Reproductive and behavioral functions of progesterone receptors (PRs) in males were assessed by examining consequences of PR gene deletion. Basal hormone levels were measured in male progesterone receptor knockout (PRKO) mice and compared to wild-type (WT) counterparts. RIA of serum LH, testosterone, and progesterone levels revealed no significant differences. Levels of FSH were moderately but significantly lower and inhibin levels were higher in PRKOs; these differences were not accompanied by gross differences in testicular weight or morphology. PRKOs exhibited significant alterations in sexual behavior. In initial tests PRKOs exhibited reduced latency to mount, compared with WT. In second sessions, PRKOs again showed a significantly reduced latency to mount and increased likelihood of achieving ejaculation. RU486 treatment in WT produced increased mount and intromission frequency and decreased latency to intromission. In anxiety-related behavior tests, PRKO mice exhibited intermediate anxiety levels, compared with WT, suggesting that enhanced sexual behavior in PRKOs is not secondary to reduced anxiety. Immunohistochemical analysis revealed significantly enhanced androgen receptor expression in the medial preoptic nucleus and bed nucleus of the stria terminalis of PRKO. We conclude that testicular development and function and homeostatic regulation of the hypothalamic-pituitary testicular axis are altered to a lesser extent by PR gene deletion. In contrast, PR appears to play a substantial role in inhibiting the anticipatory/motivational components of male sexual behavior in the mouse. The biological significance of this inhibitory mechanism and the extent to which it is mediated by reduced androgen receptor expression remain to be clarified.
THE PHYSIOLOGICAL ROLES of intracellular progesterone receptors (PRs) have been studied intensively in female mammals, yet in male mammals their functions remain unclear. Levels of PR expression are only slightly lower in males than females (1), and their distributions throughout target tissues such as the brain and pituitary gland are similar in the two sexes (2–4). Treatment with estrogen, moreover, induces expression of PRs to a similar extent in males and females (1, 5, 6). Serum progesterone (P) levels are appreciable in male animals, albeit at the low end of the physiological range of values observed in females (7). The presence of both ligand and receptor strongly suggest a physiological role for progesterone and PRs in male reproductive physiology and/or behavior, yet few studies have directly assessed this possibility.
The reproductive functions of the intracellular PRs have previously been assessed in female mice bearing deletions of the PR gene. Deletion of both PR isoforms, PRB and the N terminally truncated PRA, renders female mice infertile with ovarian deficits that prevent ovulation (8, 9) and neuroendocrine deficiencies that preclude release of preovulatory gonadotropin surges (10, 11). These mice also exhibit impaired mammary gland and uterine development (9, 12) and improved glucose homeostasis (13). Induction of lordosis behavior by P and neurotransmitter stimuli is also compromised in these animals (9, 14, 15). Taken together, these studies reveal that PR expression and activation are integral to female fertility at a variety of physiological levels.
Far less is known of the reproductive consequences of PR disruption or blockade in male animals. Previous studies have provided strong circumstantial evidence that P and its receptor have an important role in the neural and behavioral development of males. Male brains are thought to be substantially more susceptible to the effects of maternal P during development as a result of higher PR immunoreactivity in the medial preoptic area and other highly sexually dimorphic brain structures (16). Progestins given during pregnancy have been linked to altered sexual behavior in the adult offspring (17), and blockade of PR in infancy interferes with the normal development of the hypothalamo-pituitary-gonadal axis (18) and sexual behavior (19). P has antiandrogenic effects in a variety of species (20, 21), and in the guinea pig it has been demonstrated to inhibit androgen-dependent sexual behavior (22, 23). Whereas it has been demonstrated that P administration in specific contexts can affect male reproductive behavior, it remains largely unknown whether PR gene expression and PR activation is required for the normal onset and regulation of adult male sexual behavior.
Using male progesterone receptor knockout (PRKO) mice, the present studies examined the effects of PR gene deletion on the activity of the hypothalamic-pituitary testicular axis as well as on male sexual behavior. We recently found that male PRKO mice exhibit markedly decreased infant-directed aggression and increased paternal responsiveness, implicating PR activation in the inhibition of parental behavior and enhancement of male aggression toward young (24). These results prompted us to test the hypothesis that PRKO mice may additionally exhibit decreased sexual motivation. Our hypothesis was predicated on the observations in a number of species that adaptive trade-offs can occur between sexual and paternal behaviors so as to maximize overall reproductive success. Accordingly, we predicted that the highly paternal PRKO mice would exhibit reduced levels of sexual behavior. Surprisingly, our results indicate that the male PRKO mice exhibit increased, rather than decreased levels of sexual behavior. Moreover, a similar increase, rather than a decrease in male sexual behavior can be induced by pharmacological blockade of PRs in wild-type mice. The results of this study suggest an inhibitory role for PR in male reproductive behaviors.
