- Email: [email protected]
Differential effects of chronic escapable versus inescapable stress on male syrian hamster (Mesocricetus auratus) reproductive behavior
Available online at www.sciencedirect.com R
Hormones and Behavior 43 (2003) 381–387
www.elsevier.com/locate/yhbeh
Differential effects of chronic escapable versus inescapable stress on male syrian hamster (Mesocricetus auratus) reproductive behavior Haley K. Holmer, Janice E. Rodman, Dana L. Helmreich, and David B. Parfitt* Department of Biology and Neuroscience Program, Middlebury College, Middlebury, VT 05753, USA Received 1 March 2002; revised 10 October 2002; accepted 14 October 2002
Abstract Stress decreases sexual activity, but it is uncertain which aspects of stress are detrimental to reproduction. This study used an escapable/inescapable stress paradigm to attempt to dissociate physical from psychological components of stress, and assess each component’s impact on reproductive behavior in the male Syrian hamster (Mesocricetus auratus). Two experiments were completed using this protocol where two animals receive the same physical stressor (an electric footshock) but differ in the psychological aspect of control. One group (executive) could terminate the shock for themselves as well as a second group (yoked) by pressing a bar. Experiment 1 demonstrated a significant increase in plasma glucocorticoids at the end of a single 90-min stress session with no difference in glucocorticoid levels between the executive and yoked groups at any time point. Experiment 2 quantified male reproductive behavior prior to and immediately following 12 days of escapable or inescapable stress in executive, yoked, and no-stress control hamsters (n ⫽ 12/group). Repeated-measures analysis of variance revealed a number of significant changes in reproductive behavior before and after stress in the three treatment groups. The most striking difference was a decrease in hit rate observed only in the animals that could not control their stress (yoked group). Hit rate in the executive males that received the exact same physical stressor but could terminate the shock by pressing a bar was nearly identical to control animals that never received any foot shock. Therefore, we conclude that coping or control can ameliorate the negative effects of stress on male reproductive behavior. © 2003 Elsevier Science (USA). All rights reserved. Keywords: Sexual behavior; Cortisol; Corticosterone; Testosterone; Coping; Control
Introduction Stress can disrupt all aspects of reproduction from motivation to performance in multiple species, including humans. Because reproductive processes are sensitive to stress, the fertility of any individual is susceptible to a variety of insults. In females, stress disrupts estrous cyclicity and causes decreases in ovarian hormone levels (NegroVilar, 1993; Rivier, 1995; Young and Korszun, 1998) and in males, stress decreases testosterone levels and reproductive behavior. In male animal models, negative effects on both appetitive and consummatory aspects of reproductive behavior can be seen after either acute or chronic stress. More
* Corresponding author. Department of Biology, Middlebury College, Middlebury, VT 05753, USA. Fax: ⫹1-802-443-2072. E-mail address: [email protected] (D.B. Parfitt).
specifically, after 1 day of stress, male rats exhibit an increased mount and ejaculation latency, suggesting impaired sexual motivation, and even greater deficits are observed after 20 days of stress (D’Aquila et al., 1994; Gorzalka et al., 1998; Menendez-Patterson et al., 1980; Retana-Marquez et al., 1996). These rats also display increased mount frequency, decreased hit rate, and decreased ejaculations—all interpreted as disrupted sexual performance (MenendezPatterson et al., 1980; Retana-Marquez et al., 1996). Most stressors, such as immobilization and swim, used by previous investigators to study the impact of stress on reproductive behavior, have both physical and psychological components that impinge on the animal. To understand how stress negatively impacts reproductive behavior, it is important to establish the pertinent aspects of a stressor that lead to decreases in reproductive processes. In the current study, we attempted to dissociate the effects of physical
0018-506X/03/$ – see front matter © 2003 Elsevier Science (USA). All rights reserved. doi:10.1016/S0018-506X(03)00009-6
382
H.K. Holmer et al. / Hormones and Behavior 43 (2003) 381–387
versus psychological components of stress on reproductive behavior by using an escapable/inescapable stress paradigm. In this model, two animals experience the same amplitude, duration, and intensity of a physical stressor (a mild-electric foot shock), but differ in the psychological aspect of control. Two animals are placed in identical chambers and experience an identical physical stressor. One animal, the executive, can terminate his stress by pressing a lever, while his yoked partner cannot. The executive animal’s bar press also terminates the stress in his yoked partner’s chamber. Thus, both animals receive the same duration of stress, but differ only in the psychological aspect of coping. This paradigm has been used by many investigators in many different experimental models, including humans, to demonstrate that the unfavorable effects of stress, such as learned helplessness (Seligman and Issacowitz, 2000), gastric ulceration (Weiss et al., 1981), and exacerbation of tumor growth (Sklar and Anisman, 1979; Visintainer et al., 1982), occur only in the yoked animals, while executive animals appear similar to no-stress controls. The current study is the first that we know of to use this escapable/inescapable stress paradigm to dissociate the physical from the psychological effects of stress on reproductive behavior. The male Syrian hamster (Mesocricetus auratus) is a useful model for investigating the relationship between stress and reproduction because it is a robust mater, its copulatory behavior is well characterized (Bunnell et al., 1977; Huck and Lisk, 1985), and the neural circuitry governing this behavior is also well understood (Wood and Swann, 2000). We predicted that the yoked animal would have a decrease in sexual behavior and performance compared to the executive and no-stress control males. Additionally, we hypothesized that yoked animals would have lower levels of testosterone than the executive and control animals. We examined reproductive behavior and testosterone levels in male hamsters exposed to 12 days of escapable or inescapable stress. Furthermore, an additional study characterized the glucocorticoid response to a single escapable/ inescapable stress session in hamsters, as the glucocorticoid response to this particular stress paradigm has not been reported in this species.
Materials and methods Experiment 1: glucocorticoid response to acute escapable or inescapable stress Animals Forty-two adult male Syrian Hamsters (⬎90 days of age) were obtained from Harlan (Indianapolis, IN) and used to determine the glucocorticoid response to the stress session. Thirty-six animals were housed two per cage and randomly assigned to one of two groups, executive or yoked (n ⫽ 18/group). The remaining six no-stress control hamsters were housed three per cage and not assigned a stress con-
dition. All hamsters were maintained on a reversed long day photoperiod (14h:10 h light:dark; lights off at 12:30 pm) and food and water were supplied ad libitum. The Middlebury College Institutional Animal Care and Use Committee approved all experimental protocols, and all animals were treated in accordance with the NIH Guide for the Care and Use of Laboratory Animals. Experimental stress paradigm All stress sessions were conducted within the first 3 h of the dark cycle under dim red light; this time period was chosen because this is when maximum reproductive behavior occurs in the hamster. Males were removed from their home cage and placed in the stress chambers (Med Associates, St. Albans, VT). Each of the boxes (20 ⫻ 25 ⫻ 21 cm) contained a stainless steel grid floor connected to a scrambled shock generator. Each box also contained a lever wired such that when the hamster in the executive box pressed the lever, the electric footshock would terminate in both the executive and the yoked boxes. The executive animal could therefore control the duration of the stress, while the yoked animal could not. The stimulus intensity used was 0.8 mA and each stress session consisted of 80 trials. If the executive hamster failed to press the lever within 30 s, the shock was automatically terminated. The intertrial interval varied between 5 and 115 s, and the mean duration of the interval was 60 s. The entire 80 trial stress session lasted approximately 90 min. Hormone determination A trunk blood sample was collected via rapid decapitation from the executive and yoked hamsters at one of three time points, 90, 120, and 180 min from the start of the stress session. To obtain an average baseline total glucocorticoid level, six no-stress control hamsters were removed from their home cages and trunk blood samples were immediately collected (these were called time 0). Plasma cortisol and corticosterone were quantified in all 42 animals via radioimmunoassay using the Coat-a-Count Cortisol and Rat Corticosterone kits obtained from Diagnostic Products Corporation (DPC, Los Angeles, CA). Finally, all samples were run together in a single assay for either cortisol or corticosterone. Statistical analysis was done with a one-way analysis of variance (ANOVA) followed by a Fisher PLSD post hoc test to determine statistically significant differences among groups. Statview computer software (SAS Institute, Cary, NC) was used for all statistical analyses. Experiment 2: male reproductive behavior after chronic escapable or inescapable stress Animals Thirty-six adult male Syrian hamsters (⬎60 days of age and weighing 98 –146 g) were obtained from Harlan (Indianapolis, IN) and used for this experiment. Animals were
H.K. Holmer et al. / Hormones and Behavior 43 (2003) 381–387 Table 1 Daily duration of shock and number of escapes of 80 trialsa
a
Stress day
Shock duration (s)
Escapes
1 2 3 4 5 6 7 8 9 10 11 12
435.7 ⫾ 49.4 217.0 ⫾ 13.7 165.5 ⫾ 8.0 184.8 ⫾ 13.2 170.5 ⫾ 10.6 168.8 ⫾ 10.4 164.2 ⫾ 11.0 181.6 ⫾ 19.0 171.4 ⫾ 11.0 207.8 ⫾ 31.8 210.2 ⫾ 40.6 228.7 ⫾ 53.5
77.0 ⫾ 0.9 80.0 ⫾ 0.0 79.9 ⫾ 0.1 79.8 ⫾ 0.2 80.0 ⫾ 0.0 80.0 ⫾ 0.0 79.9 ⫾ 0.1 79.6 ⫾ 0.4 80.0 ⫾ 0.0 78.8 ⫾ 0.9 78.3 ⫾ 1.1 78.3 ⫾ 1.6
All measures are mean ⫾ SEM.
