- Email: [email protected]
Chronic fluoxetine suppresses circulating estrogen and the enhanced spatial learning of estrogen-treated ovariectomized rats
Psychoneuroendocrinology (2004) 29, 1241–1249
www.elsevier.com/locate/psyneuen
Chronic fluoxetine suppresses circulating estrogen and the enhanced spatial learning of estrogen-treated ovariectomized rats George T. Taylora,*, Susan Farra,b, Klaus Klingac, Juergen Weissd a
Behavioral Neuroscience Group, University of Missouri—St. Louis, 8001 Natural Bridge Road, St. Louis, MO 63121, USA b Department of Medicine, School of Medicine, St. Louis University, St. Louis, MO 63110, USA c ¨ tsstoerungen, Universitaets-Frauenklinik, Abt. Gynaekologische Endokrinologie und Fertilita Heidelberg 6900, Germany d Zentrum fur Molekular Biologie, Heidelberg University, Heidelberg 6900, Germany Received 18 August 2003; received in revised form 5 March 2004; accepted 6 March 2004
KEYWORDS Estrogen; SSRI; Fluoxetine; Serotonin; Spatial memory
*
Summary We are interested in developing animal models to evaluate cognitive processes as influenced by the interplay of steroidal hormones and drugs commonly used in psychotherapy. Two experiments with female rats were conducted to evaluate the interaction of estrogen with the serotonin specific reuptake inhibitor (SSRI) fluoxetine on spatial learning and memory and on the endocrine system. In experiment 1, estrogen (50 lg estradiol benzoate/kg body weight) was administered SC to young adult, ovariectomized (OVX) rats either alone or in combination with fluoxetine (2 mg/kg SC). After a month, the groups were compared with appropriate OVX and gonadally intact controls on trials to criterion in a hole board spatial memory task using massed training trials. Experiment 2 was a dose–response study of the influence of fluoxetine (0.5–5 mg/kg) on circulating estrogen in OVX, estrogen treated females. Results were that the OVX females administered estrogen only reached the learning criterion significantly faster than the other groups. All other groups, including the estrogen þ fluoxetine animals, performed no better than the controls. Combining fluoxetine with estrogen also lowered circulating estrogen titers, with the least estrogen reductions being in the group receiving the highest dosage of fluoxetine. No differences among groups were found on measures of activity in an open field or for anxiety in a plus maze. Conclusions were that administration of estrogen improved spatial learning and memory in OVX rats, whereas concurrent fluoxetine exposure suppressed the levels of estrogen in circulation and eliminated the gains in spatial performance obtained from chronic estrogen exposure. # 2004 Elsevier Ltd. All rights reserved.
Corresponding author. Fax: +1-314-516-5392. E-mail address: [email protected] (G.T. Taylor).
0306-4530/$ - see front matter # 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.psyneuen.2004.03.001
1242
1. Introduction There is now substantial evidence that gonadal steroids act on a range of cortical and subcortical brain regions underlying behaviors not directly related to reproduction. Of particular interest have been findings that estrogen serves as a neuroactive steroid to influence cognitive function, likely by modifying neural morphology and modulating neurotransmitter systems involved in learning and memory (Woolley, 1998; Luine et al., 1998). Our interest is the behavior of hypoestrogenic animal models administered estrogen concurrently with drugs commonly used in psychotherapy. Here, the drug was fluoxetine, the prototypical serotonin specific reuptake inhibitor (SSRI) anti-depressant medication (Taylor et al., 1996a). The literature suggests reasons to suspect interactions among estrogen, serotonin and cognitive behaviors (Fink et al., 1998). However, even the simple two-way interactions between variables have proven resistant to straightforward generalizations. For example, does estrogen improve or disrupt the spatial learning and memory of experimental animals? There are reports with gonadally intact females that spatial abilities are worse at proestrus, when both endogenous estrogen and progesterone are elevated, than at other stages of the estrous cycle (Frye, 1995; Warren and Juraska, 1997; Daniel et al., 1999; Chesler and Juraska, 2000). Yet, exogenous estrogen administered to ovariectomized (OVX) animals is often reported to enhance spatial memory (Daniel et al., 1997; Luine et al., 1998; Gibbs, 1999; cf. Daniel et al., 1999). The suggestion is that chronic, continuous estrogen improves performance whereas shortterm modulation of estrogen during the estrous cycle is detrimental or has no effect on performance (Sandstrom and Williams, 2001). Conclusions on the influence of serotonergic agents on cognitive behaviors have also proven complex. For example, serotonin agonists are often reported to disrupt normal acquisition and recall (Santucci et al., 1996; Luciana et al., 1998; Farr et al., 2000). Yet, the findings with fluoxetine and other SSRI agents that also are presumed to increase serotonergic activity suggest little or no effects on learning and memory (Bammer, 1982; Lee et al., 1992; Stewart and Reid, 2000). Fluoxetine may even improve cognitive behavior in some paradigms (Flood and Cherkin, 1987; Nowakowska et al., 1996), perhaps due to the anxiolytic properties of SSRIs. One notable feature is that the studies of serotonin–cognitive relations have typically used males as subjects. Yet, the possibility of a differ-
G.T. Taylor et al.
ent response by females is suggested by data from both intact females and OVX, hormone-restored animals. There are changes in serotonin synthesis, release, reuptake and metabolism during the estrous cycles of intact animals (Rubinow et al., 1998; Maswood et al., 1999). OVX decreases serotonin receptor numbers in various brain regions and estrogen replacement restores receptor concentrations to those of intact females (Cyr et al., 2000). The current experiments were suggested by the possibility of different fluoxetine–cognitive behavior interactions in females with or without significant levels of circulating estrogen. In the first experiment, groups of OVX female rats were exposed chronically to estrogen or fluoxetine or both estrogen and fluoxetine. Controls were groups of vehicle-only OVX females and gonadally intact females. Testing of cognitive function employed a modified hole board paradigm designed to assess acquisition rates in a spatial memory task with massed training trials. There is evidence that the hole board is sensitive both to estrogen treatments (Lannert et al., 1998) and to normal hippocampal function (Ohl et al., 2000).
2. Experiment 1 2.1. Methods 2.1.1. Subjects Long–Evans female rats (N ¼ 40) were 3–4 months of age and had been housed individually since weaning in hanging wire cages measuring 20:5 23:5 29:5 cm. Water and Richmond Standard Lab Diet 5001 were available ad libitum except during periods of food restriction. Lighting in colony rooms was on a reversed cycle of 12 h light/dark; room temperature (20–22 C) and relative humidity (55 5%) were controlled automatically. Animal care and experimental methodologies were approved by the campus Institutional Animal Care and Use Committee and were in accordance with NIH guidelines. 2.1.2. Materials The hole board apparatus was a 66:5 cm 66:5 cm box. The walls were constructed of clear Plexiglas 50.5 cm high. The gray wooden floor had four holes, 4 cm in diameter and 4.5 cm in depth, located 12 cm from each corner. A small circular section of wire-mesh screen was located halfway into each hole. This construction allowed food to be placed either below the screen, rendering the food inaccessible, or above the screen to be
Spatial learning of estrogen-treated ovariectomized rats
1243
eaten. A cylinder, 42 cm tall and 15 cm in diameter, positioned in the center of the apparatus served as the ‘‘start box’’. The apparatus sat on the floor in the center of the room to reveal various extra-maze cues to the animal. Examples of extra-maze stimuli were various posters mounted on the walls, a set of shelves on another wall, and the experimenter standing to one side of the apparatus. The elevated plus maze consisted of two open arms and two closed arms, each 86.25 cm in length and 13.75 cm wide. The closed arms had walls 16.25 cm in height. The apparatus was constructed of wood, painted black and elevated to a height of 63 cm above the floor. The 90 120 cm open field apparatus has been described in detail earlier (Taylor et al., 1996a). Estradiol benzoate (EB) was purchased from Sigma Chemical Company (St. Louis, MO) and was suspended in olive oil for injection. Fluoxetine hydrochloride was donated by Eli Lilly & Co. (Indianapolis, IN) and was solubilized in 0.9% saline solution.
before a behavioral session. Habituation sessions in the hole board apparatus began on week 4 with testing being conducted during week 5. Elevated plus maze and open field testing were done during the final week. Animals were sacrificed by decapitation with a guillotine at the end of week 6, and trunk blood was obtained for hormone assays.
2.1.3. Surgery The animals were either ovariectomized (OVX,n ¼ 32) or sham operated (Intacts, n ¼ 8) 1 week prior to the start of drug treatment. Vaginal smears were used to assess that the ovariectomies were successful and that the intact females continued to cycle normally.
2.1.5.1. Habituation. The hole board paradigm requires extensive exposure to the apparatus and procedures prior to testing. Beginning during week 4 of injections, each rat received habituation sessions of 15–30 min distributed over different days. The initial sessions consisted of allowing the rat to roam freely about the apparatus to eat pieces of a sweet breakfast cereal (Kellog’s Froot Loops) cut into quarters located in each of the four holes. When the animal was reliably finding and eating the food, a final habituation session was given on a single day to adapt the rat to repeated trials in the apparatus. All the holes were baited with accessible food, but otherwise the 10 trials administered were the same as described below in the testing phase.
