— — — 2 — = — — —_ — m, — — — — LIBRARY " ‘ Michigan State - University IlllllllllflflljflllfllljflljlflflIlllllLlfllllill This is to certify that the thesis entitled THE ROLE OF AROMATIZATION IN THE DEVELOPMENT OF SEXUAL BEHAVIOR IN THE HAMSTER presented by Patricia Hilley Ruppert has been accepted towards fulfillment of the requirements for _Eh_..D_._degree in _Es¥chology_ Major professor Date 527/ /~'/:/ 7 9’ 0-7539 OVERDUE FINES ARE 25¢ PER DAY PER ITEM Return to book drop to remove this checkout from your record. T___~_. THE ROLE OF AROMATIZATION IN THE DEVELOPMENT OF SEXUAL BEHAVIOR IN THE HAMSTER (MESOCRICETUS AURATUS) By Patricia Hilley Ruppert A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Psychology and Neuroscience Program 1979 ABSTRACT THE ROLE OF AROMATIZATION IN THE DEVELOPMENT OF SEXUAL BEHAVIOR IN THE HAMSTER (MESOCRICETUS AURATUS) By Patricia Hilley Ruppert A major factor determining the potential for masculine or feminine sexual behavior is the action of gonadal hormones during early life. Males exposed to testicular hormones during sexual differentiation are masculinized and defeminized in their behavior as adults. Female hamsters, or males castrated on the day of birth (Day 1), if exposed to androgens which can be converted to estrogens or estrogens them— selves during early develOpment, also show an increase in masculine behavior and a decrease in feminine behavior. The present study investigated whether aromatization of testos- terone to estradiol is necessary for behavioral masculinization and defeminization. An aromatization inhibitor, ATD (l,4,6-Androstatrien- 3,l7-dione) was administered to intact male and female hamsters, and to Day l castrate males along with either testosterone or estradiol. On days 2 through 4 after birth, all animals received either 1.0 or 0.5 mg ATD or pr0pylene glycol along with either 50 or 100 ug testos- terone, 2 ug estradiol or sesame oil. As adults, all were tested for both masculine and feminine sexual behavior in counter—balanced order. In Experiment 1, both testosterone and estradiol masculinized the behavior of females and Day l castrate males. Neither animals receiving ATD alone nor the control animals were masculinized. The Patricia Hilley Ruppert high dose of ATD (1.0 mg) completely blocked masculinization produced by the low dose of testosterone (50 ug) for females and Day 1 castrate males. The high ddse of ATD (1.0 mg) partially blocked masculinization produced by the high dose of testosterone (100 mg) in females but not Day l castrate males; ATD, as predicted, did not block behavioral masculinization produced by estradiol. For intact male hamsters given postnatal hormones, males receiving a combination of 1.0 mg ATD + 100 ug T intromitted and ejaculated more than males just receiving 1.0 mg ATD + oil; ATD did not block the combined effect of endogenous and exo- genous testosterone in intact males. For female behavior in Experiment 1, control females receiving ATD + oi1 showed longer mean lordosis durations than females receiving a high dose of testosterone or estradio1, with or without ATD. Feminine sexual behavior subsequent to treatment with 1.0 mg ATD + 50 ug T was intermediate between the lordosis durations of the controls and females defeminized by testosterone or estradiol. Although both intact males and Day 1 castrate males were more defeminized than female hamsters, regardless of hormone treatment, there were no significant differences within male treatment groups. Since, for each anima1 the order of testing for masculine and feminine behavior was counter-balanced as a control procedure, the data were analyzed for order effects; significant differences were found for order of testing. For some treatment groups, female hams- ters, intact males or Day 1 castrate males who received masculine behavior tests prior to feminine behavior tests showed higher masculine behavior scores than animals tested in the reverse sequence. Patricia Hilley Ruppert For feminine behavior in females, order effects were also seen. While higher masculine scores for postnatally intact males receiving masculine behavior tests first may reflect a residual effect of testi- cular hormone secretion by the adult gonad, the same finding in females and Day 1 castrate males, however, poses a challenge to the present concept of sexual differentiation. In Experiment 2, females receiving either testosterone or estra- diol postnatally, either with or without ATD, were tested for feminine behavior as adults to determine their sensitivity to ovarian hormones. With acute doses of estradiol ranging from 1.5 to 12 ug estradiol benzoate, females showed little or no lordosis in the absence of progesterone. Progesterone in doses ranging from 50 to 400 ug facilitated receptivity for all groups when combined with priming doses of estrogen; no differences were seen between postnatal treat- ment groups. In Experiment 2, as in previous behavioral studies in the hamster, evidence does not indicate that early hormone treatments influence adult hormone sensitivity. ACKNOWLEDGMENTS I would like to thank Dr. Lynwood G. C1emens, my advisor, for providing insight, support and encouragement throughout my graduate career. I am also grateful to Dr. Glenn 1. Hatton, Dr. Lawrence I. 0'Ke11y and Dr. Gail D. Riegle for their advice and suggestions. And, to my fellow graduate students, I thank a11 of you for the experiences we shared. This research was supported by USPHS Grant HD-06760 to Dr. Lynwood G. Clemens. 11° TABLE OF CONTENTS Page LIST OF TABLES ......................... iv LIST OF FIGURES ........................ V INTRODUCTION .......................... 1 Adult Masculine Sexual Behavior .............. 3 Adult Feminine Sexual Behavior .............. 8 Sexual Differentiation .................. 10 Development of Sexual Behavior in Hamsters ........ l4 Aromatization ....................... 21 GENERAL METHODS ........................ 27 Animals .......................... 27 Postnatal Treatments ................... 27 General Testing Procedure ................. 28 EXPERIMENT 1 .......................... 30 Results .......................... 32 Masculine Behavior - Females ............. 33 Masculine Behavior - Day 1 Castrate Males ...... 33 Masculine Behavior - Intact Males .......... 36 Order of Testing Effects for Masculine Behavior . . . 38 Feminine Behavior - Females ............. 41 Feminine Behavior - Males .............. 46 Morphological Data .................. 46 EXPERIMENT 2 .......................... 54 Results .......................... 57 DISCUSSION ........................... 61 Masculinization ...................... 61 Defeminization ...................... 62 Order of Testing Effects ................. 64 Estrogen and Progesterone Sensitivity ........... 67 Morphological Measures .................. 70 Conclusion ........................ 70 LIST OF REFERENCES ....................... 71 iii Table 10 LIST OF TABLES Masculine sexual behavior (X MALE) for female postnatal treatment groups .............. Masculine sexual behavior (X MALE) for Day 1 castrate male postnatal treatment groups ....... Rear mounts for intact male postnatal treatment groups ........................ Intromissions for intact male postnatal treatment groups ........................ Ejaculations for intact male postnatal treatment groups . . . . .................... Feminine sexual behavior (X TLD) for female postnatal treatment groups .............. Body weight (gm) for all postnatal treatment groups . . Morphological measures for all postnatal treatment groups: anogenital distance (A- G), penile bone and cartilage length (PB & C) and testes weight (Testes ' Wt) .......................... Feminine sexual behavior (X TLD) for female postnatal treatment groups tested with varying doses of progesterone (P) ................... Morphological measures for female postnatal treat- ment groups tested for estrogen and progesterone sensitivity ...................... iv Page 40 42 43 44 47 51 52 58 60 LIST OF FIGURES Figure Page 1 Schematic diagram of masculine sexual behavior in the hamster showing mounts (thin lines), intromissions (medium lines) and ejaculations (thick lines). The temporal relationships are depicted above the time line; ML (mount latency), IL (intromissions latency), EL (ejaculation latency) and PEI (post-ejaculatory interval) are shown .................. 4 2 ATD blocks the conversion of testosterone to estradiol (figure courtesy of Brian A. Gladue) ..... 24 3 Mean masculine behavior scores for female postnatal treatment groups; each treatment of ATD (mg), hormone (ug) or control vehicle was given on Days 2-4 after birth .................... 34 4 Mean masculine behavior scores for Day 1 castrate male postnatal treatment groups; each treatment of ATD (mg), hormone (ug) or control vehicle was given on Days 2-4 after birth ................ 35 5 Rear mounts, intromissions and ejaculations for intact male postnatal treatment groups; each treat- ment of ATD (mg). hormone (ug) or control vehicle was given on Days 2-4 after birth ............. 37 6 Mean lordosis durations for female postnatal treat- ment groups; each treatment of ATD (mg), hormone (pg) or control vehicle was given on Days 2-4 after birth ......................... 45 7 Mean lordosis durations for Day 1 castrate male postnatal treatment groups; each treatment of ATD (mg), hormone (ug), or control vehicle was given on Days 2-4 after birth .................. 48 8 Mean lordosis durations for intact male postnatal treatment groups; each treatment of ATD (mg), hormone (ug) or control vehicle was given on Days 2-4 after birth ......................... 49 INTRODUCTION For the hamster, estrogens as well as aromatizable androgens induce the potential for masculinization and defeminization of behavior when present during the first few days after birth. These behavioral studies led to the hypothesis that testosterone is a “prohormone”, achieving its biological potency via its metabolites of estradiol and dihydrotestosterone. The purpose of the present study was to inves- tigate whether testosterone must be metabolized to estradiol for hamsters to be behaviorally masculinized and defeminized. This hypothesis was tested by giving ATD (1,4,6-Androstatrien-3,l7-dione), which inhibits the aromatization of testosterone to estradiol, to male and female hamsters, in combination with either testosterone or estradiol. The specific hypotheses of Experiment 1 were: 1) Behavioral masculinization of male and female hamsters is accomplished via conversion of testosterone to estradiol. Therefore, ATD, which blocks aromatization, will block the behavioral masculinization of male and female hamsters by testosterone but will not block estradiol induced masculini- zation. 