HORMONES
AND
BEHAVIOR
25,
382-393 (1991)
Reciprocal Relationships between Puisatile Androgen Secretion and the Expression of Mating Behavior in Adult Male Ferrets G. M. LAMBERT AND M. J. BAUM Department of Biology,
Boston Universi@, Boston, Massachussets 02215
The pulsatile secretion of androgen was similar over a 12-hr period in breeding male ferrets implanted with jugular catheters which either achieved an intromission with an estrous female or received no socio-sexual contact. This negative result contrasts with the previous demonstration (Carroll, Erskine, and Baum, 1987, Endocrinology 121, 1349-1359) of a significant, delayed rise in mean plasma androgen concentrations in breeding male ferrets 5-12 hr after mating. Males used in that previous study had lower initial mean plasma levels of androgen and smaller testis diameters than the present males. We therefore asked whether differences in circulating androgen levels, characteristic of males in different phases of the seasonal breeding cycle, might affect the expression of mating behavior. Castrated males given 0, 0.2, 2.0, or 5.0 mg/kg of testosterone propionate (TP) showed dose-related increases in the expression of different components of sexual behavior, including neck gripping, mounting, and intromitting. Surprisingly, intromissive performance was significantly better in intact breeding males than in castrates given even the highest dosage of TP. These results suggest that ferrets’ mating performance may vary with seasonal variations in androgen availability, and that the ability of males to exhibit a postcoital increase in the testicular secretion of androgen may be limited to the beginning or end of the breeding season, when circulating levels of androgen are relatively low. Mating-induced increments in androgen secretion at these times may enhance subsequent reproductive successby facilitating males’ intromissive capacity, which is required for the induction of ovulation and optimal sperm transport in female partners. o 1991 Academic
Press, Inc.
Engaging in mating behavior stimulates the rapid release of testicular androgens in males of several species in which the female normally ovulates spontaneously, including the rat (Kamel and Frankel, 1978), pig (Ellendorph, Parvizi, Pomerantz, Hartjen, Konig, Smidt, and Elsaesser, 1975), bull (Katongole, Naftolin, and Short, 1971) and rhesus monkey (Herndon, Perachio, Turner, and Collins, 1981). This androgen rise results from the action of luteinizing hormone (LH) on the testes, and has been hypothesized to enhance subsequent masculine mating performance (Hart, 1983). Surprisingly, in the male ferret, a species in which the female 382 0018-506X/91 $1.50 Copyright 6 1991 by Academic Press, Inc. All rights of reproduction in any form reserwd.
ANDROGENS
AND MATING IN MALE FERRETS
383
ovulates reflexively in response to mating, the pulsatile secretion of LH was significantly inhibited, and mean androgen levels were slightly depressed l-5 hr following coitus (Carroll, Erskine, and Baum, 1987). In that study a delayed elevation in mean plasma androgen secretion occurred in males 5-12 hr after an intromission, even though the secretion of LH was not augmented prior to or during this period. Secretion of testosterone (T) is known to occur in a pulsatile manner in the male ferret (Sisk and Desjardins, 1986); however, blood samples were not collected frequently enough to allow pulsatile analysis of androgen secretion in the study by Carroll et al. (1987). A primary aim of the present study was to characterize more precisely the effects of mating on androgen secretion in the male of a species in which the female is an induced ovulator. Blood samples were collected at lo-min intervals from male ferrets in breeding condition both after mating and in the absence of a female in order to allow an analysis of the pulsatile secretion of androgen. Contrary to expectations, no significant effects of achieving an intromission on mean levels or pulses of androgen were seen. Further examination of these data showed that the initial mean plasma levels of androgen were twice as high in males used in the present study as in males used previously by Carroll et al. (1987). We hypothesized that this difference in initial plasma androgen levels could affect the expression of mating behaviors by the male ferret and the resultant postcoital secretion of androgen. Sexual behaviors are displayed by male ferrets in a specific sequence, beginning with neck gripping of the dorsal surface of the female’s neck. The male ferret eventually mounts the female and exhibits episodic pelvic thrusting which ceasesonce intromission is achieved (Baum and Schretlen, 1975). The expression of neck gripping and mounting behaviors by male ferrets increased in duration with increasing plasma levels of testosterone (Baum and Tobet, 1988). However, the effect of circulating testosterone on intromissive performance has not been studied in the ferret. It has recently been suggested that the duration of receipt of penile intromission is positively correlated with the probability of fertilization in the female ferret (Miller and Anderson, 1989). Thus, higher circulating androgen levels may stimulate intromissions of a longer duration in males. The androgen rise that occurs after mating in ferrets with initially low plasma androgen may have evolved as a mechanism to increase intromissive capacity for future mating encounters, thereby extending the breeding season and increasing the probability of fertilization. A second aim of this study was to examine whether plasma androgen levels, resulting from the exogenous administration of increasing doses of testosterone propionate (TP) to castrated male ferrets, are positively correlated with the expression of masculine coital behaviors, especially intromission.
