HORMONES
AND
BEHAVIOR
24,
71-88 (1990)
Estradiol Administration and the Sexual Activity of Castrated Male Rhesus Monkeys (Macaca mulatta) RICHARD P. MICHAEL, Department 30322;
of Psychiatry, and the Georgia
DORIS ZUMPE,
AND ROBERT W. BONSALL
Emory University School of Medicine, Mental Health Institute, 1256 Briarcliff Atlanta, Georgia 30306
Atlanta, Road,
Georgia N.E.,
To examine whether estradiol might be effective in maintaining sexual behavior after castration or after testosterone withdrawal, we have observed male rhesus monkeys during daily I-hr tests alternately with each of two ovariectomized, estradiol-treated females (four males, four females, eight male-female pairs, 798 tests). Estradiol (2-5 pg/kg se/day) or vehicle was administered in counterbalanced order immediately after castration and again immediately after withdrawal of testosterone propionate treatments (800 pg and 1.6 mg se/day). There were no significant differences in behavior during vehicle and estradiol treatments to indicate that estradiol helped to maintain male sexual activity. Instead, estradiol treatment tended to interfere with the capacity to intromit. This supported the results of other studies, namely, that the systemic administration of estradiol does not enhance the sexual behavior of castrated male macaques, and raises questions about the role of both aromatization and estrogen receptors in the male primate brain. 0 1990 Academic press, Inc.
Castration reduces or even abolishes the sexual behavior of many mammals, although the time course and extent of the behavioral deficits vary widely not only between species but also between individuals of the same species. Testosterone (T) administration restores male sexual activity and has therefore long been regarded as the major steroid responsible for male libido and potency. Since T is partly metabolized in brain by aromatization to 17p-estradiol (EJ (Naftolin, Ryan, Davies, Reddy, Flores, Petro, Kuhn, White, Takaoka, and Wolin, 1975) and is partly metabolized in both central and peripheral tissues by reduction to Sa-dihydrotestosterone (DHT) (Bruchovsky and Wilson, 1968; Martini, 1982), studies with castrated rodents and lagomorphs have compared the behavioral effects of all three steroids. With the exception of the male guinea pig (Alsum and Goy, 1974), DHT and dihydrotestosterone propionate (DHTP) were generally less effective than T or testosterone propionate (TP) in maintaining or restoring male sexual activity after castration, although there are species and strain differences (McDonald, 71 0018-506x/90 $1.50 Copyrighl Q 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.
72
MICHAEL,
ZUMPE,
AND
BONSALL
Beyer, Newton, Brien, Baker, Tan, Sampson, Kitching, Greenhill, and Pritchard, 1970; Whalen and Luttge, 1971; Beyer and Rivaud, 1973; Luttge and Hall, 1973; Olsen, 1979; Yahr, 1979; Meisel, O’Hanlon, and Sachs, 1984; Olsen and Whalen, 1984) (see review, Whalen, Yahr, and Luttge, 1985). However, when DHT or DHTP was administered to rodents in conjunction with Ez or estradiol benzoate (EB), sexual activity was often restored to intact levels (Baum and Vreeburg, 1973; Larsson, Sodersten, and Beyer, 1973; Feder, Naftolin, and Ryan, 1974; Beyer, de la Torre, Larsson, and Perez-Palacios, 1975; DeBold and Clemens, 1978; Baum, Tobet, Starr, and Bradshaw, 1982). Indeed, the systemic administration of estradiol alone to castrated male rats is sufficient to activate mounting behavior (Beach, 1942; Davidson, 1969; Pfaff, 1970; Sodersten, Eneroth, Hansson, Mode, Johansson, Naslund, Liang, and Gustafsson, 1986; McGinnis and Dreifuss, 1989). In macaques, as in other mammals, castration reduces or even abolishes male sexual activity (Michael, Wilson, and Plant, 1973; Phoenix, Slob, and Goy, 1973; Michael and Wilson, 1974), but there is much variation in the magnitude and timing of effects in different individuals. Doses of TP producing plasma hormone levels in the normal range for intact males were significantly more effective than those of DHTP in restoring the ejaculatory activity of castrated male cynomolgus monkeys (Michael, Zumpe, and Bonsall, 1986). However, doses of DHTP in the pharmacological range also increased the sexual activity of both castrated rhesus (Phoenix, 1974) and castrated cynomolgus monkeys (Michael, Bonsall, and Zumpe, 1987). It has been shown that 