INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIAL AND METHODS RESULTS DISCUSSION REFERENCES

MALE SEXUAL BEHAVIOR IS stimulated by both testosterone and estradiol; however, the mechanism of their interaction is not understood (23). Testosterone appears to act through specific intracellular receptors and induction of protein synthesis. This is supported by studies showing that the androgen receptor antagonist flutamide and protein synthesis inhibitors inhibit male sexual behavior (10, 11). The time course of testosterone's behavioral effects is also consistent with the need for new protein synthesis, requiring several days of testosterone treatment to restore behavior in castrated males (12). Testosterone also acts, in part, through in situ conversion to estradiol by aromatase in the preoptic area (3). However, the mechanism of estradiol's action on male sexual behavior is not as well understood as that of testosterone. Previous studies using antiestrogens have yielded mixed results, with no study clearly demonstrating a role for estrogen receptors in male copulatory behavior (2, 8). Moreover, male sexual behavior is disrupted, but not eliminated, in estrogen receptor knockout mice (16, 21). These data raise the question of whether estradiol facilitates copulatory behavior, in part, through nongenomic mechanisms. There is substantial evidence that estradiol is capable of acting through transcription-independent signaling pathways in brain. Recent studies demonstrate the presence of immunostaining for aromatase in axons and synaptic boutons, suggesting that locally formed estrogen may be active at the synaptic level (15). Other evidence indicates that specific receptor sites for estradiol and other steroid hormones exist in neural membranes (19). Estradiol modulates ion channels and second messenger cascades with a latency (i.e., <30 min) that is generally accepted to be too rapid to involve genomic activation (7, 28). However, up until now, the possibility that steroid hormones exert short-latency, i.e., nontranscriptional, effects on male sexual behavior has not been systematically addressed in any species. Therefore, the purpose of this study was to determine whether testosterone or estradiol is capable of exerting rapid effects on copulatory behavior in sexually experienced castrated male rats.

	MATERIAL AND METHODS

TOP ABSTRACT INTRODUCTION MATERIAL AND METHODS RESULTS DISCUSSION REFERENCES

Animals. Twenty-three male Sprague-Dawley rats (350-450 g) were obtained from Simonsen Laboratories (Gilroy, CA) and housed in pairs in a vivarium under a reversed lighting schedule (12 h light, 12 h dark), with lights off at 0900. Food and water were available ad libitum, and room temperature was maintained at 22°C. To induce behavioral estrus in stimulus females, ovariectomized Fischer 344 female rats (n = 12) were primed with 10 µg estradiol benzoate administered subcutaneously in sesame oil on days 1 and 2, and 0.5 mg progesterone was administered subcutaneously in sesame oil 4 h before the onset of behavioral testing on day 3. Surgeries were performed with rats under ketamine (55 mg/kg im)-xylazine (5 mg/kg im) anesthesia. All procedures used in these experiments adhere to the National Institutes of Health Guide for the Care and Use of Laboratory Animals and were approved by the Institutional Animal Care and Use Committee at the Oregon Health Sciences University.

Sexual behavior tests. Sexual behavior tests were conducted during the dark phase of the lighting cycle under dim red light illumination. Testing was conducted in a semicircular arena (radius 30.5 cm) with cob bedding covering the floor. Each subject was placed alone in a testing arena to acclimate for 5 min. To begin a test, a sexually receptive stimulus female rat was introduced into the arena and the subjects were observed for 20 min. The following measures of male rat sexual behavior were recorded during the testing period that preceded the first ejaculation: genital sniffs or chemoinvestigation, the total number of sniffs and licks of the female genital area by the male; thrustless mounts, the total number of times that the subject approached the stimulus female from the back and clasped her sides with front paws; mounts with thrusts, the total number of times the subject approached the stimulus female from the back, clasped her sides with front paws, and rhythmically moved the pelvis back and forth without intromission; total mounts, the sum of mounts with and without thrusts; mount latency, the time elapsed from when the female was introduced and the first mount; intromissions, the total number of mounts with intromission and without ejaculation; intromission latency, the time from introduction of female to the first intromission; ejaculations, the total number of intromissions with ejaculation; and ejaculation latency, the time elapsed from the first intromission to the first ejaculation.