384
LAMBERT
AND
BAUM
METHODS Subjects Adult male and female ferrets were purchased directly from Marshall Farms, (North Rose, NY), between the months of January and June. Once in our laboratory these subjects were housed individually in modified stainless-steel rabbit cages under a long day photoperiod (lights on 06:OO to 22:00 hr). Water was available ad libitum, and a bowl of fresh Beef and Liver Cat Food mixed with dry Kat Krunchies (Old Mother Hubbard) was provided once daily. Ferrets are brought into breeding condition throughout the year at Marshall Farms using artificial light. Males were received from Marshall Farms in breeding condition after exposure to a long-day light cycle and these subjects successfully intromitted with stimulus females when tested l-2 days after their arrival. Some adult male ferrets were castrated via a midline scrotal incision, and/or fitted with intrajugular catheters as described previously (Carroll et al., 1987). All surgical procedures were performed under sterile conditions and using xylazine (Rompun: HaverMiles Labs, Shawnie, KS; 4.0 mg/kg) and ketamine (Ketaset: Bristol Labs, Syracuse, NY; 35 mg/kg) anesthesia. The castrated males were later injected with testosterone propionate prior to mating sessions as described below. Adult females were ovariectomized via a single midline abdominal incision and were made sexually receptive by daily SCinjection of 12-15 pg/kg of estradiol benzoate (EB) in sesame oil. These animals served as stimuli for testing the mating behaviors of male ferrets. Androgen Assay Plasma androgen levels were estimated by radioimmunoassay (Erskine and Baum, 1982). The assay standards ranged from 10 to 400 pg, and the interassay and intraassay coefficients of variation were 18.6 and 13.5%, respectively. Steroids were extracted from the plasma samples using ether, but were not separated further using chromatography. Since the testosterone antiserum (T-ll-BSA S#250; Dr GD Niswender, CO) cross-reacts with dihydrotestosterone (26.0%) and other testosterone metabolites, assay values were expressed as nanograms androgen/milliter. Concentrations of androgen were calculated using the four parameter logistic method (Rodbard and Hutt, 1974). Androgen pulses were identified using the PULSAR computer program (Merriam and Wachter, 1982) with G values defined previously for androgen pulses in the ferret by Sisk and Desjardins (1986). Mating Sessions Experimental males were paired with a stimulus female in a mating cage (30 x 67 x 51 cm) equipped with a Plexiglas front to facilitate
ANDROGENS
AND MATING
IN MALE FERRETS
385
observations. The frequency and duration of mating behaviors displayed by the male ferret were recorded using a Toshiba laptop computer. The behaviors of interest were neck gripping, mounting, pelvic thrusting, and penile intromission. The occurrence of an intromission was confirmed by the experimenter with a manual check. This procedure did not seem to disturb the subjects and was essential in that the occurrence of intromissions cannot always be detected with visual observation. Experimental Procedures Pulsatile androgen secretion in breeding males. One day after catheterization, half of the males were paired individually with a stimulus female in mating cages, while the other subjects were placed alone into empty cages of a comparable size. Mating behaviors were recorded in the paired subjects for 1 hr. At the end of this testing period, males which had mated and those which were not paired with females were placed back into their home cages. Polyvinyl extensions were attached to the catheter ends and blood samples were withdrawn every 10 min for the next 11 hr. A blood replacement mixture (Sisk and Desjardins, 1986) was infused every 30 min through the jugular catheter in order to maintain subjects’ blood hematocrit and fluid balance. Blood samples were kept on ice until being centrifuged at 5°C 1006g, for 20 min. Plasma was stored at -20°C until the androgen radioimmunoassay was performed. Expression of sexual behavior in male ferrets. The mating behaviors of sexually experienced male ferrets which had been castrated 4 weeks earlier were recorded for a 1-hr time period. Castrated male ferrets were then injected SCdaily with 0.2 mg/kg of TP. After 10 days of steroid administration the mating behaviors of these males were recorded for 2 hr. In all mating tests that followed TP treatment, the observation period was extended to 4 hr if an intromission occurred within the first 2 hr. Castrated male ferrets then received an injection of 2.0 mg/kg TP and the durations of mating behaviors which followed in the next 2 hr in response to pairing with a stimulus female showed that a single pulse of plasma testosterone failed to enhance the expression of reproductive behavior. The injections of this dose of TP (2.0 mg/kg) continued for 9 more days, and male ferrets were retested. Lastly, a similiar mating test was conducted in male ferrets after 10 days of injections with 5.0 mg/kg TP. Some males were fitted with intrajugular catheters after being injected for 10 days, including at 10 AM on the day of surgery, with one of the three doses of TP. On the day following surgery at 8 AM, a blood sample was drawn (22 hr after the previous TP injection), another TP injection of the same dose was given, and blood samples were collected 1, 2, 4, 8, 12, and 16 hr later. Blood samples were centrifuged at 5”C, and the plasma was stored at -20°C until the androgen RIA was conducted.
386
LAMBERT
AND BAUM
TABLE I Parameters of Pulsatile Androgen Secretion in Gonadally Intact Male Ferrets in a Breeding Condition Which Achieved an Intromission or Which Were Not Exposed to a Stimulus Female Parameter Pulse amplitude (nshl)
Pulse duration (min)
No female present (7) 0.34 * 0.06
21.48 + 2.19
65.48 f 13.60
Intromission with female (8) 0.35 I 0.04
23.69 f 0.94
76.48 f 06.42
Group (n)
Pulse frequency (No. pulses/hr)
Note. All data are expressed as means 2 SEM.
Statistical Analyses
The overall mean androgen pulse frequencies, amplitudes, and durations for males which achieved an intromission and for control males in breeding condition which were not paired with a female were calculated. In order to make hourly group comparisons of the pulsatile release of androgens over the total blood collection period, the number of minutes included in any part of a pulse was counted for each hour, regardless of the overall duration of each pulse. Hourly group means for androgen concentration and the number of minutes included in a pulse were compared using a two-way analysis of variance (ANOVA) with repeated measures on one factor. The behavioral data for castrated males treated with different doses of TP and for intact males in breeding condition were adjusted for time in order to examine the mean duration per hour for each variable. The differences among the hourly group means for each reproductive behavior were analyzed using a one-way ANOVA. Post hoc analyses of different group means were made using Duncan’s Multiple-Range tests. RESULTS AND DISCUSSION Pulsatile Androgen Secretion in Breeding Male Ferrets
There were no significant effects of mating on androgen pulse frequency, amplitude, or duration over the total blood collection period (Table 1). Also, both the mean androgen levels per hour (Fig. 1, bottom panel, F(l, 13) = 0.48, P < 0.05), and the number of minutes per hour included in a pulse (Fig. 1, top panel, F(l, 13) = 0.01, P < 0.05) were similar in mated and unpaired male ferrets. Although the duration of intromission (28.6 2 3.6 min vs 25.6 2 2.6 min) and other behavioral parameters (data not shown) displayed by males
ANDROGENS
AND MATING IN MALE FERRETS
381
0 Na Female Present (7) m lntramission with Female (8)
t-4 4-5 i-6 6-7 m.ft?s A ‘ER cJ%!SU OF 7-EST FIG. 1. Hourly (means f SEM) plasma levels of androgen (rig/ml; lower panel) and number of minutes (means f SEM) included in a pulse each hour (top panel) in male ferrets in breeding condition. Blood samples were collected for 11 hr following a 1-hr mating test during which an intromission occurred with a stimulus female (striped bars) or during which males were housed in a cage with no female present for 1 hr (open bars).