2.1.4. Experimental design Rats were assigned at random to one of five treatment conditions (n ¼ 8 per group) to be injected SC daily with doses of estradiol benzoate (50 lg EB/kg body weight), fluoxetine (2 mg Fluox/kg body weight) or vehicle only. Group 1 females (Intact, veh-only) were gonadally intact and the other four groups were ovariectomized. Groups 2–4 females were administered, respectively, only vehicle (OVX, veh-only), only estrogen (OVX, EBonly) or only fluoxetine (OVX, Fluox-only). Group 5 animals received both estradiol and fluoxetine (OVX, EB þ Fluox). The 2 mg dosage of fluoxetine is at the low end of the range of doses typically administered to rats (Holtzman and Steinfels, 1994; Matuszczyk et al., 1998). We have used the 50 lg EB dosage in past studies to restore function to OVX females (e.g.,Taylor et al., 1993). All substances were administered as a daily SC injection of 0.2 ml of either an oil or saline solution. To equate vehicle exposure, all animals received both oil and saline vehicles. Drug treatments continued for 6 weeks, with daily injections administered to the animal 1 h
2.1.5. Hole board procedures The hole board paradigm is a spatial learning task relying mostly on reference memory. However, it is unique among reference memory tasks in that the paradigm employs massed trials, and the focus is on acquisition rates. Specifically, it represents a form of learning in which a hungry rat must learn and remember in a single session the extra-maze cues signaling the location of food (BrosnanWatters and Wozniak, 1997; Douma et al., 1998). The animals were food restricted for habituation or testing by having food removed from its home cage for the 23 h prior to a session. Typically, this food restriction regimen produces minimal loss of body weight over the course of an experiment.
2.1.5.2. Testing. At least 2 days elapsed after the final habituation session before a test session was conducted. Testing was conducted on one day for most animals, although some rats required two consecutive days to complete the test. A food-baited hole was identified by extra-maze cues. To equate food odor cues, cereal pieces were placed in three of the holes under the wiremesh screen that rendered the food inaccessible. The food was placed on top of the screen for the fourth, correct hole. Location of the correct hole was the same on every trial for an animal, but the hole serving as the correct choice was counterbalanced within and between groups. A trial began by introducing the rat into the apparatus via the cylindrical start box. After 5 s,
1244
the cylinder was lifted and the rat was allowed to move about the apparatus until it found the hole containing the accessible food. A correct choice was defined by the animal selecting first, by clearly inserting its head into the hole, the foodbaited hole before searching in one of the other three holes. The trial was defined as incorrect when the animal clearly inserted its head into an unbaited hole prior to choosing the baited hole. On these incorrect trials, the animal was allowed to visit other holes until the accessible cereal was discovered. After eating, the rat was removed to a holding cage for a 1 min ITI during which time the apparatus was cleaned. To further obscure odor cues, the apparatus was rotated 90 for the next trial, while leaving the position of the correct hole the same relative to the extra-maze cues. Testing continued until the rat met a criterion of choosing the correct hole first on eight of nine consecutive trials (Brosnan-Watters and Wozniak, 1997) or until 60 trials had been given. Most often, criterion or the 60 maximum trials was achieved on 1 test day. However, if the rat failed to search for the food for 3 min on three consecutive trials, the session was terminated and continued the next day. 2.1.6. Elevated plus maze and open field procedure During the last week of injections and on different days, each animal was also tested in the elevated plus maze and in the open field as measures of anxiety and non-specific general activity, respectively. In neither test was the animal food restricted prior to testing. A rat was placed in the center of the elevated plus maze and allowed to move about the maze freely for 5 min. Time spent in each arm was recorded. Numbers of squares crossed in the open field during a 5 min session were recorded. 2.1.7. Endocrine measurements All females were sacrificed at the end of 6 weeks of drug treatments and trunk blood was collected and analyzed. On the final day of drug treatments, animals were injected with their normal dosages and sacrificed via guillotine 1 h later to correspond to the time after drug exposure that the animals were tested in the behavioral paradigms. Concentrations of estradiol were determined via radioimmunoassays by a method similar to one already published (Taylor et al., 1996b). Serum was obtained by centrifugation and frozen at 20 C until estrogen assays were performed with a Pantex extraction 125I kit (Catalog no. 047, Boeringer, Mannheim, Germany).
G.T. Taylor et al.
2.1.8. Statistical analyses A series of one-way analyses of variance (ANOVA) were used for statistical assessments, with Tukey’s HSD tests for post hoc comparisons of groups. Data analyzed from the hole board paradigm were numbers of trials required to achieve a criterion of eight correct choices on nine consecutive trials. Also analyzed were the numbers of squares crossed in the open field apparatus and the percentage of time spent in the open arms of the elevated plus maze. Endocrine data obtained were analyzed as pg estradiol/ml serum. Body weights for each group obtained at the end of the experiment were analyzed. The SPSS statistical program for Macintosh computers was used for all calculations. Probability value for all analyses was p < 0:05.
2.2. Results 2.2.1. Behavior The behavioral data of primary interest were the trials required to reach criterion in the hole board apparatus. Those findings are depicted in Fig. 1. Results of the one-way ANOVA on those data revealed a statistically significant value [Fð4; 35Þ ¼ 6:31, p < 0:05]. Post hoc comparisons indicated that the EB-only group required significantly fewer trials to criterion than all of the other groups. Neither Fluox-only or the EB þ Fluox groups differed from the control groups or from each other.
Fig. 1. Mean trials S:E:M: to achieve a criterion of eight of nine consecutive correct trials in the females of experiment 1. Control groups were either gonadally intact or ovariectomized (OVX) and administered vehicle only. Experimental groups were administered only estradiol benzoate (50 lg EB/kg), only fluoxetine (2 mg Fluox/kg) or both EB and Fluox. Statistically significant differences from the control group are indicated by an asterisk (*).
Spatial learning of estrogen-treated ovariectomized rats
1245
Table 1 Means S:E:M: for findings of activity in the open field and elevated plus maze, circulating estrogen and final body weights from experiment 1 with gonadally intact or ovariectomized (OVX) young adult rats administered either vehicle only or estradiol benzoate (50 lg EB/kg) and/or fluoxetine (2 mg Fluox/kg). The overall ANOVA achieved statistical significance (p < 0:05) only on the endocrine measure. Superscript letters indicate statistically reliable group differences revealed by the post hoc Tukey’s test. That is, different superscript letters indicate groups that were reliably different (p < 0:05), with the same letter indicating that the groups were not reliably different Group
N
Open field squares crossed
% time in open arms of elevated plus maze
Circulating estrogen titers (pg/ml)
Body weights (g)
Intact, veh-only OVX, veh-only OVX, EB-only OVX, Fluox-only OVX, EB þ Fluox
8 8 8 8 8
117 24 105 13 88 7 106 20 96 13
12:13 7 14:00 6 9:62 4 6:75 4 10:50 3
25 6b <10a 152 14d 15 7ab 57 4c
290 11 281 9 269 9 278 4 272 8
Group means and standard errors of the means on the open field and elevated plus maze appear in Table 1, along with the estrogen measure. Results of the ANOVA on both sets of these behavioral data revealed no statistically significant differences among the groups. The ANOVA on body weights at necropsy also did not differ significantly for the five groups. 2.2.2. Endocrinology Analyses of the circulating estrogen levels yielded a significant value [Fð4; 35Þ ¼ 33:34, p < 0:05]. Both intra-assay and inter-assay coefficients of variation were less than 5%. Post hoc comparisons indicated that the OVX animals receiving daily injections of estradiol benzoate (groups 3 and 5), naturally, had higher systemic estrogen than the OVX animals not receiving EB treatments (groups 2 and 4). Estrogen titers of the intact controls (group 1) were in between those two extremes. Notably, the OVX, EB-only (group 3) group had significantly higher estrogen levels than the OVX animals administered both estrogen and fluoxetine (group 5). Finally, vaginal smears indicated that the intact females continued to show normal 4–5 day estrous cycles throughout the study. The OVX females did not cycle, confirming successful ovariectomies.
3. Experiment 2 3.1. Methods The endocrine findings of experiment 1, coupled with reports that reuptake inhibitors are capable of suppressing circulating steroids (Rehavi et al., 2000), neurosteroids (Griffin and Mellon, 1999) and steroid-sensitive behaviors (Matuszczyk et al., 1998; Ho et al., 2001) suggested a second experi-
ment. A dose–response assessment of fluoxetine on estrogen titers was conducted using OVX female rats. No behavioral tests were conducted. The sole outcome measure was estrogen titers in OVX–estrogen restored rats after 6 weeks of daily injections with different fluoxetine dosages. 3.1.1. Subjects The experimentally naive Long–Evans female rats (N ¼ 30) were of similar age and were housed similarly to the females of experiment 1. All rats were ovariectomized and allowed 1 week recovery before treatments began. 3.1.2. Experimental design Females were assigned at random to one of five conditions (n ¼ 6 per group). Group 1 (OVX, veh only) animals received vehicle only and groups 2–5 were administered a daily dose of estradiol benzoate (50 lg EB/kg). Group 2 females (OVX, EB-only) received only the estrogen treatments. The other three groups were administered EB plus fluoxetine, either 0.5, 2.0 or 5 mg/kg. That is, group 3 (OVX, EB þ 0:5 mg Fluox) received estrogen and a low dose of fluoxetine, group 4 (OVX, EB þ 2 mg Fluox) replicated the dosage used in experiment 1, and group 5 (OVX, EB þ 5 mg Fluox) animals received a higher dose of fluoxetine. 3.1.3. Endocrine measurements Similarly to experiment 1, all animals were sacrificed 1 h after their final injection at the end of 6 weeks of drug treatments. Blood was collected and assayed for estradiol as described in experiment 1. 3.1.4. Statistical analyses As in the first experiment, the pg estradiol/ml serum findings were analyzed with a one-way
1246
ANOVA and a Tukey’s post hoc test. Probability value for both analyses was p < 0:05.
3.2. Results The outcome measure in the second experiment was circulating estrogen in the five groups of OVX females. Results of the one-way ANOVA of those data revealed a significant value [Fð4; 25Þ ¼ 52:98, p < 0:05]. Post hoc comparisons of group means indicated that every group differed from every other group, except for groups 3 and 4 (see Fig. 2). Specifically, the veh-only animals (group 1) had the lowest levels and the EB-only females (group 2) had the highest levels of estrogen. The EB þ Fluox groups had estrogen levels in between the veh-only and the EB-only groups. There were statistically reliable differences among the three EB þ Fluox groups. The EB þ 5 mg Fluox animals (group 5) had higher levels of circulating estrogen than either the EB þ 0:5 mg Fluox (group 3) or the EB þ 2 mg Fluox (group 4) conditions, which did not differ.