2) Defeminization of behavior is also mediated by estradiol. Male and female hamsters given a combination of ATD and testosterone will show normal feminine behavior while those 2 receiving ATD and estradiol will show a reduced duration of lordosis. Defeminization of behavior may involve changes in sensitivity _either to estrogen, to progesterone or to both hormones in adulthood. In Experiment 1 of this study, and in other developmental studies, female hamsters were tested for lordosis with hormone treatments which would maximize behavioral responding. The purpose of Experiment 2 of this study was to obtain dose response data for lordosis in both normal and defeminized hamsters using adult hormone treatments of estrogen alone and estrogen plus progesterone. These dose response data for lordosis in the female hamster should provide information on whether behavioral defeminization involves a change in estrogen sensitivity, progesterone sensitivity, or whether both are changed. The golden hamster is a good model for studies of reproductive biology and behavior. Because of its short gestation period, only 16 days, the hamster is born at an immature stage. Therefore, develop- mental processes which occur prenatally in most species occur post- natally in the hamster. As a result of this immaturity, hormone treatments which would reach the fetus indirectly in other species, via the mother, can be given directly to hamster pups. The hamster, then, offers a comparative model for determining physiological pro- cesses regulating behavior. In this study, the goal was to elucidate one aSpect of the mechanism of early hormone action, using the hamster as a model. In the introduction which follows, the rationale for these experiments is described. The first section contains a description of the behavior patterns which characterize masculine and feminine 3 sexual behavior in the adult hamster. Then the organizational model of sexual differentiation is presented for behavior, morphology and patterns of gonadotrOpin release. Further discussion summarizes experiments on the development of masculine and feminine sexual behavior in the hamster. Then I review the physiological evidence for aromatization. and show how ATD can be used to test the hypo- theses of the present experiments. Adult Masculine Sexual Behavior Masculine sexual behavior in the hamster consists of a series of mounts, intromissions and ejaculations; hamsters are classified as a multiple intromission, multiple ejaculation Species. Capulatory behavior in the hamster has been well described by Reed and Reed (1946), Beach and Rabedeau (1959) and by Bunnell, Boland and Dews- bury (1977). The account presented here follows from their behavioral studies. Also, Figure l is a schematic diagram of masculine sexual behavior in the hamster, showing the behavioral components of c0pula- tion, and the temporal pattern of behavior. When paired with a receptive female, the first reSponse of the male is to sniff and lick the head, body and particularly the genital region of the female. Then, during a rear mount, the male clasps the female with his forelegs, elevates his pelvis and thrusts rapidly. Occasionally, males especially inexperienced ones, mount the female from the head and side; in this case, the male either reorients him- self Spontaneously on subsequent mounts or the female shifts her position so he is oriented to her rear. After a few mounts, the male intromits, i.e., the penis penetrates the vagina while the male is .czocm men APm>cmu=v xgoumpzomnmuumoav Hum new Raccoon, :owumpzumnmv Au .Aaocmump :owmm_eocpcwv 4H .Axucmump uczosv a: mm:w_ we?» as» m>ono umuuwaou men mavgmcowumpmg Pocoasmu one .Amwcp— xuwcuv mcopumpzuuwm can Amway, Ezwumsv m:o_mm_soeucw .Amocpp cwguv muczos mcwzocm Layman; as» cw Lop>msma szxmm mcwpzomms mo Emgmovu opumemcom ._ mesmwu Al ms. F tme A. 1:2 .: 4w 5 mounting. During intromission, the male thrusts rapidly and then brings his pelvis firmly against the female's perineum. After about 15 intromissions, spaced 10 seconds apart, the male ejaculates. Behavioral criteria for ejaculation (intromission plus sperm emission) are subtle in the hamster. The best indication of ejaculation is an increase in the rate of pelvic thrusting prior to intromissions. An additional behavioral criterion is a spasmodic flexion and extension of the rear leg during ejaculation. While intromissions and ejaculations differ in average duration (2.4 vs. 3.4 seconds), this is difficult to distinguish behaviorally. Figure 1 also indicates the temporal patterning of the male's behavior. The mount latency is the interval from the introduction of the female to the first mount, and the intromission latency is the interval from the introduction of the female to the first intromission. These time periods are quite variable, both between animals and on different tests with the same animal. The preliminary exploratory activity of both partners, and the initial orientation of the partners to one another contribute to this variability. The median mount latency is about one minute, with an intromission latency of 1.5 minutes. Once the male begins to copulate, his behavior is very channelized. Prior to the first ejaculation (the ejaculation latency is the inter- val from the first intromission to the first ejaculation), 80% of the male's behavior is Spent in pursuit-mount of the female or in genital grooming; males lick their penis and groom following intro- mission. During the post-ejaculatory interval (the time from ejacu- lation to the next intromission), which is about 30 seconds, males 6 groom in longer bouts (at least 6 seconds), and then eXplore the surrounding area or sniff and lick the female. These post-ejaculatory behaviors are additional indicators that ejaculation has occurred. After the first ejaculation, the temporal pattern of behavior changes. The ejaculatory latency decreases from about 3 minutes for the first ejaculation to 30 seconds for subsequent ejaculations. The number of intromissions decreases dramatically from 15 to 3, and the time between intromissions decreases slightly (the male shows shorter bouts of grooming). As the male approaches satiation, after 9-10 ejaculations, the time between intromissions increases Slightly, while the post-ejaculatory interval increases progressively during the test period. By the last series, only 52% of the male's activity is Spent in pursuit-mount or genital grooming; locomotory-eXploratory behavior, licking the female, biting the female, lying down, grooming and scratching are more prevalent. Female hamsters have a very limited capacity to diSplay components of masculine sexual behavior unless treated postnatally with gonadal hormones. In all developmental studies where control females received oil injections, none of the females mounted as adults with any of the adult hormone treatments. Tiefer (1970) tested normal adult females for masculine behavior under three conditions; natural estrus, estrogen plus progesterone and testosterone treatment. Only 1/10 females mounted and showed the intromittive pattern. The conclusion drawn from these studies was that female hamsters are not bisexual; they do not behave as males when given apprOpriate hormone treatments as adults. 7 Noble has challenged this conclusion. He tested females for masculine behavior using several combinations of adult hormone treat- ments (Noble, 1974). While neither estradiol benzoate, dihydrotes- tosterone pr0pionate or testosterone pr0pionate alone induced mounting in females (0, 29 and 33% mounted, reSpectively), with combined treat- ment with estradiol benzoate and dihydrotestosterone pr0pionate, all females mounted and most intromitted. Hormone doses were extremely high: 6 ug estradiol benzoate plus 1-2 mg dihydrotestosterone pro- pionate were given daily, and the median latency to onset of mounting was 21-31 days after the beginning of hormone treatment. Nonetheless, adult females can perform the major components of masculine copula- tory behavior except for the ejaculatory pattern. They require high doses of combined hormone treatment for an extended period of time, however, to elicit masculine behavior. With Silastic implants of hormones, producing a more physiologi- cal pattern of hormone release, Noble (1977) found that either tes- tosterone alone or a combination of either testosterone, dihydrotes- tosterone or androstenedione with estradiol was effective in inducing mounting and intromitting in females. Again, latencies to the first mount were long, 31-35 days. Noble hypothesized that the neural structures underlying masculine c0pulatory behavior are organized in both males and females; females, though, are more refractory to hormones. They may be less able to convert testosterone to its metabolites. Extended, high doses of hormones in adult females might be " ..... producing changes similar to those which typically occur during exposure to gonadal steroids early in development" (Noble, 1977, p. 520). Adult Feminine Sexual Behavior The feminine sexual behavior of hamsters is unique to rodents. When a receptive female is placed with a male, he usually sniffs or licks her a few times. She then takes a few steps away from his and assumes the lordosis posture. "She extends her forelegs but flexes them slightly at the elbow, extends the hind legs and elevates the pelvis, elongates the body and straightens the back, spreads the hind feet wide apart and raises her tail. The eyes are glazed, fixed and may be half closed" (Reed & Reed, 1946, p. 8). Beach has shown that female hamsters do behave proceptively, i.e., females in estrus solicit male attention. Receptive females Spend more time near a caged male than diestrous females, and they remain in tonic immobility or full lordosis in the presence of sensory cues from the male (Beach, Stern, Carmichael & Ranson, 1976). In a mating test, contact with the male occurs almost immediately and the female does not have the Opportunity to solicit. Since tactile cues alone are sufficient for females to show lordosis (i.e., manual stimulation, Murphy, 1974), and females will Show lordosis to the auditory, visual and olfactory cues of a caged male, the sensory cues available in a mating test are redundant and the female readily shows lordosis. Females remain in a trancelike state for long periods of time while the male copulates. Even during intervals when the male is not actively copulating (i.e., during genital grooming), females remain immobile. During the first 5 ejaculation series of the male, lordosis and walking lordosis are the only behaviors of the female (Bunnell et al., 1977). Measures of feminine receptivity in the hamster are 9 total lordosis duration (total time during a test Spent in lordosis), lordosis frequency (the number of times the female shows lordosis), and the longest duration of lordosis (the longest single lordosis episode during a test). After prolonged genital stimulation, or following the maximum period of behavioral estrus, females become less receptive to the male (Carter, Landauer, Tierney & Jones, 1976; Carter & Schein, 1971). Intact females are receptive for 14 hours if not mated; proceptive behaviors are characteristic of the middle 8 hours of estrus (Beach et al., 1976). AS the female becomes unreceptive, she shows lordosis only while the male is actively copulating; during pauses in copula- tion, She explores the testing arena. With additional genital stimu- lation or passage of time, the female actively resists the male's attempts to mount by turning on her side or biting. Females who are initially unreceptive also attack if the male persists in attempting to mount. Since male hamsters are normally bisexual, feminine sexual behavior is relatively easy to elicit, even in the intact male. Kow, Malsbury and Pfaff (1976) investigated lordosis in intact males using manual stimulation. The lordosis potential of males is greater when receiving manual stimulation than when mounted by another male hamster; the extent of tactile contact is greater and the attractiveness of the eXperimental male is not a factor. (Since the male showing lordosis lacks a vagina and cannot receive intromissions, he is not a prized partner; the stimulus male may cease mounting or become aggressive after many unsuccessful attempts to c0pulate.) The total lordosis duration of intact males in these tests with manual stimulation was 10 half that of receptive females'(57.2 vs. 109.2 seconds). When males Showed lordosis, the testes were withdrawn toward the body cavity, furthering the similarity to the female. While castration did not decrease the lordosis shown by males, treatment with estrogen and progesterone