in the present study resembled those of males studied by Carroll et al. (1987), the males used in the present study failed to show a postcoital rise in mean androgen levels. While Carroll et al. (1987) reported a significant increase in mean plasma androgen levels 5-12 hr following an intromission, we found that mean androgen levels were similar over a 12hr period in males which achieved an intromission and in males which were not paired with a female. The present results corroborate findings reported for another induced ovulating species, the rabbit (Blake, Blake, Thorneycroft, and Thorneycroft, 1978). Serum T was elevated in males with low initial plasma levels of testosterone after exposure to a female rabbit, while subjects with higher plasma T showed no change or even decreased androgen levels after exposure to female conspecifics. This phenomenon is not limited to induced ovulating species. A rise in plasma T occurred after exposure to an estrous female in male mice, a sponta-
388
LAMBERT
AND
BAUM
neously ovulating species, only when initial plasma levels of testosterone were low (Batty, 1978). Presumably, testicular secretions were already maximal in subjects with initially high plasma T levels and could not be stimulated further by exposure to a female. In addition to having higher initial mean plasma levels of androgen (12.09 + 1.30 vs 5.16 _+ 1.22 rig/ml), the male ferrets used in the present study had larger testis diameters (15.00 + 0.23 vs 11.5 2 0.26 mm) and shorter thrusting durations (4.24 + 1.11 min vs 11.46 + 2.43 min) than males used previously by Carroll et al. (1987). This suggests that the present males were in the height of the breeding condition and achieved intromission with estrous females more easily than the subjects in Carroll et al. ‘S study (1987). Males used in the present study may also have had the capacity to achieve intromisions of longer duration than did Carroll’s subjects, had mating tests not been limited to 1 hr in both studies. Thus, a postcoital rise in androgen secretion may occur selectively in ferrets which are marginally in breeding condition. The functional implications of a postcoital rise in androgens for males at the end of their breeding season may include extending the time period during which reproductive behaviors, especially intromission, are displayed. Males’ fertility is thereby enhanced, in so far as intromission is required to trigger the preovulatory LH surge in estrous females, and longer intromissions increase the likelihood of sperm transport and successful fertilization of the female (Miller and Anderson, 1989). Expression of Mating Behaviors in Male Ferrets
Administration of TP to castrated male ferrets resulted in dose-dependent increases in plasma testosterone levels (Fig. 2). In each instance there was a small increment in testosterone immediately following the daily SCinjection; however, plasma levels of testosterone then remained relatively stable for the next 24 hr between injections. There were group differences in the durations of neck gripping (F(4, 27) = 57.77, P < 0.05) and mounting (F(4, 27) = 39.36, P < 0.05) shown by male ferrets, depending upon the type of androgenic stimulation which prevailed (Fig. 3). The durations of neck gripping and mounting were significantly lower (P < 0.05) in castrated males given either no TP or 0.2 mg/kg TP than in castrates given either 2.0 or 5.0 mg/kg of TP or in intact males in breeding condition. There were also significant group differences in thrusting duration (F(4,27) = 12.03, P < 0.05). Less pelvic thrusting was displayed by castrated male ferrets given no TP or 0.2 mg/kg TP and by intact subjects than by castrated males given 2.0 or 5.0 mg/kg TP. Finally, there were group differences in intromission duration (F(4, 27) = 10.73, P < 0.05). Intromission duration was lowest in castrated males given no TP, increased significantly in males given 5.0 mg/kg, and
ANDROGENS
I
4
389
AND MATING IN MALE FERRETS
I 6
I
I
I
I
12
16
20
24
FIG. 2. Plasma androgen concentrations (rig/ml) resulting from daily subcutaneous injections of testosterone propionate (TP) to castrated male ferrets for 12 days. Blood samples were withdrawn from intrajugular catheters at 1, 2,4, 8, 12, 16, and 22 hr following a daily injection of one of three doses of TP. The number of animals sampled for each TP dose at each time point is given above the error bars.
was significantly greater in intact breeding males than in all castrated, androgen-treated groups. These data raise the possibility that the duration of an intromission in male ferrets is dependent upon the plasma level of testosterone and/or the ongoing profile of androgen secretion. This differs from results obtained in males of several species in which the female ovulates spontaneously where mating performance was not highly correlated with plasma levels of testosterone. For example, in the castrated male rat, a minimal replacement of circulating testosterone was sufficient to sustain mating performances at levels qualitatively and quantitatively similar to those of gonadally intact males (Albert, Jonik, Watson, Gorzalka, and Walsh, 1990). Castrated male rats maintained a normal pattern of ejaculatory performance with only 10% of the circulating testosterone normally present in gonadally intact males (Damassa, Smith, Tennent, and Davidson, 1977). In male sheep, administration of 4 mg/kg TP, a dose which yielded mean plasma T levels well below those naturally occurring in intact, breeding subjects, induced the full display of mating behavior (D’Occhio
390
LAMBERT
AND BAUM
TP (mg/kg) Elj 0.0 Castrated E Fi I 5.0 1 q 0.0~Intact, Breeding
Neck Gripptng Mounting Pelvic Thrusting lntromitting FIG. 3. Mean durations (minutes/hour) of components of masculine sexual behavior displayed by male ferrets which were castrated for 4 weeks and subsequently treated with
0, 0.2, 2.0, or 5.0 mg/kg TP for 10 days or which were tested while gonadally intact and in breeding condition. All data are expressed as means + SEM. The proportion of subjects in each group that expressed each behavior is given above the error bars.
and Brooks, 1982). Lastly, in male guinea pigs, plasma T levels were not correlated with individuals’ mating performance (Harding and Feder, 1976). In contrast, the durations of most mating behaviors shown by castrated male ferrets were positively related to the dose of TP injected. The duration of pelvic thrusting appears to be an exception to this relationship in the ferret. Gonadally intact male ferrets used by Carroll et al. (1987) had lower plasma androgens but longer thrusting durations than the subjects in the present study; however, a shorter thrusting duration is probably indicative of an enhanced ability to achieve an intromission. Thus, a different sequence and pattern of mating behaviors occurs in male ferrets as a result of different endogenous circulating levels of testosterone. These differences in sexual behavior may affect postcoital sensitivity of hypothalmic-pituitary-testicular axis to further activation. Pulsatile exposure of the nervous system to testosterone may be more effective than tonic exposure for the stimulation of mating behaviors in male ferrets. While gonadally intact, breeding males had mean androgen levels (5-15 rig/ml) similar to those obtained after injecting 0.2 mg/kg TP into castrated male ferrets (5-10 rig/ml), the durations of all mating behaviors, except for pelvic thrusting, were significantly longer in the intact group, which had been exposed to a pulsatile profile of androgenic stimulation. Intact males achieved intromission durations that were even longer than those of castrated males given the highest dose (5.0 mg/kg)
ANDROGENS
AND MATING
IN MALE FERRETS
391
of TP (50-70 rig/ml plasma androgen). Additional research is needed to assessthe importance of pulsatile exposure to testosterone in the male ferret for the expression of mating behaviors. An alternative explanation for the shorter intromission durations seen in the castrated, TP-treated subjects, as compared with gonadally intact breeding males, is that the castrates had been housed under a long-day photoperiod for up to 10 months longer than the intact animals prior to behavioral testing. In ferrets of both sexes, reproductive activity is normally induced by seasonal increases in daily photoperiod (Bissonnette, 1934). However, after prolonged exposure to long days, the neuroendocrine axis eventually loses its ability to respond to the light stimulus, and ferrets come out of breeding condition; i.e., they become photorefractory. Behavioral responsiveness to androgen can vary with season, as has been documented in another seasonally breeding animal, the hamster. The copulatory performance of castrated male hamsters, housed under a stimulatory (long-day) photoperiod, increased when they were given 8-, 20-, or lOO-mm capsules of testosterone, while only NO-mm T capsules were effective in castrates housed under a nonstimulatory (short-day) photoperiod (Campbell, Finkestein, and Turek, 1978). An equivalent study has not been conducted using male ferrets. In females, however, estrogen was equally effective in stimulating sexual receptivity after ovariectomy, regardless of whether subjects were housed under a long or short photoperiod (Baum and Schretlen, 1978). It remains to be determined whether male ferrets, or males of any other seasonally breeding species, gradually lose their coital responsiveness to a particular plasma level of androgen after they become refractory to a stimulatory photoperiod. ACKNOWLEDGMENTS This work was supported by U.S. Public Health Grants ROl HD21094 and R02 MHO0392 (M.J.B.) and fellowship MH09812 (G.M.L.). The authors thank Dr. G. D. Niswender (Colorado State University) for providing the antiserum to testosterone, and Dr. J. Glitxen (University of Illinois) for the PULSAR computer program for the IBM-PC. We also thank Drs. May Erskine and Cheryl Sisk for their technical advice, and Mark Basham and Eva Kornberg for their assistance. We are grateful to the staff of the Boston University Animal Care Facility for their care of the ferrets.