Materials and Methods
Animals
All surgical and behavioral experimental procedures were conducted in accordance with the policies of Northwestern University’s Animal Care and Use Committee. Wild-type (WT) (C57BL/6; Charles River Laboratories, Inc., Wilmington, MA) and PRKO (C57/BL/6X 129SvEv) mice were group housed under a 14:10 (0500–1900 h) light cycle (except for behavioral testing; see below) and fed lab chow ad libitum. 129SvEv (Charles River Laboratories) males were received at 3 wk of age and singly caged until sexual maturity and subsequent behavioral testing. In some tests we included males that were the same combination of strains present in the PRKOs (C57BL/6 and 129SvEv) but express wild-type PR [denoted as isogenic (ISO) control]. ISO mice were generated by mating heterozygous littermates. No more than two litters from each pair were used to control for any possible changes in allelic composition from the PRKO breeding colony. PRKO mice were genotyped at age 3 wk. DNA was isolated from digestion of tail tissue and subjected to phenol/chloroform extraction followed by ethanol precipitation. PCR primer sequences were: P1, 5′-TAG ACA GTG TCT TAG ACT CGT TGT TG-3′; P3, 5′-GAT GGG CAC ATG GAT GAA ATC-3′; and N2, 5′-GCA TGC TCC AGA CTG CCT TGG GAA A-3′.
Hormone measurements
Animals were deeply anesthetized via methoxyflurane inhalation, and 2 cm horizontal incisions were made with sharp scissors just below the xiphoid process. The rib cage was bisected, exposing the heart. A 21-gauge needle was inserted into the right ventricle, and blood was withdrawn. Blood was centrifuged and plasma was frozen at −20 C for later RIA. Differences in plasma hormone level were analyzed using student’s t test or one-way ANOVA where appropriate.
Sperm counts
Vas deferens were dissected from 5.5- to 17.5-wk-old male mice and placed in M2 medium (Sigma, St. Louis, MO). Sperm were removed using watchmaker’s no. 5 forceps and incubated at 37 C for 15–30 min in warmed M2 medium. A 1:10 dilution was used for hemocytometric count. Data were analyzed by Kruskal-Wallis one-way ANOVA for effects of genotype or treatment differences, followed, if applicable, by post hoc pairwise comparisons.
Sex behavior testing
Intact male mice were isolated at weaning and individually caged throughout the extent of the behavioral testing. These mice were maintained on a 12-h light,12-h dark cycle with lights off at 1000 h. In a protocol modified from Ogawa et al. (25), each male was tested twice, with an interval of 4–7 d, for sexual behavior. Tests consisted of placing a female in the male’s home cage for 30 min. Mice were tested during the dark phase of the light-dark cycle starting 2 h after lights off. All tests were videotaped using dim red light to provide illumination. Analysis of the tests was done by an observer blind to genotype or treatment group. All females were ovariectomized and sc injected with estradiol benzoate and P to ensure maximal sexual receptivity. Estradiol benzoate was injected 48 h (10 μg) and 24 h (5 μg) before the test; P (500 μg) was injected 4–7 h before onset of testing. For each male, the latency and number of mounts (including head mounts), intromissions, ejaculation, and ejaculation duration (where applicable) were recorded. No aggressive behaviors toward the female were observed in any of the tests. Sexual behavior data were analyzed by Kruskal-Wallis one-way ANOVA for effects of genotype or treatment differences, followed, if applicable, by post hoc pair wise comparisons.
P and RU486 administration
Males 8 wk of age were implanted with a SILASTIC capsule (Dow Corning Corp., Midland, MI) containing P (0.5 mg; Sigma) suspended in 20 μl sesame oil (Sigma). SILASTIC medical-grade tubing was cut into 1-cm segments and filled with either sesame oil vehicle or hormone. The ends of the implant were sealed with SILASTIC medical adhesive (Silicone type A, Dow Corning). Capsules were allowed to cure overnight before implantation. These treatments produce sustained elevations of circulating P levels of approximately 40–50 ng/ml. The RU486 pellets were purchased from Innovative Research of America (Sarasota, FL). The pellets are designed to release 0.5 mg/d and were implanted sc. Males were tested for behavior starting 12 d after implantation. Tests were scored by an observer blind to treatment group using the same sexual behavior paradigm described above.