increased lordosis duration in males to the equivalent of the estrus females . When adult males are castrated and their feminine sexual behavior compared to females', males behave qualitatively Similar but differ in the frequency and duration of lordosis. While females hold a single lordosis during the male's mounts and dismounts, males tend to remain in lordosis only while the stimulus male is actually mounting. This results in a more frequent but shorter duration of lordosis (an increase in lordosis frequency but decrease in lordosis duration). Males castrated without further hormone treatment do not Show lordosis in mating tests. With short term exposure to estrogen alone (1 or 2 injections) the data conflict on whether males show lordosis (Carter, Michael & Morris, 1973; Tiefer & Johnson, 1971); lordosis, if shown to short term estrogen alone, is very brief. Long term exposure to estrogen (6 days or more) is much more effective in inducing behavioral receptivity (Carter et al., 1973), and proges- terone facilitates receptivity with either Short term estrogen (Carter et al., 1973; Tiefer & Johnson, 1971) or long term estrogen treatment (Carter et al., 1973). Sexual Differentiation In a now classic study in behavioral endocrinology, Phoenix, Goy, Gerall and Young (1959) pr0posed that hormonal conditions during 11 early life organize sexual behavior in a masculine or feminine pattern, just as hormones organize the reproductive system and gonadotrOpin release into a masculine or feminine pattern. This inference was based on the behavior of guinea pigs treated with testosterone propionate during development. According to this hypothesis, during a restricted, critical period of early life, gonadal hormones organize neural tissues to facilitate the sexual responses of the genetic male and to inhibit or suppress the sexual response patterns characteristic of the genetic female. This is the condition of the normal male and of females receiving hormones during the critical period. In the absence of gonadal hormones, the condition of the genetic female or the post- natally castrated male, the potential for feminine behavior develops. This hypothesis initially recognized that behavioral masculini- zation and defeminization were independent and separable processes. Beach has defined masculinization and defeminization broadly to include a range of sexual differences: masculinization refers to the induction of anatomical, physiological or behavioral characteristics typical of males but lacking or poorly developed in females, while defeminization is an inhibition of characteristics well developed in females but not in males (Beach, 1975). For each Species or strain investigated, either the male or the female in each case is likely to be bisexual; for hamsters, for example, female hamsters rarely mount as adults but male hamsters easily show feminine sexual behavior. Rather than a "critical period“ when hormones can modify the potential for bisexual- ity in each species, early life can more aptly be considered a "period of maximum sensitivity", during which gonadal hormones exert their organizational effects (Goy & Goldfoot, 1975). 12 Besides behavioral sexual differentiation, morphological and neuroendocrine sexual differentiation are also influenced by the action of gonadal hormones during early life. The successive steps in morphological sexual differentiation are: genetic sex, gonadal sex and then morphological sex. Jost (1953, 1972) first obtained evidence supporting this model of development by removing the gonads from male and female rabbit fetuses before the stage of morphological differentiation. ‘Regardless of genetic sex, all gonadectomized fetuses developed as females. Jost's interpretation, that testicular secretions induce the male phenotype, and that the female phenotype results from the absence of testicular hormones, has been supported by subsequent gonadal transplant studies and by studies using exo- genous gonadal hormones. At the stage during which the undifferentiated gonad is developing into an ovary or testis, both sexes have a urogenital system composed of two parts: 1) a dual duct system (mullerian and wolffian) which is the anlage of the accessory sex organs and the upper vagina, and 2) the urogenital turbercle which is the anlage of the external genitalia. The fetal testis produces two substances necessary for male development (Wilson, 1978). The first is an unknown peptide (mullerian inhibiting substance) which causes the mullerian ducts to regress. The second is testosterone which induces maturation and development of the wolffian duct system (forming the vas deferens, seminal vesicles, epididymis, and accessory glands), the external genitalia and the testis itself. Jost (1953) found that differentiation of the mullerian duct system developed in the absence of the fetal ovary, and concluded 13 that ovarian secretions were not involved in female development. In the absence of fetal testicular secretions, the wolffian ducts regress and the mullerian ducts develOp to form the uterus, oviducts and upper vagina; the genital tubercle forms the clitoris and external vagina. However, the embryonic ovary can synthesize estrogen (Milewich, George & Wilson, 1977) and local estrogen formation may aid on the maturation of the ovary. In the hamster, morphological sexual differentiation begins on Day 10 of gestation and continues throughout the early postnatal period. First indications of testicular differentiation begin at Day 11 3/4, while the ovary begins to develop a day later. The wolffian ducts appear at Day 10 and continue to develOp in males and females until Day 13 3/4 when degeneration begins in the female. The mullerian duct is formed in both sexes by Day 11 3/4 and begins to degenerate 24 hours later in the male. Although the genital tubercle first appears at Day 9 3/4, development of the external genitalia continues postnatally (Ortiz, 1945). Just as behavioral and morphological sexual differentiation are determined early in life by the action of gonadal hormones, hypothala- mic control of gonadotrOpin release in rodents is programmed in a masculine (acyclic) or feminine (cyclic) pattern by the hormonal events in early life. Gorski (1971, 1973) has reviewed the evidence from his laboratory and others leading to this conclusion. In the absence of active stimulation by gonadal hormones, as in the normal female, the ovariectomized female and the castrate male rat, it has been proposed that the preOptic area-anterior hypothalamus differ- entiates into a region controlling cyclic release of gonadotropin. 14 If gonadal hormones are present during early life, as in the normal male or androgenized female, this cyclic center is suppressed and the surge of gonadotropin which induces ovulation does not occur in the adult. In both males and females the control system of the mediobasal hypo- thalamus is operative, maintaining testicular function and ovarian follicular development. But only in males and females who have not been exposed to gonadal hormones postnatally does the cyclic control system of the prepptic area-anterior hypothalamus develop. Develgpment of Sexual Behavior in Hamsters Behavioral studies on the development of sexual behavior in hamsters have sought to define the role of hormones in early life. These studies have shown that testosterone, other aromatizable andro- gens and estrogens masculinize and defeminize the behavior of male and female hamsters, while non-aromatizable androgens do not have this behavioral potency. The apprOpriate hormones must be present during the first few days of postnatal life to achieve these behavioral effects. Behavioral studies in the hamster have provided the first evidence that aromatization of testosterone to estradiol is involved in sexual differentiation. This diScovery led to the aromatization hypothesis, stating that estradiol is the active hormone in behavioral sex differentiation. The experiments leading up to this hypothesis and to the present study are examined below. Testicular hormones are necessary for inducing masculinization during early life. Castration of male hamsters on Day 1 of postnatal life eliminates postnatal eXposure to gonadal secretions. Thus, males are only exposed to their own testicular secretions during prenatal 15 development when they are presumably not behaviorally sensitive to the action of hormones (Nucci & Beach, 1971). Castration of male hamsters on Day 1 of postnatal life eliminates the potential for masculine behavior, even if replacement hormones are given in adulthood (Carter, Clemens & Hoekema, 1972; Eaton, 1970; Johnson, 1975; Noble, 1973; Swanson, 1970, 1971). Noble (1973) found that some Day 1 castrates mounted and intromitted while in other studies, males never mounted. The capacity of Day 1 castrates to behave as males in adulthood can be restored by testosterone replacement during the first few days after birth. This treatment increases the mounting and intromitting of Day 1 castrates (Coniglio & Clemens, 1976; Coniglio, Paup & Clemens, 1973a; Eaton, 1970; Swanson, 1971; Tiefer & Johnson, 1975). None of these studies with either a single injection or multiple injections of hormone over a short time period were able to replicate the full sexual repertoire of the normal male. Tiefer and Johnson (1975), however, gave either testosterone or androstenedione on Days 1-20; their Day 1 castrates resembled adult castrates in mounting and intromitting, but these males ejaculated infrequently. Although it is difficult to induce mounting behavior in normal female hamsters, postnatal exposure to gonadal hormones induces the potential for masculine behavior in the female. The development of the male has been "mimicked" by injecting female hamsters after birth with either testosterone or testosterone propionate, which has a longer duration of action. For both forms of testosterone, the females so treated have been masculinized, i.e., have mounted as adults (Carter et al., 1972; Coniglio & Clemens, 1976; DeBold & 16 Whalen, 1975; Johnson, 1975; Paup, Coniglio & C1emens, 1972; Swanson, 1971; Tiefer & Johnson, 1975; Whitsett & Vandenbergh, 1975). The time period for hormonal exposure and the dose of hormone have been varied to determine the most sensitive period for behavioral masculinization. ‘In comparing different days of castration, Carter et a1. (1972) showed that males castrated on Day 6 or 25 mounted and intromitted more than males castrated on Day 1. Noble (1973) found that males castrated on Day 1 only mounted, while Day 5 castrates also intromitted and only Day 10 castrates ejaculated as adults. For females also, the sensitive period for behavioral masculinization by exogenous gonadal hormones is restricted to a fairly short time following birth: 0-72 hours (Swanson, 1971), 1-4 days (Coniglio & C1emens, 1976) or 1-7 days (DeBold & Whalen, 1975). Behavioral masculinization, then, is accomplished within the first week after birth in the hamster. To determine threshold quantities of hormones needed for behavior- al masculinization, Coniglio and Clemens (1976) varied the timing and dose of testosterone given to Day 1 castrate males. 100 ug testoster- one either on Days 1-2 or 3-4 increased the mounting and intromitting of castrate males. With a lower dose of testosterone (50 ug), hormone had to be present continuously from Day 1-10 to increase adult mounting and intromitting. DeBold and Whalen (1975) found that one ug testosterone propionate was as effective as 250 ug testosterone propionate in inducing mounting. In another dose response study, Whitsett and Vandenbergh (1975) found that 90% of females mounted when treated with either 3, 30, 300 or 600 ug testosterone pr0pionate; the number of mounts increased with increasing dose of hormone. 