3H-T administered systemically to castrated rhesus and cynomolgus males is taken up by neuronal nuclei in hypothalamus, preoptic area, and amygdala, and the major form of this radioactivity is 3H-E, (Bonsall, Rees, and Michael, 1983; Michael, Bonsall, and Zumpe, 1987; Bonsall, Rees, and Michael, 1989). This raises the question of whether testosterone in monkeys, as in rodents, might exert some of its behavioral effects via aromatization to estradiol in certain critical brain regions. However, injections of estradiol benzoate did not increase the ejaculatory activity of either longcastrated rhesus monkeys (Phoenix and Chambers, 1982) or castrated cynomolgus monkeys given either DHTP or TP (Michael et al., 1987). In this regard, the primate and the rodent clearly differed, although differences in dose rates make comparisons somewhat difficult. None of these studies has tested the possibility that estradiol might play a role in maintaining sexual activity in gonadectomized male macaques, namely, when gonadotropin. secretion and steroid receptors have been little affected by diminished plasma testosterone levels. We now report on the sexual activity of male rhesus monkeys given daily SC injections of either EB or vehicle immediately after either castration or androgen withdrawal, namely, when somatic changes due to androgen withdrawal
ESTROGEN
AND
MALE
MACAQUE
BEHAVIOR
73
(e.g., erectile failure) and to estrogen administration (e.g., penile edema) have not developed sufficiently to disrupt intromission and ejaculation. METHODS
Animals Four intact male (weighing 7.7-10.4 kg) and four ovariectomized female (weighing 4.7-6.8 kg) rhesus monkeys were housed in individual cages in a colony room where artificial lighting maintained a constant 14-hr day between 0615 and 2015 hr. Temperature was maintained between 20 and 24°C. Food consisted of commercial chow supplemented with vitamins, fresh fruit, and vegetables, and water was available ad lib. All maintenance and surgical and experimental procedures were as previously described and were in accordance with institutional regulations and with the NIH Guide for the Care and Use of Laboratory Animals (NIH Publication No. 85-23, revised 1985). Surgery Animals were deeply anesthetized and, using sterile procedures, castration was performed on all males via bilateral 2.5-cm scrotal incisions. The testes were replaced with sterile Silastic testicular prostheses to maintain the appearance and stimulus qualities of the scrotum. Bilateral ovariectomy was performed on all females via 5-cm mid-line, subumbilical incisions using standard procedures. Hormone Treatments After 2 months of testing to familiarize partners and to stabilize behavior, males 1 and 2, and males 3 and 4, were matched on the basis of body weight, levels of sexual activity, and the anticipated response to castration. After 2 weeks of data collection while intact, all males were castrated within a 2-day period and received, immediately postoperatively, the first daily SC injection of either estradiol benzoate in 0.2 ml oil or 0.2 ml oil vehicle. Males 1 and 2 were treated successively with (1) estradiol benzoate, 2 pg/kg body wt for 4 weeks, (2) testosterone propionate, 800 pg/animal for 4 weeks, (3) vehicle for 6 weeks, (4) testosterone propionate, 1.6 mg/animal for 8 weeks, (5) vehicle for 4 weeks, (6) testosterone propionate, 1.6 mg/animal for 8 weeks, and (7) estradiol benzoate 2 pg/kg body wt for 4 weeks. For males 3 and 4, estradiol and vehicle treatments were reversed to produce the following treatment order: (1) vehicle for 4 weeks, (2) testosterone propionate, 800 pg/animal for 4 weeks, (3) estradiol benzoate, 2 pg/kg body wt for 4 weeks, (4) estradiol benzoate, 5 pg/kg body wt for 2 weeks, (5) testosterone propionate, 1.6 mg/animal for 8 weeks, (6) estradiol benzoate, 2 pg/kg body wt for 4 weeks, (7) testosterone propionate, 1.6 mg/animal
74
MICHAEL,
ZUMPE,
AND
BONSALL
