INFLUENCE OF NEONATAL CASTRATION OR NEONATAL ANTI-GONADOTROPIN TREATMENT ON FERTILITY, PHALLUS DEVELOPMENT, AND MALE SEXUAL BEHAVIOR IN THE MOUSE*

Transcription

1 FERTILITY AND STERILITY Copyright 1975 The American Fertility Society Vol. 26, No.9. September 1975 Printed in U.SA. INFLUENCE OF NEONATAL CASTRATION OR NEONATAL ANTI-GONADOTROPIN TREATMENT ON FERTILITY, PHALLUS DEVELOPMENT, AND MALE SEXUAL BEHAVIOR IN THE MOUSE* DAVID M. QUADAGNO, PH.D., H. GLENN WOLFE, PH.D., GOERGE KAN WHA HO, AND BRUCE D. GOLDMAN, PH.D. Department of Physiology and Cell Biology, University of Kansas, Lawrence, Kansas 66045, and Department of Biobehavioral Science, University of Connecticut, Storrs, Connecticut In the rat, it is generally accepted that cyclic gonadotropin retion is characteristic of the female, and tonic gonadotropin retion is characteristic of the male. The critical factor in normal physiologic development of gonadotropin retion is the presence or absence of a testicular product, presumably, androgen, during the perinatal period. The presence of androgen or androgenic metabolites suppresses the capacity for cyclic gonadotropin release. Similarly, the development of systems which regulate patterns of sexual behavior are greatly influenced by neonatal testicular hormones. l It has been shown that a pituitarygonadal feedback system is operating shortly after birth in the male rat. Serum concentrations of luteinizing hormone and follicle-stimulating hormone, while low in intact neonatal males, show a rapid increase following castration. 2 This postcastration rise in gonadotropin levels can be blocked with exogenous steroid treatment. 3 When anti-gonadotropin serum was administered to male rats on days 1, 3, and 5 (day 1 was the day of birth), the animals were unable to impregnate females after reaching adulthood. The testes ofthese antiserumtreated males appeared normal in size and, since sperm were present, it was Received November 14, * Supported in part by Grants HD (to H. G. W.) and MH (to D. M. Q.) from the United States Public Health Service, and by Biomedical Sciences Support Grant RR to the University of Kansas. 939 suggested that (1) gonadotropin function was not permanently impaired by neonatal treatment with anti-gonadotropin serum and (2) some disruption of masculine sexual behavior might be responsible for the failure to impregnant females.4 Recently it was demonstrated that endogenous gonadotropic hormones are required during early postnatal life for masculinization of the neural systems which suppress female sexual behavior, and for normal penile development in the male rat.5 6 It was also shown that anti-gonadotropin serum in infancy does not interrupt masculinization of gonadotropin control centers.5,6 However, as in the earlier study,4 the male rats receiving anti-gonadotropin serum in infancy failed to impregnate females. 5 They showed fewer intromissions and ejaculations in sex behavior tests, compared with controls. Since the phallus was underdeveloped in anti-gonadotropin-treated males, these data do not distinguish between a peripheral or central neural cause for the behavioral deficit. 7 The present study was undertaken to determine whether pituitary gonadotropins are important during neonatal sexual differentiation in a ond species of rodent, the mouse. MATERIALS AND METHODS All mice used in the following experiments were genetically uniform Fl hybrids, produced by crossing mice of inbred strains 129/ReWI and C57BL/6JWe.