17 Collectively, these studies have shown that the period of maximum sensitivity can be influenced by the dose of testosterone and the form of eXposure (alcohol form vs. testosterone pr0pionate). Small doses of testosterone propionate can masculinize behavior while free testosterone must be present in higher doses or for a longer time period to achieve the same effect. The develOpment of feminine sexual behavior is believed to be an anhormonal process. Since behavioral sexual differentiation presumably does not occur prenatally in the hamster (Nucci & Beach, 1971), the ovaries can be removed on Day 1 after birth in the female hamster to determine the possible influence of postnatal ovarian secretions on the development of feminine behavior. This experiment has been done and behavioral effects are slight (Gerall & Thiel, 1975; Swanson, 1970). Females ovariectomized on Day 43 maintained the lordosis posture longer than females ovariectomized on Day 1, which indicates enhanced receptivity in post-pubertally ovariectomized females. But, the frequency of lordosis was also greater in females ovariectomized on Day 43, and increased frequency of lordosis indicates reduced receptivity. One problem is that the maximally sensitive period for feminization may be very early (hours) after birth, so any delay in ovariectomy might bypass the critical period. As for masculinization of females, the first approach in studying defeminization has been testosterone administration during early life to “mimic" the hormonal condition of the normal male hamster. The general conclusion from these studies is that testosterone defeminizes females, i.e., females given testosterone during early life Show more frequent but Shorter lordosis postures (Carter et al., 1972; Coniglio 18 & Clemens, 1976; Coniglio, Paup & Clemens, 1973b; DeBold & Whalen, 1975; Gerall, McMurray & Farrell, 1975; Gottlieb, Gerall & Thiel, 1974; Johnson, 1975; Swanson, 1971; Tiefer & Johnson, 1971; Whitsett & Vandenbergh, 1975). Coniglio et al. (1973b) obtained defeminiza- tion with testosterone pr0pionate but not free testosterone (25 or 100 ug on Days 244). Subsequent studies have shown that when the duration of exposure is lengthened by daily injection, either with 50 ug testosterone on Days 1-10 (Coniglio & Clemens, 1976) or via Silastic implants on Days 1-10 (Gerall et al., 1975), free testos- terone also defeminizes. Since testosterone suppresses lordosis in females, castration of males soon after birth Should result in a feminine pattern of lordosis. Males castrated on Day 1 show longer lordosis durations than males castrated on Day 3 or 5 (Gerall & Thiel, 1975), Day 6 or 25 (Carter et al., 1972) or on Day 10 (Noble, 1973). When males, either neonatally castrated or not, however, receive supplementary testosterone, their lordosis responding is decreased (Coniglio & C1emens, 1976; Coniglio et al., 1973a; DeBold & Whalen, 1975; Gerall et al., 1975; Swanson, 1971; Tiefer & Johnson, 1975). While Coniglio and Clemens (1976) found no defeminization in Day 1 castrates given two daily injections of 100 ug testosterone, 50 ug testosterone on Days 1-10 decreased total lordosis duration. DeBold and Whalen (1975) found that as little as 5 ug testosterone propionate further defeminized intact males. These studies clearly show that defeminization results from postnatal exposure to testosterone, and that this process can be blocked by neonatal castration of males. 19 The time period which is maximally effective for suppressing lordosis is the first few days after birth. Coniglio and Clemens (1976) found that 100 ug testosterone on Days 1-2 or 3-4 increased lordosis frequency (defeminization). DeBold and Whalen (1975) found that when graded doses of testosterone pr0pionate were given 24 hours after birth, 50 ug maximally inhibited lordosis while 5 and 10 ug were less effective. As little as 10 ug testosterone propionate decreased total lordosis duration on either Day 1 or 3 but not later in development. The time period for defeminization, then, is more restricted than the time period of Day 1-7 for the masculinization of females (DeBold & Whalen, 1975). Swanson (1971) found that 0-48 hours was the maximum sensitive period for defeminization, while at least 0-72 hours was maximal for masculinization. One of the most exciting and heuristic findings of research in the development of sexual behavior has been that testosterone is not the only hormone capable of masculinizing and defeminizing behavior in hamsters. Coniglio et al. (1973a) found that testosterone, testosterone pr0pionate, estradiol, estradiol benzoate or diethyl- stilbestrol induced mounting potential in Day 1 castrate male hamsters; dihydrotestosterone and androsterone, non-aromatizable androgens, were not effective. Tiefer and Johnson (1975) also found that with extended hormone treatment, from Day 1-20, 10 ug androstenedione was equivalent to 10 ug testosterone in masculinizing males. These same two hormones were effective in inducing mounting in females; for equal treatments, however, females were less masculinized in their behavior than castrated males. 20 Pautiet a1. (1972) found that the synthetic estrogen, diethyl- stilbestrol was actually more effective in inducing mounting in female hamsters than either testosterone or testosterone propionate, while androsterone was ineffective. Whalen and Etgen (1978) compared the ability of testosterone propionate, estradiol benzoate, and a synthetic estrogen, RU-2858, to masculinize female hamsters. Although the dose- response was not linear, as little as 500 ng estradiol benzoate or 50 pg RU-2858 induced mounting in females; the only dose of testos- terone pr0pionate used, 1 ug, also masculinized females. These studies have shown that estorgen is very potent in masculinizing behavior. Defeminization of behavior can also be achieved by estrogens and androgens which can be converted to estrogens. Coniglio et al. (1973b) found that diethylstilbestrol decreased total lordosis dura- tion in female hamsters, while in the same study, androsterone, a non- aromatizable androgen, and 25 ug testosterone were not effective. Gerall et a1. (1975) did obtain weak suppression of lordosis in females given dihydrotestosterone or androsterone but the suppression was minimal compared to that produced by testosterone. In studying the dose response relationship of defeminization in female hamsters, Whalen and Etgen (1978) found that as little as 50 ng estradiol benzoate or 100 pg RU-2858 decreased mean lordosis duration in females; in the same experiment, 1 ug testosterone propionate did not defeminize. Thus either natural or synthetic estrogens are more potent than androgens in defeminizing females. Males are similarly affected. Coniglio et al. (1973a) found decreased total lordosis duration in male hamsters receiving either testosterone pr0pionate, estradiol, estradiol benzoate or diethylstilbestrol but not 21 testosterone, dihydrotestosterone or androsterone. Gerall et a1. (1975) found that silastics of testosterone implanted on Days 2-10 (with castration on Day 5) suppressed lordosis strongly; dihydro- testosterone was weak and androsterone had no effect. Aromatization Physiological studies have confirmed not only that aromatization is important in sexual differentiation, but have also localized the aromatization enzymes. These are localized in limbic areas which also bind steroid hormones, regulate gonadotrOpin release, and are involved in the neural control of sexual behavior (Naftolin, Ryan, Davies, Reddy, Flores, Petro, Kuhn, White, Takaoka & Wolin, 1975; Reddy, Naftolin & Ryan, 1974). Moreover, these enzymes are functional in the brain of the neonatal rat. When 3H-testosterone is injected in neonatal male and female rates, 3H-estradiol is recovered from cell nuclei in the pre0ptic area, hypothalamus and amygdala (Lieberburg & McEwen, 1975; Weisz & Gibbs, 1974). Neonatal rat brains also have receptors which bind estrogen. When tritium labelled estradiol or diethylstilbestrol is injected into Day 3 rats, estrogen is recovered from cell nuclei in the pre0ptic area, hypothalamus, amygdala and cerebral cortex; binding capacity on Day 3 is about 1/3 the adult levels (McEwen, Plapinger, Chaptal, Gerlach & Wallach, 1975). Other studies have reported high levels of estrone and estradiol in the plasma of neonatal rats, both male and female (see Gorski, Harlan & Christensen, 1977). While the source of these estrogens is controversial, plasma levels of estrogen may be higher in the neonate than in the adult. 22 Estrogens normally in the plasma of postnatal female rats do not actually masculinize these females. Neonatal rat plasma and brain contain a Specific estrogen binding protein, alpha-fetoprotein (Ray- naud, Mercier-Bodard & Baulieu, 1971). Because of this binding protein, the estrone and estradiol found in the neonatal rat may be effectively inactivated (McEwen, Lieberburg, Maclusky & Plapinger, 1976). This protein disappears from the blood at 3 weeks of age, when the adult estrogen binding capacity increases (McEwen et al., 1975; Plapinger & McEwen, 1973). When exogenous estrogen masculinizes and defeminizes, it probably does so by overloading this protective estrogen binding system. These studies still do not Show that estrogen is the hormone that normally masculinizes and defeminizes behavior. One critical experi- ment is to block the aromatization of testosterone to estradiol; if masculinization and defeminization of behavior are also blocked, then aromatization is a necessary step in normal sexual differentiation. Aromatization of testosterone to estradiol is a multiple reaction process with several enzyme systems involved; the pathways for this conversion have not been worked out completely but several intermediate products have been identified. Two major reactions are first the hydroxylation of C-19 and finally conversion of the A ring to an aromatic structure (Engel, 1975). One compound which inhibits estrogen biosynthesis from precursors in human placental microsomes and in the neonatal rat brain is ATD (1,4,6-Androstatrien-3,l7-dione). Figure 2 shows a diagram of the testosterone, estradiol and ATD molecules. ATD was first tested in a placental microsome system, where it was tested for competitive 23 Figure 2. ATD blocks the conversion of testosterone to estradiol (figure courtesy of Brian A. Gladue). 24 .N assay; 2:5 szE -5...” - zm_mh .05). Thus, females not receiving hormones were not masculinized (controls), ATD by itself did not masculinize behavior, and 1.0 mg ATD completely blocked the masculin- izing action of 50 pg 1. While all other hormone treatment combina- tions of testosterone or estradiol masculinized behavior to some extent, the high dose of ATD (1.0 mg) partially blocked the masculini- zation produced by 100 pg T (p < .05), while the low dose of A10 (0.5 mg) did not block masculinization in females. Masculine Behavior - Day 1 Castrate Males. The mean masculine behavior scores for Day 1 castrate males are shown in Figure 4. For Day 1 castrate males, a comparison of all groups via one way analysis of variance showed significant differences between postnatal treatments 34 15 FEMALES MEAN MALE BEHAVIOR PG .SATD lATD PG lATD PG .SATD IATD PG .SATD lATD Jul 011 oil SOT SOT 100T 100T 100T 2E 26 2E Figure 3. Mean masculine behavior scores for female postnatal treat- ment groups; each treatment of ATD (mg), hormone (pg) or control vehicle was given on Days 2-4 after birth. 35 I 9 ‘5 DAYi CASTRATE MALES E m“ 10 I “’0 1.1.10 .