for 8 weeks, and (8) vehicle for 4 weeks. This counterbalanced design both during the first part (EB immediately after castration or 800 pg TP) and during the second part (EB after 1.6 mg TP) helped control for treatment order, season, and the effects of long-term testing. The 800 pg dose of TP was chosen because, 16 hr after injection, it produced plasma testosterone levels within the range of those of intact males (Michael et al., 1984). The 2 pug/kg EB dose was selected because previous experience had shown that higher doses produced edema of the perineum, scrotum, and penis. To simulate the diurnal plasma testosterone rhythm, injections were given at 1600 hr, and the first injection of each new treatment was given on the afternoon (Friday) after the last behavioral test of the previous treatment. Thus, three injections were given before the first behavioral test of a new treatment period (Monday). Females (A-D) received daily SC injections of estradiol benzoate, 10 pg/female, throughout the entire study. Plasma
Samples
and Hormone
Assays
Samples of 3 ml blood were obtained once per week at 0800 hr (16 hr after injection) from each of the four untranquilized males that had been adapted previously to the procedure, and alternate samples were assayed for both testosterone and estradiol. Plasma testosterone levels were measured by a coated-tube radioimmunoassay method using a highly specific antiserum that obviated the need for extraction (Diagnostic Products Corporation). The only significant cross-reactivity was with Sa-dihydrotestosterone (3.3%). The sensitivity of the assay (blank + 2 SD) was 40 ng/lOO ml, the intraassay coefficient of variation was 6.7%, and the interassay coefficient was 10.4%. Plasma estradiol levels were measured in ether extracts of plasma by radioimmunoassay using an antiserum provided by Dr. D. C. Collins (Wright, Collins, & Preedy, 1978). The sensitivity of the assay was 29.3 pg/ml, the intraassay coefficient of variation was 7.3%, and the interassay coefficient was 12.1%. Behavioral
Testing
Observations were made on oppositely sexed pairs of animals in quiet, isolated rooms from behind one-way vision mirrors in special observation cages 1.19 m wide by 1.07 m deep by 1.14 m high into which first the male and then the female was introduced at the start of each test session. Observers were aware that males had been castrated but were unaware of the contents of the daily injections which were number coded. Tests were of 60 min duration and were conducted 5 days a week, Mondays through Fridays. On consecutive test days, males 1 and 3 were paired with females A and B in turn and males 2 and 4 were paired with females C and D in turn in a crossover pairing design. Thus, each of the eight pairs was tested five times every 2 weeks. All four tests on a given day
ESTROGEN
AND
MALE
MACAQUE
were conducted simultaneously between husbandry reasons, two tests had to be and female B and another involved male results were therefore based on 798 I-hr four males and four females. Dejinitions
BEHAVIOR
75
0900 and 1200 hr. For animal canceled: one involved male 3 2 and female D. The numerical tests with eight pairs involving
and Terminology
We report here on the following behavioral measures: (a) number of ejaculations per test (potency)-the characteristic behavioral response, typical of ejaculation in intact males, performed in coitu while mounting the female; (b) time to first ejaculation-time in seconds from the start of the test to the first ejaculation (a default value of 3600 was given where ejaculation did not occur); (c) number of male mounting attempts per test-sum of the number of successful attempts (male-initiated mounts) and unsuccessful attempts (female refusals) by the male to initiate a mount; (d) latency to the first male mounting attempt-time in seconds from the start of the test to the first mounting attempt of the test (a default value of 3600 was given where no mounting attempts occurred); (e) percentage intromitted thrusts-number of intromitted thrusts expressed as a percentage of the total number of thrusts per test; (f) number of male yawns per test; (g) male grooming time-time in seconds spent by the male in grooming the female; (h) female grooming time-time in seconds spent by the female in grooming the male. Numerical
Treatment
of Results
and Statistical
Analysis