4. Discussion Results of experiment 1 revealed that chronic exposure to EB in OVX female rats enhanced
Fig. 2. Mean levels of circulating estrogen after 6 weeks of daily injections to the OVX females of experiment 2 with either vehicle only (veh only) or estradiol benzoate (50 g EB/kg) alone or in combination with fluoxetine (Fluox). The EB þ Fluox groups were administered fluoxetine at one of three doses, either 0.5, 2, or 5 mg Fluox/kg. Error bars are S.E.M. Statistically significant group differences are indicated by letters, with different letters indicating the groups that were reliably different (Tukey’s post hoc test, p < 0:05) and groups with the same letter indicating no significant difference.
G.T. Taylor et al.
learning and memory of a task that relies on spatial abilities. The cognitive gains with EB exposure were eliminated, however, in OVX animals administered the same EB dose in combination with the SSRI anti-depressant agent, fluoxetine. At the same time, there was the suggestion that the 2 mg fluoxetine dosage administered to the EB þ Fluox females reduced their circulating estrogen levels relative to the EB-only animals. Results of experiment 2 confirmed the capacity of various dosages of fluoxetine to suppress estrogen titers in OVX, estrogen treated rats. Compared to an EB-only group of OVX females, concomitant fluoxetine from 0.5–5 mg/kg suppressed circulating estrogen in a pattern suggesting a relation between fluoxetine dose and circulating estrogen.
4.1. Spatial abilities following estrogen treatments with or without fluoxetine The literature suggests the influence of estrogen on cognition is complex. Variables influencing whether estrogen enhances, disrupts or has no effect on spatial learning and memory include the presence or absence of functional ovaries and the nature of the task (Berry et al., 1997; Sandstrom and Williams, 2001). In ovariectomized female rodents tested in a spatial task, estrogen treatments are reported to improve working memory whereas the influence on reference memory is more uncertain (Luine et al., 1998; Wilson et al., 1999). The hole board task is a spatial reference memory task but a unique one in that learning occurs in massed trials, and acquisition rates are the primary outcome measure. Here, chronic estrogen treatments enhanced learning rates in the hole board paradigm in that the OVX EB-only group required the fewest trials to criterion. As with a previous report of similar locomotor activity in the open field by OVX female rats untreated or treated chronically with estrogen (Luine et al., 1998), our EB-only animals were similarly active in the open field to the other groups. The EB-only females also spent similar time in the open arms of the elevated plus maze. Although not conclusive, these data point to the enhanced performance on the hole board being specific to learning and memory rather than to non-specific behavioral activation or anxiolytic properties of the treatments (Scott et al., 1994). The other OVX groups all performed similarly in the hole board paradigm. Specifically, OVX rats administered fluoxetine only or, more notably, estrogen plus fluoxetine did not differ significantly from each other or from OVX controls. These data are in agreement with previous stu-
Spatial learning of estrogen-treated ovariectomized rats
dies, generally using higher doses, reporting that fluoxetine had no reliable influence on acquisition and recall of a spatial task (Lee et al., 1992; Jansen and Andrews, 1994; McEwen et al., 1997; Stewart and Reid, 2000). It may appear contradictory that the spatial performance of the group with functional ovaries also did not differ from the vehicle-only OVX controls. However, the intact animals were tested at random times during their estrous cycle, and it is possible that the presence of endogenous progesterone attenuated any gains from the presence of endogenous estrogen (Loscher et al., 1992; Farr et al., 1995; Laconi et al., 2001).
4.2. Relation of exogenous estradiol to estrogen titers Circulating estrogen levels were measured after 6 weeks of drug treatments. In experiment 1, the gonadally intact control group naturally had higher estrogen titers than the vehicle-only OVX controls. The intact females, however, had significantly lower estrogen levels than the two OVX groups (EB-only and EB þ Fluox) administered exogenous estrogen. The implication is that the estradiol benzoate dosage employed here (50 lg EB/kg) for OVX rats produced supraphysiologic levels of circulating estrogen. However, we suggest that the estrogen values for those two OVX groups may be only moderately higher than the circulating estrogen of intact females at proestrus. Our intact group was tested for behaviors and for hormones at random times of their estrous cycle. As a result, the mean estrogen value reported in Table 1 for the intact group represents females at different stages of their estrous cycles rather than proestrous values. In addition, there is no clear agreement in the literature on normal estrogen values in intact rats. The range reported for estrogen titers at proestrus is from slightly over 20 pg/ml to near 100 pg/ml serum (Butcher et al., 1973; Schwartz, 1974; Watanabe et al., 1990; Loscher et al., 1992; Freeman, 1994; Bimonte and Denenberg, 1999). Nonetheless, it is important to acknowledge the limitations of comparing our females to intact, cycling females. Females here were OVX and exposed daily for weeks to the same, high dosage of estradiol and with no progesterone.
4.3. Influence of fluoxetine on systemic estrogen The endocrine results from experiment 1 also indicated that the EB þ Fluox group had significantly lower estrogen titers than the EB-only group.
1247
Because the two groups were administered the same dose of estrogen, the suggestion was of fluoxetine suppressing circulating estrogen. Experiment 2 was conducted to further examine this suggestion using groups administered different doses of fluoxetine. The goal was to determine (a) if the estrogen findings of the EB þ Fluox group in experiment 1 were anomalous, and (b) if a higher or a lower fluoxetine dosage would influence estrogen levels differently. Experimentally naive groups of OVX rats were administered the same 50 lg EB/kg dose and exposure duration as used for experiment 1. The difference was concurrent injections of fluoxetine at dosages of 0.5, 2.0 or 5.0 mg/kg. No behavioral tests were conducted in experiment 2, and blood was obtained at the end of the experiment for hormone analyses. Results confirmed the endocrine findings of experiment 1. The EB-only animals of experiment 2 had significantly higher circulating estrogen than all three EB þ Fluox groups. Moreover, the EB þ 5 mg Fluox females had significantly higher estrogen levels than the other two EB þ Fluox groups. The suggestion is of a dose–response relation of fluoxetine to estrogen, however, one in which the highest dose of fluoxetine produced the smallest decreases in circulating estrogen. There is at least one other report of a reverse dose–response relation between estrogen and serotonin (Jackson and Uphouse, 1998) in which a smaller estrogen amount induced a greater behavioral change in serotonin-sensitive lordosis than larger doses. There are more reports that serotonergic agents can modify estrogen titers, and vice versa (Maswood et al., 1999; Raap et al., 2000). Because changes in circulating estrogen in the periphery are reflected in similar changes in estrogen concentrations in brain tissues (Morissette et al., 1992; Woolley et al., 1993), the suggestion is that serotonergic drugs may attenuate the expression of estrogen-sensitive behaviors. Indirect evidence includes reports that fluoxetine decreased estrogen-sensitive sociosexual responses (Matuszczyk et al., 1998; Ho et al., 2001). More direct evidence is that chronic administration of the SSRI fluvoxamine substantially reduced circulating estrogen levels in intact females (Rehavi et al., 2000). An unknown mechanism(s) is responsible for estrogen being reduced in EB treated animals co-administered fluoxetine. Rehavi et al. (2000) speculated that their SSRI-induced suppression of estrogen in intact animals was from inhibition of luteinizing hormone and, subsequently, reduced
1248
ovarian release of estrogen. Our females were ovariectomized, focusing attention instead on data linking SSRIs and estrogen to modifications in liver metabolism. For example, SSRIs have been reported to influence cytochrome P450, a class of hepatic enzymes responsible for metabolizing many endogenous and exogenous agents, including steroids (Brosen, 1995; Harvey and Preskorn, 1996). Here, fluoxetine may have modified activity of cytochrome P450 enzymes in such a manner to reduce estrogen titers before the steroid could reach the brain and influence behavior.
4.4. Conclusions Fluoxetine suppressed circulating estrogen and estrogen-enhanced learning in OVX females treated with exogenous estrogen. These findings suggest a surprising level of complexity in drug– hormone interactions that point to the difficulty in predicting behavioral outcomes, and cognitive outcomes in particular, in females undergoing psychotherapeutic drug therapies.
Acknowledgements The research was supported in part by grants from the Alexander von Humboldt Foundation (Bonn, Germany) and the UM–St. Louis Research Incentive Fund.
References Bammer, G., 1982. Pharmacological investigations of neurotransmitter involvement in passive avoidance responding: a review and some new results. Neurosci. Biobehav. Rev. 6, 247–296. Berry, B., McMahan, R., Gallagher, M., 1997. Spatial learning and memory at defined points of the estrous cycle: effects on performance of a hippocampal-dependent task. Behav. Neurosci. 111, 267–274. Bimonte, H.A., Denenberg, V.H., 1999. Estradiol facilitates performances as working memory load increases. Psychoneuroendocrinology 24, 161–173. Brosen, K., 1995. Drug interactions and the cytochrome P450 system: the role of cytochrome P450 1A2. Clin. Pharmacokinet. 29 (Suppl 1), 20–25. Brosnan-Watters, G., Wozniak, D.F., 1997. A rotating hole board procedure for testing drug effects on spatial learning and memory in mice. Brain Res. Prot. 1, 331–338. Butcher, R., Collins, W., Fugo, N., 1973. Plasma concentrations of LH, FSH, prolactin, progesterone and estradiol-17 beta throughout the four day estrous cycle of the rat. Endocrinology 94, 1704–1708. Chesler, E.J., Juraska, J.M., 2000. Acute administration of estrogen and progesterone impairs the acquisition of the spatial Morris water maze in ovariectomized rats. Horm. Behav. 38, 234–242. Cyr, M., Landry, M., DiPaolo, T., 2000. Modulation by estrogenreceptor directed drugs of 5-hydroxytryptamine-2A receptors in rat brain. Neuropsychopharmacology 23, 69–78.