(m 5 2 z. 3‘: o 2 1 I I ! I I I 1ATD 1ATD TATD 1ATD 01? oil SOT 501 1001 100T 2E 25 Figure 4. Mean masculine behavior scores for Day 1 castrate male postnatal treatment groups; each treatment of ATD (mg), hormone (pg) or control vehicle was given on Days 2-4 after birth. 36 [F(7,68) = 4.15, p < .0008]. Day 1 castrate male hamsters receiving 1.0 mg ATD + 50 pg T were not different from groups receiving no hormone (p > .05), while 1.0 mg ATD did not block 100 pg T, the high dose of testosterone (p > .05). Day 1 castrate males receiving 1.0 mg ATD + oil alone mounted more than expected, and males receiving PG + 2 pg E mounted less than expected (p < .05): Otherwise, these results agree with the data from the female groups. Masculine Behavior - Intact Males. For intact males, a more detailed analysis was made of masculine behavior scores since more masculine behavior was shown. Figure 5 Shows rear mounts, intro- missions and ejaculations for all postnatal treatment groups. Since there was no difference for head and side mounts for any groups (p > .05), these data are not listed. The range of head and Side mounts was 2.6 i 0.5 (1.0 mg ATD + oil) to 5.0 i 1.1 (PG + 100 pg T). Rear mounts did differ between postnatal treatment groups [F(8,91) = 2.89, p < .006]. Males receiving PG + 2 pg E or 1.0 mg ATD + 2 pg E showed more rear mounts than males receiving just 1.0 mg ATD + oil (p < .05); so, the addition of postnatal estradiol to intact males increased the number of rear mounts. Intromission frequency also differed between postnatal treatment groups [F(8,91)] = 2.75, p < .009]. Males receiving 1.0 mg ATD + 100 pg T intromitted more than males receiving either dose of ATD + oil (p < .05). This shows that ATD was not able to block the combined effect of endogenous and exogenous testosterone. A Similar pattern was shown for ejaculation frequency, which also differed between postnatal treatments [F(8,91) = 2.70, p < .01]. Males receiving 1.0 mg ATD + 100 pg T ejaculated more than males receiving either dose of ATD + oil (p < .05). Figure 5. 37 INTACT MALES iflnfinnnnnn EJACU‘A I leS lNlHOMlbblONo g EI— + —1- :1— «[1111 10‘ 5* ! I I ! I PG SAID IATD 5.ATD lATD SATD IATO Oll ml 01! 100T 1007 1007 RE AR MOUNTS O A 0 Rear mounts, intromissions and ejaculations for intact male postnatal treatment groups; each treatment of ATD (mg), hormone (pg) or control vehicle was given on Days 2-4 after birth. 38 Order of Testigg Effects for Masculine Behavior. In the experi- mental design for Experiment 1, the test sequence of animals within each group was randomized, so approximately half would receive feminine behavior tests first, and half would receive masculine behavior tests first. Although counterbalancing was used as a control procedure, no independent effects due to order of testing were expected. In the initial analysis, all scores for animals within each postnatal treatment group were pooled, regardless of order of testing. However, while recording the data, I noticed differences within groups that seemed related to the order of testing (all behavior tests were con- ducted without awareness of the treatment group being tested). A two way analysis of variance was used to test for the significance of this order effect. For masculine behavior shown by females, order of testing was a significant variable [F(l,106) = 4.41, p < .038]. This means that the masculine behavior scores of females were influenced by whether females received their four weekly masculine behavior tests first, or after feminine behavior tests were completed. Table 1 shows the mean : standard error of masculine behavior scores for females, broken down by order of testing; the last column indicates significant differences within groups as shown by t-tests. Females receiving postnatal treat- ments of 1.0 mg ATD + 100 pg T had Significantly higher masculine behavior scores if male tests were given first; the same was true for females receiving 1.0 mg ATD + 2 pg E postnatally. For masculine behavior shown by Day 1 castrate males, an inter- action between postnatal treatment and order of testing was signifi- cant [F(7,60) = 2.69, p < .017]. Table 2 shows the masculine behavior 39 Table 1. Masculine Sexual Behavior (X MALE) for Female Postnatal Treatment Groups. X MALE Female Test Male Test Groups All First First t-tests PG + 011 0 1 i 0 1 0.0 4 0.0 0.1 i 0.1 n = (12) (5) (7) 0.5 ATD + 011 0.2 i 0. 0.4 i 0. 0.0 i 0.0 n = (8) (5) (3) 1.0 ATD + 011 0.4 i 0 0.2 i 0. 1.1 i 0.6, n = (14) (11) (3) PG + SOT 7.1 i 1 6.9 i 1. 7.1 i 1.4 n = (10) (3) (7) 1.0 ATD + 50T 0.6 i 0 0.7 i 0. 0.5 i 0.3 n = (12) (5) (5) PG + 100T 9.6 i 1 9.4 i 1 10.0 i 3.0 n = (15) (9) (5) 0.5 ATD + 100T 10.5 i 2 9.8 i 2 11.5 i 3.7 n = (12) (7) (5) 1.0 ATD + 100T 5.1 i 1. 1 7 i 0 9 1 i 2.2 +(11) = 3.28, n = (13) (7) (6) p < .007 PG + 2E 6.7 t 1. 5.9 i 1 8.3 i 3.1 n = (14) (9) (5) 0.5 ATD + 2E 8.0 i 1. 9.2 i 2 6.1 i 1.0 n = » (10) (5) (4) 1.0 ATD + 2E 11.8 i 1.8 2.9 i 0.6 14.0 i 1.2 +(8) = 4.32, n = (10) (2) (8) p < .003 40 Table 2. Masculine Sexual Behavior (X MALE) for Day 1 Castrate Male Postnatal Treatment Groups. 2 MALE Female Test Male Test Groups All First First t-tests PGI+ oil 0.5 i 0.5 0.0 2 0.0 1.1 s 1.0 n = (6) (3) (3) 1.0 ATD + oil 2.9 i 1.3 0.5 i 0.2 6.0 i 1.7 +(5) = 3.95, n = (7) (4) (3) p < .011 PG + 50T 3.2 i 0.9 3.4 i 1.3 3.0 i 1.3 n = (13) (7) (5) 1.0 ATD + SOT 1.2 i 0.3 1.3 i 0.5 1.1 i 0.5 n = (11) (4) (7) PG + 100T 5.6 i 1.3 3 9 i 1 5 7.3 i 2.1 n = (10) (5) (5) 1.0 ATD + 100T 6.6 i 2.0 8 1 i 2 2 1.4 i 1.4 n = (9) (7) (2) PG + 25 1 o i o 4 0.4 t 0.4 2.1 t 0.9 +(7) = 2.73, n = (9) (6) (3) p < .029 1.0 ATD + 2E 4.1 i 0.7 5.1 i 1.1 3.2 i 0.9 n = (11) (5) (5) 41 scores of Day 1 castrate males broken down by order of testing. Masculine behavior scores were significantly higher when masculine behavior was tested before feminine behavior for males treated post- natally with PG + 2 pg E or 1.0 mg ATD + oil. Although the order of testing for 1.0 mg ATD + 100 pg T was not Significant [t(7) = 1.56, p < .162], masculine behavior scores were higher when maSculine behavior tests followed female behavior tests; this difference probably accounted for the significant interaction. Again, for intact males receiving postnatal treatments, order of testing was a significant variable for rear mounts [F(1,82) = 26.63, p < .0001] as seen in Table 3; for intromissions [F(1,82) = 92.81, p < .0001] as seen in Table 4; and for ejaculations [F(l,82) = 79.19, p < .0001] as seen in Table 5. By visual inspection of Tables 3-5, and by examining the t-test results, it is obvious that there are large differences due to order of testing for intact males. Although most of these are in the direction of higher masculine behavior scores if male tests are given first, significant interaction effects for rear mounts [F(8,82) = 2.25, p < .03] and for ejaculations [F(8,82) = 4.33, p < .0001] indicate that for some postnatal treat- ments, the direction of difference for order effects was reversed. Feminine Behavior - Females. Feminine behavior was analyzed in females using mean total lordosis duration (XTLD) as the behavioral measure; this is the total lordosis duration divided by lordosis frequency. Figure 6 shows the mean i the standard error for feminine behavior scores for all postnatal treatment groups. A Significant difference was found between treatment groups in a one way comparison [F(10,119) = 5.82, p < .0001]. Further tests revealed that females 42 Table 3. Rear Mounts for Intact Male Postnatal Treatment Groups Rear Mounts FemaTé Test Male Test Groups All First First t-tests PG + oil 8.6 i 1.9 5.0 i 0.9 14.7 t 0.6 +(6) = 7.74, n = (8) (5) (3) p < .0001 0.5 ATD + oil 4.1 i 1.0 4.5 i 1.1‘ 3.6 i 2.0 n = (10) (6) (4) 1.0 ATD + oil 3.8 i 0.9 2.4 i 0.9 5.9 t 1.5 n = (10) (5) (4) PG + 100T 6.2 i 1.1 4.5 t 1.2 9.2 t 1.5 +(9) = 2.43, n = (11) (7) (4) p < .038 0.5 ATD + 100T 7.0 i 0.9 5.5 i 1.5 8.2 i 1.1 n = (14) (6) (8) 1.0 ATD + 100T 6.5 i 0.8 5.4 i 1.3 7.4 i 1.0 n = (11) (5) (6) PG + 2E 8.9 i 0.8 8.6 i 0.7 9.2 t 1.5 n = (11) (5) (5) 0.5 ATD + 2E 8.0 i 1.1 6.2 i 0.8 8.7 i 1.5 n = (11) (3) (8) 1.0 ATD + 2E 8.7 i 1.3 5.4 i 1.8 11.2 t 1.1 +(12) = 2.86, n = (14) (6) (8) p < .014 43 Table 4. Intromissions for Intact Male Postnatal Treatment GrOUpS Intromissions Female Test Male Test Groups All First First t-tests PG + oil 9.8 i 2. 5.8 i 1.7 16.4 i 1.8 +(6) = 4.06, n = (8) (5) (3) p < .007 0.5 ATD + oil 6.5 i 2 5.1 i 2.5 8.7 i 6.2 n = (10) (5) (4) 1.0 ATD + oil 7.1 i 2 3.6 i 2.1 12.4 i 4.8 n = (10) (6) (4) PG + 100T 13.0 i 3 7.1 i 2.3 23.2 i 4.5 +(9) = 3.58, n = (11) (7) (4) p < .005 0.5 ATD + 100T 15.8 i 2 10.3 i 3.7 19.8 i 2.7 n = (14) (6) (8) 1.0 ATD + 1001 20.4 i 3 8.8 i 2.6 30.0 i 2.2 +(9) = 6.29, n = (11) (5) (6) p < .0001 PG + 2E 16.7 i 3. 8.8 i 1.7 26.0 i 2.3 +(9) = 6.30, n = (11) (6) (5) p < .0001 0.5 ATD + 2E 16.7 i 2 5.8 i 0.9 20.8 i 1.7 +(9) = 5.17, n = (11) (3) (8) p < .001 1.0 ATD + 2E 10.5 i 2 3.2 i 1.5 16.1 i 2.4 +(12) = 4.16, n = (14) (6) (8) p < .001 44 Table 5. Ejaculations for Intact Male Postnatal Treatment Groups. Ejaculations ema e est Male Test Groups All First First t-tests PG + oil 1.4 i 0.4 0.7 i 0.2 2.6 i 0.4 +(6) = 4.34, n = (9) (5) (3) p < .005 0.5 ATD + oil 0.5 i 0.3 0.6 t 0.4 0.5 i 0.5 n = (10) (6) (4) 1.0 ATD + oil 1.0 i 0.5 0.4 i 0.3 1.8 i 1.0 n = (10) (5) (4) PG + 100T 2.8 i 0.3 O 9 i 0.4 3.1 i O 9 +(9) = 2.48, n = (11) (7) (4) p < .035 0.5 ATD + 100T 1.8 i O 4 l 5 i 0.5 2 1 i O 6 n = (14) (5) (3) 1.0 ATD + 100T 3.3 i O 8 O 8 i 0.4 5 4 i O 4 +(9) = 8.66, n = (11) (5) (6) p < .0001 PG + 2E 2.4 i 0 6 0 8 i 0.2 4 2 i 0 6 +(9) = 5.79, n = (11) (6) (5) p < .0001 0.5 ATD + 2E 2.0 i O 4 O 2 i 0.2 2 6 i 0 4 +(9) = 3.96, n = (11) (3) (8) p < .003 1.0 ATD + 2E 1.2 i 0 4 0 O i 0.0 2 0 i O 5 +(12) = 3.93, n = (14) (6) (8) p < .002 45 FEMALES N O O + +- 100 MEAN LORDOSIS DURATHDN 0| 0 + 1;].- ::i- 0‘ O . PG .SATO lATD PG lATD PG .5ATO lATD PG .5ATD lATD oil oil oil SOT SOT lOOT 100T 100T 2E 2E 2E Figure 6. Mean lordosis durations for female postnatal treatment groups; each treatment of ATD (mg), hormone (pg) or control vehicle was given on Days 2-4 after birth. 46 receiving 1.0 mg ATD + oil or PG + oil showed longer mean lordosis durations than the most defeminized groups: PG + 100 pg T, 0.5 ATD + 100 pg T, 1.0 mg ATD + 2 pg E and 0.5 mg ATD + 2 pg E (p < .05). Although 0.5 mg ATD did not block the defeminization produced by 100 pg T, a higher dose of ATD (1.0 mg), or lower dose of testosterone (50 pg), produced less defeminization (no statistical tests). Order of testing was a significant variable in a two way analysis of variance [F(1,106) = 32.29, p < .0001] and interaction effects were also significant [F(10,106) = 2.95, p < .003]. Table 6 shows the feminine behavior scores for females, broken down by order of testing; the last column indicates significant differences between groups as Shown by t-tests. Females receiving postnatal treatments of PG + 100 pg T, 1.0 mg ATD + 100 pg T, PG + 2 pg E or PG + oil Showed signi- ficantly longer mean total lordosis durations if feminine behavior tests were preceded by masculine behavior tests. Feminine Behavior - Males. Neither male group, Day 1 castrate males or intact males, differed between postnatal treatments on mean total lordosis duration. Figure 7 shows feminine behavior scores for Day 1 castrate males and Figure 8 shows feminine behavior scores for postnatally intact males. Male hamsters did Show Shorter mean lordosis durations than females, including Day 1 castrates receiving no additional hormone treatments. Possibly, the doses of hormone used in this experiment were too low to further defeminize males. Morphological Data. At the time of sacrifice, body weights and ano-genital distance measurements were obtained for animals in all postnatal treatment groups; in addition, the length of the penile bone and cartilage was measured in males. Testes weights were 47 Lav Luv flopv A c o.m. A m.mw m.om A o.