To study the time course of the behavioral changes, all data were analyzed in successive 2-week treatment periods. Whenever possible, the estradiol and vehicle treatments of males 1 and 2 were counterbalanced with those of males 3 and 4 to control for any effects of longterm testing and of time since castration. However, it was not possible to maintain complete experimental symmetry during the first part of the study because (a) immediately after castration, males 1 and 2 received EB while males 3 and 4 received vehicle (castration could not be repeated), and (b) males 3 and 4 received EB at a dose of 5 pg/kg not given males 1 and 2 because this dose produced penile edema. Therefore, the combined data of males 1 and 2 were analyzed separately from those of males 3 and 4. However, data from the final 12 treatment periods involving the higher TP dose could in addition be analyzed for all four males combined. Data from all TP treatment periods preceding EB in the final 12 treatment periods were combined, and those preceding vehicle were combined separately. Each of the two sets of combined TP data were balanced because each comprised tests following EB treatment in one pair of males and vehicle treatment in the other pair. Results were subjected to two-way analyses of variance for repeated measures
76
MICHAEL,
ZUMPE,
AND
BONSALL
(SPSS/PC + , SPSS Inc., Chicago, IL) using a treatment period x malefemale pair design or a hormone treatment x male design, and the Scheffe test was applied to all posr hoc comparisons between treatment means using a value of P < 0.1 (equivalent to P < 0.05 with other post hoc tests) to indicate a significant difference. RESULTS Effects of Administering Estradiol to Castrated Males Immediately after Withdrawing 1.6 mg TP/Day There were highly significant treatment effects for all measures of behavior in the combined data for all four males [F( 11, 384) = 6.4 21.2, P < O.OOOl]. However, the majority of significant differences between treatment periods was during the administration and withdrawal of TP, but it was the difference between EB and vehicle treatment periods that is the critical comparison here. Figure 1 shows the changes in two measures of copulatory activity, numbers of ejaculations per test (upper) and latencies to the first ejaculation (lower), in contiguous 2-week periods during the administration of TP and EB and of TP and vehicle. Copulatory activity increased and decreased significantly with TP treatment and withdrawal. There were no significant differences (Scheffe, P > 0.1) between EB and vehicle treatments either for the first 2-week periods (Fig. 1, A columns) or for the second 2-week periods (Fig. 1, B columns). Ejaculations were significantly higher during TP treatments that preceded vehicle administration compared with those preceding EB administration, but this could not be attributed to an experimental intervention (see Table 1). Figure 2 shows changes in two behavioral measures often considered to reflect male sexual motivation, male mounting attempts (upper) and latencies to the first mounting attempt (lower), during the administration of TP, EB, and vehicle. In general, these two indices showed changes similar to those in ejaculations and ejaculatory latencies. Mounting attempts increased and mounting attempt latencies decreased with TP treatments while TP withdrawal reversed these effects. Again, there were no significant differences (P > 0.1) between EB and vehicle treatments during either the first (A) or second (B) 2-week periods. Two additional behavioral measures (of the many available) are also given attention here. The percentage of intromitted thrusts (Fig. 3, upper) is sometimes used as an index of erectile capacity, and the number of yawns (Fig. 3, lower) is a good indicator of an androgen effect. The percentage of intromitted thrusts paradoxically increased after TP withdrawal during the first 2-week period of both EB and vehicle treatments, but this was not statistically significant in either case. There were no significant differences (P > 0.1) between EB and vehicle treatments
ESTROGEN
AND MALE
MACAQUE
EFFECT OF 2 /q/kg EB AFTER TP WITHDRAWAL
1.6 m g TP
2 pg/kg
77
BEHAVIOR
EFFECT OF VEHICLE AFTER TP WITHDRAWAL
EB
1.6 m g TP
J
Vehicle
FIG. 1. Changes in numbers of ejaculations (upper) and in ejaculatory latencies (lower) by castrated male rhesus monkeys in successive f-week periods of treatment with estradiol benzoate (EB) or vehicle after withdrawal of testosterone propionate (TP). There were no significant differences between EB and vehicle treatments during either the first (A) or the second (B) 2-week treatment periods. Each column gives the mean + SEM of 40 I-hr behavior tests (four males, eight pairs) in this and subsequent figures.