divided into 12 cohorts (n ⫽ 3/cohort) and each hamster was randomly assigned to one of three groups, executive, yoked, or control (n ⫽ 12/group). In addition, 15 ovariectomized adult female hamsters were used as stimulus animals during the mating sessions. All hamsters were housed and treated as described above. Sexual experience Prior to the experiment, all males were given three sessions of sexual experience on three consecutive days. Each male was paired with a sexually receptive female until three ejaculations were observed or for 15 min, whichever came first. Sexual receptivity in the ovariectomized female hamsters was induced by subcutaneous injections of 10 g of estradiol benzoate 48 h and 24 h prior to, and 250 g of progesterone 4 h before the mating sessions. The sexual experience sessions and the experimental behavior testing were conducted during the dark cycle, under dim red lights, and in glass aquariums (51 ⫻ 25 ⫻ 30.5 cm) in a room separate from the home colony. Experimental stress paradigm The escapable/inescapable stress paradigm used identical parameters to Experiment 1. All stress sessions were conducted within the first 3 h of the dark cycle under dim red light. Each male received 12 consecutive days of this stress paradigm, with animals remaining in their same stress conditions (executive or yoked) throughout the 12 days. In this experiment, a third, no-stress control hamster was placed in a box similar to the executive and yoked stress chambers, but received no electric foot shock during the 12 days. For the 12 cohorts used in this experiment, the executives hit the bar to escape the foot shock an average of 79.3 ⫾ 0.3 times (of 80 trials), and the average total duration of footshock per day was 208.8 ⫾ 14.3 s. The duration of shock and number of escapes was averaged for all 12 cohorts each day and is presented in Table 1. For a complete description of the escapable/inescapable paradigm and the stress conditions, see Experiment 1.
383
Reproductive behavior testing Two days prior to the onset of 12 consecutive days of stress, all males were mated to sexual satiety to determine a prestress baseline. For this single prestress mating session, each male was paired with a separate, sexually receptive female and mated until long intromissions [a behavioral measure for the onset of sexual satiety (Arteaga et al., 2000; Parfitt and Newman, 1998)], or 1 h, whichever came first. Immediately after the mating sessions, all hamsters were returned to their cages in the home colony. The poststress mating session was performed 1 h after the 12th stress session. Both pre- and poststress mating sessions were videotaped for future analysis of all components of male hamster reproductive behavior. The following 10 behaviors were quantified on a secondby-second basis: no interaction, grooming, investigation, anogenital investigation, ectopic mount, mount, intromission, ejaculation, long intromission, and fighting. Each of these actions was assigned a number and every second the hamster engaged in one of the behaviors, the corresponding number was entered into a Microsoft Excel spreadsheet. From these spreadsheets, the following behavioral components were calculated: duration of anogenital investigation before first ejaculation, number of ejaculations, mount latency, intromission latency, total number of mounts, total number of intromissions, hit rate, and time to reach long intromissions. In addition, averages of the following measures were calculated per ejaculation including: average number of mounts, average number of intromissions, average ejaculation latency, average interintromission interval, and average postejaculatory interval. The components of reproductive behavior listed above were defined by using the criteria of Meisel and Sachs (1994). Mount latency, intromission latency, and time to reach long intromissions were defined as the length of time for each behavior to occur after introduction of the female. The ejaculation latency, interintromission interval, and postejaculatory interval were calculated for each ejaculatory series (from first mount to ejaculation). Ejaculation latency was identified as the time from the first intromission to ejaculation. Interintromission interval was determined by dividing the ejaculation latency by the number of intromissions in that ejaculatory series. Postejaculatory interval was the time elapsed between ejaculation and the first mount of the next series. Hit rate was calculated by dividing the total number of intromissions in the entire mating session by the total number of mounts plus the total number of intromissions. Last, the average ejaculation latency, interintromission interval, and postejaculatory interval was calculated for each hamster by summing each of those measures and dividing by the number of ejaculations achieved in the mate session. Statistical analyses on all reproductive behavior measures were done by using repeated-measures ANOVA using the computer program Statview. If this repeated-measures ANOVA revealed a significant interaction between treat-
384
H.K. Holmer et al. / Hormones and Behavior 43 (2003) 381–387
of the dark cycle. There was a significant increase in glucocorticoids in both the executive and yoked groups 90 min after the onset of stress compared to time 0 controls [F(6,35) ⫽ 31.37; P ⱕ 0.0001]. Plasma glucocorticoids then significantly decreased in both executive and yoked groups at 120 and 180 min compared to time 0 controls (P ⱕ 0.01). Cortisol and corticosterone were not significantly different between the executive and yoked groups at any time points. Experiment 2: male reproductive behavior after chronic escapable or inescapable stress Fig. 1. Mean (⫾ SEM) total glucocorticoid levels following escapable or inescapable stress. Lowercase letters indicate significant differences between the groups; groups that share a common letter do not differ.
ment group and time, a post hoc paired t test comparison was made between the pre- and poststress measurement within an experimental group. If a male did not mate for either the pre- or poststress mate session it was not included in the statistical analysis. Hormone determination The following morning after the poststress mate session, animals were removed from their home cages and administered an overdose of sodium pentobarbital (130 mg/kg of body weight). A blood sample was collected via cardiac puncture. Testosterone was quantified via radioimmunoassay using the Coat-a-Count Total Testosterone kit obtained from DPC.
Results Experiment 1: glucocorticoid response to acute escapable or inescapable stress Total plasma glucocorticoid levels were determined in each animal after one escapable/inescapable stress session. As shown in Fig. 1, plasma glucocorticoid levels at time 0 were approximately 10 g/dl; this slight elevation reflects the diurnal rhythm as samples were taken at the beginning
Reproductive behavior The entire pattern of male hamster reproductive behavior was quantified on a second-by-second basis before and after 12 days of stress. Ninety-four percent (34 of 36) of the males ejaculated during the prestress mate session (12 control, 11 executive, and 11 yoked). All 36 animals ejaculated in the poststress mate session. Repeated-measures ANOVA revealed a significant main effect of time on the number of mounts and ejaculations [F(2,30) ⫽ 10.762; P ⱕ 0.005; F(2,30) ⫽ 16.724; P ⱕ 0.001; respectively; Table 2]. No differences were observed in the number of intromissions achieved in control, executive, and yoked males after stress (Table 2). Because of this increase in mounts and no change in intromissions, there were significant changes in hit rate over time [F(2,31) ⫽ 6.783; P ⱕ 0.01] and a significant interaction between treatment group and time [Table 2; F(2,31) ⫽ 4.490; P ⫽ 0.0194]. A post hoc paired t test revealed the yoked group had a significant decrease in hit rate following stress (P ⱕ 0.005). No significant differences were observed in hit rate between the executive and control groups. Hormone concentration Terminal blood samples were collected in 10 of 12 triads of control, executive, and yoked animals. The average testosterone levels in the control, executive, and yoked groups were 5.2 ⫾ 0.8, 4.5 ⫾ 1.0, and 3.7 ⫾ 0.6 ng/ml, respectively.
