Physiology & Behavior, Vol. 21, pp. 409-415. Pergamon Press and Brain Research Publ., 1978. Printed in the U.S.A.
The Interaction of Photoperiod and Testosterone on the Development of Copulatory Behavior in Castrated Male Hamsters CONSTANCE
S. C A M P B E L L ,
J O E L S. F I N K E L S T E I N
2 AND FRED W. TUREK
D e p a r t m e n t o f Biological Sciences, N o r t h w e s t e r n University, Evanston, I L 60201 ( R e c e i v e d 19 D e c e m b e r 1977) CAMPBELL, C. S., J. S. FINKELSTEIN AND F. W. TUREK. The interaction ofphotoperiod and testosterone on the development of copulatory behavior in castrated male hamsters. PHYSIOL. BEHAV. 21(3) 409-415, 1978.--Castrated adult male golden hamsters were maintained on either a stimulatory (LD 14:10) or a nonstimulatory (LD 6:18) light cycle for 10 weeks, and then were implanted subcutaneously with empty or testosterone-filled Silastic capsules of various lengths. Animals were tested for copulatory behavior prior to capsule implantation and 10, 20 and 40 days after implantation. Androgen-treated LD 14:10 hamsters showed a higher incidence of ejaculation on the final trial than did similarly treated LD 6:18 animals. When serum androgen levels were maintained at physiological levels (about 3 ng/ml, from a 20 mm capsule), significantly greater numbers of LD 14:10 hamsters intromitted and ejaculated compared to LD 6:18 animals. Examination of the development of copulation over trials revealed an interaction of photoperiod and androgen: LD 14:10 animals showed significant improvement over trials if stimulated with 8, 20 or 100 mm long testosterone capsules, while LD 6:18 animals showed increased copulation over trials only if they were implanted with 100 mm capsules. These results indicate that exposure to short days for 10 weeks renders copulatory behavior of the castrate male hamster less responsive to the stimulatory effects of testosterone. This alteration in sensitivity to androgen may be one way in which the photoperiod acts to decrease copulation in seasonally breeding animals. Photoperiodicity
Copulatory behavior
Testosterone
IN A W I D E variety of animals exposure to a seasonal shortening of daylength results in gonadal regression and a cessation of reproductive activity [4,16]. Although considerable information on the physiology of this system has been obtained, there has been remarkably little investigation of the changes in copulatory behavior which might accompany gonadal changes. Studies of Japanese quail have documented a decline in sexual behavior in short day photoperiods [18], as have studies of the red deer stag [13] and the lizard [6]. Serum testosterone levels in the male golden hamster decline after exposure to short-day photoperiods [4] and it is possible that this decrease in androgen could be the major factor leading to a decrease in copulatory behavior. However, even though exogenous testosterone results in a dosedependent restoration of copulation in castrate male hamsters [26], endogenous serum testosterone levels have rarely been shown to be reliably correlated with levels of sexual performance [8]. Furthermore, there is considerable evidence indicating that there are changes in the responsiveness of central nervous system thresholds to androgen stimu-
Castration
Photoperiod and androgen interaction
lation in short-day photoperlods [15, 21, 25]. Exposure to short days for ten weeks renders castrate hamsters extremely sensitive to the negative feedback effect of testosterone such that serum L H and F S H are considerably more suppressed than they are in animals exposed to photostimulatory light cycles [25]. The purpose of the present study was to determine whether photoperiod length can induce changes in sexual behavior of the male golden hamster apart from its effects on circulating testosterone. Thus, castrate animals maintained in stimulatory and nonstimulatory light cycles were administered different doses of exogenous testosterone in order to identify differences in thresholds of responsiveness. Since it was important that all animals were deprived of androgen for a similar length of time before restoration, animals were castrated before being exposed to different photoperiods. A second goal was to identify the controlling factors of the precise parameters of male copulatory behavior through an examination of the impact of photoperiod and testosterone on those parameters. Finally, it was felt to be of considerable interest to monitor the gradual development of copulatory
~We express our most sincere appreciation to Shirley Snerling, Brig,iRe G. Mann and Julian A. Alvarez for their expert technical assistance, and Dr. G. D. Niswender for contributing antiserum for the steroid assay. This study was supported by NIH grants HD--09885 and NSF grant PCM 76-09955 to F. W. Turek and by NIH grant HD--10050 to C. S. Campbell. SNow at Washington University School of Medicine, St. Louis, MO 63110.