Anxiety-related behavior testing
For these behavioral tests, male mice were isolated at weaning and singly caged throughout the experiment. Each 5-min test was performed 5–9 h after lights off under dim red light illumination. All tests were performed in a room isolated from outside disturbances such as motion or noise. Animals were housed in this room for 1 wk before the start of the experiments and throughout the testing. To eliminate olfactory stimuli, the testing apparatus was cleaned thoroughly with 95% ethanol between each test. Animals aged 11–15 wk were tested first in the elevated plus maze (EPM) followed by the open field (OF) test 1–3 d later. Both genotypes were tested on any given test day and each test was videotaped to assist in data collection. Data were analyzed by Kruskal-Wallis one-way ANOVA for effects of genotype or treatment differences, followed, if applicable, by post hoc pairwise comparisons.
EPM
The EPM is a widely used procedure based on a rodent’s unconditioned aversion to heights and open spaces. The EPM apparatus is held above the ground on a central pole and consists of a cross with intersecting pairs of arms, two of which are open and two are enclosed. In this test, the rodent has the choice to hide in the walled arms of the maze or explore the exposed, open arms of the maze. More time spent exploring the open arms and/or an increase in the number of times the rodent enters the open arms is believed to correspond to a lack of anxiety (26, 27). Each arm was 33 cm in length and 7.5 cm wide. The closed arms opposite to each other had 18-cm-high black walls. At the start of the EPM, the mouse was placed on the center platform oriented toward an open arm and allowed to explore the maze for 5 min. The number of open arm entries, number of closed arm entries, number of self-grooming bouts, number of rearings (lifting of forelimbs off the ground to sit on hindlimbs), number of head dippings (below the plane of the open arm), and the amount of time spent in either closed or open arms was recorded. A separate self-grooming bout was defined as one occurring 3 sec or more after the end of the previous bout.
OF test
This test is based on a rodent’s natural tendency to walk alongside the walls of the apparatus when anxious; the amount of time spent in the open field is inversely correlated with anxiety levels. The OF testing apparatus is a chamber made of plywood with 31-cm-high walls surrounding a 61 × 52 cm open area. The OF was divided into nine equally sized squares. The center area was defined as the area more than 2.5 cm away from the wall. Animals were placed in a corner square with the head facing the corner at the start of each test and allowed to ambulate freely for 5 min. The variables recorded included latency to first line crossing, number of leans (lifting forelimbs off the ground to touch the walls), number of rearings, bouts of self-grooming, time spent in the center area, time spent near the edge, and total number of grid line crossings. The total number of grid line crossings and the time spent in motion are indicative of exploratory activity in a novel environment.
RIA
Plasma samples were assayed for several hormones using the following reagents. LH and FSH standards, RP-3, and RP-2, respectively, were provided by the National Institute of Diabetes and Digestive and Kidney Diseases. Inhibin levels reported were α-inhibin subunit. Intraassay coefficients of variance were as follows: LH, 6.1%; FSH, 9.1%; inhibin, 10.7%; interassay coefficients of variance were: LH, 10.3%; FSH, 7.8%.
Histology
Animals were anesthetized via methoxyflurane inhalation and then transcardially perfused with ice-cold 4% paraformaldehyde in phosphate buffer (pH 7.6). Brains were removed and postfixed for 2 h and placed into 30% sucrose Tris-buffered saline solution overnight at 4 C. The following day, perfused brains were cut in a coronal plane with a sliding microtome and sections (40 μm) stored at −20 C in a cryoprotectant solution [PVP-40, sucrose, ethylene glycol, and 0.1 m phosphate buffer (pH 7.4)].
Antisera and immunochemicals
The following antisera were used: primary antiandrogen receptor (1 μg/ml concentration); PG-21, rabbit immunoaffinity purified IgG, PG-21 (Upstate Biotechnology, Lake Placid, NY); secondary antibody, biotinylated goat antirabbit (1:250 dilution; Jackson ImmunoResearch, West Grove, PA).