REFERENCES Albert, D. J., Jonik, R. H., Watson, N. V., Gorxalka, B. B., and Walsh, M. L. (1990). Hormone-dependent aggression in male rats is proportional to serum testosterone concentration but sexual behavior is not. Physiol. Behav. 48, 409-416. Batty, J. (1978). Acute changes in plasma testosterone levels and their relation to measures of sexual behaviour in the male house mouse (Mw muscdus). Anim. Behav. X,349357. Baum, M. J., and Schretlen, P. J. M. (1978). Oestrogenic induction of sexual behaviour
392
LAMBERT
AND BAUM
in ovariectomized ferrets housed under short or long photoperiods. J. Endocrinol. 78, 295-296. Baum, M. J., and Schretlen, P. (1975). Neuroendocrine effects of perinatal androgenization in the male ferret. In Progress in Brain Research: Hormones, Homeostasis and the Brain, Vol. 42, pp. 343-355. Elsevier Science, Amsterdam. Baum, M. J., and Tobet, S. A. (1988). Endocrine control of coital sexual differentiation in the ferret: A model for higher animals. In J. M. A. Sitsen (Ed.), Handbook of Sexology: The Pharmacology and Endocrinology of Sexual Function, Vol. 6, pp. 193208. Elsevier Science, Amsterdam. Bissonnette, T. H. (1934). Modification of mammalian sexual cycles. II. Effects upon young male ferrets (Putorius vulgaris) of constant eight and one-half hour days and of six hours of illumination after dark, between November and June. Biol. Bull. 68, 300313. Blake, C. A., Blake, P. K., Thorneycroft, N. K., and Throneycroft, I. H. (1978). Effects of mating and injection of luteinizing hormone releasing hormone on serum luteinizing hormone and testosterone concentrations in rabbits. Endocrinology 76, 417-425. Campbell, C. S., Finkelstein, J. S., and Turek, F. W. (1978). The interaction of photoperiod and testosterone on the development of copulatory behavior in castrated male hamsters. Physiol. Behav.
21, 409-415.
Carroll, R. S., Erskine, M. S., and Baum, M. J. (1987). Sex difference in the effect of mating on the pulsatile release of luteinizing hormone in a reflex ovulator, the ferret. Endocrinology 121, 1349-1359. Damassa, D. A., Smith, E. R., Tennent, B., and Davidson, J. M. (1977). The relationship between circulating testosterone levels and male sexual behavior in rats. Horm. Behav. 8, 275-286. D’Occhio, M. J., and Brooks, D. E. (1982). Threshold of plasma testosterone required for normal mating activity in male sheep. Horm. Behav. 16, 383-394. Ellendorph, F., Parvizi, N., Pomerantz, D. K., Hartjen, A., Konig, A., Smidt, D., and Elsaesser, F. (1975). Plasma luteinizing hormone and testosterone in the adult male pig: 24 hour fluctuation and the effect of copulation. J. Endocrinol. 67, 403-410. Erskine, M. S., and Baum, M. J. (1982). Plasma concentrations of testosterone and dihydrotestosterone dumg perinatal development in male and female ferrets. Endocrinology
111, 767-772. Harding, C. F., and Feder, H. H. (1976). Relation between individual differences in sexual behavior and plasma testosterone levels in the guinea pig. Endocrinology 98, 11981205. Hart, B. L. (1983). Role of testosterone secretion and penile reflexes in sexual behavior and sperm competition in male rats: A theoretical contribution. Physiol Behav. 31, 823-827. Hemdon, J. G., Perachio, A. A., Turner, J. J., and Collins, D. C. (1981). Fluctuation in testosterone levels of male rhesus monkeys during copulatory activity. Physiol. Behav. 26, 525-528. Kamel, F., and Frankel, A. I. (1978). Hormone release during mating in the male rat: Time course, relation to sexual behavior, and interaction with handling procedures. Endocrinology 103, 2172-2179. Katongole, C. B., Naftolin, F., and Short, R. V. (1971). Relationship between blood levels of luteinizing hormone and testosterone in bulls, and the effects of sexual stimulation. J. Endocrinol.
SO, 457-466.
Merriam, G. R., and Wachter, K. W. (1982). Algorithms for the study of episodic hormone secretion. Am J. Physiol. 243, E310-E318. Miller, B. J., and Anderson, S. H. (1989). Failure of fertilization following abbreviated copulation in the ferret (Mustela putorius furo) J. Exp. Zool. 249, 85-89.
ANDROGENS
AND MATING IN MALE FERRETS
393
Rodbard, D., and Hutt, D. M. (1974). Statistical analysis of radioimmunoassay and immunoradiometric assays: A generalized, weighted, iterative least squares method of logistic curve fitting. In International Atomic Energy Agency Symposium on R&h munoassay and Related Procedures in Medicine, p. 165. Uniput, New York. Sisk, C. L., and Desjardins, C. (1986). Pulsatile release of luteinizing hormone and testosterone in male ferrets. Endocrinology 119, 11951203.