G.T. Taylor et al.
Daniel, J.M., Fader, A.J., Spencer, A.L., Dohanich, G.P., 1997. Estrogen enhances performance of female rats during acquisition of a radial arm maze. Horm. Behav. 32, 217– 225. Daniel, J.M., Roberts, S.L., Dohanich, G.P., 1999. Effects of ovarian hormones and environment on radial maze and water maze performance of female rats. Physiol. Behav. 66, 11–20. Douma, B.R., Korte, S., Buwalda, B., La Fleur, S., Bohus, B., Luiten, P., 1998. Repeated blockade of mineralocorticoid receptors, but not of glucocorticoid receptors impairs food rewarded spatial learning. Psychoneuroendocrinology 23, 33–44. Farr, S., Flood, J., Scherrer, J., Kaiser, F., Taylor, G.T., Morley, J., 1995. Effect of ovarian steroids on footshock avoidance learning and retention in female mice. Physiol. Behav. 58, 715–723. Farr, S.A., Uezu, K., Creonte, T.A., Flood, J.F., Morley, J.E., 2000. Modulation of memory processing in the cingulate cortex of mice. Pharmacol. Biochem. Behav. 65, 363–368. Fink, G., Sumner, B.E., McQueen, J.K., Wilson, H., Rosie, R., 1998. Sex steroid control of mood, mental state and memory. Clin. Exp. Pharmacol. Physiol. 25, 764–775. Flood, J.F., Cherkin, A., 1987. Fluoxetine enhances memory processing in mice. Psychopharmacology 93, 36–43. Freeman, M.E., 1994. The neuroendocrine control of the ovarian cycle of the rat. In: Knobil, E., Neill, J.D. (Eds.), The Physiology of Reproduction. Raven Press, New York, pp. 613–650. Frye, C.A., 1995. Estrus-associated decrements in a water maze task are limited to acquisition. Physiol. Behav. 57, 5–14. Gibbs, R.B., 1999. Estrogen replacement enhances acquisition of a spatial memory task and reduces deficits associated with hippocampal muscarinic receptor inhibition. Horm. Behav. 36, 222–233. Griffin, L.D., Mellon, S.H., 1999. Selective serotonin reuptake inhibitors directly alter activity of neurosteroidogenic enzymes. Proc. Nat. Acad. Sci. USA 96, 13512–13517. Harvey, A., Preskorn, S., 1996. Cytochrome P450 enzymes: interpretation of their interactions with selective serotonin reuptake inhibitors. Part II. J. Clin. Psychopharmacol. 16, 345–355. Ho, H.P., Olsson, M., Westberg, L., Melke, J., Eriksson, E., 2001. The serotonin reuptake inhibitor fluoxetine reduces sex steroid-related aggression in female rats: an animal model of premenstrual irritability? Neuropsychopharmacology 24, 502–510. Holtzman, S., Steinfels, G., 1994. Monoamine systems in the discriminative effects of spiradoline, a kappa-opioid agonist. Pharmacol. Biochem. Behav. 47, 575–578. Jackson, A., Uphouse, L., 1998. Dose-dependent effects of estradiol benzoate on 5-HT1A receptor agonist action. Brain Res. 796, 299–302. Jansen, J.H.M., Andrews, J.S., 1994. The effects of serotonergic drugs on short-term spatial memory in rats. J. Psychopharmacol. 8, 157–163. Laconi, M.R., Casteller, G., Gargiulo, P.A., Bregonzio, C., Cabrera, R.J., 2001. The anxiolytic effect of allopregnanolone is associated with gonadal hormonal status in female rats. Eur. J. Pharmacol. 417, 111–116. Lannert, H., Wirtz, P., Schuhmann, V., Galmbacher, R., 1998. Effects of estradiol on learning, memory and cerebral energy in male rats after intracerebroventricular administration of streptozocin. J. Neural Transm. 105, 1045–1063. Lee, E.H., Lin, W.R., Chen, H.Y., Shiu, W.H., Liang, K.C., 1992. Fluoxetine and 8-OH-DPAT in the lateral septum
Spatial learning of estrogen-treated ovariectomized rats
enhances and impairs retention of an inhibitory avoidance response in rats. Physiol. Behav. 51, 681–688. Loscher, W., Wahnschaffe, U., Rundfeldt, C., Honack, D., Hoppen, H., 1992. Regional alterations in brain amino acids during the estrous cycle of the rat. Neurochem. Res. 17, 973–977. Luciana, M., Collins, P.F., Depue, R., 1998. Opposing roles for dopamine and serotonin in the modulation of human spatial working memory functions. Cereb. Cortex 8, 218–226. Luine, V.N., Richards, S.T., Wu, V.Y., Beck, K.D., 1998. Estradiol enhances learning and memory in a spatial memory task and effects levels of monoaminergic neurotransmitters. Horm. Behav. 34, 149–162. Maswood, S., Truitt, W., Hotema, M., Caldarola-Pastuszka, M., Uphouse, L., 1999. Estrous cycle modulation of extracellular serotonin in mediobasal hypothalamus: role of the serotonin transporter and terminal autoreceptors. Brain Res. 831, 146–154. Matuszczyk, J.V., Larsson, K., Eriksson, E., 1998. Subchronic administration of fluoxetine impairs estrous behavior in intact female rats. Neuropsychopharmacology 19, 492–498. McEwen, B.S., Conrad, C.D., Kuroda, Y., Frankfurt, M., Magarinos, A., McKittrick, C., 1997. Prevention of stressinduced morphological and cognitive consequences. Eur. Neuropsychopharmacol. 7, S323–S328. Morissette, M., Garcia-Segura, L.M., Belanger, A., DiPaolo, T., 1992. Changes of rat striatal neuronal membrane morphology and steroid content during the estrous cycle. Neuroscience 49, 893–902. Nowakowska, E., Chodera, A., Kus, K., 1996. Anxiolytic and memory improving activity of fluoxetine. Polish J. Pharmacol. 48, 255–260. Ohl, F., Michaelis, T., Vollmann-Honsdorf, G.K., Kirschbaum, C., Fuchs, E., 2000. Effect of chronic psychosocial stress and long-term cortisol treatment on hippocampus-mediated memory and hippocampal volume: a pilot-study in tree shrews. Psychoneuroendocrinology 25, 357–363. Raap, D.K., DonCarlos, L., Garcia, F., Muma, N.A., Wolf, W.A., Battaglia, G., Van de Kar, L.D., 2000. Estrogen desensitizes 5-HT1A receptors and reduces levels of Gz, Gi1 and Gi3 proteins in the hypothalamus. Neuropharmacology 39, 1823– 1832. Rehavi, M., Attali, G., Gil-Ad, I., Weizman, A., 2000. Suppression of serum gonadal steroids in rats by chronic treatment with dopamine and serotonin reuptake inhibitors. Eur. Neuropsychopharmacol. 10, 145–150.
1249
Rubinow, D.R., Schmidt, P.J., Roca, C.A., 1998. Estrogen–serotonin interactions: implications for affective regulation. Biol. Psychiat. 44, 839–850. Sandstrom, N.J., Williams, C.L., 2001. Memory retention is modulated by acute estradiol and progesterone replacement. Behav. Neurosci. 115, 384–393. Santucci, A.C., Knott, P.J., Haroutunian, V., 1996. Excessive serotonin release, not depletion, leads to memory impairments in rats. Eur. J. Pharmacol. 295, 7–17. Schwartz, N.B., 1974. The role of FSH and LH and of their antibodies on follicle growth and on ovulation. Biol. Reprod. 10, 236–272. Scott, M., Ettenberg, A., Olster, D., 1994. Effects of alcohol on the sexual motivation of the male rat. Pharmacol. Biochem. Behav. 48, 929–934. Stewart, C.A., Reid, I.C., 2000. Repeated ECS and fluoxetine administration have equivalent effects on hippocampal synaptic plasticity. Psychopharmacology 148, 217–223. Taylor, G.T., Bardgett, M., Farr, S., Womack, S., Komitowski, D., Weiss, J., 1993. Steroidal interactions in the ageing endocrine system: absence of suppression and pathology in reproductive systems of old males from a mixed-sex, socially stressful rat colony. J. Endocrinol. 137, 115–122. Taylor, G., Bardgett, M., Csernansky, J., Early, T., Haller, J., Scherrer, J., Womack, S., 1996a. Male rat reproductive systems under chronic fluoxetine or trimipramine treatment. Physiol. Behav. 59, 479–485. Taylor, G., Bardgett, M., Farr, S., Humphrey, W., Womack, S., Weiss, J., 1996b. Aging of the brain–testicular axis: reproductive systems of healthy old males with or without endocrine stimulation. Proc. Soc. Exp. Biol. Med. 211, 69–75. Warren, S.G., Juraska, J.M., 1997. Spatial and nonspatial learning across the rat estrous cycle. Behav. Neurosci. 111, 259–266. Watanabe, G., Taya, K., Sasamoto, S., 1990. Dynamics of ovarian inhibin secretion during the oestrous cycle of the rat. J. Endocrinol. 126, 151–157. Wilson, I.A., Puolivali, J., Heikkinen, T., Riekkinen, P., 1999. Estrogen and NMDA receptor antagonism: effects upon reference and working memory. Eur. J. Pharmacol. 381, 93–99. Woolley, C.S., 1998. Estrogen-mediated structural and functional synaptic plasticity in the female rat hippocampus. Horm. Behav. 34, 140–148. Woolley, C.S., McEwen, B.S., Beatty, W.W., 1993. Roles of estradiol and progesterone in regulation of hippocampal dendritic spine density during the estrous cycle in the rat. J. Comp. Neurol. 336, 293–306.