Immunocytochemistry and quantification
Briefly, free-floating sections were rinsed three times in a solution of PBS, Triton X-100, and normal goat serum and then incubated 48 h in primary antibody at 4 C. After rinsing, secondary antibody was applied for 2 h a room temperature. After further rinsing, tissue was stained using the ABC peroxidase staining kit (Vectastain Elite; Vector Laboratories, Burlingame, CA), and diaminobenzidine (0.25 mg/ml in H2O with 0.01% H2O2 and 0.04% nickel; Sigma) was used as a chromogen. For quantization of immunocytochemical results, the optical dissector procedure was used to determine the numerical density of androgen receptor (AR) immunopositive cells in the medial preoptic nucleus (MPN) and bed nucleus of the stria terminalis (BST) of PRKO and isogenic mice (n = 4 for both groups). Structures were defined and counted using StereoInvestigator (version 4.37; MicroBrightField Inc., Williston, VT) from tissue sections taken at rostrocaudal stereotaxic coordinate −0.22 mm from bregma. For BST a 30 × 50 μm grid was used. Counting frame thickness for both brain regions was 15.0 μm. Average immunoreactive cell number per area for each group was derived, and comparisons were made between the groups for both tissues using Student’s t tests.
Results
Basal hormone levels in adult mice
Among the hormones examined, PRKO and WT males did not vary significantly except for basal levels of FSH and inhibin. As depicted in Fig. 1A, PRKO males exhibited significantly lower (P < 0.0001) FSH levels (9.76 ± 0.92 ng/ml; n = 14) than WT counterparts (C57BL/6 15.64 ± 0.33 ng/ml; n = 10 and ISO 20.5 ± 2.9 ng/ml; n = 8), whereas inhibin levels (Fig. 1B) were significantly higher in PRKO males (PRKO, 121.8 ± 22.7, n = 10 vs. WT, 70.2 ± 5.14 pg/ml, n = 10 vs. ISO 75.8 ± 6.3 pg/ml; n = 5; P < 0.01). Basal LH levels (Fig. 1C) were not significantly changed in male PRKO mice (1.08 ± 0.46 ng/ml; n = 9), compared with WT (0.28 ± 0.08 ng/ml; n = 10) and ISO (1.4 ± 0.39 ng/ml; n = 7). We have previously reported that basal testosterone (T) levels are not altered in adult PRKO males, compared with levels in WT animals (24). Testicular weights were not different between the genotypes (PRKO, 81.3 ± 2.9 mg; n = 20; WT, 87.4 ± 3.1 mg n = 36; P > 0.05), and no gross differences in testicular morphology were observed (data not shown). Developmental analysis of sperm production revealed a delay in full spermatogenesis in the PRKO animals (Fig. 2). However, equivalent levels were achieved in PRKO and WT by 9 wk of age.
Plasma FSH, inhibin, and LH levels in C57BL/6, PRKO, and ISO control mice. A, Plasma FSH is significantly lower in PRKO males than C57BL/6 (a, P < 0.0001). B, Inhibin levels are significantly higher in PRKO males (P < 0.04). C. Basal LH levels are not significantly changed in male PRKO and C57BL/6 mice.
Plasma FSH, inhibin, and LH levels in C57BL/6, PRKO, and ISO control mice. A, Plasma FSH is significantly lower in PRKO males than C57BL/6 (a, P < 0.0001). B, Inhibin levels are significantly higher in PRKO males (P < 0.04). C. Basal LH levels are not significantly changed in male PRKO and C57BL/6 mice.
Sperm counts from the vas deferens of PRKO and WT mice. Sperm counts were significantly increased in PRKO mice at 6.5, 7.5, and 8.5 wk, compared with WT of the same age. a, P < 0.002; b, P < 0.02; c, P < 0.0001.
Sperm counts from the vas deferens of PRKO and WT mice. Sperm counts were significantly increased in PRKO mice at 6.5, 7.5, and 8.5 wk, compared with WT of the same age. a, P < 0.002; b, P < 0.02; c, P < 0.0001.