AND
BEHAVIOR
25,
382-393 (1991)
Reciprocal Relationships between Puisatile Androgen Secretion and the Expression of Mating Behavior in Adult Male Ferrets G. M. LAMBERT AND M. J. BAUM Department of Biology,
Boston Universi@, Boston, Massachussets 02215
The pulsatile secretion of androgen was similar over a 12-hr period in breeding male ferrets implanted with jugular catheters which either achieved an intromission with an estrous female or received no socio-sexual contact. This negative result contrasts with the previous demonstration (Carroll, Erskine, and Baum, 1987, Endocrinology 121, 1349-1359) of a significant, delayed rise in mean plasma androgen concentrations in breeding male ferrets 5-12 hr after mating. Males used in that previous study had lower initial mean plasma levels of androgen and smaller testis diameters than the present males. We therefore asked whether differences in circulating androgen levels, characteristic of males in different phases of the seasonal breeding cycle, might affect the expression of mating behavior. Castrated males given 0, 0.2, 2.0, or 5.0 mg/kg of testosterone propionate (TP) showed dose-related increases in the expression of different components of sexual behavior, including neck gripping, mounting, and intromitting. Surprisingly, intromissive performance was significantly better in intact breeding males than in castrates given even the highest dosage of TP. These results suggest that ferrets’ mating performance may vary with seasonal variations in androgen availability, and that the ability of males to exhibit a postcoital increase in the testicular secretion of androgen may be limited to the beginning or end of the breeding season, when circulating levels of androgen are relatively low. Mating-induced increments in androgen secretion at these times may enhance subsequent reproductive successby facilitating males’ intromissive capacity, which is required for the induction of ovulation and optimal sperm transport in female partners. o 1991 Academic
Press, Inc.
Engaging in mating behavior stimulates the rapid release of testicular androgens in males of several species in which the female normally ovulates spontaneously, including the rat (Kamel and Frankel, 1978), pig (Ellendorph, Parvizi, Pomerantz, Hartjen, Konig, Smidt, and Elsaesser, 1975), bull (Katongole, Naftolin, and Short, 1971) and rhesus monkey (Herndon, Perachio, Turner, and Collins, 1981). This androgen rise results from the action of luteinizing hormone (LH) on the testes, and has been hypothesized to enhance subsequent masculine mating performance (Hart, 1983). Surprisingly, in the male ferret, a species in which the female 382 0018-506X/91 $1.50 Copyright 6 1991 by Academic Press, Inc. All rights of reproduction in any form reserwd.
ANDROGENS
AND MATING IN MALE FERRETS
383
ovulates reflexively in response to mating, the pulsatile secretion of LH was significantly inhibited, and mean androgen levels were slightly depressed l-5 hr following coitus (Carroll, Erskine, and Baum, 1987). In that study a delayed elevation in mean plasma androgen secretion occurred in males 5-12 hr after an intromission, even though the secretion of LH was not augmented prior to or during this period. Secretion of testosterone (T) is known to occur in a pulsatile manner in the male ferret (Sisk and Desjardins, 1986); however, blood samples were not collected frequently enough to allow pulsatile analysis of androgen secretion in the study by Carroll et al. (1987). A primary aim of the present study was to characterize more precisely the effects of mating on androgen secretion in the male of a species in which the female is an induced ovulator. Blood samples were collected at lo-min intervals from male ferrets in breeding condition both after mating and in the absence of a female in order to allow an analysis of the pulsatile secretion of androgen. Contrary to expectations, no significant effects of achieving an intromission on mean levels or pulses of androgen were seen. Further examination of these data showed that the initial mean plasma levels of androgen were twice as high in males used in the present study as in males used previously by Carroll et al. (1987). We hypothesized that this difference in initial plasma androgen levels could affect the expression of mating behaviors by the male ferret and the resultant postcoital secretion of androgen. Sexual behaviors are displayed by male ferrets in a specific sequence, beginning with neck gripping of the dorsal surface of the female’s neck. The male ferret eventually mounts the female and exhibits episodic pelvic thrusting which ceasesonce intromission is achieved (Baum and Schretlen, 1975). The expression of neck gripping and mounting behaviors by male ferrets increased in duration with increasing plasma levels of testosterone (Baum and Tobet, 1988). However, the effect of circulating testosterone on intromissive performance has not been studied in the ferret. It has recently been suggested that the duration of receipt of penile intromission is positively correlated with the probability of fertilization in the female ferret (Miller and Anderson, 1989). Thus, higher circulating androgen levels may stimulate intromissions of a longer duration in males. The androgen rise that occurs after mating in ferrets with initially low plasma androgen may have evolved as a mechanism to increase intromissive capacity for future mating encounters, thereby extending the breeding season and increasing the probability of fertilization. A second aim of this study was to examine whether plasma androgen levels, resulting from the exogenous administration of increasing doses of testosterone propionate (TP) to castrated male ferrets, are positively correlated with the expression of masculine coital behaviors, especially intromission.
