrenal mass in comparison to saline. Exercise had no significant effect on number of pups, ratio of male:female pups, or mass of the litter at the time of parturition. At the final week of the postpartum period, both maternal postpartum CORT as well as FLX independently reduced litter mass in comparison to oil and saline, respectively. CORT: Corticosterone; FLX: Fluoxetine; G0: day 1 of pregnancy (conception); G21: day 21 of pregnancy. * denotes p < 0.05. Exercise and fluoxetine for postpartum depression  33   755  756   757 Endophenotype of Postpartum Depression CORT FLX only Exercise only FLX + Exercise Body mass ↓ ↓ ↔ ↔ Loss of interest in litter (anhedonia) ↑ ↓ ↔ ↓ (high running) HPA axis activity ↑ ↓ (recovery) ↔ ↔ Depressive-like behaviour (forced swim test) ↑ ↑ ↓ ↓ Hippocampal neurogenesis (density of DCX-expressing cells) ↓ ↓ ↑ (high running) ↑ Table 6. Summary of the effects of maternal postpartum CORT on endophenotypes of postpartum depression and how FLX, exercise, and FLX + exercise affected these measures under CORT-conditions. Maternal postpartum FLX prevented disruptions in maternal care and serum CORT recovery after FST1. Maternal exercise also prevented disruptions in maternal care and depression-like behaviour in FST2.  Further, FLX + exercise or high running alone increased hippocampal neurogenesis in comparison to controls.  CORT: Corticosterone; DCX: doublecortin; FLX: Fluoxetine; FST1: forced swim test, day 1; FST2: forced swim test, day 2; HPA: hypothalamic-pituitary-adrenal ↑: increased; ↓: decreased; ↔: no significant change Exercise and fluoxetine for postpartum depression  34   758  759   760 Figure 1. Exercise and fluoxetine for postpartum depression  35  Figure 2.  761   762 Exercise and fluoxetine for postpartum depression  36  Figure 3.  763   764 Exercise and fluoxetine for postpartum depression  37  Figure 4.  765   766 Exercise and fluoxetine for postpartum depression  38  Figure 5.  767  768 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items