Sexual behavior
Many measures of sexual behavior were enhanced in PRKO males. In the first of two consecutive tests, virgin PRKO males exhibited a significant reduction in latency to first mount (Fig. 3A; C57BL/6, 597.2 ± 52.8 sec, n = 10; PRKO, 299.8 ± 48.5 sec, n = 10; 129SvEv, 615.7 ± 92.2 sec, n = 10; P < 0.004) but no significant difference in number of mountings/test or number of intromissions (Fig. 3, B and C). In the second test, conducted 1 wk after the first, PRKO males again exhibited a significant decrease in latency to mount (Fig. 4A; C57BL/6, 541.7 ± 150 sec; PRKO, 103 ± 16.8 sec; 129SvEv, 386.7 ± 57.6 sec; P < 0.0009) and tended to exhibit increased ejaculatory behavior, although this trend did not reach statistical significance. The mean number of mounts and intromissions did not differ between genotypes, most likely due to the behavioral effects of ejaculation; after ejaculating, males will not mount or intromit for approximately 20–30 min.
Sexual behavior of male mice during initial test with a receptive female. A, In initial tests PRKO males exhibited reduced latency to mount, compared with both WT strains (a, P < 0.004). B, No significant differences in number of mounts were observed. C, No significant differences in number of intromissions were observed.
Sexual behavior of male mice during initial test with a receptive female. A, In initial tests PRKO males exhibited reduced latency to mount, compared with both WT strains (a, P < 0.004). B, No significant differences in number of mounts were observed. C, No significant differences in number of intromissions were observed.
Sexual behavior of male mice during the second test with a receptive female. A, PRKO males again exhibited reduced latency to mount, compared with both WT strains (a, P < 0.0009). B, No significant differences in number of mounts were observed. C, No significant differences in number of intromissions were observed.
Sexual behavior of male mice during the second test with a receptive female. A, PRKO males again exhibited reduced latency to mount, compared with both WT strains (a, P < 0.0009). B, No significant differences in number of mounts were observed. C, No significant differences in number of intromissions were observed.
Pharmacological blockade of PR in WT males enhanced several aspects of male mouse sexual behavior. Whereas the RU486-treated males showed a tendency for reduced latency to mount that did not achieve statistical significance (Fig. 5A), they did exhibit a significant increase in the number of mounts (Fig. 5B) and intromissions (Fig. 5C). This increase in sexual behavior was observed only in virgin males during their first encounter with a female; in a subsequent test, both oil- and P-treated males exhibited behavior similar to males treated with the PR antagonist (data not shown). P treatments produced no significant effects on any aspect of sexual behavior in any of the tests.
Males treated with RU486 exhibited increased sexual behavior in the first encounter with a female. A, The decreased mount latency observed in RU486-treated C57BL/6 males was not statistically significant. B, RU486-treated WT males exhibited increased numbers of mounts (a, P < 0.05 vs. control; b, P < 0.01 vs. P-treated males). C, RU486-treated males exhibited increased number of intromissions (a, P < 0.05 vs. controls).
Males treated with RU486 exhibited increased sexual behavior in the first encounter with a female. A, The decreased mount latency observed in RU486-treated C57BL/6 males was not statistically significant. B, RU486-treated WT males exhibited increased numbers of mounts (a, P < 0.05 vs. control; b, P < 0.01 vs. P-treated males). C, RU486-treated males exhibited increased number of intromissions (a, P < 0.05 vs. controls).
Anxiety-related behavior
Differences in anxiety-related behavior could possibly be a factor in producing differences in sexual behavior among the mouse genotypes. To determine whether the observed increases in sexual behavior in PRKO males are secondary to differences in innate anxiety levels, we measured their anxiety-related behaviors in the EPM and OF tests. In the EPM, PRKO males displayed a level of anxiety-related behavior that was intermediate between the levels observed in the two WT background strains. The number of entries into the closed arms did not vary significantly between C56BL/6 and PRKO males. 129SvEV males, however, displayed significantly fewer closed arms entries (Table 1), compared with PRKO. The number of open arm entries did not vary between C57BL/6 and PRKO males, but 129SvEv males displayed significantly fewer open arms entries. PRKO males spent significantly less time in the open arm of the EPM apparatus. PRKO males displayed similar levels of exploratory behaviors, compared with C57BL/6 males, and 129SvEv males displayed significantly fewer rearing behaviors and head dipping behaviors (Table 1).