384
LAMBERT
AND
BAUM
METHODS Subjects Adult male and female ferrets were purchased directly from Marshall Farms, (North Rose, NY), between the months of January and June. Once in our laboratory these subjects were housed individually in modified stainless-steel rabbit cages under a long day photoperiod (lights on 06:OO to 22:00 hr). Water was available ad libitum, and a bowl of fresh Beef and Liver Cat Food mixed with dry Kat Krunchies (Old Mother Hubbard) was provided once daily. Ferrets are brought into breeding condition throughout the year at Marshall Farms using artificial light. Males were received from Marshall Farms in breeding condition after exposure to a long-day light cycle and these subjects successfully intromitted with stimulus females when tested l-2 days after their arrival. Some adult male ferrets were castrated via a midline scrotal incision, and/or fitted with intrajugular catheters as described previously (Carroll et al., 1987). All surgical procedures were performed under sterile conditions and using xylazine (Rompun: HaverMiles Labs, Shawnie, KS; 4.0 mg/kg) and ketamine (Ketaset: Bristol Labs, Syracuse, NY; 35 mg/kg) anesthesia. The castrated males were later injected with testosterone propionate prior to mating sessions as described below. Adult females were ovariectomized via a single midline abdominal incision and were made sexually receptive by daily SCinjection of 12-15 pg/kg of estradiol benzoate (EB) in sesame oil. These animals served as stimuli for testing the mating behaviors of male ferrets. Androgen Assay Plasma androgen levels were estimated by radioimmunoassay (Erskine and Baum, 1982). The assay standards ranged from 10 to 400 pg, and the interassay and intraassay coefficients of variation were 18.6 and 13.5%, respectively. Steroids were extracted from the plasma samples using ether, but were not separated further using chromatography. Since the testosterone antiserum (T-ll-BSA S#250; Dr GD Niswender, CO) cross-reacts with dihydrotestosterone (26.0%) and other testosterone metabolites, assay values were expressed as nanograms androgen/milliter. Concentrations of androgen were calculated using the four parameter logistic method (Rodbard and Hutt, 1974). Androgen pulses were identified using the PULSAR computer program (Merriam and Wachter, 1982) with G values defined previously for androgen pulses in the ferret by Sisk and Desjardins (1986). Mating Sessions Experimental males were paired with a stimulus female in a mating cage (30 x 67 x 51 cm) equipped with a Plexiglas front to facilitate
ANDROGENS
AND MATING
IN MALE FERRETS
385
observations. The frequency and duration of mating behaviors displayed by the male ferret were recorded using a Toshiba laptop computer. The behaviors of interest were neck gripping, mounting, pelvic thrusting, and penile intromission. The occurrence of an intromission was confirmed by the experimenter with a manual check. This procedure did not seem to disturb the subjects and was essential in that the occurrence of intromissions cannot always be detected with visual observation. Experimental Procedures Pulsatile androgen secretion in breeding males. One day after catheterization, half of the males were paired individually with a stimulus female in mating cages, while the other subjects were placed alone into empty cages of a comparable size. Mating behaviors were recorded in the paired subjects for 1 hr. At the end of this testing period, males which had mated and those which were not paired with females were placed back into their home cages. Polyvinyl extensions were attached to the catheter ends and blood samples were withdrawn every 10 min for the next 11 hr. A blood replacement mixture (Sisk and Desjardins, 1986) was infused every 30 min through the jugular catheter in order to maintain subjects’ blood hematocrit and fluid balance. Blood samples were kept on ice until being centrifuged at 5°C 1006g, for 20 min. Plasma was stored at -20°C until the androgen radioimmunoassay was performed. Expression of sexual behavior in male ferrets. The mating behaviors of sexually experienced male ferrets which had been castrated 4 weeks earlier were recorded for a 1-hr time period. Castrated male ferrets were then injected SCdaily with 0.2 mg/kg of TP. After 10 days of steroid administration the mating behaviors of these males were recorded for 2 hr. In all mating tests that followed TP treatment, the observation period was extended to 4 hr if an intromission occurred within the first 2 hr. Castrated male ferrets then received an injection of 2.0 mg/kg TP and the durations of mating behaviors which followed in the next 2 hr in response to pairing with a stimulus female showed that a single pulse of plasma testosterone failed to enhance the expression of reproductive behavior. The injections of this dose of TP (2.0 mg/kg) continued for 9 more days, and male ferrets were retested. Lastly, a similiar mating test was conducted in male ferrets after 10 days of injections with 5.0 mg/kg TP. Some males were fitted with intrajugular catheters after being injected for 10 days, including at 10 AM on the day of surgery, with one of the three doses of TP. On the day following surgery at 8 AM, a blood sample was drawn (22 hr after the previous TP injection), another TP injection of the same dose was given, and blood samples were collected 1, 2, 4, 8, 12, and 16 hr later. Blood samples were centrifuged at 5”C, and the plasma was stored at -20°C until the androgen RIA was conducted.
386
LAMBERT
AND BAUM
TABLE I Parameters of Pulsatile Androgen Secretion in Gonadally Intact Male Ferrets in a Breeding Condition Which Achieved an Intromission or Which Were Not Exposed to a Stimulus Female Parameter Pulse amplitude (nshl)
Pulse duration (min)
No female present (7) 0.34 * 0.06
21.48 + 2.19
65.48 f 13.60
Intromission with female (8) 0.35 I 0.04
23.69 f 0.94
76.48 f 06.42
Group (n)
Pulse frequency (No. pulses/hr)
Note. All data are expressed as means 2 SEM.
Statistical Analyses
The overall mean androgen pulse frequencies, amplitudes, and durations for males which achieved an intromission and for control males in breeding condition which were not paired with a female were calculated. In order to make hourly group comparisons of the pulsatile release of androgens over the total blood collection period, the number of minutes included in any part of a pulse was counted for each hour, regardless of the overall duration of each pulse. Hourly group means for androgen concentration and the number of minutes included in a pulse were compared using a two-way analysis of variance (ANOVA) with repeated measures on one factor. The behavioral data for castrated males treated with different doses of TP and for intact males in breeding condition were adjusted for time in order to examine the mean duration per hour for each variable. The differences among the hourly group means for each reproductive behavior were analyzed using a one-way ANOVA. Post hoc analyses of different group means were made using Duncan’s Multiple-Range tests. RESULTS AND DISCUSSION Pulsatile Androgen Secretion in Breeding Male Ferrets
There were no significant effects of mating on androgen pulse frequency, amplitude, or duration over the total blood collection period (Table 1). Also, both the mean androgen levels per hour (Fig. 1, bottom panel, F(l, 13) = 0.48, P < 0.05), and the number of minutes per hour included in a pulse (Fig. 1, top panel, F(l, 13) = 0.01, P < 0.05) were similar in mated and unpaired male ferrets. Although the duration of intromission (28.6 2 3.6 min vs 25.6 2 2.6 min) and other behavioral parameters (data not shown) displayed by males
ANDROGENS
AND MATING IN MALE FERRETS
381
0 Na Female Present (7) m lntramission with Female (8)
t-4 4-5 i-6 6-7 m.ft?s A ‘ER cJ%!SU OF 7-EST FIG. 1. Hourly (means f SEM) plasma levels of androgen (rig/ml; lower panel) and number of minutes (means f SEM) included in a pulse each hour (top panel) in male ferrets in breeding condition. Blood samples were collected for 11 hr following a 1-hr mating test during which an intromission occurred with a stimulus female (striped bars) or during which males were housed in a cage with no female present for 1 hr (open bars).