www.elsevier.com/locate/psyneuen
Chronic fluoxetine suppresses circulating estrogen and the enhanced spatial learning of estrogen-treated ovariectomized rats George T. Taylora,*, Susan Farra,b, Klaus Klingac, Juergen Weissd a
Behavioral Neuroscience Group, University of Missouri—St. Louis, 8001 Natural Bridge Road, St. Louis, MO 63121, USA b Department of Medicine, School of Medicine, St. Louis University, St. Louis, MO 63110, USA c ¨ tsstoerungen, Universitaets-Frauenklinik, Abt. Gynaekologische Endokrinologie und Fertilita Heidelberg 6900, Germany d Zentrum fur Molekular Biologie, Heidelberg University, Heidelberg 6900, Germany Received 18 August 2003; received in revised form 5 March 2004; accepted 6 March 2004
KEYWORDS Estrogen; SSRI; Fluoxetine; Serotonin; Spatial memory
*
Summary We are interested in developing animal models to evaluate cognitive processes as influenced by the interplay of steroidal hormones and drugs commonly used in psychotherapy. Two experiments with female rats were conducted to evaluate the interaction of estrogen with the serotonin specific reuptake inhibitor (SSRI) fluoxetine on spatial learning and memory and on the endocrine system. In experiment 1, estrogen (50 lg estradiol benzoate/kg body weight) was administered SC to young adult, ovariectomized (OVX) rats either alone or in combination with fluoxetine (2 mg/kg SC). After a month, the groups were compared with appropriate OVX and gonadally intact controls on trials to criterion in a hole board spatial memory task using massed training trials. Experiment 2 was a dose–response study of the influence of fluoxetine (0.5–5 mg/kg) on circulating estrogen in OVX, estrogen treated females. Results were that the OVX females administered estrogen only reached the learning criterion significantly faster than the other groups. All other groups, including the estrogen þ fluoxetine animals, performed no better than the controls. Combining fluoxetine with estrogen also lowered circulating estrogen titers, with the least estrogen reductions being in the group receiving the highest dosage of fluoxetine. No differences among groups were found on measures of activity in an open field or for anxiety in a plus maze. Conclusions were that administration of estrogen improved spatial learning and memory in OVX rats, whereas concurrent fluoxetine exposure suppressed the levels of estrogen in circulation and eliminated the gains in spatial performance obtained from chronic estrogen exposure. # 2004 Elsevier Ltd. All rights reserved.
Corresponding author. Fax: +1-314-516-5392. E-mail address: [email protected] (G.T. Taylor).
0306-4530/$ - see front matter # 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.psyneuen.2004.03.001
1242
1. Introduction There is now substantial evidence that gonadal steroids act on a range of cortical and subcortical brain regions underlying behaviors not directly related to reproduction. Of particular interest have been findings that estrogen serves as a neuroactive steroid to influence cognitive function, likely by modifying neural morphology and modulating neurotransmitter systems involved in learning and memory (Woolley, 1998; Luine et al., 1998). Our interest is the behavior of hypoestrogenic animal models administered estrogen concurrently with drugs commonly used in psychotherapy. Here, the drug was fluoxetine, the prototypical serotonin specific reuptake inhibitor (SSRI) anti-depressant medication (Taylor et al., 1996a). The literature suggests reasons to suspect interactions among estrogen, serotonin and cognitive behaviors (Fink et al., 1998). However, even the simple two-way interactions between variables have proven resistant to straightforward generalizations. For example, does estrogen improve or disrupt the spatial learning and memory of experimental animals? There are reports with gonadally intact females that spatial abilities are worse at proestrus, when both endogenous estrogen and progesterone are elevated, than at other stages of the estrous cycle (Frye, 1995; Warren and Juraska, 1997; Daniel et al., 1999; Chesler and Juraska, 2000). Yet, exogenous estrogen administered to ovariectomized (OVX) animals is often reported to enhance spatial memory (Daniel et al., 1997; Luine et al., 1998; Gibbs, 1999; cf. Daniel et al., 1999). The suggestion is that chronic, continuous estrogen improves performance whereas shortterm modulation of estrogen during the estrous cycle is detrimental or has no effect on performance (Sandstrom and Williams, 2001). Conclusions on the influence of serotonergic agents on cognitive behaviors have also proven complex. For example, serotonin agonists are often reported to disrupt normal acquisition and recall (Santucci et al., 1996; Luciana et al., 1998; Farr et al., 2000). Yet, the findings with fluoxetine and other SSRI agents that also are presumed to increase serotonergic activity suggest little or no effects on learning and memory (Bammer, 1982; Lee et al., 1992; Stewart and Reid, 2000). Fluoxetine may even improve cognitive behavior in some paradigms (Flood and Cherkin, 1987; Nowakowska et al., 1996), perhaps due to the anxiolytic properties of SSRIs. One notable feature is that the studies of serotonin–cognitive relations have typically used males as subjects. Yet, the possibility of a differ-
G.T. Taylor et al.
ent response by females is suggested by data from both intact females and OVX, hormone-restored animals. There are changes in serotonin synthesis, release, reuptake and metabolism during the estrous cycles of intact animals (Rubinow et al., 1998; Maswood et al., 1999). OVX decreases serotonin receptor numbers in various brain regions and estrogen replacement restores receptor concentrations to those of intact females (Cyr et al., 2000). The current experiments were suggested by the possibility of different fluoxetine–cognitive behavior interactions in females with or without significant levels of circulating estrogen. In the first experiment, groups of OVX female rats were exposed chronically to estrogen or fluoxetine or both estrogen and fluoxetine. Controls were groups of vehicle-only OVX females and gonadally intact females. Testing of cognitive function employed a modified hole board paradigm designed to assess acquisition rates in a spatial memory task with massed training trials. There is evidence that the hole board is sensitive both to estrogen treatments (Lannert et al., 1998) and to normal hippocampal function (Ohl et al., 2000).
2. Experiment 1 2.1. Methods 2.1.1. Subjects Long–Evans female rats (N ¼ 40) were 3–4 months of age and had been housed individually since weaning in hanging wire cages measuring 20:5 23:5 29:5 cm. Water and Richmond Standard Lab Diet 5001 were available ad libitum except during periods of food restriction. Lighting in colony rooms was on a reversed cycle of 12 h light/dark; room temperature (20–22 C) and relative humidity (55 5%) were controlled automatically. Animal care and experimental methodologies were approved by the campus Institutional Animal Care and Use Committee and were in accordance with NIH guidelines. 2.1.2. Materials The hole board apparatus was a 66:5 cm 66:5 cm box. The walls were constructed of clear Plexiglas 50.5 cm high. The gray wooden floor had four holes, 4 cm in diameter and 4.5 cm in depth, located 12 cm from each corner. A small circular section of wire-mesh screen was located halfway into each hole. This construction allowed food to be placed either below the screen, rendering the food inaccessible, or above the screen to be
Spatial learning of estrogen-treated ovariectomized rats
1243
eaten. A cylinder, 42 cm tall and 15 cm in diameter, positioned in the center of the apparatus served as the ‘‘start box’’. The apparatus sat on the floor in the center of the room to reveal various extra-maze cues to the animal. Examples of extra-maze stimuli were various posters mounted on the walls, a set of shelves on another wall, and the experimenter standing to one side of the apparatus. The elevated plus maze consisted of two open arms and two closed arms, each 86.25 cm in length and 13.75 cm wide. The closed arms had walls 16.25 cm in height. The apparatus was constructed of wood, painted black and elevated to a height of 63 cm above the floor. The 90 120 cm open field apparatus has been described in detail earlier (Taylor et al., 1996a). Estradiol benzoate (EB) was purchased from Sigma Chemical Company (St. Louis, MO) and was suspended in olive oil for injection. Fluoxetine hydrochloride was donated by Eli Lilly & Co. (Indianapolis, IN) and was solubilized in 0.9% saline solution.
before a behavioral session. Habituation sessions in the hole board apparatus began on week 4 with testing being conducted during week 5. Elevated plus maze and open field testing were done during the final week. Animals were sacrificed by decapitation with a guillotine at the end of week 6, and trunk blood was obtained for hormone assays.
2.1.3. Surgery The animals were either ovariectomized (OVX,n ¼ 32) or sham operated (Intacts, n ¼ 8) 1 week prior to the start of drug treatment. Vaginal smears were used to assess that the ovariectomies were successful and that the intact females continued to cycle normally.
2.1.5.1. Habituation. The hole board paradigm requires extensive exposure to the apparatus and procedures prior to testing. Beginning during week 4 of injections, each rat received habituation sessions of 15–30 min distributed over different days. The initial sessions consisted of allowing the rat to roam freely about the apparatus to eat pieces of a sweet breakfast cereal (Kellog’s Froot Loops) cut into quarters located in each of the four holes. When the animal was reliably finding and eating the food, a final habituation session was given on a single day to adapt the rat to repeated trials in the apparatus. All the holes were baited with accessible food, but otherwise the 10 trials administered were the same as described below in the testing phase.