Elevated plus maze
| Genotype | Entries, closed arm | Entries, open arm | Time in open | Rearing | Head dipping |
|---|---|---|---|---|---|
| C57BL/6 | 7.4 ± 1.6 | 5.8 ± 0.6 | 170.5 ± 21.5 | 11.4 ± 2.8 | 19.6 ± 1.9 |
| PRKO | 10.1 ± 1.3 | 4.5 ± 1.0 | 67.5 ± 14.1a | 8.7 ± 2.1 | 12.7 ± 1.8 |
| 129SV | 3.0 ± 1.1a | 1.0 ± 0.2a,b | 90.8 ± 33.3 | 0.08 ± 0.08a,b | 9.7 ± 1.8c |
| Genotype | Entries, closed arm | Entries, open arm | Time in open | Rearing | Head dipping |
|---|---|---|---|---|---|
| C57BL/6 | 7.4 ± 1.6 | 5.8 ± 0.6 | 170.5 ± 21.5 | 11.4 ± 2.8 | 19.6 ± 1.9 |
| PRKO | 10.1 ± 1.3 | 4.5 ± 1.0 | 67.5 ± 14.1a | 8.7 ± 2.1 | 12.7 ± 1.8 |
| 129SV | 3.0 ± 1.1a | 1.0 ± 0.2a,b | 90.8 ± 33.3 | 0.08 ± 0.08a,b | 9.7 ± 1.8c |
PRKO male mice display intermediate levels of anxiety-related behavior in the elevated plus maze. The number of entries into the closed arms did not vary significantly between C56BL/6 and PRKO males. 129SvEV males displayed significantly fewer closed arms entries (aP < 0.01), compared with PRKO. The number of open arm entries did not vary between C57BL/6 and PRKO males. 129SvEV males displayed significantly fewer open arms entries (aP < 0.001 vs. C57BL/6; bP < 0.01 vs. PRKO). PRKO males spent significantly less time in the open arm of the EPM apparatus (aP < 0.05 vs. C57BL/6). PRKO males displayed similar levels of exploratory behaviors, compared with C57BL/6 males. 129SvEv males displayed significantly fewer rearing behaviors (aP < 0.001 vs. C57BL/6; bP < 0.01 vs. PRKO) and head dipping behavior (cP < 0.01 vs. C57BL/6).
Elevated plus maze
| Genotype | Entries, closed arm | Entries, open arm | Time in open | Rearing | Head dipping |
|---|---|---|---|---|---|
| C57BL/6 | 7.4 ± 1.6 | 5.8 ± 0.6 | 170.5 ± 21.5 | 11.4 ± 2.8 | 19.6 ± 1.9 |
| PRKO | 10.1 ± 1.3 | 4.5 ± 1.0 | 67.5 ± 14.1a | 8.7 ± 2.1 | 12.7 ± 1.8 |
| 129SV | 3.0 ± 1.1a | 1.0 ± 0.2a,b | 90.8 ± 33.3 | 0.08 ± 0.08a,b | 9.7 ± 1.8c |
| Genotype | Entries, closed arm | Entries, open arm | Time in open | Rearing | Head dipping |
|---|---|---|---|---|---|
| C57BL/6 | 7.4 ± 1.6 | 5.8 ± 0.6 | 170.5 ± 21.5 | 11.4 ± 2.8 | 19.6 ± 1.9 |
| PRKO | 10.1 ± 1.3 | 4.5 ± 1.0 | 67.5 ± 14.1a | 8.7 ± 2.1 | 12.7 ± 1.8 |
| 129SV | 3.0 ± 1.1a | 1.0 ± 0.2a,b | 90.8 ± 33.3 | 0.08 ± 0.08a,b | 9.7 ± 1.8c |
PRKO male mice display intermediate levels of anxiety-related behavior in the elevated plus maze. The number of entries into the closed arms did not vary significantly between C56BL/6 and PRKO males. 129SvEV males displayed significantly fewer closed arms entries (aP < 0.01), compared with PRKO. The number of open arm entries did not vary between C57BL/6 and PRKO males. 129SvEV males displayed significantly fewer open arms entries (aP < 0.001 vs. C57BL/6; bP < 0.01 vs. PRKO). PRKO males spent significantly less time in the open arm of the EPM apparatus (aP < 0.05 vs. C57BL/6). PRKO males displayed similar levels of exploratory behaviors, compared with C57BL/6 males. 129SvEv males displayed significantly fewer rearing behaviors (aP < 0.001 vs. C57BL/6; bP < 0.01 vs. PRKO) and head dipping behavior (cP < 0.01 vs. C57BL/6).