in the present study resembled those of males studied by Carroll et al. (1987), the males used in the present study failed to show a postcoital rise in mean androgen levels. While Carroll et al. (1987) reported a significant increase in mean plasma androgen levels 5-12 hr following an intromission, we found that mean androgen levels were similar over a 12hr period in males which achieved an intromission and in males which were not paired with a female. The present results corroborate findings reported for another induced ovulating species, the rabbit (Blake, Blake, Thorneycroft, and Thorneycroft, 1978). Serum T was elevated in males with low initial plasma levels of testosterone after exposure to a female rabbit, while subjects with higher plasma T showed no change or even decreased androgen levels after exposure to female conspecifics. This phenomenon is not limited to induced ovulating species. A rise in plasma T occurred after exposure to an estrous female in male mice, a sponta-
388
LAMBERT
AND
BAUM
neously ovulating species, only when initial plasma levels of testosterone were low (Batty, 1978). Presumably, testicular secretions were already maximal in subjects with initially high plasma T levels and could not be stimulated further by exposure to a female. In addition to having higher initial mean plasma levels of androgen (12.09 + 1.30 vs 5.16 _+ 1.22 rig/ml), the male ferrets used in the present study had larger testis diameters (15.00 + 0.23 vs 11.5 2 0.26 mm) and shorter thrusting durations (4.24 + 1.11 min vs 11.46 + 2.43 min) than males used previously by Carroll et al. (1987). This suggests that the present males were in the height of the breeding condition and achieved intromission with estrous females more easily than the subjects in Carroll et al. ‘S study (1987). Males used in the present study may also have had the capacity to achieve intromisions of longer duration than did Carroll’s subjects, had mating tests not been limited to 1 hr in both studies. Thus, a postcoital rise in androgen secretion may occur selectively in ferrets which are marginally in breeding condition. The functional implications of a postcoital rise in androgens for males at the end of their breeding season may include extending the time period during which reproductive behaviors, especially intromission, are displayed. Males’ fertility is thereby enhanced, in so far as intromission is required to trigger the preovulatory LH surge in estrous females, and longer intromissions increase the likelihood of sperm transport and successful fertilization of the female (Miller and Anderson, 1989). Expression of Mating Behaviors in Male Ferrets
Administration of TP to castrated male ferrets resulted in dose-dependent increases in plasma testosterone levels (Fig. 2). In each instance there was a small increment in testosterone immediately following the daily SCinjection; however, plasma levels of testosterone then remained relatively stable for the next 24 hr between injections. There were group differences in the durations of neck gripping (F(4, 27) = 57.77, P < 0.05) and mounting (F(4, 27) = 39.36, P < 0.05) shown by male ferrets, depending upon the type of androgenic stimulation which prevailed (Fig. 3). The durations of neck gripping and mounting were significantly lower (P < 0.05) in castrated males given either no TP or 0.2 mg/kg TP than in castrates given either 2.0 or 5.0 mg/kg of TP or in intact males in breeding condition. There were also significant group differences in thrusting duration (F(4,27) = 12.03, P < 0.05). Less pelvic thrusting was displayed by castrated male ferrets given no TP or 0.2 mg/kg TP and by intact subjects than by castrated males given 2.0 or 5.0 mg/kg TP. Finally, there were group differences in intromission duration (F(4, 27) = 10.73, P < 0.05). Intromission duration was lowest in castrated males given no TP, increased significantly in males given 5.0 mg/kg, and
ANDROGENS
I
4
389
AND MATING IN MALE FERRETS
I 6
I
I
I
I
12
16
20
24
FIG. 2. Plasma androgen concentrations (rig/ml) resulting from daily subcutaneous injections of testosterone propionate (TP) to castrated male ferrets for 12 days. Blood samples were withdrawn from intrajugular catheters at 1, 2,4, 8, 12, 16, and 22 hr following a daily injection of one of three doses of TP. The number of animals sampled for each TP dose at each time point is given above the error bars.
was significantly greater in intact breeding males than in all castrated, androgen-treated groups. These data raise the possibility that the duration of an intromission in male ferrets is dependent upon the plasma level of testosterone and/or the ongoing profile of androgen secretion. This differs from results obtained in males of several species in which the female ovulates spontaneously where mating performance was not highly correlated with plasma levels of testosterone. For example, in the castrated male rat, a minimal replacement of circulating testosterone was sufficient to sustain mating performances at levels qualitatively and quantitatively similar to those of gonadally intact males (Albert, Jonik, Watson, Gorzalka, and Walsh, 1990). Castrated male rats maintained a normal pattern of ejaculatory performance with only 10% of the circulating testosterone normally present in gonadally intact males (Damassa, Smith, Tennent, and Davidson, 1977). In male sheep, administration of 4 mg/kg TP, a dose which yielded mean plasma T levels well below those naturally occurring in intact, breeding subjects, induced the full display of mating behavior (D’Occhio
390
LAMBERT
AND BAUM
TP (mg/kg) Elj 0.0 Castrated E Fi I 5.0 1 q 0.0~Intact, Breeding
Neck Gripptng Mounting Pelvic Thrusting lntromitting FIG. 3. Mean durations (minutes/hour) of components of masculine sexual behavior displayed by male ferrets which were castrated for 4 weeks and subsequently treated with
0, 0.2, 2.0, or 5.0 mg/kg TP for 10 days or which were tested while gonadally intact and in breeding condition. All data are expressed as means + SEM. The proportion of subjects in each group that expressed each behavior is given above the error bars.
and Brooks, 1982). Lastly, in male guinea pigs, plasma T levels were not correlated with individuals’ mating performance (Harding and Feder, 1976). In contrast, the durations of most mating behaviors shown by castrated male ferrets were positively related to the dose of TP injected. The duration of pelvic thrusting appears to be an exception to this relationship in the ferret. Gonadally intact male ferrets used by Carroll et al. (1987) had lower plasma androgens but longer thrusting durations than the subjects in the present study; however, a shorter thrusting duration is probably indicative of an enhanced ability to achieve an intromission. Thus, a different sequence and pattern of mating behaviors occurs in male ferrets as a result of different endogenous circulating levels of testosterone. These differences in sexual behavior may affect postcoital sensitivity of hypothalmic-pituitary-testicular axis to further activation. Pulsatile exposure of the nervous system to testosterone may be more effective than tonic exposure for the stimulation of mating behaviors in male ferrets. While gonadally intact, breeding males had mean androgen levels (5-15 rig/ml) similar to those obtained after injecting 0.2 mg/kg TP into castrated male ferrets (5-10 rig/ml), the durations of all mating behaviors, except for pelvic thrusting, were significantly longer in the intact group, which had been exposed to a pulsatile profile of androgenic stimulation. Intact males achieved intromission durations that were even longer than those of castrated males given the highest dose (5.0 mg/kg)
ANDROGENS
AND MATING
IN MALE FERRETS
391
of TP (50-70 rig/ml plasma androgen). Additional research is needed to assessthe importance of pulsatile exposure to testosterone in the male ferret for the expression of mating behaviors. An alternative explanation for the shorter intromission durations seen in the castrated, TP-treated subjects, as compared with gonadally intact breeding males, is that the castrates had been housed under a long-day photoperiod for up to 10 months longer than the intact animals prior to behavioral testing. In ferrets of both sexes, reproductive activity is normally induced by seasonal increases in daily photoperiod (Bissonnette, 1934). However, after prolonged exposure to long days, the neuroendocrine axis eventually loses its ability to respond to the light stimulus, and ferrets come out of breeding condition; i.e., they become photorefractory. Behavioral responsiveness to androgen can vary with season, as has been documented in another seasonally breeding animal, the hamster. The copulatory performance of castrated male hamsters, housed under a stimulatory (long-day) photoperiod, increased when they were given 8-, 20-, or lOO-mm capsules of testosterone, while only NO-mm T capsules were effective in castrates housed under a nonstimulatory (short-day) photoperiod (Campbell, Finkestein, and Turek, 1978). An equivalent study has not been conducted using male ferrets. In females, however, estrogen was equally effective in stimulating sexual receptivity after ovariectomy, regardless of whether subjects were housed under a long or short photoperiod (Baum and Schretlen, 1978). It remains to be determined whether male ferrets, or males of any other seasonally breeding species, gradually lose their coital responsiveness to a particular plasma level of androgen after they become refractory to a stimulatory photoperiod. ACKNOWLEDGMENTS This work was supported by U.S. Public Health Grants ROl HD21094 and R02 MHO0392 (M.J.B.) and fellowship MH09812 (G.M.L.). The authors thank Dr. G. D. Niswender (Colorado State University) for providing the antiserum to testosterone, and Dr. J. Glitxen (University of Illinois) for the PULSAR computer program for the IBM-PC. We also thank Drs. May Erskine and Cheryl Sisk for their technical advice, and Mark Basham and Eva Kornberg for their assistance. We are grateful to the staff of the Boston University Animal Care Facility for their care of the ferrets.