2 940 QUADAGNO ET AL. September 1975 Rabbit anti-ovine gonadotropic hormone immune serum (AGS) that crossreacted with rat gonadotropins was obtained from rabbits immunized with NIH LH-S12 by a procedure similar to that described in an earlier communication. 8 The AGS cross-reacted with both luteinizing hormone and follicle-stimulating hormone of rat origin, as determined by radioimmunoassay. The AGS used in the present study binds iodinated rat luteinizing hormone and follicle-stimulating hormone obtained from the NIAMD Rat Pituitary Program. Experiment 1. To determine whether the AGS would cross-react with mouse gonadotropin, adult F 1 hybrid male mice received subcutaneous injections of AGS. Five adult males (high-ags) received 0.43 ml of AGS over a 5-day period, and five adult males (high-nrs) received 0.43 ml of normal rabbit sera (NRS) over a 5-day period. In addition, four adult males (low AGS) received 0.20 ml of AGS over a 2-day period, and four adult males (low-nrs) received 0.20 ml of NRS over a 2-day period. The animals receiving injections over a 5-day period were killed 2 days after the last injection, whereas animals receiving injections over a 2-day period were killed 5 days after the last injection. At the time of autopsy the seminal vesicles and testes were removed from each animal and weighed. The body weight of each animal was also recorded. Experiment 2. Newborn mice were produced from hormonally primed, mature females of inbred strain C57BLl6JWe mated to males of inbred strain 129/ ReW!. The females were primed with an intraperitoneal injection of 2 IU of pregnant mare's serum gonadotropin followed 40 hours later by an intraperitoneal injection of 2 IU of human chorionic gonadotropin. Young were born over a 2- day period. The animals were maintained on a 13-hour light and ll-hour dark cycle (lights on at 7:00 A.M., off at 8:00 P.M.; water and mouse laboratory chow were available ad libitum. Pregnant females were checked daily, and the day on which litters were found was designated day 1. Neonatal males and females received injections of either anti-gonadotropic immune serum (AGS) or normal rabbit serum (NRS) on day 1 (0.02 mi), day 3 (0.03 mi), and day 5 (0.05 mi). This male treatment group was termed either low-ags or low-nrs. Another group of neonatal males each received an injection of either AGS or NRS on day 1 (0.04 mi), day 3 (0.06 mi), and day 5 (0.1 mi). This treatment group was termed either high-ags or high-nrs. The injections were administered subcutaneously. Both NRS- and AGS-treated mice were weaned at 30 days and housed individually. When NRS- and AGS-treated females reached 42 days of age, a single adult hybrid male was introduced into each cage. The pairs were left together for 4 months and the number of litters and size of each litter were recorded. When NRS- and AGS-treated males reached approximately 75 days of age, each was presented for 30 minutes with a mature hybrid female induced into estrus by intraperitoneal injection of 2 IU of pregnant mare's serum gonadotropin followed 40 hours later by intraperitoneal injection of 2 IU of human chorionic gonadotropin (first sex behavior test). After 7 days each male was presented with a ond female for 30 minutes, injected as before (ond sex behavior test). Each male was scored for sexual behavior by observing him in his home cage with the stimulus females on both presentations. All females were found to be sexually receptive. The following sexual behavior patterns, modified from those described by McGill and Blight,9 were recorded for each male during each test: mount latency-the number of onds from the introduction of the

3 Vol. 26, No.9 PHALLUS DEVELOPMENT AND SEX BEHAVIOR 941 female until the male first mounts; number of mounts without intromission; intromission latency-the number of onds from the introduction of the female until the male gains first intromission; number of intromissions preceding ejaculation; ejaculation latencythe number of onds from the introduction of the female until the male ejaculates. Next, after a 1-week interval, two adult hybrid females were introduced into each cage and left for 2 weeks. The females were removed and placed in individual cages for 3 weeks. The females were checked frequently, and births were recorded. The number of