C o p y r i g h t © 1978 B r a i n R e s e a r c h P u b l i c a t i o n s Inc.--0031-9384/78/090409-07502.00/0
410
CAMPBELL, FINKELSTEIN AN D TUREK
patterns in the naive male hamster and to examine the extent to which testosterone levels and photoperiod interact to elicit this development. MATERIALS AND METHOD
Animals Ninety-six 9-week old male Golden Hamsters (Mesocricetus auratus) were obtained from Lakeview Hamster Colony, Newfield, NJ. Animals were initially group-housed in LD 14:10 with food and water ad lib.
in hamsters [3], we verified our ability to detect its occurrence in pilot tests by examining the vaginas of stimulus females for the presence of sperm. Aggressive acts were defined as pushing, biting, chasing or boxing between a male and the stimulus female with which he was paired [22]. During the course of the experiment, five animals died from cardiac puncture. All data are expressed with an indication of the number of animals in each group at each time: means and standard errors of the data in the tables were computed only on animals which showed the behavior in question.
General Procedure One week after adaptation to the laboratory, animals were anesthetized with sodium pentobarbital (60 mg/kg IP) and bilaterally castrated. Immediately after surgery, animals were placed into one of two photoperiodic regimes: LD 14:10 (lights on at 0700 and off at 2100) and LD 6:18 (lights on at 0800 and off at 1400). After 9 weeks exposure to these photoperiods, all animals were tested for sexual behavior (see below). Five days following the behavioral tests, the animals were implanted with subcutaneous Silastic capsules; animals in both photoperiods received either empty or testosterone-filled capsules that were 2, 4, 8, 20 or 100 m long. Further behavioral tests were conducted 10, 20 and 40 days after implantation. Cardiac punctures were performed at the time of implantation as well as 21 and 42 days later for measurement of serum LH and FSH; these results are reported elsewhere [25]. Animals were autopsied 42 days after implantation; blood was collected and the length of the right flank gland was measured.
Statistics The frequency of ejaculation and intromission on the last trial was analyzed using the G test of independence adapted for multiway tables [20]. Individual comparisons between photoperiod groups at each dose of testosterone were made using Fisher's exact test incorporating Tocher's modification [19]. Analysis of variance was used to determine the effects of testosterone doses on the specific components of copulatory behavior and on flank gland length. Differences between treatments were tested for significance using the StudentNewman-Keuls sequential range test [20]. Analysis of variance was also used to determine the effects of photoperiod on components of copulatory behavior on the last trial. Changes in the incidence of mounting, intromission and ejaculation over trials were analyzed using Cochran's Q test [19]. RESULTS
Impact of Photoperiod on Copulatory. Behavior Testosterone: Treatment and Assays Crystalline testosterone was placed into Silastic tubing (3.18 mm OD; 1.98 mm ID, Dow-Corning Co.) by methods previously described [24]. Capsule size is presented as the length of tubing filled with testosterone crystals; the amount of testosterone released is relatively constant and proportional to capsule length [15]. Serum testosterone levels were measured by radioimmunoassay methods previously reported [25]. Behavioral Tests Mating tests were conducted using ovariectomized females rendered highly estrous by subcutaneous injections of 10/zg of estradiol cypionate 42 hr before testing and 0.5 mg of progesterone 6 hr before testing. Tests were run beginning at 2300, which was approximately 15-16 hr after lights onset for both photoperiod groups. Thus, for both groups, this testing was done during their "subjective night" [11] at their time of maximal behavioral responsiveness [9]. Each male was introduced to the observation arenas for 5 min before the stimulus females were added. The following parameters were recorded: intromission latency (IL), time from introduction of the female to the first intromission; ejaculatory latency (EL), time from first intromission to ejaculation; post-ejaculatory period (PEP), time from ejaculation to the next intromission; intromission frequency (IF); and mount frequency (MF) [2]. Tests were terminated ff IL was greater than 15 rain, if EL was greater than 30 rain, or when the first intromission of the second series had occurred. Due to the difficulty of reliably detecting ejaculation
Of the total number of animals tested under both photoperiods, there was a higher incidence of ejaculation on the final test among animals exposed to long days (14/46) compared to animals exposed to short days (6/45) independent of testosterone dose. This was also the case with respect to the incidence of intromission (18/46 vs 12/45). These differences were significant for incidence of ejaculation (G(t)=3.98; p0.1). When animals were being stimulated with physiological levels of testosterone (20 mm capsule) there were signifi, cantly higher incidences of both ejaculation 60