during either the first (A) or second (B) 2-week periods. However, the percentage of intromitted thrusts declined significantly from the first to the second EB treatment period (P < 0.05) and this did not occur with vehicle. Male yawns were abruptly and significantly reduced during both EB (fourth TP period to second EB period, P < 0.01) and vehicle treatments (third TP period to second vehicle period, P < O.l), particularly in males 3 and 4 (Table 1). Effects of Administering Estradiol Immediately after Castration or Immediately after Withdrawing 800 pg TP/Day
Because the effects of castration in the four pairs involving males 1 and 2 were so dramatically different from those in the four pairs involving males 3 and 4, the numerical data for each pair of males were analyzed separately. Of the 32 possible comparisons between EB and vehicle
P
F(11, 192) =
2 e/kg EB 2 I.cg/kg EB
TP TP TP TP
2 0.11 rt 0.08 2 0.09 k 0.12 2 0.15 5 0.11 +- 0.11 2 0.10 f 0.11 2 0.10 f 0.12 + 0.11 6.0
AND
BEHAVIOR
24,
71-88 (1990)
Estradiol Administration and the Sexual Activity of Castrated Male Rhesus Monkeys (Macaca mulatta) RICHARD P. MICHAEL, Department 30322;
of Psychiatry, and the Georgia
DORIS ZUMPE,
AND ROBERT W. BONSALL
Emory University School of Medicine, Mental Health Institute, 1256 Briarcliff Atlanta, Georgia 30306
Atlanta, Road,
Georgia N.E.,
To examine whether estradiol might be effective in maintaining sexual behavior after castration or after testosterone withdrawal, we have observed male rhesus monkeys during daily I-hr tests alternately with each of two ovariectomized, estradiol-treated females (four males, four females, eight male-female pairs, 798 tests). Estradiol (2-5 pg/kg se/day) or vehicle was administered in counterbalanced order immediately after castration and again immediately after withdrawal of testosterone propionate treatments (800 pg and 1.6 mg se/day). There were no significant differences in behavior during vehicle and estradiol treatments to indicate that estradiol helped to maintain male sexual activity. Instead, estradiol treatment tended to interfere with the capacity to intromit. This supported the results of other studies, namely, that the systemic administration of estradiol does not enhance the sexual behavior of castrated male macaques, and raises questions about the role of both aromatization and estrogen receptors in the male primate brain. 0 1990 Academic press, Inc.
Castration reduces or even abolishes the sexual behavior of many mammals, although the time course and extent of the behavioral deficits vary widely not only between species but also between individuals of the same species. Testosterone (T) administration restores male sexual activity and has therefore long been regarded as the major steroid responsible for male libido and potency. Since T is partly metabolized in brain by aromatization to 17p-estradiol (EJ (Naftolin, Ryan, Davies, Reddy, Flores, Petro, Kuhn, White, Takaoka, and Wolin, 1975) and is partly metabolized in both central and peripheral tissues by reduction to Sa-dihydrotestosterone (DHT) (Bruchovsky and Wilson, 1968; Martini, 1982), studies with castrated rodents and lagomorphs have compared the behavioral effects of all three steroids. With the exception of the male guinea pig (Alsum and Goy, 1974), DHT and dihydrotestosterone propionate (DHTP) were generally less effective than T or testosterone propionate (TP) in maintaining or restoring male sexual activity after castration, although there are species and strain differences (McDonald, 71 0018-506x/90 $1.50 Copyrighl Q 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.
72
MICHAEL,
ZUMPE,
AND
BONSALL
Beyer, Newton, Brien, Baker, Tan, Sampson, Kitching, Greenhill, and Pritchard, 1970; Whalen and Luttge, 1971; Beyer and Rivaud, 1973; Luttge and Hall, 1973; Olsen, 1979; Yahr, 1979; Meisel, O’Hanlon, and Sachs, 1984; Olsen and Whalen, 1984) (see review, Whalen, Yahr, and Luttge, 1985). However, when DHT or DHTP was administered to rodents in conjunction with Ez or estradiol benzoate (EB), sexual activity was often restored to intact levels (Baum and Vreeburg, 1973; Larsson, Sodersten, and Beyer, 1973; Feder, Naftolin, and Ryan, 1974; Beyer, de la Torre, Larsson, and Perez-Palacios, 1975; DeBold and Clemens, 1978; Baum, Tobet, Starr, and Bradshaw, 1982). Indeed, the systemic administration of estradiol alone to castrated male rats is sufficient to activate mounting behavior (Beach, 1942; Davidson, 1969; Pfaff, 1970; Sodersten, Eneroth, Hansson, Mode, Johansson, Naslund, Liang, and Gustafsson, 1986; McGinnis and Dreifuss, 1989). In macaques, as in other mammals, castration reduces or even abolishes male sexual activity (Michael, Wilson, and Plant, 1973; Phoenix, Slob, and Goy, 1973; Michael and Wilson, 1974), but there is much variation in the magnitude and timing of effects in