Table 2 Measurements of reproductive behavior before and after 12 days of stressa Control (n ⫽ 12)
b
No. of mounts No. of intromissions Hit ratec No. of ejaculationsb
Executive (n ⫽ 10)
Yoked (n ⫽ 11)
Prestress
Poststress
Prestress
Poststress
Prestress
Poststress
52.0 ⫾ 6.1 32.9 ⫾ 3.3 0.39 ⫾ 0.02 7.8 ⫾ 0.8
60.7 ⫾ 8.8 37.3 ⫾ 3.0 0.40 ⫾ 0.02 10.0 ⫾ 0.6
63.7 ⫾ 7.8 38.6 ⫾ 2.4 0.39 ⫾ 0.02 8.6 ⫾ 0.8
71.8 ⫾ 9.1 38.6 ⫾ 2.7 0.37 ⫾ 0.02 9.6 ⫾ 0.6
46.4 ⫾ 3.1 31.5 ⫾ 2.8 0.40 ⫾ 0.02 8.3 ⫾ 0.9
76.3 ⫾ 9.4 34.3 ⫾ 3.3 0.33 ⫾ 0.03d 8.9 ⫾ 0.6
All measures are mean ⫾ SEM. Significant main effect of time with poststress greater than prestress collapsed across all three treatment groups, P ⱕ 0.005. c Significant interaction between treatment group and time, P ⫽ 0.0194. d Significant difference between pre- and poststress, P ⱕ 0.005. a
b
H.K. Holmer et al. / Hormones and Behavior 43 (2003) 381–387
Discussion In the current study, we examined changes in glucocorticoid levels and male mating behavior in the Syrian hamster following escapable or inescapable stress. Experiment 1 demonstrated that acute intermittent foot shock, either escapable or inescapable, elicited a marked increase in glucocorticoid secretion following a single 90-min stress session. Experiment 2 demonstrated that chronic intermittent foot shock caused a slight detriment in sexual behavior only in males who could not escape their stress; animals that could escape the stress did not experience any decrease in copulation. These data suggested, first, that the foot shock stressor elicits a dramatic glucocorticoid response in the Syrian hamster, and that over repeated presentation, this stressor can lead to a mild inhibition of reproductive behavior. Second, this study demonstrated that the psychological variable of control may prevent the detrimental effects of stress on male reproduction. Previous observations using the escapable/inescapable stress paradigm have demonstrated that these two psychologically distinct stress conditions elicit similar patterns of ACTH (adrenocorticotropin) and glucocorticoid secretion (Maier et al., 1986; Prince and Anisman, 1990; Seligman and Weiss, 1980; Swenson and Vogel, 1983). Therefore, it was not surprising that no differences in plasma glucocorticoid levels were observed between executive and yoked male hamsters following a single stress session. However, dysregulation of the hypothalamic-pituitary-adrenal (HPA) axis can be observed in rodents using a single escapable/ inescapable stress session and administration of dexamethasone. Dexamethasone, a synthetic glucocorticoid agonist, should inhibit glucocorticoid secretion through negative feedback mechanisms. However, in the escapable/inescapable stress model, dexamethasone administration suppresses corticosterone secretion in the executive (escapable stress) but not the yoked (inescapable stress) animals (Haracz et al., 1988; Helmreich, D., personal observation). Because this is the first time this model system has been utilized in the Syrian hamster, it was important to demonstrate the endocrine response elicited by the stress stimuli. Clearly, the stress parameters used in this experiment (shock duration, intensity, route of administration, time of day, and so on) produce a robust stimulation of the hamster hypothalamicpituitary-adrenal axis, and future studies can utilize this paradigm to determine the differential regulation of the hypothalamic-pituitary-adrenal axis by escapable or inescapable stress in the Syrian hamster. While it was clear from Experiment 1 that acute stress provoked an endocrine response in the male hamster, previous studies in mice and rats demonstrated that chronic stress is required to inhibit male reproductive behavior (Chrousos and Gold, 1992; D’Aquila et al., 1994; Matsuda et al., 1996; Retana-Marquez et al., 1996). Despite receiving 12 days of foot shock stress, the male hamsters only demonstrated subtle changes in reproductive
385
behavior. And, most important to the current study, detrimental changes occurred only in the yoked animals. For this study, to increase the power of statistical analysis, animals served as their own prestress controls to combine longitudinal studies with comparisons between groups. This led to a significant main effect of time on a number of behavioral measures regardless of stress condition. We interpret the significant increase in the number of ejaculations, decreased number of intromissions/ejaculation, and decreased postejaculatory interval/ejaculation as due to the increased experience received by all males over the multiple mating tests of this study. Previous experiments in the male hamster have documented that sexual experience has profound effects on both copulatory behavior as well as the brain regions controlling the behavior. Sexual experience can delay the decline in reproductive behavior following castration (Lisk and Heimann, 1980). In addition, sexual experience alone results in increased neuronal activation (as detected by Fos immunocytochemistry) within the medial nucleus of the amygdala and the medial preoptic nucleus— two brain regions known to regulate male mating behavior (Kollack-Walker and Newman, 1997). However, in addition to the effects of experience on male mating behavior, there were also changes specifically attributed to chronic inescapable but not escapable stress. The most striking effect of inescapable, but not escapable, stress was a difference in hit rate. All animals displayed an increase in mounts during the poststress mating session. However, we interpret the increased number of mounts in the no-stress control and executive males to coincide with an experience-induced increase in ejaculations; the yoked animals did not exhibit as large of an experienced-induced increase in ejaculations, but yet had the largest poststress increase in mounts. This increase in mounts with no poststress increase in intromissions or ejaculations is reflected as a marked decrease in hit rate only in the yoked animals. In addition, this complex interaction between experience and stress might also explain the lack of a significant interaction between treatment group and time for other measures. All groups were demonstrating an increase in mounts. However, only the no-stress control and executive males were increasing ejaculations. Therefore, the yoked males would continue mounting, but could not finish other consummatory aspects of the behavior. This observed effect of inescapable stress on male reproductive behavior is similar to studies in the male rat where different types of chronic stressors (immobilization, swim, and foot shock) all resulted in an increase in mounts and a decrease in hit rate (Retana-Marquez et al., 1996). Therefore, it would seem that in the hamster, like the rat, the motor components of copulation culminating with intromission are those aspects most sensitive to the inhibitory effects of stress. Unlike previous studies that have demonstrated a stressinduced decrease in testosterone secretion (Bardin and Peterson, 1967; Free and Tillson, 1973; Sapolsky, 1982, 1985), no change in plasma testosterone was detected in our
386
H.K. Holmer et al. / Hormones and Behavior 43 (2003) 381–387
animals. There are a number of explanations for this finding. First, it could be that changes in reproductive behavior occur independently of any change in peripheral testosterone secretion. Second, 12 days of this stress paradigm could be just beginning to suppress testicular function in the yoked males, and lengthening the stress paradigm would result in a more pronounced change in testosterone secretion. Third, animals could be habituating to the stressor. Both the executive and yoked animals received the same intensity of foot shock (0.8 mA) for 12 consecutive days. This intensity was chosen to use the lowest amount of shock possible that will still elicit a response in the animals (both in terms of pressing the lever and inhibiting reproduction). Over the 12 days of testing, the executive males become extremely adept at lever pressing to terminate the shock— oftentimes in 0.8 s (the minimum shock duration). Over the 80 trials the animals receive an average of 208.8 ⫾ 14.3 s of shock or approximately 2.6 s of shock per trial. If you disregard the initial day of stress where the executive animals are learning to press the bar to turn off the shock, the average shock per day decreases to 188.2 ⫾ 16.0 s or approximately 2.3 s of shock per trial. Drugan and colleagues (1997) have commented that in the rat at least 2 s of shock is required to prevent habituation to the stressor. Therefore, the hamsters in this study are receiving a low intensity stimulus for a short duration, possibly allowing for habituation or tolerance to the stress effects to occur. However, with this caveat in mind, we still observe differential effects of this stress paradigm on copulatory performance. Ongoing studies in this laboratory have attempted to diminish the effect of habituation to the stressor by increasing the number of bar presses required to turn off the shock as well as increasing the intensity of the footshock over the 12 days of stress. Preliminary results suggest eliminating the effect of habituation and/or tolerance exacerbates the effects of inescapable stress on male reproductive behavior, and also demonstrates that the subtle effects of stress reported here are reliable and repeatable (Cordner et al., 2002). Future work will use this model to characterize how the psychological variable of coping alters the brains of executive animals to protect them from the damaging effects of stress.