~o o.m_ A ~._m mm + oe< o.P Lev Lev Ao_v A e A._P A A.AA m.o A ~.me e.m A o.m~ AN + oe< m.o Amv Amv AAPV A e _ooo. v a .AP.A A AN_V+ A.NN A ~.AAN o.mp A _.Aw A.AN A _.Am_ mm + we 1 Amy ARV AMFV N : mmo. v a .mm.~ A A__v+ o.Am A A.m- m.op A m.Am m.m~ A m.o__ poop + QA< o._ “my fihv .NPV A e m.o_ A o.mm m.m~ A A.m~ m.m_ A m.mm Loo_ + DL< m.o . Lev Amy Am_v A e mmo. v a Am.N A Am_v+ o.m_ A _.~__ A.mp A m.me o.AF A w.AA Loo_ A me “my “my AN_V A : «.mm A A.om_ o.m~ A _.mN_ o.o~ A m.ee_ pom A oe< o._ Rev any Aopv A c m.mp A m.mpp m.om A m.mm o.e_ A m.mop Pom + as .mv APLV AA_V A c A.Ae A A.AoN m.- A o.mm~ m.e~ A m.FmN _Ao + oh< 0.. Amy Amy Amy A e m.mm A ~.m_~ A.m~ A o.e__ A.Am A A.mm. FLO + op< m.o . ARV Amy A~_v A e «co. v a ~o.¢ A Ao_v+ o.m~ A A.om~ o.~_ A m.A.F A.A~ A m.mm_ __o + on AAAAA-A AALAE AALAE _p< masoem amok o—ez pmm_ «Page; one x .mazocw pewsumoep ~mum=umo¢ «Foam; LoA and» my Lov>mgom —m=xom mcvcwsmm .o o—nmh MEAN LORDOSIS OUR AT 1 O N 48 N 01 O DAYl CASTRATE MALES N O O 150 100 01 O lATD lATD lATD lATD oil Oil SOT SOT lOOT lOOT 2E 2E Figure 7. Mean lordosis durations for Day 1 castrate male postnatal treatment groups; each treatment of ATD (mg), hormone (pg) or control vehicle was given on Days 2-4 after birth. 49 z 250 ‘ 9 '— EEOC)" INTACT MALES 8 9150- m 0 D 51001 4 z s so- PG .SATD lATD SATD lATD SATD lATD oil oil oil 100T 100T 100T 2E 2E 2E Figure 8. Mean lordosis durations for intact male postnatal treatment groups; each treatment of ATD (mg), hormone (pg) or control vehicle was given on Days 2 -4 after birth. 50 obtained from intact males at the time of adult castration (Day 60). Body weights for all animals are shown in Table 7 and morphological measurements are shown in Table 8. For body weight in females, there was a significant difference between postnatal treatment groups [F(10,ll7) = 4.705, p < .0001]. Females receiving either PG + 100 pg T, 1.0 mg ATD + oil or PG + 2 pg E weighed more than females receiving PG + 50 pg T or 1.0 mg ATD + 50 pg T postnatally (p < .05). There were no differences in body weights for Day 1 castrate males. For intact male postnatal treat- ment groups, body weight did differ [F(8,91) = 2.066, p < .0472]. Males receiving PG + 100 pg T postnatally weighed more than males receiving 1.0 mg ATD + 2 pg E (p < .05). For females in all postnatal treatment groups, the ano-genital distance was shorter than for any male postnatal treatment group; there were no Significant differences between treatment groups. Ana-genital distance did differ for Day 1 castrate males [F(7,68) = 4.807, p < .0002] and for intact male postnatal treatment groups [F(8,91) = 2.482, p < .0176]. For Day 1 castrate males, males receiving PG + 100 pg T postnatally had a longer ano-genital distance than males receiving either PG + 50 pg T, 1.0 mg ATD + 50 pg T, 1.0 mg ATD + 2 pg E or PG + oil (p < .05).. For intact males, males receiving either 0.5 mg ATD + oil or PG + 2 pg E postnatally had longer ano-genital distances than males receiving 1.0 mg ATD + 2 pg E (p < .05). For males, when the length of the penile bone and cartilage was measured, there were Significant differences between postnatal treat- ment groups for Day 1 castrate males [F(7,68) = 6.846, p < .0001] and Table 7. Body Weight (gm) for All Postnatal Treatment Groups. Day 1 Groups Females Castrate Males Intact Males PG + oil 126.4 i 4.5 114.9 i 5.2 134.5 i 4.7 0.5 ATD + oil 118.1 i 4.9 --- 141.5 2 6.1 1.0 ATD + oil 139.6 i 5.3 118.9 i 4.4 132.3 2 3.1 PG + 50T 114.7 i 3.1 119.5 i 2.5 --- 1.0 ATD + 50T 113.5 i 4.0 126.8 i 3.9 --- PG + 100T 141.0 i 5.7 128.7 i 3.9 144.5 i 9.8 0.5 ATD + 100T 132.5 i 3.9 --- 132.1 i 4.3 1.0 ATD + 100T 130.5 i 2.8 133.2 1 4.5 127.2 i 5.5 PG + 2E 142.4 i 4.7 130.3 1 4.2 136.0 i 0.6 0.5 ATD + 2E 130.6 i 6.1 --- 140.7 i 4.2 1.0 ATD + 2E 133.5 i 4.2 127.1 4 5.4 '120.9 i 3.5 52 m.o A m.~ F.o A ¢.m m.o A m.e— _.o A ~.¢ ¢.o A m.m— m.o A m.o~ mm + oh< o.~ _.o A m.m N.o A e.m m.o A o.m_ AAA 1-- m.o A p.op mm + oh< m.o m.o A m.m To A ad ed A 93 No A m.m Yo A 93 m6 A m6 mm + an. ~.o A m.m _.o A m.m v.o A m.m— F.o A m.¢ «.0 A ~.m~ ~.o A ¢.op hoop + op< o._ m.o A n.~ —.o A m.m ¢.o A w.m_ AAA AAA m.o A ~.o_ poop + ah< m.o m.o A m.~ _.o A ~.m m.o A m.m_ _.o A m.¢ n.o A m.mp m.o A ~.o— poo_ + we 11 1| 11 No A AA A... A Z: No A 3: SA + a: o; 1-- AA- -1- _.o A ~.¢ m.o A m.mp m.o A m.o_ Pom + on «.0 A m.m ~.o A ¢.m m.o A o.mp ~.o A m.¢ m.o A «.mp ~.o A m.o_ Fwo + oh< c.~ p.o A m.m P.o A e.m «.0 A m.m— AA- AAA ¢.o A m.m __o + ah< m.o AAA p.o A _.m 5.0 A o.m— m.o A e.¢ n.o A m.m_ m.o A m.o_ .wo + on £3 A: 2?: E: u A E 225% E5 0 A E 35 .3 Es RF mm—ez pomucH mmpmz mumeummu _ awe mopmswm mazoew .huz mmummkv agave: mmumop use Au a mav gumcm4 mumpwucmu use mcom mpwcma .Aou saw: umummp maaogw “cospmmch Peamcumoa upmeom com Ange xv Low>ocmm pcsxom mcwcwemu .m «Pack 59 doses (50, 100, 200 and 400 pg progesterone). Females from all post- natal treatment groups showed a definite facilitation of feminine sexual behavior in response to progesterone combined with a priming dose of estradiol benzoate. However, there were no differences between postnatal treatment groups; facilitation was equal for all groups at all dose levels of progesterone. Ana-genital distance did not differ for any postnatal treatment groups, but there was a significant difference in body weights [F(8,57) = 2.5664, p < .0183]. Normal females weighed more than females receiving 0.5 mg ATD + oil postnatally (p < .05). The means and standard errors for these morphological measures are shown in Table 10. 60 Table 10. Morphological Measures for Female Postnatal Treatment Groups Tested for Estrogen and Progesterone Sensitivity. Groups Body Weight (gm) Anogenital Distance (mm) Normal 145.1 i 8.8 10.8 i 0.4 0.5 ATD + oil 117.0 1 5.3 10.0 i 0.2 1.0 ATD + 011 131.2 1 4.9 10.7 t 0.3 PG + 100T 121.3 i 2.6 9.6 t 0.1 0.5 ATD + 100T 121.2 i 2.6 9.2 i 0.2 1.0 ATD + 100T 117.8 2 5.4 8.8 t 0.2 PG + 2E 112.7 i 6.4 ' 10.1 t 0.5 0.5 ATD + 2E 133.8 2 4.7 10.0 t 0.4 1.0 ATD + 2E 133.9 2 8.8 9.9 t 0.4 DISCUSSION This present study extends the concept of the aromatization hypothesis by Showing that inhibition of aromatization blocked behavioral masculinization and defeminization in the hamster. This is in agreement with previous work by Coniglio, Paup and Clemens (Coniglio et al., 1973a,b; Paup et al., 1972, 1974) which provided a data base and suggestion for this interpretation of hormone action during early life. This is further evidence that for the hamster, at least, estrogen is the major active hormone involved in behavioral sexual differentiation. The effects of inhibition of aromatization on the two behavioral processes examined in this study, masculiniza- tion and defeminization will be considered separately. Masculinization In the present study, both testosterone and estradiol induced the potential for adult mounting in females and Day 1 castrate males given hormones on Days 2-4 after birth. Although the testosterone doses were 50 or 25 times greater than the estradiol dose, the effects on induction of mounting were equivalent. This correSponds to an estimated in vivo aromatization rate of 1-2% for testoSterone conver- sion to estradiol (Naftolin et al., 1975). The degree of masculini- zation produced by either hormone treatment was, however, minimal; most animals mounted but did not intromit or ejaculate. With higher 61 62 hormone doses, a more extended time period of administration, or with the pr0pionate or benzoate form of these hormones, more complete masculinization of behavior probably would have resulted. The 20:1 ratio of ATD to testosterone (1.0 mg ATD + 50 pg T) completely blocked behavioral masculinization in females and Day 1 castrate male hamsters, while the 10:1 ratio (1.0 mg ATD + 100 pg T) only achieved a partial block of masculinization in females; the 10:1 ratio did not block masculinization in Day 1 castrate males. The 5:1 ratio (0.5 mg ATD + 100 pg T) did not inhibit masculinization for either group. In addition, ATD had no independent or synergistic effects on behavior; animals receiving ATD alone were not masculinized and ATD had no effect on masculinization produced by estradiol. These results conform to the idea that aromatization of testosterone to estradiol is essential for behavioral masculinization. Intact males were also tested for inhibition of behavioral mascu- linization by ATD. Those males which received ATD Showed fewest intro- missions and ejaculations. The highest scores for intromissions and ejaculations were obtained by males treated postnatally with 1.0 mg ATD + 100 pg T or PG + 100 pg T. With a combined exposure to endogenous and exogenous testosterone, the amount of ATD present was not suffi- cient to influence behavior. Defeminization For feminine sexual behavior, the mean total lordosis duration (XTLD) was compared for the various postnatal treatment groups. This behavioral measure incorporates two aspects of feminine behavior which are altered in the hamster by postnatal exposure to gonadal steroids: 63 the total lordosis duration which decreases as a result of exposure to gonadal steroids, and the frequency of lordosis, which increases as a result of early steroid exposure. In brief, defeminized females show short lordosis responses, hence the frequency is increased. For female hamsters in the present eXperiment, control animals not receiving hormones showed the longest mean duration of lordosis. Feminine behavior in female hamsters was suppressed by estradiol, testosterone and the 5:1 ratio of ATD to testosterone (0.5 mg ATD + 100 pg T). The 20:1 ratio (1.0 mg ATD + 50 pg T) partially blocked the defeminizing effects of testosterone. ATD had no independent effects on behavior. A11 male groups, regardless of postnatal treatment, were defemi- nized in comparison to females, but there were no differences in feminine behavior between the postnatal treatment groups. The doses of hormone used in the present study may have been insufficient to further defeminize the behavior of males. Since, for Day 1 castrate males, development occurs in the absence of postnatal gonadal hormones, these males should be equivalent to normal females in their lordosis responding. However, other studies have shown that Day 1 castrate male hamsters are not equivalent to normal females. Coniglio et al. (1973a,b) found a lordosis duration of 418 seconds for females but only 108 seconds for males castrated on Day 1. These authors suggested several possible explanations for the difference. First, since the maximum sensitive period for defeminization is within the first 3 days after birth, perinatal exposure to testicular androgen prior to castration may be sufficient to defeminize. Second, the stimulus male may react differently to a male partner than to a 64 female partner. Kow et a1. (1976), using manual stimulation, found a greater degree of receptivity in males than is usually found in mating tests. Third, vaginal stimulation, which is not available to the male, may influence lordosis duration; in female hamsters at least, prolonged genital stimulation can be a signal to terminate lordosis (Carter et al., 1976). Order of Testing Effects In the present study, all animals were tested for both masculine and feminine sexual behavior. Since no data are available on whether adult eXposure to particular hormones (either testosterone or estrogen + progesterone) or whether exposure to a particular testing situation (masculine or feminine behavior) can affect subsequent sexual behavior, order of testing was counterbalanced in this experiment. An exact 50:50 ratio was not maintained for all postnatal treatment groups, in part because of the logistics of housing and testing animals, and in part because of animal mortality. When the data were analyzed for order of testing effects, signi- ficant differences were found within postnatal treatment groups for some female and Day 1 castrate male treatment groups, and for most intact male treatment groups. These differences generally meant that animals which had received masculine behavior tests before feminine behavior tests had higher masculine behavior scores; for female treat- ment groups, females from some postnatal treatment groups had higher feminine scores if feminine sexual behavior was tested last. Several interpretations of these findings are possible. 