different individuals. Doses of TP producing plasma hormone levels in the normal range for intact males were significantly more effective than those of DHTP in restoring the ejaculatory activity of castrated male cynomolgus monkeys (Michael, Zumpe, and Bonsall, 1986). However, doses of DHTP in the pharmacological range also increased the sexual activity of both castrated rhesus (Phoenix, 1974) and castrated cynomolgus monkeys (Michael, Bonsall, and Zumpe, 1987). It has been shown that 3H-T administered systemically to castrated rhesus and cynomolgus males is taken up by neuronal nuclei in hypothalamus, preoptic area, and amygdala, and the major form of this radioactivity is 3H-E, (Bonsall, Rees, and Michael, 1983; Michael, Bonsall, and Zumpe, 1987; Bonsall, Rees, and Michael, 1989). This raises the question of whether testosterone in monkeys, as in rodents, might exert some of its behavioral effects via aromatization to estradiol in certain critical brain regions. However, injections of estradiol benzoate did not increase the ejaculatory activity of either longcastrated rhesus monkeys (Phoenix and Chambers, 1982) or castrated cynomolgus monkeys given either DHTP or TP (Michael et al., 1987). In this regard, the primate and the rodent clearly differed, although differences in dose rates make comparisons somewhat difficult. None of these studies has tested the possibility that estradiol might play a role in maintaining sexual activity in gonadectomized male macaques, namely, when gonadotropin. secretion and steroid receptors have been little affected by diminished plasma testosterone levels. We now report on the sexual activity of male rhesus monkeys given daily SC injections of either EB or vehicle immediately after either castration or androgen withdrawal, namely, when somatic changes due to androgen withdrawal
ESTROGEN
AND
MALE
MACAQUE
BEHAVIOR
73
(e.g., erectile failure) and to estrogen administration (e.g., penile edema) have not developed sufficiently to disrupt intromission and ejaculation. METHODS
Animals Four intact male (weighing 7.7-10.4 kg) and four ovariectomized female (weighing 4.7-6.8 kg) rhesus monkeys were housed in individual cages in a colony room where artificial lighting maintained a constant 14-hr day between 0615 and 2015 hr. Temperature was maintained between 20 and 24°C. Food consisted of commercial chow supplemented with vitamins, fresh fruit, and vegetables, and water was available ad lib. All maintenance and surgical and experimental procedures were as previously described and were in accordance with institutional regulations and with the NIH Guide for the Care and Use of Laboratory Animals (NIH Publication No. 85-23, revised 1985). Surgery Animals were deeply anesthetized and, using sterile procedures, castration was performed on all males via bilateral 2.5-cm scrotal incisions. The testes were replaced with sterile Silastic testicular prostheses to maintain the appearance and stimulus qualities of the scrotum. Bilateral ovariectomy was performed on all females via 5-cm mid-line, subumbilical incisions using standard procedures. Hormone Treatments After 2 months of testing to familiarize partners and to stabilize behavior, males 1 and 2, and males 3 and 4, were matched on the basis of body weight, levels of sexual activity, and the anticipated response to castration. After 2 weeks of data collection while intact, all males were castrated within a 2-day period and received, immediately postoperatively, the first daily SC injection of either estradiol benzoate in 0.2 ml oil or 0.2 ml oil vehicle. Males 1 and 2 were treated successively with (1) estradiol benzoate, 2 pg/kg body wt for 4 weeks, (2) testosterone propionate, 800 pg/animal for 4 weeks, (3) vehicle for 6 weeks, (4) testosterone propionate, 1.6 mg/animal for 8 weeks, (5) vehicle for 4 weeks, (6) testosterone propionate, 1.6 mg/animal for 8 weeks, and (7) estradiol benzoate 2 pg/kg body wt for 4 weeks. For males 3 and 4, estradiol and vehicle treatments were reversed to produce the following treatment order: (1) vehicle for 4 weeks, (2) testosterone propionate, 800 pg/animal for 4 weeks, (3) estradiol benzoate, 2 pg/kg body wt for 4 weeks, (4) estradiol benzoate, 5 pg/kg body wt for 2 weeks, (5) testosterone propionate, 1.6 mg/animal for 8 weeks, (6) estradiol benzoate, 2 pg/kg body wt for 4 weeks, (7) testosterone propionate, 1.6 mg/animal
74
MICHAEL,
ZUMPE,
AND
BONSALL