Acknowledgments This study was supported by grants from the VTEPSCoR Small College Research Development Fund (D.B.P. and D.L.H.), an NSF-Research Opportunity Award (D.B.P.), the Howard Hughes Medical Institute (H.K.H.), and the Middlebury College Roddy Fund (J.E.R.). We are indebted to Mr. Michael Betourney and Ms. Vicki Major for their excellent maintenance and care of the hamster colony.
References Arteaga, M., Motte-Lara, J., Velazquez-Moctezuma, J., 2000. Temporal pattern of recovery from sexual satiety in male golden hamster (Mesocricetus auratus). Physiol. Behav. 68, 591–594. Bardin, C., Peterson, R., 1967. Studies of androgen production in the rat: testosterone and androstenedione content of blood. Endocrinology 80, 38 – 44. Bunnell, B., Boland, B., Dewsbury, D., 1977. Copulatory behavior of golden hamsters. Behaviour 61, 180 –206. Chrousos, G., Gold, P., 1992. The concepts of stress and stress system disorders: overview of physical and behavioral homeostasis. JAMA 267, 1244 –1252. Cordner, A., Helmreich, D., Parfitt, D., 2002. Antidepressant administration prevents the effect of uncontrollable stress on male reproductive behavior in the Syrian hamster (Mesocricetus auratus [abstract]). Soc. Behav Neuroendocrinol. 41, 462. D’Aquila, P., Brain, P., Willner, P., 1994. Effects of chronic mild stress on performance in behavioral tests relevant to anxiety and depression. Physiol Behav. 56, 861– 867. Drugan, R., Basile, A., Ha, J., Healy, D., Ferland, R., 1997. Analysis of the importance of controllable versus uncontrollable stress on subsequent behavioral and physiological functioning. Brain Res. Protocols 2, 69 –74. Free, M., Tillson, S., 1973. Secretion rate of testicular steroids in conscious and halothane-anesthetized rat. Endocrinology 93, 874 – 879. Gorzalka, B., Hanson, L., Brotto, L., 1998. Chronic stress effects on sexual behavior in male and female rats: mediation by 5-HT2A receptors. Pharmacol Biochem. Behav. 61, 405– 412. Haracz, J., Minor, T., Wilkins, J., Zimmerman, E., 1988. Learned helplessness: an experimental model of the DST in rats. Biol. Psychiatry 23, 388 –396. Huck, U., Lisk, R., 1985. Determinants of mating success in the golden hamster (Mesocricetus auratus): I. Male capacity. J. Comp. Psychol. 99, 98 –107. Kollack-Walker, S., Newman, S., 1997. Mating-induced expression of c-fos in the male Syrian hamster brain: role of experience, pheromones, and ejaculations. J. Neurobiol. 32, 481–501. Lisk, R., Heimann, J., 1980. The effects of sexual experience and frequency of testing on retention of copulatory behavior following castration in the male hamster. Behav. Neural Biol. 28, 156 –171. Maier, S., Ryan, S., Barksdale, C., Kalin, N., 1986. Stressor controllability and the pituitary-adrenal system. Behav. Neurosci. 100, 669 – 674. Matsuda, S., Peng, H., Yoshimura, H., Wen, T., Fukuda, T., Sakanaka, M., 1996. Persistent c-fos expression in the brains of mice with chronic social stress. Neurosci. Res. 26, 157–170. Meisel, R., Sachs, B., 1994. The physiology of male sexual behavior, in: Knobil, E., Neill, J. (Eds.), The Physiology of Reproduction, Raven Press, New York, pp. 3– 84. Menendez-Patterson, A., Florez-Lozano, J., Fernandez, S., Marin, B., 1980. Stress and sexual behavior in male rats. Physiol. Behav. 24, 403– 406. Negro-Vilar, A., 1993. Stress and other environmental factors affecting fertility in men and women: overview. Environ. Health Perspect. 101 (S2), 59 – 64. Parfitt, D., Newman, S., 1998. Fos-immunoreactivity within the extended amygdala is correlated with the onset of sexual satiety. Horm. Behav. 34, 17–29. Prince, C., Anisman, H., 1990. Situation specific effects of stressor controllability on plasma corticosterone changes in mice. Pharmacol. Biochem Behav. 37, 613– 621. Retana-Marquez, S., Dominguez-Salazar, E., Velazquez-Moctezuma, J., 1996. Effect of acute and chronic stress on masculine sexual behavior in the rat. Psychoneuroendocrinology 21, 39 –50. Rivier, C., 1995. Luteinizing hormone releasing hormone, gonadotropins, and gonadal steroids in stress. Ann. NY Acad. Sci. 771, 187–191.
H.K. Holmer et al. / Hormones and Behavior 43 (2003) 381–387 Sapolsky, R., 1982. The endocrine stress-response and social status in the wild baboon. Horm. Behav. 16, 279 –292. Sapolsky, R., 1985. Stress-induced suppression of testicular function in the wild baboon: role of glucocorticoids. Endocrinology 116, 2273–2278. Seligman, M., Issacowitz, D., 2000. Learned helplessness, in: Fink, G. (Ed.), The Encyclopedia of Stress, Vol. 2, Academic Press, San Diego, pp. 599 – 602. Seligman, M., Weiss, J., 1980. Coping behavior: learned helplessness, physiological change, and learned activity. Behav. Res. Ther. 18, 459–512. Sklar, L., Anisman, H., 1979. Stress and coping factors influence tumor growth. Science 205, 513–515. Swenson, R., Vogel, W., 1983. Plasma catecholamine and corticosterone as well as brain catecholamine changed during coping in rats exposed to stressful footshock. Pharmacol. Biochem. Behav. 18, 689 – 693.
387
Visintainer, M., Volpicelli, J., Seligman, M., 1982. Tumor rejection in rats after inescapable or escapable shock. Science 216, 437– 439. Weiss, J., Goodman, P., Losito, B., Corrigan, S., Charry, J., Bailey, W., 1981. Behavioral depression produced by an uncontrollable stressor: relationship to norepinephrine, dopamine, and serotonin levels in various regions of the rat brain. Brain Res. Rev. 3, 167–205. Wood, R., Swann, J., 2000. Neuronal integration of chemosensory and hormonal signals in the control of male sexual behavior, in: Waller, K., Schneider, J. (Eds.), Reproduction in Context: Social and Environmental Influences on Reproductive Physiology and Behavior, MIT Press, Cambridge, MA, pp. 423– 444. Young, E., Korszun, A., 1998. Psychoneuroendocrinology of depression: hypothalamic-pituitary-gonadal axis, in: Nemeroff, C. (Ed.), Psychiatric Clinics: Psychoneuroendocrinology, Vol. 21, W.B. Saunders, Philadelphia, PA, pp. 309 –324.