Experimental design. Males were given 20-min exposures to females on a weekly basis for 5 wk to gain sexual experience and screen for noncopulators. Copulatory performance was measured, and only sexually vigorous males that had achieved ejaculation by the end of this pretest period were used in the experiment. The males were then castrated and isolated from further contact with females. Three weeks later, the sexually experienced castrated males were divided into the following treatment groups (n = 7 or 8 rats per group): group 1, vehicle control; group 2, low-dose steroid hormone; group 3, high-dose steroid hormone. The rats in these groups exhibited equivalent levels of sexual behavior on the final pretest as verified by ANOVA (Table 1). To determine whether testosterone and/or estradiol can exert rapid effects on copulatory behavior, the castrated rats were given sexual behavior tests that were timed to begin 15 min after intraperitoneal injection of hormone or 2-hydroxypropyl--cyclodextrin vehicle (20% wt/vol in saline; Research Biochemical International, Natick, MA). 2-Hydroxypropyl--cyclodextrin is a macro ring structure composed of seven glucopyranase units forming a lipophilic cavity and is used as a solubility-enhancing carrier for steroid hormones (25). All rats were given two tests that were separated by 1 wk. Those assigned to steroid treatments were given testosterone before the first test and estradiol before the second test. Testosterone was administered at doses of 2 and 10 mg/kg. Estradiol was given at doses of 20 and 100 µg/kg. We chose to use this experimental design because it had been demonstrated that testosterone, incorporated into a -cyclodextrin complex, has a half-life of 70 min in plasma and is therefore rapidly cleared within <24 h (18, 25). On the day of testing, the rats received an injection of vehicle or hormone in a volume of 1 µl/g body wt and were returned to their home cage. After 10 min, they were placed in the testing apparatus and allowed to adapt for 5 min. The test was then started with the introduction of a sexually receptive female into the arena.

Hormone measurements. One week after the behavioral tests were completed, the males were randomly reassigned to five groups, which received the following intraperitoneal injections: vehicle control (n = 4), 2 mg/kg testosterone (n = 4), 10 mg/kg testosterone (n = 4), 20 µg/kg estradiol (n = 5), and 100 µg/kg estradiol (n = 5). The rats were decapitated 15 min after injection, and trunk blood was collected. Serum levels of testosterone and estradiol were measured by radioimmunoassay in samples chromatographed on Sephadex LH-20 according to previously published procedures (20). For each steroid measured, all samples were measured in the same assay. The mean percentages of recovery, water blanks, and intra-assay coefficients of variation were, for testosterone, 79.3%, 2.4 pg, and 7.3% and, for estradiol, 67.3%, 0.4 pg, and 12.3%, respectively.

Statistical analysis. The chi-square test was used to compare the proportions of animals mounting in each treatment group. Other behavioral measures were analyzed by one-way analysis of variance, followed by Fisher's protected least squares difference test. When necessary, log₁₀ transformation of the data was used to obtain homogeneous variances before analysis of variance was performed. Differences between groups were considered significant at a level of P < 0.05 (two tailed).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIAL AND METHODS RESULTS DISCUSSION REFERENCES

None of the castrated males intromitted or ejaculated within the two weekly time-limited tests that were administered. On the first week of testing, the percentage of rats mounting with or without thrust after vehicle injection (75%) was not different from the percentage of rats that mounted after injection of 2 mg/kg (100%) or 10 mg/kg (100%) testosterone [²(2) = 2.6, P > 0.05]. At the doses tested, testosterone did not have short-latency effects on the number of genital sniffs or mounts or on the mount latency (Fig. 1, A-C). Estradiol treatment also did not affect the percentage of rats that mounted. On the second week of testing, the percentage of rats mounting after vehicle injection (100%) was not different from the number of rats that mounted after injection of 20 µg/kg (100%) or 100 µg/kg (100%) estradiol. However, estradiol did have significant short-latency effects on chemoinvestigatory and mounting behaviors (Fig. 2, A-C). At a dose of 20 µg/kg, estradiol significantly increased the total number of mounts but not the number of genital sniffs and it had no significant effect on mount latency. At the highest dose of 100 µg/kg, estradiol significantly increased the number of genital sniffs and the total number of mounts and it significantly decreased mount latency.

View larger version (12K): [in this window] [in a new window]   Fig. 1.   Mean (±SE) number of genital sniffs (A) and mounts (B) and mean (±SE) mount latency (C) exhibited by castrated adult male rats treated 15 min before testing with -cyclodextrin vehicle without testosterone (0 mg/kg control), low-dose testosterone (2 mg/kg), or high-dose testosterone (10 mg/kg).

View larger version (13K): [in this window] [in a new window]   Fig. 2.   Mean (±SE) number of genital sniffs (A) and mounts (B) and mean (±SE) mount latency (C) exhibited by castrated adult male rats treated 15 min before testing with -cyclodextrin vehicle without estradiol (0 µg/kg control), low-dose estradiol (20 µg/kg), or high-dose estradiol (100 µg/kg). * P < 0.05 compared with control group (0 µg/kg).