2.1.4. Experimental design Rats were assigned at random to one of five treatment conditions (n ¼ 8 per group) to be injected SC daily with doses of estradiol benzoate (50 lg EB/kg body weight), fluoxetine (2 mg Fluox/kg body weight) or vehicle only. Group 1 females (Intact, veh-only) were gonadally intact and the other four groups were ovariectomized. Groups 2–4 females were administered, respectively, only vehicle (OVX, veh-only), only estrogen (OVX, EBonly) or only fluoxetine (OVX, Fluox-only). Group 5 animals received both estradiol and fluoxetine (OVX, EB þ Fluox). The 2 mg dosage of fluoxetine is at the low end of the range of doses typically administered to rats (Holtzman and Steinfels, 1994; Matuszczyk et al., 1998). We have used the 50 lg EB dosage in past studies to restore function to OVX females (e.g.,Taylor et al., 1993). All substances were administered as a daily SC injection of 0.2 ml of either an oil or saline solution. To equate vehicle exposure, all animals received both oil and saline vehicles. Drug treatments continued for 6 weeks, with daily injections administered to the animal 1 h
2.1.5. Hole board procedures The hole board paradigm is a spatial learning task relying mostly on reference memory. However, it is unique among reference memory tasks in that the paradigm employs massed trials, and the focus is on acquisition rates. Specifically, it represents a form of learning in which a hungry rat must learn and remember in a single session the extra-maze cues signaling the location of food (BrosnanWatters and Wozniak, 1997; Douma et al., 1998). The animals were food restricted for habituation or testing by having food removed from its home cage for the 23 h prior to a session. Typically, this food restriction regimen produces minimal loss of body weight over the course of an experiment.
2.1.5.2. Testing. At least 2 days elapsed after the final habituation session before a test session was conducted. Testing was conducted on one day for most animals, although some rats required two consecutive days to complete the test. A food-baited hole was identified by extra-maze cues. To equate food odor cues, cereal pieces were placed in three of the holes under the wiremesh screen that rendered the food inaccessible. The food was placed on top of the screen for the fourth, correct hole. Location of the correct hole was the same on every trial for an animal, but the hole serving as the correct choice was counterbalanced within and between groups. A trial began by introducing the rat into the apparatus via the cylindrical start box. After 5 s,
1244
the cylinder was lifted and the rat was allowed to move about the apparatus until it found the hole containing the accessible food. A correct choice was defined by the animal selecting first, by clearly inserting its head into the hole, the foodbaited hole before searching in one of the other three holes. The trial was defined as incorrect when the animal clearly inserted its head into an unbaited hole prior to choosing the baited hole. On these incorrect trials, the animal was allowed to visit other holes until the accessible cereal was discovered. After eating, the rat was removed to a holding cage for a 1 min ITI during which time the apparatus was cleaned. To further obscure odor cues, the apparatus was rotated 90 for the next trial, while leaving the position of the correct hole the same relative to the extra-maze cues. Testing continued until the rat met a criterion of choosing the correct hole first on eight of nine consecutive trials (Brosnan-Watters and Wozniak, 1997) or until 60 trials had been given. Most often, criterion or the 60 maximum trials was achieved on 1 test day. However, if the rat failed to search for the food for 3 min on three consecutive trials, the session was terminated and continued the next day. 2.1.6. Elevated plus maze and open field procedure During the last week of injections and on different days, each animal was also tested in the elevated plus maze and in the open field as measures of anxiety and non-specific general activity, respectively. In neither test was the animal food restricted prior to testing. A rat was placed in the center of the elevated plus maze and allowed to move about the maze freely for 5 min. Time spent in each arm was recorded. Numbers of squares crossed in the open field during a 5 min session were recorded. 2.1.7. Endocrine measurements All females were sacrificed at the end of 6 weeks of drug treatments and trunk blood was collected and analyzed. On the final day of drug treatments, animals were injected with their normal dosages and sacrificed via guillotine 1 h later to correspond to the time after drug exposure that the animals were tested in the behavioral paradigms. Concentrations of estradiol were determined via radioimmunoassays by a method similar to one already published (Taylor et al., 1996b). Serum was obtained by centrifugation and frozen at 20 C until estrogen assays were performed with a Pantex extraction 125I kit (Catalog no. 047, Boeringer, Mannheim, Germany).
G.T. Taylor et al.
2.1.8. Statistical analyses A series of one-way analyses of variance (ANOVA) were used for statistical assessments, with Tukey’s HSD tests for post hoc comparisons of groups. Data analyzed from the hole board paradigm were numbers of trials required to achieve a criterion of eight correct choices on nine consecutive trials. Also analyzed were the numbers of squares crossed in the open field apparatus and the percentage of time spent in the open arms of the elevated plus maze. Endocrine data obtained were analyzed as pg estradiol/ml serum. Body weights for each group obtained at the end of the experiment were analyzed. The SPSS statistical program for Macintosh computers was used for all calculations. Probability value for all analyses was p < 0:05.
2.2. Results 2.2.1. Behavior The behavioral data of primary interest were the trials required to reach criterion in the hole board apparatus. Those findings are depicted in Fig. 1. Results of the one-way ANOVA on those data revealed a statistically significant value [Fð4; 35Þ ¼ 6:31, p < 0:05]. Post hoc comparisons indicated that the EB-only group required significantly fewer trials to criterion than all of the other groups. Neither Fluox-only or the EB þ Fluox groups differed from the control groups or from each other.
Fig. 1. Mean trials S:E:M: to achieve a criterion of eight of nine consecutive correct trials in the females of experiment 1. Control groups were either gonadally intact or ovariectomized (OVX) and administered vehicle only. Experimental groups were administered only estradiol benzoate (50 lg EB/kg), only fluoxetine (2 mg Fluox/kg) or both EB and Fluox. Statistically significant differences from the control group are indicated by an asterisk (*).
Spatial learning of estrogen-treated ovariectomized rats
1245
Table 1 Means S:E:M: for findings of activity in the open field and elevated plus maze, circulating estrogen and final body weights from experiment 1 with gonadally intact or ovariectomized (OVX) young adult rats administered either vehicle only or estradiol benzoate (50 lg EB/kg) and/or fluoxetine (2 mg Fluox/kg). The overall ANOVA achieved statistical significance (p < 0:05) only on the endocrine measure. Superscript letters indicate statistically reliable group differences revealed by the post hoc Tukey’s test. That is, different superscript letters indicate groups that were reliably different (p < 0:05), with the same letter indicating that the groups were not reliably different Group
N
Open field squares crossed
% time in open arms of elevated plus maze
Circulating estrogen titers (pg/ml)
Body weights (g)
Intact, veh-only OVX, veh-only OVX, EB-only OVX, Fluox-only OVX, EB þ Fluox
8 8 8 8 8
117 24 105 13 88 7 106 20 96 13
12:13 7 14:00 6 9:62 4 6:75 4 10:50 3
25 6b <10a 152 14d 15 7ab 57 4c
290 11 281 9 269 9 278 4 272 8
Group means and standard errors of the means on the open field and elevated plus maze appear in Table 1, along with the estrogen measure. Results of the ANOVA on both sets of these behavioral data revealed no statistically significant differences among the groups. The ANOVA on body weights at necropsy also did not differ significantly for the five groups. 2.2.2. Endocrinology Analyses of the circulating estrogen levels yielded a significant value [Fð4; 35Þ ¼ 33:34, p < 0:05]. Both intra-assay and inter-assay coefficients of variation were less than 5%. Post hoc comparisons indicated that the OVX animals receiving daily injections of estradiol benzoate (groups 3 and 5), naturally, had higher systemic estrogen than the OVX animals not receiving EB treatments (groups 2 and 4). Estrogen titers of the intact controls (group 1) were in between those two extremes. Notably, the OVX, EB-only (group 3) group had significantly higher estrogen levels than the OVX animals administered both estrogen and fluoxetine (group 5). Finally, vaginal smears indicated that the intact females continued to show normal 4–5 day estrous cycles throughout the study. The OVX females did not cycle, confirming successful ovariectomies.
3. Experiment 2 3.1. Methods The endocrine findings of experiment 1, coupled with reports that reuptake inhibitors are capable of suppressing circulating steroids (Rehavi et al., 2000), neurosteroids (Griffin and Mellon, 1999) and steroid-sensitive behaviors (Matuszczyk et al., 1998; Ho et al., 2001) suggested a second experi-
ment. A dose–response assessment of fluoxetine on estrogen titers was conducted using OVX female rats. No behavioral tests were conducted. The sole outcome measure was estrogen titers in OVX–estrogen restored rats after 6 weeks of daily injections with different fluoxetine dosages. 3.1.1. Subjects The experimentally naive Long–Evans female rats (N ¼ 30) were of similar age and were housed similarly to the females of experiment 1. All rats were ovariectomized and allowed 1 week recovery before treatments began. 3.1.2. Experimental design Females were assigned at random to one of five conditions (n ¼ 6 per group). Group 1 (OVX, veh only) animals received vehicle only and groups 2–5 were administered a daily dose of estradiol benzoate (50 lg EB/kg). Group 2 females (OVX, EB-only) received only the estrogen treatments. The other three groups were administered EB plus fluoxetine, either 0.5, 2.0 or 5 mg/kg. That is, group 3 (OVX, EB þ 0:5 mg Fluox) received estrogen and a low dose of fluoxetine, group 4 (OVX, EB þ 2 mg Fluox) replicated the dosage used in experiment 1, and group 5 (OVX, EB þ 5 mg Fluox) animals received a higher dose of fluoxetine. 3.1.3. Endocrine measurements Similarly to experiment 1, all animals were sacrificed 1 h after their final injection at the end of 6 weeks of drug treatments. Blood was collected and assayed for estradiol as described in experiment 1. 3.1.4. Statistical analyses As in the first experiment, the pg estradiol/ml serum findings were analyzed with a one-way
1246
ANOVA and a Tukey’s post hoc test. Probability value for both analyses was p < 0:05.
3.2. Results The outcome measure in the second experiment was circulating estrogen in the five groups of OVX females. Results of the one-way ANOVA of those data revealed a significant value [Fð4; 25Þ ¼ 52:98, p < 0:05]. Post hoc comparisons of group means indicated that every group differed from every other group, except for groups 3 and 4 (see Fig. 2). Specifically, the veh-only animals (group 1) had the lowest levels and the EB-only females (group 2) had the highest levels of estrogen. The EB þ Fluox groups had estrogen levels in between the veh-only and the EB-only groups. There were statistically reliable differences among the three EB þ Fluox groups. The EB þ 5 mg Fluox animals (group 5) had higher levels of circulating estrogen than either the EB þ 0:5 mg Fluox (group 3) or the EB þ 2 mg Fluox (group 4) conditions, which did not differ.