The EPM and OF tests are generally considered complementary tests of anxiety. Indeed, the intermediate level of anxiety-related behaviors of PRKO males observed in the EPM test were reflected in the results of the OF test. C57BL/6 males spent significantly more time in the center of the OF apparatus (Table 2.). PRKO males exhibited significantly fewer grid crossings than C57BL/6 males and significantly more than 129SvEv males. Latency to first grid crossing was similar between C57BL/6 and PRKO males, whereas 129SvEv males displayed significantly increased latency, compared with C57BL/6 and PRKO males. PRKO males engaged in intermediate levels of both leaning and rearing behaviors (Table 2). As in the EPM, bouts of self-grooming did not significantly differ between the genotypes. Thus, in OF and EPM tests, PRKO mice exhibited intermediate anxiety levels, compared with the two WT strains, and the results presented here do not substantiate the hypothesis that the increased sexual behavior of PRKO males is due to a reduction of anxiety in a novel situation.
Open field test
| Genotype | Center time (sec) | Grid crossings | Latency | Leans | Rears |
|---|---|---|---|---|---|
| C57BL/6 | 246 ± 8a | 108.3 ± 6.5a,b | 4.9 ± 0.8 | 16.8 ± 2.1a,b | 15.9 ± 1.4a,b |
| PRKO | 88.5 ± 12.7 | 78.7 ± 5.4c | 16.4 ± 2.1 | 6.7 ± 2.1c | 3.8 ± 1.4 |
| 129SV | 81.9 ± 31.4 | 8.33 ± 4 | 151.3 ± 39.4a,b | 0.8 ± 0.08 | 0.3 ± 0.3 |
| Genotype | Center time (sec) | Grid crossings | Latency | Leans | Rears |
|---|---|---|---|---|---|
| C57BL/6 | 246 ± 8a | 108.3 ± 6.5a,b | 4.9 ± 0.8 | 16.8 ± 2.1a,b | 15.9 ± 1.4a,b |
| PRKO | 88.5 ± 12.7 | 78.7 ± 5.4c | 16.4 ± 2.1 | 6.7 ± 2.1c | 3.8 ± 1.4 |
| 129SV | 81.9 ± 31.4 | 8.33 ± 4 | 151.3 ± 39.4a,b | 0.8 ± 0.08 | 0.3 ± 0.3 |
PRKO males exhibit intermediate levels of anxiety-related behaviors in the open field test. C57BL/6 males spent significantly more time in the center of the OF apparatus (aP < 0.001 vs. PRKO and 129). PRKO males exhibited significantly fewer grid crossings than C57BL/6 males (aP < 0.01) and significantly more than 129SvEv males (bP < 0.001). C57BL/6 males had significantly more grid crossings than 129SvEv males (cP < 0.01). Latency to first grid crossing was similar between C57BL/6 and PRKO males. 129SvEv males displayed significantly increased latency, compared with C57BL/6 (aP < 0.001) and PRKO males (bP < 0.01). PRKO males engaged in intermediate levels of both leaning and rearing behaviors (aP < 0.001 vs. PRKO; bP < 0.001 vs. 129SvEv; cP < 0.05 vs. 129SvEv).
Open field test
| Genotype | Center time (sec) | Grid crossings | Latency | Leans | Rears |
|---|---|---|---|---|---|
| C57BL/6 | 246 ± 8a | 108.3 ± 6.5a,b | 4.9 ± 0.8 | 16.8 ± 2.1a,b | 15.9 ± 1.4a,b |
| PRKO | 88.5 ± 12.7 | 78.7 ± 5.4c | 16.4 ± 2.1 | 6.7 ± 2.1c | 3.8 ± 1.4 |
| 129SV | 81.9 ± 31.4 | 8.33 ± 4 | 151.3 ± 39.4a,b | 0.8 ± 0.08 | 0.3 ± 0.3 |
| Genotype | Center time (sec) | Grid crossings | Latency | Leans | Rears |
|---|---|---|---|---|---|
| C57BL/6 | 246 ± 8a | 108.3 ± 6.5a,b | 4.9 ± 0.8 | 16.8 ± 2.1a,b | 15.9 ± 1.4a,b |
| PRKO | 88.5 ± 12.7 | 78.7 ± 5.4c | 16.4 ± 2.1 | 6.7 ± 2.1c | 3.8 ± 1.4 |
| 129SV | 81.9 ± 31.4 | 8.33 ± 4 | 151.3 ± 39.4a,b | 0.8 ± 0.08 | 0.3 ± 0.3 |
PRKO males exhibit intermediate levels of anxiety-related behaviors in the open field test. C57BL/6 males spent significantly more time in the center of the OF apparatus (aP < 0.001 vs. PRKO and 129). PRKO males exhibited significantly fewer grid crossings than C57BL/6 males (aP < 0.01) and significantly more than 129SvEv males (bP < 0.001). C57BL/6 males had significantly more grid crossings than 129SvEv males (cP < 0.01). Latency to first grid crossing was similar between C57BL/6 and PRKO males. 129SvEv males displayed significantly increased latency, compared with C57BL/6 (aP < 0.001) and PRKO males (bP < 0.01). PRKO males engaged in intermediate levels of both leaning and rearing behaviors (aP < 0.001 vs. PRKO; bP < 0.001 vs. 129SvEv; cP < 0.05 vs. 129SvEv).