REFERENCES Albert, D. J., Jonik, R. H., Watson, N. V., Gorxalka, B. B., and Walsh, M. L. (1990). Hormone-dependent aggression in male rats is proportional to serum testosterone concentration but sexual behavior is not. Physiol. Behav. 48, 409-416. Batty, J. (1978). Acute changes in plasma testosterone levels and their relation to measures of sexual behaviour in the male house mouse (Mw muscdus). Anim. Behav. X,349357. Baum, M. J., and Schretlen, P. J. M. (1978). Oestrogenic induction of sexual behaviour
392
LAMBERT
AND BAUM
in ovariectomized ferrets housed under short or long photoperiods. J. Endocrinol. 78, 295-296. Baum, M. J., and Schretlen, P. (1975). Neuroendocrine effects of perinatal androgenization in the male ferret. In Progress in Brain Research: Hormones, Homeostasis and the Brain, Vol. 42, pp. 343-355. Elsevier Science, Amsterdam. Baum, M. J., and Tobet, S. A. (1988). Endocrine control of coital sexual differentiation in the ferret: A model for higher animals. In J. M. A. Sitsen (Ed.), Handbook of Sexology: The Pharmacology and Endocrinology of Sexual Function, Vol. 6, pp. 193208. Elsevier Science, Amsterdam. Bissonnette, T. H. (1934). Modification of mammalian sexual cycles. II. Effects upon young male ferrets (Putorius vulgaris) of constant eight and one-half hour days and of six hours of illumination after dark, between November and June. Biol. Bull. 68, 300313. Blake, C. A., Blake, P. K., Thorneycroft, N. K., and Throneycroft, I. H. (1978). Effects of mating and injection of luteinizing hormone releasing hormone on serum luteinizing hormone and testosterone concentrations in rabbits. Endocrinology 76, 417-425. Campbell, C. S., Finkelstein, J. S., and Turek, F. W. (1978). The interaction of photoperiod and testosterone on the development of copulatory behavior in castrated male hamsters. Physiol. Behav.
21, 409-415.
Carroll, R. S., Erskine, M. S., and Baum, M. J. (1987). Sex difference in the effect of mating on the pulsatile release of luteinizing hormone in a reflex ovulator, the ferret. Endocrinology 121, 1349-1359. Damassa, D. A., Smith, E. R., Tennent, B., and Davidson, J. M. (1977). The relationship between circulating testosterone levels and male sexual behavior in rats. Horm. Behav. 8, 275-286. D’Occhio, M. J., and Brooks, D. E. (1982). Threshold of plasma testosterone required for normal mating activity in male sheep. Horm. Behav. 16, 383-394. Ellendorph, F., Parvizi, N., Pomerantz, D. K., Hartjen, A., Konig, A., Smidt, D., and Elsaesser, F. (1975). Plasma luteinizing hormone and testosterone in the adult male pig: 24 hour fluctuation and the effect of copulation. J. Endocrinol. 67, 403-410. Erskine, M. S., and Baum, M. J. (1982). Plasma concentrations of testosterone and dihydrotestosterone dumg perinatal development in male and female ferrets. Endocrinology
111, 767-772. Harding, C. F., and Feder, H. H. (1976). Relation between individual differences in sexual behavior and plasma testosterone levels in the guinea pig. Endocrinology 98, 11981205. Hart, B. L. (1983). Role of testosterone secretion and penile reflexes in sexual behavior and sperm competition in male rats: A theoretical contribution. Physiol Behav. 31, 823-827. Hemdon, J. G., Perachio, A. A., Turner, J. J., and Collins, D. C. (1981). Fluctuation in testosterone levels of male rhesus monkeys during copulatory activity. Physiol. Behav. 26, 525-528. Kamel, F., and Frankel, A. I. (1978). Hormone release during mating in the male rat: Time course, relation to sexual behavior, and interaction with handling procedures. Endocrinology 103, 2172-2179. Katongole, C. B., Naftolin, F., and Short, R. V. (1971). Relationship between blood levels of luteinizing hormone and testosterone in bulls, and the effects of sexual stimulation. J. Endocrinol.
SO, 457-466.
Merriam, G. R., and Wachter, K. W. (1982). Algorithms for the study of episodic hormone secretion. Am J. Physiol. 243, E310-E318. Miller, B. J., and Anderson, S. H. (1989). Failure of fertilization following abbreviated copulation in the ferret (Mustela putorius furo) J. Exp. Zool. 249, 85-89.
ANDROGENS
AND MATING IN MALE FERRETS
393
Rodbard, D., and Hutt, D. M. (1974). Statistical analysis of radioimmunoassay and immunoradiometric assays: A generalized, weighted, iterative least squares method of logistic curve fitting. In International Atomic Energy Agency Symposium on R&h munoassay and Related Procedures in Medicine, p. 165. Uniput, New York. Sisk, C. L., and Desjardins, C. (1986). Pulsatile release of luteinizing hormone and testosterone in male ferrets. Endocrinology 119, 11951203.