65 There are three major differences for sequence of testing: proximity to time of castration, prior exogenous hormone eXposure and prior behavioral testing. For intact males castrated as adults, different time intervals between castration and testing for masculine behavior could account for the order effects. That is, males receiving masculine behavior tests first had a 10 day period with no hormones between castration and the initiation of testosterone pr0pionate re- placement. For males tested in the reverse sequence, 6 weeks with no hormone treatment preceded masculine behavior tests. Since other behavioral studies have shown that with increasing time following castration, either increasing amount of hormones or longer exposure to hormones is needed to reinstate behavior (Christensen, Coniglio, Paup & Clemens, 1973), males tested after 6 weeks of hormonal depri- vation might be expected to Show reduced masculine behavior. For masculine and feminine behavior tests in female hamsters, the interpretation used for intact males is not applicable. Although order of testing effects were not as pervasive across postnatal treat- ment groups in female hamsters, when order effects were seen, both masculine and feminine sexual behavior were facilitated if masculine behavior tests occurred first. For females, it is possible that there were some carryover effects from ovarian secretions which could faci- litate the diSplay of mounting. Several studies have shown that masculinized female hamsters mount prior to ovariectomy, and continue to mount for short periods following ovariectomy (Carter et al., 1972; Swanson & Crossley, 197; Tiefer & Johnson, 1975). Residual effects of ovarian secretions could possibly account for increased mounting in some female groups when masculine behavior was tested first. 66 The finding that, for some postnatal treatment groups, feminine behavior scores were higher in female hamsters when tested following masculine behavior tests, is counter-intuitive. Beach (1976) has applied a deprivation desensitivity interpretation, used to explain decrements in post-castration masculine behavior, to eXplain a Similar phenomenon for feminine behavior. For female rats, females are less sensitive to estrogen and progesterone with increasing time post- ovariectomy. Since the order of testing effects seen for feminine behavior in this experiment were Opposite to this, another explanation is needed. Day 1 castrate males also showed order of testing effects for masculine behavior for several postnatal treatment groups, with higher masculine behavior scores for masculine behavior tests first. Since these males were deprived of hormones from Day 1 after birth until Day 70 when adult hormone treatments began, it is difficult to see how deprivation desensitivity could account for order of testing effects in these males. 0f the three major differences for sequence of testing, proximity to time of gonadectomy offers a plausible explanation for order of testing effects for masculine behavior in postnatally intact male treatment groups and for masculine behavior in female treatment groups. For masculine behavior in Day 1 castrate males and for feminine behavior in females, though, the effects of first hormone treatment and first behavioral testing upon the second test sequence must be considered. There are no data available to suggest that performing adult masculine sexual behavior will alter an animal's ability to show feminine responses, or vice versa. However, adult hormone treatments could possibly modify the effects of perinatal hormone treatments. 67 Two examples will illustrate the possibility that adult hormone treatment could modify the process of sexual differentiation. Noble (1974, 1977) found that normal female hamsters will mount receptive females when given prolonged (4 week) very high doses of gonadal hormones as adults; he suggested that such adult hormone exposure might modify the female's ability to respond to hormones. Gorski has also described a physiological syndrome in female rats, the delayed anovulatory syndrome, where it is suspected that feedback from adult ovarian hormones can modify the pattern of gonadotropin release (Gorski et al., 1977). Female rats which are lightly masculinized by postnatal testosterone begin cycling at puberty but then become anovu- latory. Failure to ovulate, then, is initiated by postnatal exposure to testosterone and then completed by postpubertal ovarian feedback. There could be an analogous behavioral syndrome where adult testos- terone treatment or adult estrogen + progesterone treatment could be completing or modifying the process of sexual differentiation begun at birth. Estrogen and Progesterone Sensitivity Defeminization, i.e., decreased lordosis scores in response to estrogen and progesterone, may result from altered sensitivity to ovarian hormones (Clemens, 1972; Kow et al., 1974). One way to deter- mine the mechanism of behavioral defeminization is to look for changes in reSponsiveness to estrogen and progesterone in the adult. In Experiment 2, female hamsters receiving testosterone or estradiol during early life, with or without ATD, were tested for lordosis in response to varying doses of estrogen alone and in reSponse to estrogen 68 and varying doses of progesterone. With acute doses of estradiol benzoate, ranging from 1.5 to 12 pg, females did not Show measurable levels of feminine behavior. This finding is consistent with previous work showing that female hamsters are not responsive to estrogen alone, and does not indicate that postnatal hormone treatment altered sensi- tivity to estrogen. After completion of the estrogen only lordosis testing, females from all groups were tested for lordosis with 3 pg estradiol benzoate combined with progesterone in doses ranging from 50 to 400 pg. Pro- gesterone facilitated lordosis for all postnatal treatments, but there was no difference between groups in response to any dose of proges- terone. Facilitation was equal for all treatment groups, whereas a dose response relationship had been anticipated. DeBold, Martin and Whalen (1976) found that for normal female hamsters, 50 pg proges- terone was the minimal dose to facilitate lordosis, while 200 pg progesterone produced maximal responding. The priming dose of estro- gen was different for the two studies: DeBold et a1. (1976) gave a single injection of 10 pg estradiol benzoate, while in the present study, 3 daily injections of 3 pg estradiol benzoate preceded proges- terone injections. The differences in estrogen priming, however, probably were not sufficient to account for the behavioral difference. The female hamsters in this experiment appeared to be more sensitive to progesterone, and this was not altered by postnatal treatment. In Experiment 2, lordosis scores were quite variable. This can be seen in the means and standard errors for each dose of progesterone (as shown in Table 9), and also in the lack of consistent response for each postnatal treatment group with increasing doses of 69 progesterone. Two periods of hormone exposure need consideration in interpreting this variability: postnatal treatment and adult hormone treatment. In Experiment 1 of this study, females were defeminized by the same doses of testosterone and estradiol used in this second experiment. Failure to find differences between groups, then, is most likely due to differences in adult hormone treatment.' Before being tested for lordosis in reSponse to estradiol ben- zoate plus varying doses of progesterone, females from all treatment groups were eXposed to 4 weeks of hormone treatment and testing with estradiol benzoate only (3 injections per week). It is possible that this estrogen priming, although not affecting behavior at the time of testing, could have attenuated or left incomplete the process of defeminization. The variability in behavior, then, would reflect the variability in this adult action of low doses of estradiol benzoate. Again, the analogy can be made to the adult action of hormones on the mounting of normal female hamsters (Noble, 1974, 1977) and on the delayed anovulatory syndrome in the female rat (Gorski et al., 1977). This present eXperiment, though, does not provide support for the hypothesis that defeminization in the hamster involves changes in sensitivity to either estrOgen or progesterone. Furthermore, since the initiation of this study, it has been found that estrogen influences the amount of progesterone receptors available (Feder, Landau, Marrone & Walker, 1977). If so, then any altered response to progesterone would be difficult to interpret, Since it might reflect either de- creased responsiveness to progesterone or an alteration in the capacity of estrogens to induce progesterone receptors. 70 Morphological Measures For the morphological measures recorded in this study, one . difference was clear and consistent: females from any postnatal treatment groups had a Shorter ano-genital distance than any of the male treatment groups. This morphological characteristic distin- guishes between males and females. For the other morphological measures, body weight, ano-genital distance (between postnatal treat- ment groups) and penile bone and cartilage length (males only), no clear and consistent differences were found, although some statisti- cally significant differences were found. The magnitude of these differences was probably not sufficient to be functionally important. Previous developmental studies in the hamster have not found a corre- lation between behavioral masculinization and morphological masculini- zation for different postnatal treatment groups (Coniglio et al., 1973a,b; Paup et al., 1972). Conclusion This study provided experimental support for the aromatization hypothesis; ATD, which inhibits aromatization, blocked behavioral masculinization in female hamsters, Day 1 castrate males and post- natally intact males. ATD also partially blocked defeminization in female hamsters. These data also suggest that adult hormone treat- ment may modify the effects of exposure to hormones in early life; further experiments are needed to test the possible nature of this modification. Finally, defeminization in the female hamster did not involve altered sensitivity to estrogen or progesterone in adulthood. LIST OF REFERENCES LIST OF REFERENCES Beach, F. A. Hormonal modification of sexually dimorphic behavior. Psychoneuroendocrinology, 1975, 1, 3-23. Beach, F. A. Prolonged hormone deprivation and pretest cage adaptation as factors affecting the display of lordosis by female rats. Physiology and Behavior, 1976, 19, 807-808. Beach, F. A., & Rabedeau, R. G. Sexual exhaustion and recovery in the male hamster. Journal of Comparative and Physiological Psychology, 1959, 52, 56-66. Beach, F. A., Stern, 8., Carmichael, M., & Ranson, E. Comparisons of sexual receptivity and proceptivity in female hamsters. Behavioral Biology, 1976, 18, 473-488. Booth, J. E. Effects of aromatization inhibitor Androst-4-ene-3,6,l7- trione on sexual differentiation induced by testosterone