Hormones and Behavior 43 (2003) 381–387
www.elsevier.com/locate/yhbeh
Differential effects of chronic escapable versus inescapable stress on male syrian hamster (Mesocricetus auratus) reproductive behavior Haley K. Holmer, Janice E. Rodman, Dana L. Helmreich, and David B. Parfitt* Department of Biology and Neuroscience Program, Middlebury College, Middlebury, VT 05753, USA Received 1 March 2002; revised 10 October 2002; accepted 14 October 2002
Abstract Stress decreases sexual activity, but it is uncertain which aspects of stress are detrimental to reproduction. This study used an escapable/inescapable stress paradigm to attempt to dissociate physical from psychological components of stress, and assess each component’s impact on reproductive behavior in the male Syrian hamster (Mesocricetus auratus). Two experiments were completed using this protocol where two animals receive the same physical stressor (an electric footshock) but differ in the psychological aspect of control. One group (executive) could terminate the shock for themselves as well as a second group (yoked) by pressing a bar. Experiment 1 demonstrated a significant increase in plasma glucocorticoids at the end of a single 90-min stress session with no difference in glucocorticoid levels between the executive and yoked groups at any time point. Experiment 2 quantified male reproductive behavior prior to and immediately following 12 days of escapable or inescapable stress in executive, yoked, and no-stress control hamsters (n ⫽ 12/group). Repeated-measures analysis of variance revealed a number of significant changes in reproductive behavior before and after stress in the three treatment groups. The most striking difference was a decrease in hit rate observed only in the animals that could not control their stress (yoked group). Hit rate in the executive males that received the exact same physical stressor but could terminate the shock by pressing a bar was nearly identical to control animals that never received any foot shock. Therefore, we conclude that coping or control can ameliorate the negative effects of stress on male reproductive behavior. © 2003 Elsevier Science (USA). All rights reserved. Keywords: Sexual behavior; Cortisol; Corticosterone; Testosterone; Coping; Control
Introduction Stress can disrupt all aspects of reproduction from motivation to performance in multiple species, including humans. Because reproductive processes are sensitive to stress, the fertility of any individual is susceptible to a variety of insults. In females, stress disrupts estrous cyclicity and causes decreases in ovarian hormone levels (NegroVilar, 1993; Rivier, 1995; Young and Korszun, 1998) and in males, stress decreases testosterone levels and reproductive behavior. In male animal models, negative effects on both appetitive and consummatory aspects of reproductive behavior can be seen after either acute or chronic stress. More
* Corresponding author. Department of Biology, Middlebury College, Middlebury, VT 05753, USA. Fax: ⫹1-802-443-2072. E-mail address: [email protected] (D.B. Parfitt).
specifically, after 1 day of stress, male rats exhibit an increased mount and ejaculation latency, suggesting impaired sexual motivation, and even greater deficits are observed after 20 days of stress (D’Aquila et al., 1994; Gorzalka et al., 1998; Menendez-Patterson et al., 1980; Retana-Marquez et al., 1996). These rats also display increased mount frequency, decreased hit rate, and decreased ejaculations—all interpreted as disrupted sexual performance (MenendezPatterson et al., 1980; Retana-Marquez et al., 1996). Most stressors, such as immobilization and swim, used by previous investigators to study the impact of stress on reproductive behavior, have both physical and psychological components that impinge on the animal. To understand how stress negatively impacts reproductive behavior, it is important to establish the pertinent aspects of a stressor that lead to decreases in reproductive processes. In the current study, we attempted to dissociate the effects of physical
0018-506X/03/$ – see front matter © 2003 Elsevier Science (USA). All rights reserved. doi:10.1016/S0018-506X(03)00009-6
382
H.K. Holmer et al. / Hormones and Behavior 43 (2003) 381–387
versus psychological components of stress on reproductive behavior by using an escapable/inescapable stress paradigm. In this model, two animals experience the same amplitude, duration, and intensity of a physical stressor (a mild-electric foot shock), but differ in the psychological aspect of control. Two animals are placed in identical chambers and experience an identical physical stressor. One animal, the executive, can terminate his stress by pressing a lever, while his yoked partner cannot. The executive animal’s bar press also terminates the stress in his yoked partner’s chamber. Thus, both animals receive the same duration of stress, but differ only in the psychological aspect of coping. This paradigm has been used by many investigators in many different experimental models, including humans, to demonstrate that the unfavorable effects of stress, such as learned helplessness (Seligman and Issacowitz, 2000), gastric ulceration (Weiss et al., 1981), and exacerbation of tumor growth (Sklar and Anisman, 1979; Visintainer et al., 1982), occur only in the yoked animals, while executive animals appear similar to no-stress controls. The current study is the first that we know of to use this escapable/inescapable stress paradigm to dissociate the physical from the psychological effects of stress on reproductive behavior. The male Syrian hamster (Mesocricetus auratus) is a useful model for investigating the relationship between stress and reproduction because it is a robust mater, its copulatory behavior is well characterized (Bunnell et al., 1977; Huck and Lisk, 1985), and the neural circuitry governing this behavior is also well understood (Wood and Swann, 2000). We predicted that the yoked animal would have a decrease in sexual behavior and performance compared to the executive and no-stress control males. Additionally, we hypothesized that yoked animals would have lower levels of testosterone than the executive and control animals. We examined reproductive behavior and testosterone levels in male hamsters exposed to 12 days of escapable or inescapable stress. Furthermore, an additional study characterized the glucocorticoid response to a single escapable/ inescapable stress session in hamsters, as the glucocorticoid response to this particular stress paradigm has not been reported in this species.
Materials and methods Experiment 1: glucocorticoid response to acute escapable or inescapable stress Animals Forty-two adult male Syrian Hamsters (⬎90 days of age) were obtained from Harlan (Indianapolis, IN) and used to determine the glucocorticoid response to the stress session. Thirty-six animals were housed two per cage and randomly assigned to one of two groups, executive or yoked (n ⫽ 18/group). The remaining six no-stress control hamsters were housed three per cage and not assigned a stress con-
dition. All hamsters were maintained on a reversed long day photoperiod (14h:10 h light:dark; lights off at 12:30 pm) and food and water were supplied ad libitum. The Middlebury College Institutional Animal Care and Use Committee approved all experimental protocols, and all animals were treated in accordance with the NIH Guide for the Care and Use of Laboratory Animals. Experimental stress paradigm All stress sessions were conducted within the first 3 h of the dark cycle under dim red light; this time period was chosen because this is when maximum reproductive behavior occurs in the hamster. Males were removed from their home cage and placed in the stress chambers (Med Associates, St. Albans, VT). Each of the boxes (20 ⫻ 25 ⫻ 21 cm) contained a stainless steel grid floor connected to a scrambled shock generator. Each box also contained a lever wired such that when the hamster in the executive box pressed the lever, the electric footshock would terminate in both the executive and the yoked boxes. The executive animal could therefore control the duration of the stress, while the yoked animal could not. The stimulus intensity used was 0.8 mA and each stress session consisted of 80 trials. If the executive hamster failed to press the lever within 30 s, the shock was automatically terminated. The intertrial interval varied between 5 and 115 s, and the mean duration of the interval was 60 s. The entire 80 trial stress session lasted approximately 90 min. Hormone determination A trunk blood sample was collected via rapid decapitation from the executive and yoked hamsters at one of three time points, 90, 120, and 180 min from the start of the stress session. To obtain an average baseline total glucocorticoid level, six no-stress control hamsters were removed from their home cages and trunk blood samples were immediately collected (these were called time 0). Plasma cortisol and corticosterone were quantified in all 42 animals via radioimmunoassay using the Coat-a-Count Cortisol and Rat Corticosterone kits obtained from Diagnostic Products Corporation (DPC, Los Angeles, CA). Finally, all samples were run together in a single assay for either cortisol or corticosterone. Statistical analysis was done with a one-way analysis of variance (ANOVA) followed by a Fisher PLSD post hoc test to determine statistically significant differences among groups. Statview computer software (SAS Institute, Cary, NC) was used for all statistical analyses. Experiment 2: male reproductive behavior after chronic escapable or inescapable stress Animals Thirty-six adult male Syrian hamsters (⬎60 days of age and weighing 98 –146 g) were obtained from Harlan (Indianapolis, IN) and used for this experiment. Animals were
H.K. Holmer et al. / Hormones and Behavior 43 (2003) 381–387 Table 1 Daily duration of shock and number of escapes of 80 trialsa
a
Stress day
Shock duration (s)
Escapes
1 2 3 4 5 6 7 8 9 10 11 12
435.7 ⫾ 49.4 217.0 ⫾ 13.7 165.5 ⫾ 8.0 184.8 ⫾ 13.2 170.5 ⫾ 10.6 168.8 ⫾ 10.4 164.2 ⫾ 11.0 181.6 ⫾ 19.0 171.4 ⫾ 11.0 207.8 ⫾ 31.8 210.2 ⫾ 40.6 228.7 ⫾ 53.5
77.0 ⫾ 0.9 80.0 ⫾ 0.0 79.9 ⫾ 0.1 79.8 ⫾ 0.2 80.0 ⫾ 0.0 80.0 ⫾ 0.0 79.9 ⫾ 0.1 79.6 ⫾ 0.4 80.0 ⫾ 0.0 78.8 ⫾ 0.9 78.3 ⫾ 1.1 78.3 ⫾ 1.6
All measures are mean ⫾ SEM.