young in each litter was also recorded. After the females had been removed, all males were castrated (approximately 100 days of age). Approximately 5 weeks later, each male received a subcutaneous graft of one histocompatible ovary from a 20-day-old mouse. The graft was placed under the kidney capsule, and 1 month later the animal was killed. At autopsy the ovarian graft was removed, placed in Bouin's solution for approximately 72 hours, frozen, tioned at 30 p.m, and examined histologically for the presence of follicles and corpora lutea. The penis was removed and weighed, and its length was measured. The body weight for each animal was also recorded. Experiment 3. Hybrid male mice were castrated or sham-castrated on the day of birth. Those sham-castrated were later castrated at 25 days of age. At approximately 100 days of age, all males received an injection of 100 p.g of testosterone propionate (TP) on each of 7 days, followed by 200 p.g of TP on each of 7 additional days. At the end of that time, each male was placed with a sexually receptive female (primed with estradiol benzoate and progesterone following the procedure of McGill and Blight 9 ) for 30 minutes. Females that were not receptive were replaced with receptive females. The number of mounts, intromissions, and ejaculations shown by each male was recorded, as well as the latency to the first mount, intromission, and ejaculation. When the sex behavior tests were concluded, the test males were killed, the penis was removed and weighed, and the penis length was measured. The body weight for each animal was also recorded. RESULTS Experiment 1. High and low doses of AGS significantly reduced the weights of the seminal vesicles and testes, compared with those obtained with high and low doses ofnrs (Table 1). This suggests a high degree of cross-reactivity between AGS and mouse gonadotropins. The groups were compared by one-way analysis of variance (Table 1). Experiment 2. The fertility of female mice and male mice treated with AGS in infancy was not significantly different from that of NRS controls. The females treated with AGS had litter sizes (mean, S.S ± 0.3; 45 litters; 9 females treated) comparable with those of NRS females (mean, S.7 ± 0.4; 2S litters; 6 females treated). Similarly, males treated with AGS had litter sizes (low-ags mean, S.5 ± 0.4; 19 litters; 11 males treated; high-ags mean, 7.9 ± 0.4; 14 litters; 7 males treated) conparable with those of NRS males (low-nrs mean, S.4 ± 0.2; 12 males treated; high-nrs mean, 7.S ± O.S; 14 litters; 7 males treated). When male sexual behavior was analyzed with a one-way analysis of variance, there were no significant differences between high- and low-ags groups. Likewise, there were no significant differences between high-nrs and low-nrs groups. Therefore, high-ags and low-ags groups were combined and statistically compared with the combined high-nrs and low-nrs groups (Table 2). The groups

4 942 QUADAGNO ET AL. September 1975 TABLE 1. Effect of Anti-gonadotropic Hormone Immune Serum (AGS) and Normal Rabbit Serum (NRS) Administered to Adult Male Mice on Seminal Vesicle, Testis, and Body Weight'" Body wt Group Quantity of Seminal vesicle Testis wtll AGS ornrs wi! Before treatment After treatment ml mg gm Low-AGS (N = 4) ± 1.7} c ± 2.1} d 29.0 ± ± 1.2 Low-NRS (N = 4) ± ± ± ± 1.2 High-AGS (N = 5) ± 3A} c ± 1.6} d 29.9 ± ± 1.0 High-NRS(N = 5) ± ± ± ± 0.3 avalues are means ± standard error. bmean weight for single lobe of seminal vesicle or single testis. CSignificantly different (J> < 0.01). dsignificantly different (J> < 0.05). were analyzed by a one-way analysis of variance. No significant differences were found in any behavior measures. At autopsy, high-ags and low-ags groups did not differ from each other in body and penis weights or in penis length; likewise, high-nrs and low-nrs groups did not differ from each other in body and penis weights or in penis length. Thus, high-ags and low-ags groups were combined and statistically compared by oneway analysis of variance with the high NRS and low-nrs groups (Table 3). The NRS group had penises significantly larger (both in weight and length) than those of the AGS group (Table 3). This was true whether penis values were expressed in absolute terms or relative to body weight. Approximately 35% of