in neo- natally castrated rat. Journal of Endocrinolggy, 1977, 72, 69-76. Bunnell, B. N., Boland, B. 0., & Dewsbury, D. A. Copulatory behavior of golden hamsters (Mesocricetus auratus). Behavior, 1977, Q1, 180-206. Carter, C. 5., C1emens, L. G., B Hoekema, D. J. Neonatal androgen and adult sexual behavior in the golden hamster. Physiology and Behavior, 1972, 9, 89-95. Carter, 0. S., Landauer, M. R., Tierney, B. M., & Jones, T. Regulation of female sexual behavior in the golden hamster: Behavioral effects of mating and ovarian hormones. Journal of Comparative and Physiological Psychology, 1976, 99, 839-850. Carter, C. 5., Michael, S. J., & Morris, A. H. Hormonal induction of female sexual behavior in male and female hamsters. Hormones and Behavior, 1973, 4, 129-141. Carter, C. S., & Schein, M. W. Sexual receptivity and exhaustion in the female golden hamster. Hormones and Behavior, 1971, 2, 191- 200. '— 71 72 Christensen, L. W., & C1emens, L. G. Blockade of testosterone induced mounting behavior in the male rat with intracranial application of the aromatization inhibitor, androst-l,4,6-triene-3,17-dione. Endocrinology, 1975, 21, 1545-1551. Christensen, L. W., Coniglio, L. P., Paup, D. C., & C1emens, L. G. Sexual behavior of male golden hamsters receiving diverse andro- gen treatments. Hormones and Behavior, 1973, 4, 223-229. Clemens, L. G. Perinatal hormones and the modification of adult . behavior. In C. H. Sawyer & R. A. Gorski (Eds.), Steroid hormones 23d brain function. Los Angeles: University of California Press, 72. C1emens, L. G., Hiroi, M., & Gorski, R. A. Induction and facilitation of female mating behavior in rats treated neonatally with low doses of testosterone pr0pionate. Endocrinology, 1969, 24, 1430-1438. Coniglio, L. P., & C1emens, L. G. Period of maximal susceptibility to behavioral modification by testosterone in the golden hamster. Hormones and Behavior, 1976, Z, 267-282. Coniglio, L. P., Paup, D. C., & Clemens, L. G. Hormonal factors con- trolling the develOpment of sexual behavior in the male golden hamster. Physiology_and Behavior, 1973a, 12, 1087-1094. Coniglio, L. P., Paup, D. C., & Clemens, L. G. Hormonal specificity in the suppression of sexual receptivity of the female golden hamster. Journal of Endocrinology, 1973b, 2], 55-61. DeBold, J. F., Martin, J. V., & Whalen, R. E. The excitation and in- hibition of receptivity in female hamsters by progesterone: Time and dose relationships, neural localization and mechanisms of action. Endocrinology, 1976, 22, 1519-1527. DeBold, J. F., & Whalen, R. E. Differential sensitivity of mounting and lordosis control systems to early androgen treatment in male and female hamsters. Hormones and Behavior, 1975, 2, 197-209. Eaton, G. Effect of a single prepuberal injection of testosterone propionate on adult bisexual behavior of male hamsters castrated at birth. Endocrinology, 1970, 22, 934-940. Engel, L. L. The biosynthesis of estrogens. In R. O. Greep (Ed.), Handbook of physiology (Section 7, Vol. 2, Pt. 1). Washington, D. 0.: American Physiological Society, 1975. Feder, H. H., Landau, I. T., Marrone, B. L., & Walker, W. A. Inter- actions between estrogen and progesterone in neural tissues that mediate sexual behavior of guinea pigs. Psychoneuroendocrinology, 1977, 2, 237- 73 Gerall, A. A., McMurray, M. M., & Farrell, A. Suppression of the development of female hamster behavior by implants of testosterone and non-aromatizable androgens administered neonatally. Journal of Endocrinology, 1975, 22, 439-445. Gerall, A. A., & Thiel, A. R. Effects of perinatal gonadal secretions on parameters of weight gain and receptivity in hamsters. Journal of Comparative and Physiological Psychology, 1975, 22, 580-589. Gorski, R. A. Gonadal hormones and the perinatal develOpment of neuro- endocrine function. In L. Martini & W. F. Ganong (Eds.), Frontiers in neuroendocrinology. New York: Oxford University Press, Gorski, R. A. Mechanisms of androgen induced masculine differentiation of the rat brain. In K. Lissak (Ed.), Hormones and brain function. New York: Plenum Press, 1973. Gorski, R. A., Harlan, R. E., & Christensen, L. W. Perinatal hormonal exposure and the development of neuroendocrine regulatory processes. Journal of Toxicology_and Environmental Health, 1977.2, 97-121. Gottlieb, H., Gerall, A. A., & Thiel, A. Receptivity in female hamsters following neonatal testosterone, testosterone pr0pionate and MER-25. Physiology and Behavior, 1974, 12, 61-68. Goy, R. W., & Goldfoot, D. A. Neuroendocrinology: Animal models and problems of human sexuality. Archives of Sexual Behavior, 1975, 4, 405-420. Johnson, W. A. Neonatal androgenic stimulation and adult sexual be- havior in male and female golden hamsters. Journal of Comparative and Physiological Psychology, 1975, 22, 433-44Tl Jost, A. Problems of fetal endocrinology: the gonadal and hypOphyseal hormones. Recent Progress in Hormone Research, 1953, 2, 379-418. Jost, A. A new look at the mechanism controlling sex differentiation in mammals. Johns Hapkins Medical Journal, 1972, 122, 38-53. Kow, L.-M., Malsbury, C. W., & Pfaff, D. W. Effect of progesterone on female reproductive behavior in rats: Possible modes of action and role in behavioral sex differences. In W. Montagna & W. A. 2agler (Eds.), Reproductive behavior. New York: Plenum Press, 9 . Kow, L.-M., Malsbury, C. W., & Pfaff, D. W. Lordosis in the male golden hamster elicited by manual stimulation: Characteristics and hormone sensitivity. Journal of Comparative and Physiological Psychology, 1976, 22, 26-40. Lieberburg, I., & McEwen, B. S. Estradiol 17 B: A metabolite of' testosterone recovered in cell nuclei from limbic areas of neo- natal rat brain. Brain Research, 1975, 22, 165-170. 74 Lieberburg, I., Wallach, G., & McEwen, B. S. The effects of an in- hibitor of aromatization (1,4,6-androstatriene-3,17-dione) and an anti-estrogen (CI-628) on in vivo formed testosterone meta- bolites recovered from neonatal rat brain tissues and purified cell nuclei. Implications for sexual differentiation of the rat brain. Brain Research, 1977, 122, 176-181. McEwen, B. S., Lieberburg, I., Chaptal, C., & Krey, L. C. Aromatiza- tion: Important for sexual differentiation of the neonatal rat brain. Hormones and Behavior, 1977, 2, 249-263. McEwen, B. S., Lieberburg, I., Maclusky, N., & Plapinger, L. Inter- actions of testosterone and estradiol with the neonatal rat brain: Protective mechanism and possible relationship to sexual differen- tiation. Ann. Biol. anim. Bioch. BiOphys., 1976, 12, 471-478. McEwen, B. S., Plapinger, L., Chaptal, C., Gerlach, J., B Wallach, G. Role of fetoneonatal estrogen binding proteins in the association of estrogen with neonatal brain cell nuclear receptors, Brain Research, 1975, 22, 400-406. Milewich, L., George, F. W., & Wilson, J. D. Estrogen formation by the ovary of the rat embryo. Endocrinology, 1977, 122, 187-196. Morali, G., Larsson, K., B Beyer, C. Inhibition of testosterone in- duced sexual behavior in the castrated male rat by aromatization inhibitors. Hormones and Behavior, 1977, 2, 203-213. Murphy, M. R. Relative importance of tactual and nontactual stimuli in eliciting lordosis in the female golden hamster. Behavioral Biology, 1974, 12, 115-119. Naftolin, F., Ryan, K. J., Davies, I. J., Reddy, V. V., Flores, F., Petro, Z., White, R. J., Takaoka, Y., & Wolin, L. The formation of estrogens by central neuroendocrine tissues. Recent Progress in Hormone Research, 1975, 22, 295-319. Noble, R. G. The effects of castration at different intervals after birth on the copulatory behavior of male hamsters (Mesocricetus auratus). Hormones and Behavior, 1973, 4, 45-52. Noble, R. Estrogen plus androgen induced mounting in adult female hamsters. Hormones and Behavior, 1974, 2, 227-234. Noble, R. G. Mounting behavior in female hamsters: Effects of different hormone regimens. Physiology and Behavior, 1977, 12, 519-529. Noble, R. G., & Alsum, P. B. Hormone dependent sex dimorphisms in the golden hamster (Mesocricetus auratus). Physiology and Behavior, 1975, 14, 567-574. 75 Nucci, L. P., & Beach, F. A. Effects of prenatal androgen treatment on mating behavior in female hamsters. Endocrinology, 1971, 22, 1514-1515. Ortiz, E. The embryological development of the wolffian and mullerian ducts and the accessory reproductive organs of the golden hamster (Cricetus auratus). Anatomical Record, 1945, 22, 371-389. Paup, D. C., Coniglio, L. P., & Clemens, L. G. Masculinization of the female golden hamster by neonatal treatment with androgen or estrogen. Hormones and Behavior, 1972, 2, 123-131. Paup, D. C., Coniglio, L. P., & Clemens, L. G. Hormonal determinants in the develOpment of masculine and feminine behavior in the female hamster. Behavioral Biology, 1974, 12, 353-363. Phoenix, C. H., Goy, R. W., Gerall, A. A., & Young, W. C. Organizing acting of prenatally administered testosterone propionate on the tissues mediating mating behavior in the female guinea pig. Endocrinology, 1959, 22, 369-382. Plapinger, L., & McEwen, B. S. Ontogeny of estradiol binding sites in rat brain 1. Appearance of presumptive adult receptors in cytosol and nuclei. Endocrinology, 1973, 22, 1119-1128. Raynaud, J. P., Mercier-Bodard, C., & Baulieu, E. E. Rat estradiol binding plasma protein (EBP). Steroids, 1971, 12, 767-788. Reddy, V. V. R., Naftolin, F., & Ryan, K. J. Conversion of andros- tenedione to estrone by neural tissues from fetal and neonatal rats. Endocrinology, 1974, 24, 117-121. Reed, C. A., & Reed, R. The copulatory behavior of the golden hamster. Journal of Comparative Psychology. 1945, 22, 7-12. Schwarzel, W. C., Kruggel, W. G., & Brodie, H. J. Studies on the mechanisms of estrogen biosynthesis VIII. Enzyme systems in human placenta. Endocrinology, 1973, 22, 866-880. Swanson, H. H. Effects of castration at birth in hamsters of both sexes on luteinization of ovarian implants, oestrous cycles and sexual behavior. Journal of Reproduction and Fertilipy, 1970, 21, 183-186. Swanson, H. H. Determination of the sex role in hamsters by the action of sex hormones in infancy. In D. H. Ford (Ed.), Influence of hormones on the nervous system. Basel: Karger, 1971. Swanson, H. H., & Crossley, D. A. Sexual behavior in the golden hamster and its modification by neonatal administration of testosterone pr0pionate. In M. Hamburgh & E. J. W. Barrington (Eds.), Hormones in development. New York: Appleton-Century-Crofts, 1971. 76 Tiefer, L. Gonadal hormones and mating behavior in the adult golden hamster. Hormones and Behavior, 1970, 1, 25-37. Tiefer, L., B Johnson, W. A. Female sexual behavior in male golden hamsters. Journal of Endocrinology, 1971, 21, 615-620. Tiefer, L., & Johnson, W. Neonatal androstenedione and adult sexual behavior in golden hamsters, Journal of Comparative and Physio- logical Psychology, 1975, 22, 239-2471 Vreeburg, J. T. M., van der Vaart, P. D. M., G van der Schoot, P. Prevention of central defeminization but not masculinizaton in male rats by inhibition neonatally of estrogen biosynthesis. Journal of Endocrinology, 1977, 24, 375-382. Weisz, J., & Gibbs, C. Metabolism of testosterone in the brain of the newborn female rat after an injection of tritiated testos- terone. Neuroendocrinology, 1974, 14, 72-86. Whalen, R. E., & Etgen, A. M. Masculinization and defeminization induced in female hamsters by neonatal treatment with estradiol benzoate and RU-2858. Hormones and Behavior, 1978, 12, 170-177. Whitsett, J. M., & Vandenbergh, J. G. Influence of testosterone pr0pionate administered neonatally on puberty and bisexual be- havior in female hamsters. Journal of Comparative and Physiologi- cal Poychology, 1975, 22, 248-255. Wilson, J. 0. Sexual differentiation. Annual Review of Physiology, 1978, 42, 279-306. "‘TT‘TTITLTTTTTAAEs