divided into 12 cohorts (n ⫽ 3/cohort) and each hamster was randomly assigned to one of three groups, executive, yoked, or control (n ⫽ 12/group). In addition, 15 ovariectomized adult female hamsters were used as stimulus animals during the mating sessions. All hamsters were housed and treated as described above. Sexual experience Prior to the experiment, all males were given three sessions of sexual experience on three consecutive days. Each male was paired with a sexually receptive female until three ejaculations were observed or for 15 min, whichever came first. Sexual receptivity in the ovariectomized female hamsters was induced by subcutaneous injections of 10 g of estradiol benzoate 48 h and 24 h prior to, and 250 g of progesterone 4 h before the mating sessions. The sexual experience sessions and the experimental behavior testing were conducted during the dark cycle, under dim red lights, and in glass aquariums (51 ⫻ 25 ⫻ 30.5 cm) in a room separate from the home colony. Experimental stress paradigm The escapable/inescapable stress paradigm used identical parameters to Experiment 1. All stress sessions were conducted within the first 3 h of the dark cycle under dim red light. Each male received 12 consecutive days of this stress paradigm, with animals remaining in their same stress conditions (executive or yoked) throughout the 12 days. In this experiment, a third, no-stress control hamster was placed in a box similar to the executive and yoked stress chambers, but received no electric foot shock during the 12 days. For the 12 cohorts used in this experiment, the executives hit the bar to escape the foot shock an average of 79.3 ⫾ 0.3 times (of 80 trials), and the average total duration of footshock per day was 208.8 ⫾ 14.3 s. The duration of shock and number of escapes was averaged for all 12 cohorts each day and is presented in Table 1. For a complete description of the escapable/inescapable paradigm and the stress conditions, see Experiment 1.
383
Reproductive behavior testing Two days prior to the onset of 12 consecutive days of stress, all males were mated to sexual satiety to determine a prestress baseline. For this single prestress mating session, each male was paired with a separate, sexually receptive female and mated until long intromissions [a behavioral measure for the onset of sexual satiety (Arteaga et al., 2000; Parfitt and Newman, 1998)], or 1 h, whichever came first. Immediately after the mating sessions, all hamsters were returned to their cages in the home colony. The poststress mating session was performed 1 h after the 12th stress session. Both pre- and poststress mating sessions were videotaped for future analysis of all components of male hamster reproductive behavior. The following 10 behaviors were quantified on a secondby-second basis: no interaction, grooming, investigation, anogenital investigation, ectopic mount, mount, intromission, ejaculation, long intromission, and fighting. Each of these actions was assigned a number and every second the hamster engaged in one of the behaviors, the corresponding number was entered into a Microsoft Excel spreadsheet. From these spreadsheets, the following behavioral components were calculated: duration of anogenital investigation before first ejaculation, number of ejaculations, mount latency, intromission latency, total number of mounts, total number of intromissions, hit rate, and time to reach long intromissions. In addition, averages of the following measures were calculated per ejaculation including: average number of mounts, average number of intromissions, average ejaculation latency, average interintromission interval, and average postejaculatory interval. The components of reproductive behavior listed above were defined by using the criteria of Meisel and Sachs (1994). Mount latency, intromission latency, and time to reach long intromissions were defined as the length of time for each behavior to occur after introduction of the female. The ejaculation latency, interintromission interval, and postejaculatory interval were calculated for each ejaculatory series (from first mount to ejaculation). Ejaculation latency was identified as the time from the first intromission to ejaculation. Interintromission interval was determined by dividing the ejaculation latency by the number of intromissions in that ejaculatory series. Postejaculatory interval was the time elapsed between ejaculation and the first mount of the next series. Hit rate was calculated by dividing the total number of intromissions in the entire mating session by the total number of mounts plus the total number of intromissions. Last, the average ejaculation latency, interintromission interval, and postejaculatory interval was calculated for each hamster by summing each of those measures and dividing by the number of ejaculations achieved in the mate session. Statistical analyses on all reproductive behavior measures were done by using repeated-measures ANOVA using the computer program Statview. If this repeated-measures ANOVA revealed a significant interaction between treat-
384
H.K. Holmer et al. / Hormones and Behavior 43 (2003) 381–387
of the dark cycle. There was a significant increase in glucocorticoids in both the executive and yoked groups 90 min after the onset of stress compared to time 0 controls [F(6,35) ⫽ 31.37; P ⱕ 0.0001]. Plasma glucocorticoids then significantly decreased in both executive and yoked groups at 120 and 180 min compared to time 0 controls (P ⱕ 0.01). Cortisol and corticosterone were not significantly different between the executive and yoked groups at any time points. Experiment 2: male reproductive behavior after chronic escapable or inescapable stress Fig. 1. Mean (⫾ SEM) total glucocorticoid levels following escapable or inescapable stress. Lowercase letters indicate significant differences between the groups; groups that share a common letter do not differ.
ment group and time, a post hoc paired t test comparison was made between the pre- and poststress measurement within an experimental group. If a male did not mate for either the pre- or poststress mate session it was not included in the statistical analysis. Hormone determination The following morning after the poststress mate session, animals were removed from their home cages and administered an overdose of sodium pentobarbital (130 mg/kg of body weight). A blood sample was collected via cardiac puncture. Testosterone was quantified via radioimmunoassay using the Coat-a-Count Total Testosterone kit obtained from DPC.
Results Experiment 1: glucocorticoid response to acute escapable or inescapable stress Total plasma glucocorticoid levels were determined in each animal after one escapable/inescapable stress session. As shown in Fig. 1, plasma glucocorticoid levels at time 0 were approximately 10 g/dl; this slight elevation reflects the diurnal rhythm as samples were taken at the beginning
Reproductive behavior The entire pattern of male hamster reproductive behavior was quantified on a second-by-second basis before and after 12 days of stress. Ninety-four percent (34 of 36) of the males ejaculated during the prestress mate session (12 control, 11 executive, and 11 yoked). All 36 animals ejaculated in the poststress mate session. Repeated-measures ANOVA revealed a significant main effect of time on the number of mounts and ejaculations [F(2,30) ⫽ 10.762; P ⱕ 0.005; F(2,30) ⫽ 16.724; P ⱕ 0.001; respectively; Table 2]. No differences were observed in the number of intromissions achieved in control, executive, and yoked males after stress (Table 2). Because of this increase in mounts and no change in intromissions, there were significant changes in hit rate over time [F(2,31) ⫽ 6.783; P ⱕ 0.01] and a significant interaction between treatment group and time [Table 2; F(2,31) ⫽ 4.490; P ⫽ 0.0194]. A post hoc paired t test revealed the yoked group had a significant decrease in hit rate following stress (P ⱕ 0.005). No significant differences were observed in hit rate between the executive and control groups. Hormone concentration Terminal blood samples were collected in 10 of 12 triads of control, executive, and yoked animals. The average testosterone levels in the control, executive, and yoked groups were 5.2 ⫾ 0.8, 4.5 ⫾ 1.0, and 3.7 ⫾ 0.6 ng/ml, respectively.