for 8 weeks, and (8) vehicle for 4 weeks. This counterbalanced design both during the first part (EB immediately after castration or 800 pg TP) and during the second part (EB after 1.6 mg TP) helped control for treatment order, season, and the effects of long-term testing. The 800 pg dose of TP was chosen because, 16 hr after injection, it produced plasma testosterone levels within the range of those of intact males (Michael et al., 1984). The 2 pug/kg EB dose was selected because previous experience had shown that higher doses produced edema of the perineum, scrotum, and penis. To simulate the diurnal plasma testosterone rhythm, injections were given at 1600 hr, and the first injection of each new treatment was given on the afternoon (Friday) after the last behavioral test of the previous treatment. Thus, three injections were given before the first behavioral test of a new treatment period (Monday). Females (A-D) received daily SC injections of estradiol benzoate, 10 pg/female, throughout the entire study. Plasma
Samples
and Hormone
Assays
Samples of 3 ml blood were obtained once per week at 0800 hr (16 hr after injection) from each of the four untranquilized males that had been adapted previously to the procedure, and alternate samples were assayed for both testosterone and estradiol. Plasma testosterone levels were measured by a coated-tube radioimmunoassay method using a highly specific antiserum that obviated the need for extraction (Diagnostic Products Corporation). The only significant cross-reactivity was with Sa-dihydrotestosterone (3.3%). The sensitivity of the assay (blank + 2 SD) was 40 ng/lOO ml, the intraassay coefficient of variation was 6.7%, and the interassay coefficient was 10.4%. Plasma estradiol levels were measured in ether extracts of plasma by radioimmunoassay using an antiserum provided by Dr. D. C. Collins (Wright, Collins, & Preedy, 1978). The sensitivity of the assay was 29.3 pg/ml, the intraassay coefficient of variation was 7.3%, and the interassay coefficient was 12.1%. Behavioral
Testing
Observations were made on oppositely sexed pairs of animals in quiet, isolated rooms from behind one-way vision mirrors in special observation cages 1.19 m wide by 1.07 m deep by 1.14 m high into which first the male and then the female was introduced at the start of each test session. Observers were aware that males had been castrated but were unaware of the contents of the daily injections which were number coded. Tests were of 60 min duration and were conducted 5 days a week, Mondays through Fridays. On consecutive test days, males 1 and 3 were paired with females A and B in turn and males 2 and 4 were paired with females C and D in turn in a crossover pairing design. Thus, each of the eight pairs was tested five times every 2 weeks. All four tests on a given day
ESTROGEN
AND
MALE
MACAQUE
were conducted simultaneously between husbandry reasons, two tests had to be and female B and another involved male results were therefore based on 798 I-hr four males and four females. Dejinitions
BEHAVIOR
75
0900 and 1200 hr. For animal canceled: one involved male 3 2 and female D. The numerical tests with eight pairs involving
and Terminology
We report here on the following behavioral measures: (a) number of ejaculations per test (potency)-the characteristic behavioral response, typical of ejaculation in intact males, performed in coitu while mounting the female; (b) time to first ejaculation-time in seconds from the start of the test to the first ejaculation (a default value of 3600 was given where ejaculation did not occur); (c) number of male mounting attempts per test-sum of the number of successful attempts (male-initiated mounts) and unsuccessful attempts (female refusals) by the male to initiate a mount; (d) latency to the first male mounting attempt-time in seconds from the start of the test to the first mounting attempt of the test (a default value of 3600 was given where no mounting attempts occurred); (e) percentage intromitted thrusts-number of intromitted thrusts expressed as a percentage of the total number of thrusts per test; (f) number of male yawns per test; (g) male grooming time-time in seconds spent by the male in grooming the female; (h) female grooming time-time in seconds spent by the female in grooming the male. Numerical
Treatment
of Results
and Statistical
Analysis
To study the time course of the behavioral changes, all data were analyzed in successive 2-week treatment periods. Whenever possible, the estradiol and vehicle treatments of males 1 and 2 were counterbalanced with those of males 3 and 4 to control for any effects of longterm testing and of time since castration. However, it was not possible to maintain complete experimental symmetry during the first part of the study because (a) immediately after castration, males 1 and 2 received EB while males 3 and 4 received vehicle (castration could not be repeated), and (b) males 3 and 4 received EB at a dose of 5 pg/kg not given males 1 and 2 because this dose produced penile edema. Therefore, the combined data of males 1 and 2 were analyzed separately from those of males 3 and 4. However, data from the final 12 treatment periods involving the higher TP dose could in addition be analyzed for all four males combined. Data from all TP treatment periods preceding EB in the final 12 treatment periods were combined, and those preceding vehicle were combined separately. Each of the two sets of combined TP data were balanced because each comprised tests following EB treatment in one pair of males and vehicle treatment in the other pair. Results were subjected to two-way analyses of variance for repeated measures