The Interaction of Photoperiod and Testosterone on the Development of Copulatory Behavior in Castrated Male Hamsters CONSTANCE
S. C A M P B E L L ,
J O E L S. F I N K E L S T E I N
2 AND FRED W. TUREK
D e p a r t m e n t o f Biological Sciences, N o r t h w e s t e r n University, Evanston, I L 60201 ( R e c e i v e d 19 D e c e m b e r 1977) CAMPBELL, C. S., J. S. FINKELSTEIN AND F. W. TUREK. The interaction ofphotoperiod and testosterone on the development of copulatory behavior in castrated male hamsters. PHYSIOL. BEHAV. 21(3) 409-415, 1978.--Castrated adult male golden hamsters were maintained on either a stimulatory (LD 14:10) or a nonstimulatory (LD 6:18) light cycle for 10 weeks, and then were implanted subcutaneously with empty or testosterone-filled Silastic capsules of various lengths. Animals were tested for copulatory behavior prior to capsule implantation and 10, 20 and 40 days after implantation. Androgen-treated LD 14:10 hamsters showed a higher incidence of ejaculation on the final trial than did similarly treated LD 6:18 animals. When serum androgen levels were maintained at physiological levels (about 3 ng/ml, from a 20 mm capsule), significantly greater numbers of LD 14:10 hamsters intromitted and ejaculated compared to LD 6:18 animals. Examination of the development of copulation over trials revealed an interaction of photoperiod and androgen: LD 14:10 animals showed significant improvement over trials if stimulated with 8, 20 or 100 mm long testosterone capsules, while LD 6:18 animals showed increased copulation over trials only if they were implanted with 100 mm capsules. These results indicate that exposure to short days for 10 weeks renders copulatory behavior of the castrate male hamster less responsive to the stimulatory effects of testosterone. This alteration in sensitivity to androgen may be one way in which the photoperiod acts to decrease copulation in seasonally breeding animals. Photoperiodicity
Copulatory behavior
Testosterone
IN A W I D E variety of animals exposure to a seasonal shortening of daylength results in gonadal regression and a cessation of reproductive activity [4,16]. Although considerable information on the physiology of this system has been obtained, there has been remarkably little investigation of the changes in copulatory behavior which might accompany gonadal changes. Studies of Japanese quail have documented a decline in sexual behavior in short day photoperiods [18], as have studies of the red deer stag [13] and the lizard [6]. Serum testosterone levels in the male golden hamster decline after exposure to short-day photoperiods [4] and it is possible that this decrease in androgen could be the major factor leading to a decrease in copulatory behavior. However, even though exogenous testosterone results in a dosedependent restoration of copulation in castrate male hamsters [26], endogenous serum testosterone levels have rarely been shown to be reliably correlated with levels of sexual performance [8]. Furthermore, there is considerable evidence indicating that there are changes in the responsiveness of central nervous system thresholds to androgen stimu-
Castration
Photoperiod and androgen interaction
lation in short-day photoperlods [15, 21, 25]. Exposure to short days for ten weeks renders castrate hamsters extremely sensitive to the negative feedback effect of testosterone such that serum L H and F S H are considerably more suppressed than they are in animals exposed to photostimulatory light cycles [25]. The purpose of the present study was to determine whether photoperiod length can induce changes in sexual behavior of the male golden hamster apart from its effects on circulating testosterone. Thus, castrate animals maintained in stimulatory and nonstimulatory light cycles were administered different doses of exogenous testosterone in order to identify differences in thresholds of responsiveness. Since it was important that all animals were deprived of androgen for a similar length of time before restoration, animals were castrated before being exposed to different photoperiods. A second goal was to identify the controlling factors of the precise parameters of male copulatory behavior through an examination of the impact of photoperiod and testosterone on those parameters. Finally, it was felt to be of considerable interest to monitor the gradual development of copulatory
~We express our most sincere appreciation to Shirley Snerling, Brig,iRe G. Mann and Julian A. Alvarez for their expert technical assistance, and Dr. G. D. Niswender for contributing antiserum for the steroid assay. This study was supported by NIH grants HD--09885 and NSF grant PCM 76-09955 to F. W. Turek and by NIH grant HD--10050 to C. S. Campbell. SNow at Washington University School of Medicine, St. Louis, MO 63110.