Table 2 Measurements of reproductive behavior before and after 12 days of stressa Control (n ⫽ 12)
b
No. of mounts No. of intromissions Hit ratec No. of ejaculationsb
Executive (n ⫽ 10)
Yoked (n ⫽ 11)
Prestress
Poststress
Prestress
Poststress
Prestress
Poststress
52.0 ⫾ 6.1 32.9 ⫾ 3.3 0.39 ⫾ 0.02 7.8 ⫾ 0.8
60.7 ⫾ 8.8 37.3 ⫾ 3.0 0.40 ⫾ 0.02 10.0 ⫾ 0.6
63.7 ⫾ 7.8 38.6 ⫾ 2.4 0.39 ⫾ 0.02 8.6 ⫾ 0.8
71.8 ⫾ 9.1 38.6 ⫾ 2.7 0.37 ⫾ 0.02 9.6 ⫾ 0.6
46.4 ⫾ 3.1 31.5 ⫾ 2.8 0.40 ⫾ 0.02 8.3 ⫾ 0.9
76.3 ⫾ 9.4 34.3 ⫾ 3.3 0.33 ⫾ 0.03d 8.9 ⫾ 0.6
All measures are mean ⫾ SEM. Significant main effect of time with poststress greater than prestress collapsed across all three treatment groups, P ⱕ 0.005. c Significant interaction between treatment group and time, P ⫽ 0.0194. d Significant difference between pre- and poststress, P ⱕ 0.005. a
b
H.K. Holmer et al. / Hormones and Behavior 43 (2003) 381–387
Discussion In the current study, we examined changes in glucocorticoid levels and male mating behavior in the Syrian hamster following escapable or inescapable stress. Experiment 1 demonstrated that acute intermittent foot shock, either escapable or inescapable, elicited a marked increase in glucocorticoid secretion following a single 90-min stress session. Experiment 2 demonstrated that chronic intermittent foot shock caused a slight detriment in sexual behavior only in males who could not escape their stress; animals that could escape the stress did not experience any decrease in copulation. These data suggested, first, that the foot shock stressor elicits a dramatic glucocorticoid response in the Syrian hamster, and that over repeated presentation, this stressor can lead to a mild inhibition of reproductive behavior. Second, this study demonstrated that the psychological variable of control may prevent the detrimental effects of stress on male reproduction. Previous observations using the escapable/inescapable stress paradigm have demonstrated that these two psychologically distinct stress conditions elicit similar patterns of ACTH (adrenocorticotropin) and glucocorticoid secretion (Maier et al., 1986; Prince and Anisman, 1990; Seligman and Weiss, 1980; Swenson and Vogel, 1983). Therefore, it was not surprising that no differences in plasma glucocorticoid levels were observed between executive and yoked male hamsters following a single stress session. However, dysregulation of the hypothalamic-pituitary-adrenal (HPA) axis can be observed in rodents using a single escapable/ inescapable stress session and administration of dexamethasone. Dexamethasone, a synthetic glucocorticoid agonist, should inhibit glucocorticoid secretion through negative feedback mechanisms. However, in the escapable/inescapable stress model, dexamethasone administration suppresses corticosterone secretion in the executive (escapable stress) but not the yoked (inescapable stress) animals (Haracz et al., 1988; Helmreich, D., personal observation). Because this is the first time this model system has been utilized in the Syrian hamster, it was important to demonstrate the endocrine response elicited by the stress stimuli. Clearly, the stress parameters used in this experiment (shock duration, intensity, route of administration, time of day, and so on) produce a robust stimulation of the hamster hypothalamicpituitary-adrenal axis, and future studies can utilize this paradigm to determine the differential regulation of the hypothalamic-pituitary-adrenal axis by escapable or inescapable stress in the Syrian hamster. While it was clear from Experiment 1 that acute stress provoked an endocrine response in the male hamster, previous studies in mice and rats demonstrated that chronic stress is required to inhibit male reproductive behavior (Chrousos and Gold, 1992; D’Aquila et al., 1994; Matsuda et al., 1996; Retana-Marquez et al., 1996). Despite receiving 12 days of foot shock stress, the male hamsters only demonstrated subtle changes in reproductive
385
behavior. And, most important to the current study, detrimental changes occurred only in the yoked animals. For this study, to increase the power of statistical analysis, animals served as their own prestress controls to combine longitudinal studies with comparisons between groups. This led to a significant main effect of time on a number of behavioral measures regardless of stress condition. We interpret the significant increase in the number of ejaculations, decreased number of intromissions/ejaculation, and decreased postejaculatory interval/ejaculation as due to the increased experience received by all males over the multiple mating tests of this study. Previous experiments in the male hamster have documented that sexual experience has profound effects on both copulatory behavior as well as the brain regions controlling the behavior. Sexual experience can delay the decline in reproductive behavior following castration (Lisk and Heimann, 1980). In addition, sexual experience alone results in increased neuronal activation (as detected by Fos immunocytochemistry) within the medial nucleus of the amygdala and the medial preoptic nucleus— two brain regions known to regulate male mating behavior (Kollack-Walker and Newman, 1997). However, in addition to the effects of experience on male mating behavior, there were also changes specifically attributed to chronic inescapable but not escapable stress. The most striking effect of inescapable, but not escapable, stress was a difference in hit rate. All animals displayed an increase in mounts during the poststress mating session. However, we interpret the increased number of mounts in the no-stress control and executive males to coincide with an experience-induced increase in ejaculations; the yoked animals did not exhibit as large of an experienced-induced increase in ejaculations, but yet had the largest poststress increase in mounts. This increase in mounts with no poststress increase in intromissions or ejaculations is reflected as a marked decrease in hit rate only in the yoked animals. In addition, this complex interaction between experience and stress might also explain the lack of a significant interaction between treatment group and time for other measures. All groups were demonstrating an increase in mounts. However, only the no-stress control and executive males were increasing ejaculations. Therefore, the yoked males would continue mounting, but could not finish other consummatory aspects of the behavior. This observed effect of inescapable stress on male reproductive behavior is similar to studies in the male rat where different types of chronic stressors (immobilization, swim, and foot shock) all resulted in an increase in mounts and a decrease in hit rate (Retana-Marquez et al., 1996). Therefore, it would seem that in the hamster, like the rat, the motor components of copulation culminating with intromission are those aspects most sensitive to the inhibitory effects of stress. Unlike previous studies that have demonstrated a stressinduced decrease in testosterone secretion (Bardin and Peterson, 1967; Free and Tillson, 1973; Sapolsky, 1982, 1985), no change in plasma testosterone was detected in our
386
H.K. Holmer et al. / Hormones and Behavior 43 (2003) 381–387
animals. There are a number of explanations for this finding. First, it could be that changes in reproductive behavior occur independently of any change in peripheral testosterone secretion. Second, 12 days of this stress paradigm could be just beginning to suppress testicular function in the yoked males, and lengthening the stress paradigm would result in a more pronounced change in testosterone secretion. Third, animals could be habituating to the stressor. Both the executive and yoked animals received the same intensity of foot shock (0.8 mA) for 12 consecutive days. This intensity was chosen to use the lowest amount of shock possible that will still elicit a response in the animals (both in terms of pressing the lever and inhibiting reproduction). Over the 12 days of testing, the executive males become extremely adept at lever pressing to terminate the shock— oftentimes in 0.8 s (the minimum shock duration). Over the 80 trials the animals receive an average of 208.8 ⫾ 14.3 s of shock or approximately 2.6 s of shock per trial. If you disregard the initial day of stress where the executive animals are learning to press the bar to turn off the shock, the average shock per day decreases to 188.2 ⫾ 16.0 s or approximately 2.3 s of shock per trial. Drugan and colleagues (1997) have commented that in the rat at least 2 s of shock is required to prevent habituation to the stressor. Therefore, the hamsters in this study are receiving a low intensity stimulus for a short duration, possibly allowing for habituation or tolerance to the stress effects to occur. However, with this caveat in mind, we still observe differential effects of this stress paradigm on copulatory performance. Ongoing studies in this laboratory have attempted to diminish the effect of habituation to the stressor by increasing the number of bar presses required to turn off the shock as well as increasing the intensity of the footshock over the 12 days of stress. Preliminary results suggest eliminating the effect of habituation and/or tolerance exacerbates the effects of inescapable stress on male reproductive behavior, and also demonstrates that the subtle effects of stress reported here are reliable and repeatable (Cordner et al., 2002). Future work will use this model to characterize how the psychological variable of coping alters the brains of executive animals to protect them from the damaging effects of stress.