4. Discussion Results of experiment 1 revealed that chronic exposure to EB in OVX female rats enhanced
Fig. 2. Mean levels of circulating estrogen after 6 weeks of daily injections to the OVX females of experiment 2 with either vehicle only (veh only) or estradiol benzoate (50 g EB/kg) alone or in combination with fluoxetine (Fluox). The EB þ Fluox groups were administered fluoxetine at one of three doses, either 0.5, 2, or 5 mg Fluox/kg. Error bars are S.E.M. Statistically significant group differences are indicated by letters, with different letters indicating the groups that were reliably different (Tukey’s post hoc test, p < 0:05) and groups with the same letter indicating no significant difference.
G.T. Taylor et al.
learning and memory of a task that relies on spatial abilities. The cognitive gains with EB exposure were eliminated, however, in OVX animals administered the same EB dose in combination with the SSRI anti-depressant agent, fluoxetine. At the same time, there was the suggestion that the 2 mg fluoxetine dosage administered to the EB þ Fluox females reduced their circulating estrogen levels relative to the EB-only animals. Results of experiment 2 confirmed the capacity of various dosages of fluoxetine to suppress estrogen titers in OVX, estrogen treated rats. Compared to an EB-only group of OVX females, concomitant fluoxetine from 0.5–5 mg/kg suppressed circulating estrogen in a pattern suggesting a relation between fluoxetine dose and circulating estrogen.
4.1. Spatial abilities following estrogen treatments with or without fluoxetine The literature suggests the influence of estrogen on cognition is complex. Variables influencing whether estrogen enhances, disrupts or has no effect on spatial learning and memory include the presence or absence of functional ovaries and the nature of the task (Berry et al., 1997; Sandstrom and Williams, 2001). In ovariectomized female rodents tested in a spatial task, estrogen treatments are reported to improve working memory whereas the influence on reference memory is more uncertain (Luine et al., 1998; Wilson et al., 1999). The hole board task is a spatial reference memory task but a unique one in that learning occurs in massed trials, and acquisition rates are the primary outcome measure. Here, chronic estrogen treatments enhanced learning rates in the hole board paradigm in that the OVX EB-only group required the fewest trials to criterion. As with a previous report of similar locomotor activity in the open field by OVX female rats untreated or treated chronically with estrogen (Luine et al., 1998), our EB-only animals were similarly active in the open field to the other groups. The EB-only females also spent similar time in the open arms of the elevated plus maze. Although not conclusive, these data point to the enhanced performance on the hole board being specific to learning and memory rather than to non-specific behavioral activation or anxiolytic properties of the treatments (Scott et al., 1994). The other OVX groups all performed similarly in the hole board paradigm. Specifically, OVX rats administered fluoxetine only or, more notably, estrogen plus fluoxetine did not differ significantly from each other or from OVX controls. These data are in agreement with previous stu-
Spatial learning of estrogen-treated ovariectomized rats
dies, generally using higher doses, reporting that fluoxetine had no reliable influence on acquisition and recall of a spatial task (Lee et al., 1992; Jansen and Andrews, 1994; McEwen et al., 1997; Stewart and Reid, 2000). It may appear contradictory that the spatial performance of the group with functional ovaries also did not differ from the vehicle-only OVX controls. However, the intact animals were tested at random times during their estrous cycle, and it is possible that the presence of endogenous progesterone attenuated any gains from the presence of endogenous estrogen (Loscher et al., 1992; Farr et al., 1995; Laconi et al., 2001).
4.2. Relation of exogenous estradiol to estrogen titers Circulating estrogen levels were measured after 6 weeks of drug treatments. In experiment 1, the gonadally intact control group naturally had higher estrogen titers than the vehicle-only OVX controls. The intact females, however, had significantly lower estrogen levels than the two OVX groups (EB-only and EB þ Fluox) administered exogenous estrogen. The implication is that the estradiol benzoate dosage employed here (50 lg EB/kg) for OVX rats produced supraphysiologic levels of circulating estrogen. However, we suggest that the estrogen values for those two OVX groups may be only moderately higher than the circulating estrogen of intact females at proestrus. Our intact group was tested for behaviors and for hormones at random times of their estrous cycle. As a result, the mean estrogen value reported in Table 1 for the intact group represents females at different stages of their estrous cycles rather than proestrous values. In addition, there is no clear agreement in the literature on normal estrogen values in intact rats. The range reported for estrogen titers at proestrus is from slightly over 20 pg/ml to near 100 pg/ml serum (Butcher et al., 1973; Schwartz, 1974; Watanabe et al., 1990; Loscher et al., 1992; Freeman, 1994; Bimonte and Denenberg, 1999). Nonetheless, it is important to acknowledge the limitations of comparing our females to intact, cycling females. Females here were OVX and exposed daily for weeks to the same, high dosage of estradiol and with no progesterone.
4.3. Influence of fluoxetine on systemic estrogen The endocrine results from experiment 1 also indicated that the EB þ Fluox group had significantly lower estrogen titers than the EB-only group.
1247
Because the two groups were administered the same dose of estrogen, the suggestion was of fluoxetine suppressing circulating estrogen. Experiment 2 was conducted to further examine this suggestion using groups administered different doses of fluoxetine. The goal was to determine (a) if the estrogen findings of the EB þ Fluox group in experiment 1 were anomalous, and (b) if a higher or a lower fluoxetine dosage would influence estrogen levels differently. Experimentally naive groups of OVX rats were administered the same 50 lg EB/kg dose and exposure duration as used for experiment 1. The difference was concurrent injections of fluoxetine at dosages of 0.5, 2.0 or 5.0 mg/kg. No behavioral tests were conducted in experiment 2, and blood was obtained at the end of the experiment for hormone analyses. Results confirmed the endocrine findings of experiment 1. The EB-only animals of experiment 2 had significantly higher circulating estrogen than all three EB þ Fluox groups. Moreover, the EB þ 5 mg Fluox females had significantly higher estrogen levels than the other two EB þ Fluox groups. The suggestion is of a dose–response relation of fluoxetine to estrogen, however, one in which the highest dose of fluoxetine produced the smallest decreases in circulating estrogen. There is at least one other report of a reverse dose–response relation between estrogen and serotonin (Jackson and Uphouse, 1998) in which a smaller estrogen amount induced a greater behavioral change in serotonin-sensitive lordosis than larger doses. There are more reports that serotonergic agents can modify estrogen titers, and vice versa (Maswood et al., 1999; Raap et al., 2000). Because changes in circulating estrogen in the periphery are reflected in similar changes in estrogen concentrations in brain tissues (Morissette et al., 1992; Woolley et al., 1993), the suggestion is that serotonergic drugs may attenuate the expression of estrogen-sensitive behaviors. Indirect evidence includes reports that fluoxetine decreased estrogen-sensitive sociosexual responses (Matuszczyk et al., 1998; Ho et al., 2001). More direct evidence is that chronic administration of the SSRI fluvoxamine substantially reduced circulating estrogen levels in intact females (Rehavi et al., 2000). An unknown mechanism(s) is responsible for estrogen being reduced in EB treated animals co-administered fluoxetine. Rehavi et al. (2000) speculated that their SSRI-induced suppression of estrogen in intact animals was from inhibition of luteinizing hormone and, subsequently, reduced
1248
ovarian release of estrogen. Our females were ovariectomized, focusing attention instead on data linking SSRIs and estrogen to modifications in liver metabolism. For example, SSRIs have been reported to influence cytochrome P450, a class of hepatic enzymes responsible for metabolizing many endogenous and exogenous agents, including steroids (Brosen, 1995; Harvey and Preskorn, 1996). Here, fluoxetine may have modified activity of cytochrome P450 enzymes in such a manner to reduce estrogen titers before the steroid could reach the brain and influence behavior.
4.4. Conclusions Fluoxetine suppressed circulating estrogen and estrogen-enhanced learning in OVX females treated with exogenous estrogen. These findings suggest a surprising level of complexity in drug– hormone interactions that point to the difficulty in predicting behavioral outcomes, and cognitive outcomes in particular, in females undergoing psychotherapeutic drug therapies.
Acknowledgements The research was supported in part by grants from the Alexander von Humboldt Foundation (Bonn, Germany) and the UM–St. Louis Research Incentive Fund.
References Bammer, G., 1982. Pharmacological investigations of neurotransmitter involvement in passive avoidance responding: a review and some new results. Neurosci. Biobehav. Rev. 6, 247–296. Berry, B., McMahan, R., Gallagher, M., 1997. Spatial learning and memory at defined points of the estrous cycle: effects on performance of a hippocampal-dependent task. Behav. Neurosci. 111, 267–274. Bimonte, H.A., Denenberg, V.H., 1999. Estradiol facilitates performances as working memory load increases. Psychoneuroendocrinology 24, 161–173. Brosen, K., 1995. Drug interactions and the cytochrome P450 system: the role of cytochrome P450 1A2. Clin. Pharmacokinet. 29 (Suppl 1), 20–25. Brosnan-Watters, G., Wozniak, D.F., 1997. A rotating hole board procedure for testing drug effects on spatial learning and memory in mice. Brain Res. Prot. 1, 331–338. Butcher, R., Collins, W., Fugo, N., 1973. Plasma concentrations of LH, FSH, prolactin, progesterone and estradiol-17 beta throughout the four day estrous cycle of the rat. Endocrinology 94, 1704–1708. Chesler, E.J., Juraska, J.M., 2000. Acute administration of estrogen and progesterone impairs the acquisition of the spatial Morris water maze in ovariectomized rats. Horm. Behav. 38, 234–242. Cyr, M., Landry, M., DiPaolo, T., 2000. Modulation by estrogenreceptor directed drugs of 5-hydroxytryptamine-2A receptors in rat brain. Neuropsychopharmacology 23, 69–78.