76
MICHAEL,
ZUMPE,
AND
BONSALL
(SPSS/PC + , SPSS Inc., Chicago, IL) using a treatment period x malefemale pair design or a hormone treatment x male design, and the Scheffe test was applied to all posr hoc comparisons between treatment means using a value of P < 0.1 (equivalent to P < 0.05 with other post hoc tests) to indicate a significant difference. RESULTS Effects of Administering Estradiol to Castrated Males Immediately after Withdrawing 1.6 mg TP/Day There were highly significant treatment effects for all measures of behavior in the combined data for all four males [F( 11, 384) = 6.4 21.2, P < O.OOOl]. However, the majority of significant differences between treatment periods was during the administration and withdrawal of TP, but it was the difference between EB and vehicle treatment periods that is the critical comparison here. Figure 1 shows the changes in two measures of copulatory activity, numbers of ejaculations per test (upper) and latencies to the first ejaculation (lower), in contiguous 2-week periods during the administration of TP and EB and of TP and vehicle. Copulatory activity increased and decreased significantly with TP treatment and withdrawal. There were no significant differences (Scheffe, P > 0.1) between EB and vehicle treatments either for the first 2-week periods (Fig. 1, A columns) or for the second 2-week periods (Fig. 1, B columns). Ejaculations were significantly higher during TP treatments that preceded vehicle administration compared with those preceding EB administration, but this could not be attributed to an experimental intervention (see Table 1). Figure 2 shows changes in two behavioral measures often considered to reflect male sexual motivation, male mounting attempts (upper) and latencies to the first mounting attempt (lower), during the administration of TP, EB, and vehicle. In general, these two indices showed changes similar to those in ejaculations and ejaculatory latencies. Mounting attempts increased and mounting attempt latencies decreased with TP treatments while TP withdrawal reversed these effects. Again, there were no significant differences (P > 0.1) between EB and vehicle treatments during either the first (A) or second (B) 2-week periods. Two additional behavioral measures (of the many available) are also given attention here. The percentage of intromitted thrusts (Fig. 3, upper) is sometimes used as an index of erectile capacity, and the number of yawns (Fig. 3, lower) is a good indicator of an androgen effect. The percentage of intromitted thrusts paradoxically increased after TP withdrawal during the first 2-week period of both EB and vehicle treatments, but this was not statistically significant in either case. There were no significant differences (P > 0.1) between EB and vehicle treatments
ESTROGEN
AND MALE
MACAQUE
EFFECT OF 2 /q/kg EB AFTER TP WITHDRAWAL
1.6 m g TP
2 pg/kg
77
BEHAVIOR
EFFECT OF VEHICLE AFTER TP WITHDRAWAL
EB
1.6 m g TP
J
Vehicle
FIG. 1. Changes in numbers of ejaculations (upper) and in ejaculatory latencies (lower) by castrated male rhesus monkeys in successive f-week periods of treatment with estradiol benzoate (EB) or vehicle after withdrawal of testosterone propionate (TP). There were no significant differences between EB and vehicle treatments during either the first (A) or the second (B) 2-week treatment periods. Each column gives the mean + SEM of 40 I-hr behavior tests (four males, eight pairs) in this and subsequent figures.
during either the first (A) or second (B) 2-week periods. However, the percentage of intromitted thrusts declined significantly from the first to the second EB treatment period (P < 0.05) and this did not occur with vehicle. Male yawns were abruptly and significantly reduced during both EB (fourth TP period to second EB period, P < 0.01) and vehicle treatments (third TP period to second vehicle period, P < O.l), particularly in males 3 and 4 (Table 1). Effects of Administering Estradiol Immediately after Castration or Immediately after Withdrawing 800 pg TP/Day
Because the effects of castration in the four pairs involving males 1 and 2 were so dramatically different from those in the four pairs involving males 3 and 4, the numerical data for each pair of males were analyzed separately. Of the 32 possible comparisons between EB and vehicle
P
F(11, 192) =
2 e/kg EB 2 I.cg/kg EB
TP TP TP TP
2 0.11 rt 0.08 2 0.09 k 0.12 2 0.15 5 0.11 +- 0.11 2 0.10 f 0.11 2 0.10 f 0.12 + 0.11 6.0