Acknowledgments This study was supported by grants from the VTEPSCoR Small College Research Development Fund (D.B.P. and D.L.H.), an NSF-Research Opportunity Award (D.B.P.), the Howard Hughes Medical Institute (H.K.H.), and the Middlebury College Roddy Fund (J.E.R.). We are indebted to Mr. Michael Betourney and Ms. Vicki Major for their excellent maintenance and care of the hamster colony.
References Arteaga, M., Motte-Lara, J., Velazquez-Moctezuma, J., 2000. Temporal pattern of recovery from sexual satiety in male golden hamster (Mesocricetus auratus). Physiol. Behav. 68, 591–594. Bardin, C., Peterson, R., 1967. Studies of androgen production in the rat: testosterone and androstenedione content of blood. Endocrinology 80, 38 – 44. Bunnell, B., Boland, B., Dewsbury, D., 1977. Copulatory behavior of golden hamsters. Behaviour 61, 180 –206. Chrousos, G., Gold, P., 1992. The concepts of stress and stress system disorders: overview of physical and behavioral homeostasis. JAMA 267, 1244 –1252. Cordner, A., Helmreich, D., Parfitt, D., 2002. Antidepressant administration prevents the effect of uncontrollable stress on male reproductive behavior in the Syrian hamster (Mesocricetus auratus [abstract]). Soc. Behav Neuroendocrinol. 41, 462. D’Aquila, P., Brain, P., Willner, P., 1994. Effects of chronic mild stress on performance in behavioral tests relevant to anxiety and depression. Physiol Behav. 56, 861– 867. Drugan, R., Basile, A., Ha, J., Healy, D., Ferland, R., 1997. Analysis of the importance of controllable versus uncontrollable stress on subsequent behavioral and physiological functioning. Brain Res. Protocols 2, 69 –74. Free, M., Tillson, S., 1973. Secretion rate of testicular steroids in conscious and halothane-anesthetized rat. Endocrinology 93, 874 – 879. Gorzalka, B., Hanson, L., Brotto, L., 1998. Chronic stress effects on sexual behavior in male and female rats: mediation by 5-HT2A receptors. Pharmacol Biochem. Behav. 61, 405– 412. Haracz, J., Minor, T., Wilkins, J., Zimmerman, E., 1988. Learned helplessness: an experimental model of the DST in rats. Biol. Psychiatry 23, 388 –396. Huck, U., Lisk, R., 1985. Determinants of mating success in the golden hamster (Mesocricetus auratus): I. Male capacity. J. Comp. Psychol. 99, 98 –107. Kollack-Walker, S., Newman, S., 1997. Mating-induced expression of c-fos in the male Syrian hamster brain: role of experience, pheromones, and ejaculations. J. Neurobiol. 32, 481–501. Lisk, R., Heimann, J., 1980. The effects of sexual experience and frequency of testing on retention of copulatory behavior following castration in the male hamster. Behav. Neural Biol. 28, 156 –171. Maier, S., Ryan, S., Barksdale, C., Kalin, N., 1986. Stressor controllability and the pituitary-adrenal system. Behav. Neurosci. 100, 669 – 674. Matsuda, S., Peng, H., Yoshimura, H., Wen, T., Fukuda, T., Sakanaka, M., 1996. Persistent c-fos expression in the brains of mice with chronic social stress. Neurosci. Res. 26, 157–170. Meisel, R., Sachs, B., 1994. The physiology of male sexual behavior, in: Knobil, E., Neill, J. (Eds.), The Physiology of Reproduction, Raven Press, New York, pp. 3– 84. Menendez-Patterson, A., Florez-Lozano, J., Fernandez, S., Marin, B., 1980. Stress and sexual behavior in male rats. Physiol. Behav. 24, 403– 406. Negro-Vilar, A., 1993. Stress and other environmental factors affecting fertility in men and women: overview. Environ. Health Perspect. 101 (S2), 59 – 64. Parfitt, D., Newman, S., 1998. Fos-immunoreactivity within the extended amygdala is correlated with the onset of sexual satiety. Horm. Behav. 34, 17–29. Prince, C., Anisman, H., 1990. Situation specific effects of stressor controllability on plasma corticosterone changes in mice. Pharmacol. Biochem Behav. 37, 613– 621. Retana-Marquez, S., Dominguez-Salazar, E., Velazquez-Moctezuma, J., 1996. Effect of acute and chronic stress on masculine sexual behavior in the rat. Psychoneuroendocrinology 21, 39 –50. Rivier, C., 1995. Luteinizing hormone releasing hormone, gonadotropins, and gonadal steroids in stress. Ann. NY Acad. Sci. 771, 187–191.
H.K. Holmer et al. / Hormones and Behavior 43 (2003) 381–387 Sapolsky, R., 1982. The endocrine stress-response and social status in the wild baboon. Horm. Behav. 16, 279 –292. Sapolsky, R., 1985. Stress-induced suppression of testicular function in the wild baboon: role of glucocorticoids. Endocrinology 116, 2273–2278. Seligman, M., Issacowitz, D., 2000. Learned helplessness, in: Fink, G. (Ed.), The Encyclopedia of Stress, Vol. 2, Academic Press, San Diego, pp. 599 – 602. Seligman, M., Weiss, J., 1980. Coping behavior: learned helplessness, physiological change, and learned activity. Behav. Res. Ther. 18, 459–512. Sklar, L., Anisman, H., 1979. Stress and coping factors influence tumor growth. Science 205, 513–515. Swenson, R., Vogel, W., 1983. Plasma catecholamine and corticosterone as well as brain catecholamine changed during coping in rats exposed to stressful footshock. Pharmacol. Biochem. Behav. 18, 689 – 693.
387
Visintainer, M., Volpicelli, J., Seligman, M., 1982. Tumor rejection in rats after inescapable or escapable shock. Science 216, 437– 439. Weiss, J., Goodman, P., Losito, B., Corrigan, S., Charry, J., Bailey, W., 1981. Behavioral depression produced by an uncontrollable stressor: relationship to norepinephrine, dopamine, and serotonin levels in various regions of the rat brain. Brain Res. Rev. 3, 167–205. Wood, R., Swann, J., 2000. Neuronal integration of chemosensory and hormonal signals in the control of male sexual behavior, in: Waller, K., Schneider, J. (Eds.), Reproduction in Context: Social and Environmental Influences on Reproductive Physiology and Behavior, MIT Press, Cambridge, MA, pp. 423– 444. Young, E., Korszun, A., 1998. Psychoneuroendocrinology of depression: hypothalamic-pituitary-gonadal axis, in: Nemeroff, C. (Ed.), Psychiatric Clinics: Psychoneuroendocrinology, Vol. 21, W.B. Saunders, Philadelphia, PA, pp. 309 –324.