the ovarian grafts were located, and they appeared normal. However, none of the ovarian grafts in either of the AGS or NRS groups showed the presence of corpora lutea. Many of the grafts did show large follicles. Experiment 3. There were significant differences in sex behavior measures between the males castrated at birth and the males castrated at 25 days of age (Table 4). The males castrated on day 1 showed lower levels of mounting and intromission behavior. The groups were compared by one-way analysis of variance. The males castrated at 25 days of age had penis lengths and weights significantly greater than those of the males castrated at birth (Table 3). A one-way analysis of variance was used to compare the groups. DISCUSSION The present results suggest that neonatal gonadotropin(s) may not be important for regulating early testicular function and, thereby, sexual differentiation in the mouse. Male mice treated during neonatal life with AGS and studied in adulthood demonstrated normal fertility and also normal sexual behavior when paired with receptive females. These results differ from those reported for the rat. Male rats treated with AGS neonatally were infertile and showed an incomplete pattern of masculine sexual TABLE 2. Influence of Neonatal Anti-gonadotropin Treatment (AGS) Compared with Normal Rabbit Serum Treatment (NRS) on Male Sexual Behavior in the Adult Mouse when Presented with a Receptive Female Mousea Group No. of mounts Latency to No. of Latency to first first mount intromissions intromission AGS(N = 19) 4.6 ± ± ± ± ± NRS (N = 19) 4.8 ± ± ± ± ± 85.0 Latency to first No. showing No. showing ejaculation intromission ejaculation /19 "Values are means ± standard error. Treatments were administered at 1, 3, and 5 days of age

5 Vol. 26, No.9 PHALLUS DEVELOPMENT AND SEX BEHAVIOR 943 TABLE 3. Influence of Neonatal Treatment with A nti-gonadotropin (AGS),Normal RabbitSerum(NRS),and Neonatal Castration on Body Weight, Penis Weight, and Penis Length" Group Body wt Penis wt Penis length gm mg mm AGS (N = 16) 29.2 ± ± 1.43} b 12.5 ± 0.35} c NRS (N = 18) 28.7 ± ± ± 0.28 Castrated day 1 (N = 5) 35.3 ± ± 3.51} b 11.7 ± 0.54}b Castrated day 25 (N = 6) 39.0 ± ± ± 0.23 Values are means ± standard error. Treatments were administered at 1, 3, and 5 days of age. bp < cp < behavior in adulthood. 4 5 The AGS used in the present study was shown to be capable of causing a reduction in seminal vesicle weight in adult male mice (presumably by preventing gonadotropin-dependent androgen production). Thus, it does not seem likely that the lack of effect of the AGS in neonatal male mice can be attributed to any failure to crossreact with mouse gonadotropin. A total dose of 0.3 ml of AGS was administered to rat pups in previous studies,s. 6 while a total dose of 0.2 ml was administered to the mouse pups in the present study. It is thought that the large size difference between rat pups and mice pups compensates for the lower dose given to the mice pups. The AGS used in the present study was from the same preparation used in an earlier study with rats. 6 Thus, it is not likely that the doses of AGS administered were insufficient to "neutralize" endogenous gonadotropins during the neonatal period. Since AGS failed to affect male sex behavior or fertility in male mice, compared with NRS controls, we were led to investigate further the role of the testes in this species. Castration on the day of birth resulted in a decreased capacity to display masculine sexual behavior, compared with mice castrated at 25 days of age. This observation suggests that, in the mouse as in the rat, testicular hormone(s) is important for the differentiation (i.e., masculinization) of the systems required for the display of male sexual behavior. Since the penises of the castrates on day 1 were significantly smaller than the penises of castrates on day 25, these data do not distinguish between a peripheral or central neural cause. Our data (i.e., males castrated at birth having levels of male sexual behavior lower than those of shamoperated controls) are in agreement with recent findings. lo It was found that neonatal androgen stimulation in the female mouse was important for the development of the capacity to show male sexual behavior. Neonatally androgenized females with or without their ovaries and with or without exogenous androgen in adulthood exhibited the full range of mas- TABLE 4. Influence of Neonatal Castration on Sexual Behavior in the Adult Male Mouse when Presented with a Receptive Female Mouse" Group No. of mounts Latency to first mount No. of intromissions Latency to first intromi88ion Castrated day 1 (N = 10) 0.8 ± 0.59} b 690 ± ± 0.21}c 1110 ± Castrated day 25 (N = 10) 6.9 ± ± ± ± Values are means ± standard error. bp < cp < dx' analysis fp < 0.01). No. showing No. showing intromibsiods ejaculation 1/ OlIO} d 3/10