C o p y r i g h t © 1978 B r a i n R e s e a r c h P u b l i c a t i o n s Inc.--0031-9384/78/090409-07502.00/0
410
CAMPBELL, FINKELSTEIN AN D TUREK
patterns in the naive male hamster and to examine the extent to which testosterone levels and photoperiod interact to elicit this development. MATERIALS AND METHOD
Animals Ninety-six 9-week old male Golden Hamsters (Mesocricetus auratus) were obtained from Lakeview Hamster Colony, Newfield, NJ. Animals were initially group-housed in LD 14:10 with food and water ad lib.
in hamsters [3], we verified our ability to detect its occurrence in pilot tests by examining the vaginas of stimulus females for the presence of sperm. Aggressive acts were defined as pushing, biting, chasing or boxing between a male and the stimulus female with which he was paired [22]. During the course of the experiment, five animals died from cardiac puncture. All data are expressed with an indication of the number of animals in each group at each time: means and standard errors of the data in the tables were computed only on animals which showed the behavior in question.
General Procedure One week after adaptation to the laboratory, animals were anesthetized with sodium pentobarbital (60 mg/kg IP) and bilaterally castrated. Immediately after surgery, animals were placed into one of two photoperiodic regimes: LD 14:10 (lights on at 0700 and off at 2100) and LD 6:18 (lights on at 0800 and off at 1400). After 9 weeks exposure to these photoperiods, all animals were tested for sexual behavior (see below). Five days following the behavioral tests, the animals were implanted with subcutaneous Silastic capsules; animals in both photoperiods received either empty or testosterone-filled capsules that were 2, 4, 8, 20 or 100 m long. Further behavioral tests were conducted 10, 20 and 40 days after implantation. Cardiac punctures were performed at the time of implantation as well as 21 and 42 days later for measurement of serum LH and FSH; these results are reported elsewhere [25]. Animals were autopsied 42 days after implantation; blood was collected and the length of the right flank gland was measured.
Statistics The frequency of ejaculation and intromission on the last trial was analyzed using the G test of independence adapted for multiway tables [20]. Individual comparisons between photoperiod groups at each dose of testosterone were made using Fisher's exact test incorporating Tocher's modification [19]. Analysis of variance was used to determine the effects of testosterone doses on the specific components of copulatory behavior and on flank gland length. Differences between treatments were tested for significance using the StudentNewman-Keuls sequential range test [20]. Analysis of variance was also used to determine the effects of photoperiod on components of copulatory behavior on the last trial. Changes in the incidence of mounting, intromission and ejaculation over trials were analyzed using Cochran's Q test [19]. RESULTS
Impact of Photoperiod on Copulatory. Behavior Testosterone: Treatment and Assays Crystalline testosterone was placed into Silastic tubing (3.18 mm OD; 1.98 mm ID, Dow-Corning Co.) by methods previously described [24]. Capsule size is presented as the length of tubing filled with testosterone crystals; the amount of testosterone released is relatively constant and proportional to capsule length [15]. Serum testosterone levels were measured by radioimmunoassay methods previously reported [25]. Behavioral Tests Mating tests were conducted using ovariectomized females rendered highly estrous by subcutaneous injections of 10/zg of estradiol cypionate 42 hr before testing and 0.5 mg of progesterone 6 hr before testing. Tests were run beginning at 2300, which was approximately 15-16 hr after lights onset for both photoperiod groups. Thus, for both groups, this testing was done during their "subjective night" [11] at their time of maximal behavioral responsiveness [9]. Each male was introduced to the observation arenas for 5 min before the stimulus females were added. The following parameters were recorded: intromission latency (IL), time from introduction of the female to the first intromission; ejaculatory latency (EL), time from first intromission to ejaculation; post-ejaculatory period (PEP), time from ejaculation to the next intromission; intromission frequency (IF); and mount frequency (MF) [2]. Tests were terminated ff IL was greater than 15 rain, if EL was greater than 30 rain, or when the first intromission of the second series had occurred. Due to the difficulty of reliably detecting ejaculation
Of the total number of animals tested under both photoperiods, there was a higher incidence of ejaculation on the final test among animals exposed to long days (14/46) compared to animals exposed to short days (6/45) independent of testosterone dose. This was also the case with respect to the incidence of intromission (18/46 vs 12/45). These differences were significant for incidence of ejaculation (G(t)=3.98; p0.1). When animals were being stimulated with physiological levels of testosterone (20 mm capsule) there were signifi, cantly higher incidences of both ejaculation 60