The mean ± SE concentrations of serum testosterone and estradiol in vehicle-injected rats were 310 ± 89 pg/ml (range = 180-592 pg/ml) and 1.8 ± 0.5 pg/ml (range = 1-3 pg/ml), respectively. The mean ± SE concentrations of testosterone in the low- and high-steroid groups were 143.5 ± 34.8 ng/ml (range = 50-218 ng/ml) and 1,246 ± 70.6 ng/ml (range = 1,101-1,433 ng/ml), respectively. The mean ± SE concentrations of estradiol in the low- and high-steroid groups were 1.1 ± 0.16 ng/ml (range = 0.8-1.5 ng/ml) and 8.5 ± 3.0 ng/ml (range = 1.5-15.9 ng/ml), respectively. Normal levels of testosterone and estradiol in males are considered to be ~1-6 ng/ml and 1-4 pg/ml, respectively (1).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIAL AND METHODS RESULTS DISCUSSION REFERENCES

The results of the present study demonstrate that exposure of sexually experienced castrated rats to a single injection of estradiol can rapidly (i.e., in 15-35 min) increase chemoinvestigation and mounting and reduces the latency to mount. Our results are the first to show a short-latency effect of estradiol in male sexual behavior. Previous studies have shown that estrogens stimulate male sexual behavior within 1-3 days (4, 24), but none have examined its acute behavioral effects. It is well established that estradiol acts on the genome via nuclear estrogen receptors. However, the effect of estradiol that we observed on masculine reproductive behaviors occurred within a shorter time frame than that expected for transcriptional activation (27) and, as such, suggests that estradiol is acting through a nongenomic mechanism.

A substantial body of evidence exists that suggests estrogens can exert effects on neuronal activity that are different from the classical genomic (i.e., estrogen response element dependent) mechanism of estrogen receptor action. Estrogens have been shown to alter neuronal activity with latencies of seconds to minutes and to act at the cellular membrane level to hyperpolarize preoptic and hypothalamic neurons (6, 7, 14, 28). These effects are stereospecific and cannot be prevented by protein synthesis inhibitors (14). Estrogen treatment also rapidly (within 15 min) induces the phosphorylation of the cAMP response element protein in the preoptic area and bed nucleus of the stria terminalis, suggesting that nongenomic effects of estrogen may utilize protein kinase-associated signal transduction pathways initiated at the cell membrane (5, 29). Finally, crucial support for a membrane site of estrogen action is provided by the demonstration that estrogens bind with high affinity, limited capacity, and high specificity to neural membrane preparations (19).

The fact that testosterone did not acutely stimulate male sexual behavior in the current study agrees with previous studies that have shown that it takes several days of testosterone treatment to stimulate mounting in castrated male rats and up to 2 wk to restore the complete behavioral repertoire (9, 12). Protein synthesis inhibitors have been found to inhibit male sexual behavior (10), suggesting that ongoing protein synthesis is necessary for the full expression of sexual behaviors in male rats. In addition to its direct androgen effects, testosterone is believed to act after local conversion to estrogen by cytochrome P-450 aromatase in brain (3). However, aromatase is significantly decreased after castration because it is regulated pretranslationally by an androgen receptor-dependent mechanism (22). Thus the time required for testosterone to restore copulatory behavior in castrates may, in part, be needed for the induction of aromatase to levels that are sufficient enough to produce local behaviorally effective concentrations of estradiol.

An important caveat to our interpretation of the data is that under the present experimental design, in which testosterone was tested 1 wk before estradiol, it is possible that the effects we observed were due to long-latency actions of testosterone instead of short-latency actions of estradiol. This interpretation seems unlikely, because testosterone complexed to -cyclodextrin has been shown to be rapidly distributed and cleared from both plasma and brain (17, 25). Indeed, the half-life of the cyclodextrin-testosterone complex in plasma is ~70 min (18), suggesting that, even at the highest dose administered in the present study, testosterone should be cleared in <24 h. McGinnis et al. (12) demonstrated that castrated rats must receive a minimum of 16 h of testosterone exposure per day for treatment effects on mount frequency and latency to be evident after 1 wk. This level of exposure is not sustained by a single injection of -cyclodextrin-complexed testosterone (25). In a previous study that did observe a long-latency behavioral effect (9), testosterone was administered subcutaneously in oil vehicle as a propionated derivative. It is well recognized that the propionated form of testosterone extends the half-life and potency of the hormone by slowing its metabolism (13). The use of an oil vehicle further prolongs steroid hormone action by acting as a depot that slowly releases the hormone into the general circulation (26). In the present study, free testosterone was administered intraperitoneally in aqueous -cyclodextrin and would therefore be rapidly absorbed into blood capillaries and would not accumulate in tissue.

In conclusion, our results are the first to demonstrate that estradiol can exert short-latency effects on sexual activity in castrated male rats. This result suggests that estrogens may affect behavior in males through nontranscriptional mechanisms mediated at the membrane level in addition to genomic mechanisms mediated through nuclear estrogen receptors. Future experiments are planned to further test this hypothesis by examining the time course and specificity of this effect and the effect of estrogen derivatives that cannot cross the cell membrane and determining whether protein synthesis inhibitors alter the short-latency actions of estradiol on male sexual activity.