6 944 QUADAGNO ET AL. September 1975 culine responses, including the ejaculatory reflex. 10 Both neonatal castration and neonatal treatment with AGS resulted in reduced penis size in adulthood; however, the effect of neonatal castration appeared to be greater in this respect. It is possible that the neonatal mouse testis is capable of reting some androgen in the absence of gonadotropic hormone stimulation. The failure of neonatal AGS treatment to modify the masculine pattern of gonadotropin retion is similar to results obtained in the rat. 5 6 Also, the failure of neonatal AGS treatment to affect fertility in female mice is not surprising. These observations provide conclusive evidence that anti-gonadotropic activity does not persist until adulthood following injection of AGS into the neonate. SUMMARY Male mice were castrated on the day of birth or were treated with anti-gonadotropin serum (AGS) successively on days 1,3, and 5 of age. When the adult AGStreated males were compared with normal rabbit serum (NRS)-treated males with respect to fertility, male sexual behavior, gonadotropin release, body size, and penis size it was found that: (1) fertility of AGS-treated males did not differ from that of NRS-treated males; (2) AGS-treated males showed a significant reduction in penis size, compared with NRS-treated males; (3) both groups showed high levels of male sexual behavior; (4) neither group showed the capacity to support ovulation in ovarian grafts from immature mice. It was also observed that the fertility of female mice treated with AGS on days 1, 3, and 5 of age did not differ from that of control females treated with comparable doses of NRS. Male mice castrated at birth showed a significant reduction in male sexual behavior and penis size compared with males first sham-operated at birth then castrated at 25 days of age. Thus, males castrated at birth were similar to males receiving AGS in infancy, except for sex behavior. The results suggest that the neonatal pituitary in the mouse may not have an important role in the process of sexual differentiation. Previous data for the male rat were compared with those for the mouse. REFERENCES 1. Gorski RA: Gonadal hormones and perinatal development of neuroendocrine function. In Frontiers in Neuroendocrinology, 1971, Edited by L Martini, W Ganong. New York, Oxford University Press, 1971, p Goldman BD, Grazia YR, Kamberi IA, Porter JC: Serum gonadotropin concentrations in intact and castrated neonatal rats. Endocrinology 88:771, Goldman BD, Gorski RA: Effects of gonadal steroids on the retion on LH and FSH in neonatal rats. Endocrinology 89:112, Goldman BD, Mahesh VB: Induction of infertility in male rats by treatment with gonadotropin antiserum during neonatal life. BioI Reprod 2:444, Goldman BD, Quadagno DM, Shryne J, Gorski RA: Modification of phallus development and sexual behavior in rats treated with gonadotropin antiserum neonatally. Endocrinology 90:1025, McCullough J, Quadagno DM, Goldman BD: Neonatal gonadal hormone: effect on maternal and sexual behavior in the male rat. Physiol Behav 12:183, Whalen RE: Differentiation of the neural mechanisms which control gonadotropin retion and sexual behavior. In Perspectives in Reproductive and Sexual Behavior, Edited by M Diamond. Bloomington Ind, Indiana University Press, 1968, p Goldman BD, Mahesh VB: A possible role of acute FSH-release in ovulation in the hamster, as demonstrated by utilization of antibodies to LH and FSH. Endocrinology 84:236, McGill TE, Blight we: The sexual behavior of hybrid male mice compared with the sexual behavior of males of the inbred parent strains. Anim Behav 11:480, Manning A, McGill TE: Neonatal androgen and sexual behavior in female house mice. Horm Behav 5:19, 1974