G.T. Taylor et al.
Daniel, J.M., Fader, A.J., Spencer, A.L., Dohanich, G.P., 1997. Estrogen enhances performance of female rats during acquisition of a radial arm maze. Horm. Behav. 32, 217– 225. Daniel, J.M., Roberts, S.L., Dohanich, G.P., 1999. Effects of ovarian hormones and environment on radial maze and water maze performance of female rats. Physiol. Behav. 66, 11–20. Douma, B.R., Korte, S., Buwalda, B., La Fleur, S., Bohus, B., Luiten, P., 1998. Repeated blockade of mineralocorticoid receptors, but not of glucocorticoid receptors impairs food rewarded spatial learning. Psychoneuroendocrinology 23, 33–44. Farr, S., Flood, J., Scherrer, J., Kaiser, F., Taylor, G.T., Morley, J., 1995. Effect of ovarian steroids on footshock avoidance learning and retention in female mice. Physiol. Behav. 58, 715–723. Farr, S.A., Uezu, K., Creonte, T.A., Flood, J.F., Morley, J.E., 2000. Modulation of memory processing in the cingulate cortex of mice. Pharmacol. Biochem. Behav. 65, 363–368. Fink, G., Sumner, B.E., McQueen, J.K., Wilson, H., Rosie, R., 1998. Sex steroid control of mood, mental state and memory. Clin. Exp. Pharmacol. Physiol. 25, 764–775. Flood, J.F., Cherkin, A., 1987. Fluoxetine enhances memory processing in mice. Psychopharmacology 93, 36–43. Freeman, M.E., 1994. The neuroendocrine control of the ovarian cycle of the rat. In: Knobil, E., Neill, J.D. (Eds.), The Physiology of Reproduction. Raven Press, New York, pp. 613–650. Frye, C.A., 1995. Estrus-associated decrements in a water maze task are limited to acquisition. Physiol. Behav. 57, 5–14. Gibbs, R.B., 1999. Estrogen replacement enhances acquisition of a spatial memory task and reduces deficits associated with hippocampal muscarinic receptor inhibition. Horm. Behav. 36, 222–233. Griffin, L.D., Mellon, S.H., 1999. Selective serotonin reuptake inhibitors directly alter activity of neurosteroidogenic enzymes. Proc. Nat. Acad. Sci. USA 96, 13512–13517. Harvey, A., Preskorn, S., 1996. Cytochrome P450 enzymes: interpretation of their interactions with selective serotonin reuptake inhibitors. Part II. J. Clin. Psychopharmacol. 16, 345–355. Ho, H.P., Olsson, M., Westberg, L., Melke, J., Eriksson, E., 2001. The serotonin reuptake inhibitor fluoxetine reduces sex steroid-related aggression in female rats: an animal model of premenstrual irritability? Neuropsychopharmacology 24, 502–510. Holtzman, S., Steinfels, G., 1994. Monoamine systems in the discriminative effects of spiradoline, a kappa-opioid agonist. Pharmacol. Biochem. Behav. 47, 575–578. Jackson, A., Uphouse, L., 1998. Dose-dependent effects of estradiol benzoate on 5-HT1A receptor agonist action. Brain Res. 796, 299–302. Jansen, J.H.M., Andrews, J.S., 1994. The effects of serotonergic drugs on short-term spatial memory in rats. J. Psychopharmacol. 8, 157–163. Laconi, M.R., Casteller, G., Gargiulo, P.A., Bregonzio, C., Cabrera, R.J., 2001. The anxiolytic effect of allopregnanolone is associated with gonadal hormonal status in female rats. Eur. J. Pharmacol. 417, 111–116. Lannert, H., Wirtz, P., Schuhmann, V., Galmbacher, R., 1998. Effects of estradiol on learning, memory and cerebral energy in male rats after intracerebroventricular administration of streptozocin. J. Neural Transm. 105, 1045–1063. Lee, E.H., Lin, W.R., Chen, H.Y., Shiu, W.H., Liang, K.C., 1992. Fluoxetine and 8-OH-DPAT in the lateral septum
Spatial learning of estrogen-treated ovariectomized rats
enhances and impairs retention of an inhibitory avoidance response in rats. Physiol. Behav. 51, 681–688. Loscher, W., Wahnschaffe, U., Rundfeldt, C., Honack, D., Hoppen, H., 1992. Regional alterations in brain amino acids during the estrous cycle of the rat. Neurochem. Res. 17, 973–977. Luciana, M., Collins, P.F., Depue, R., 1998. Opposing roles for dopamine and serotonin in the modulation of human spatial working memory functions. Cereb. Cortex 8, 218–226. Luine, V.N., Richards, S.T., Wu, V.Y., Beck, K.D., 1998. Estradiol enhances learning and memory in a spatial memory task and effects levels of monoaminergic neurotransmitters. Horm. Behav. 34, 149–162. Maswood, S., Truitt, W., Hotema, M., Caldarola-Pastuszka, M., Uphouse, L., 1999. Estrous cycle modulation of extracellular serotonin in mediobasal hypothalamus: role of the serotonin transporter and terminal autoreceptors. Brain Res. 831, 146–154. Matuszczyk, J.V., Larsson, K., Eriksson, E., 1998. Subchronic administration of fluoxetine impairs estrous behavior in intact female rats. Neuropsychopharmacology 19, 492–498. McEwen, B.S., Conrad, C.D., Kuroda, Y., Frankfurt, M., Magarinos, A., McKittrick, C., 1997. Prevention of stressinduced morphological and cognitive consequences. Eur. Neuropsychopharmacol. 7, S323–S328. Morissette, M., Garcia-Segura, L.M., Belanger, A., DiPaolo, T., 1992. Changes of rat striatal neuronal membrane morphology and steroid content during the estrous cycle. Neuroscience 49, 893–902. Nowakowska, E., Chodera, A., Kus, K., 1996. Anxiolytic and memory improving activity of fluoxetine. Polish J. Pharmacol. 48, 255–260. Ohl, F., Michaelis, T., Vollmann-Honsdorf, G.K., Kirschbaum, C., Fuchs, E., 2000. Effect of chronic psychosocial stress and long-term cortisol treatment on hippocampus-mediated memory and hippocampal volume: a pilot-study in tree shrews. Psychoneuroendocrinology 25, 357–363. Raap, D.K., DonCarlos, L., Garcia, F., Muma, N.A., Wolf, W.A., Battaglia, G., Van de Kar, L.D., 2000. Estrogen desensitizes 5-HT1A receptors and reduces levels of Gz, Gi1 and Gi3 proteins in the hypothalamus. Neuropharmacology 39, 1823– 1832. Rehavi, M., Attali, G., Gil-Ad, I., Weizman, A., 2000. Suppression of serum gonadal steroids in rats by chronic treatment with dopamine and serotonin reuptake inhibitors. Eur. Neuropsychopharmacol. 10, 145–150.
1249
Rubinow, D.R., Schmidt, P.J., Roca, C.A., 1998. Estrogen–serotonin interactions: implications for affective regulation. Biol. Psychiat. 44, 839–850. Sandstrom, N.J., Williams, C.L., 2001. Memory retention is modulated by acute estradiol and progesterone replacement. Behav. Neurosci. 115, 384–393. Santucci, A.C., Knott, P.J., Haroutunian, V., 1996. Excessive serotonin release, not depletion, leads to memory impairments in rats. Eur. J. Pharmacol. 295, 7–17. Schwartz, N.B., 1974. The role of FSH and LH and of their antibodies on follicle growth and on ovulation. Biol. Reprod. 10, 236–272. Scott, M., Ettenberg, A., Olster, D., 1994. Effects of alcohol on the sexual motivation of the male rat. Pharmacol. Biochem. Behav. 48, 929–934. Stewart, C.A., Reid, I.C., 2000. Repeated ECS and fluoxetine administration have equivalent effects on hippocampal synaptic plasticity. Psychopharmacology 148, 217–223. Taylor, G.T., Bardgett, M., Farr, S., Womack, S., Komitowski, D., Weiss, J., 1993. Steroidal interactions in the ageing endocrine system: absence of suppression and pathology in reproductive systems of old males from a mixed-sex, socially stressful rat colony. J. Endocrinol. 137, 115–122. Taylor, G., Bardgett, M., Csernansky, J., Early, T., Haller, J., Scherrer, J., Womack, S., 1996a. Male rat reproductive systems under chronic fluoxetine or trimipramine treatment. Physiol. Behav. 59, 479–485. Taylor, G., Bardgett, M., Farr, S., Humphrey, W., Womack, S., Weiss, J., 1996b. Aging of the brain–testicular axis: reproductive systems of healthy old males with or without endocrine stimulation. Proc. Soc. Exp. Biol. Med. 211, 69–75. Warren, S.G., Juraska, J.M., 1997. Spatial and nonspatial learning across the rat estrous cycle. Behav. Neurosci. 111, 259–266. Watanabe, G., Taya, K., Sasamoto, S., 1990. Dynamics of ovarian inhibin secretion during the oestrous cycle of the rat. J. Endocrinol. 126, 151–157. Wilson, I.A., Puolivali, J., Heikkinen, T., Riekkinen, P., 1999. Estrogen and NMDA receptor antagonism: effects upon reference and working memory. Eur. J. Pharmacol. 381, 93–99. Woolley, C.S., 1998. Estrogen-mediated structural and functional synaptic plasticity in the female rat hippocampus. Horm. Behav. 34, 140–148. Woolley, C.S., McEwen, B.S., Beatty, W.W., 1993. Roles of estradiol and progesterone in regulation of hippocampal dendritic spine density during the estrous cycle in the rat. J. Comp. Neurol. 336, 293–306.