AR immunoreactive cells
The number of AR immunoreactive cells per unit area was significantly increased in the PRKO compared with the ISO tissues (Fig. 6, A and B). The MPN, an important neural substrate regulating mating, is connected with many other regions that regulate reproductive functions, including the BST (28, 29). The MPN is important for copulation in a number of species, and lesions cause profound loss of sexual behavior (30, 31). Moreover, AR blockade in the MPN produces a significant decrement in male sexual behaviors (32). Accordingly, these findings reveal that the enhanced sexual behaviors observed in the PRKO males are accompanied by a significant increase in the number of AR immunoreactive cells in the MPN (Fig. 6C) (a, P < 0.0003; n = 4) and BST (Fig. 6D) (b, P < 0.0003; n = 4).
A and B, AR immunopositive cells in PRKO and WT, respectively. C, In the MPN, PRKO males express significantly higher levels of AR immunopositive cells than ISO males (a, P < 0.0003; n = 4, both groups). D, In the BST, PRKO males express significantly higher levels of AR immunopositive cells than ISO males (b, P < 0.0003; n = 4, both groups).
A and B, AR immunopositive cells in PRKO and WT, respectively. C, In the MPN, PRKO males express significantly higher levels of AR immunopositive cells than ISO males (a, P < 0.0003; n = 4, both groups). D, In the BST, PRKO males express significantly higher levels of AR immunopositive cells than ISO males (b, P < 0.0003; n = 4, both groups).
Discussion
These studies addressed the potential functions of PRs in males by analyzing the physiological and behavioral characteristics of male PRKO mice. Mice carrying a homozygous null mutation for PR exhibited increases in some components of masculine sexual behavior, an effect that is partially recapitulated in WT mice treated with a PR antagonist. Our observations are consistent with the hypothesis that PRs exert a major inhibitory influence on the circuitry that governs anticipatory aspects of male sexual behavior. Thus, PRKO males exhibit significantly shorter latencies to mount a female in both their initial and subsequent encounters. This rapid initiation of sexual behavior is most likely not due to a reduction of anxiety-related behaviors; PRKO males exhibit an intermediate phenotype of anxiety-related behavior in both the EPM and OF tests. Further work is required to determine the role of PR, if any, on other components of arousal such as sensory alertness, motor activity, and emotional reactivity. Furthermore, why PR would be inhibitory to mating behaviors in the normal male mammal and how this would be reproductively advantageous remain to be elucidated
Deletion or antagonism of PR resulted in the enhancement of masculine sexual behavior, but in our studies exogenous administration of P had no effect on sexual behavior in WT animals. Prior studies involving administration of exogenous P in many species suggests P inhibits the expression of sexual behaviors in males (21, 22, 33, 34). Based on these studies, which often used supraphysiological doses, P has been used to control male libido. In male cynomolgus monkeys treated with P or a synthetic progestin, medroxyprogesterone acetate (MPA), experienced decreases in many aspects of sexual behavior including motivational components and ejaculatory behaviors (35). Unlike MPA, treatment with P did not change T levels but still produced profound decreases in sexual behavior. The nonhuman primates used in these studies were castrated and T replaced before being treated long term (4–11 wk) with either MPA or P. Behavior testing was frequently tested at regular intervals over the course of 12 wk. The testing paradigm used on PRKO mice in this study was by contrast acute: the mice were treated for only 12 d before testing and were tested only for behavior with a female twice over the course of 1 wk. The possibility remains that chronic, high doses of P treatment in castrated, T-replaced mice would have similar inhibitory effects as those observed in nonhuman primates.
Because the administration of supraphysiological doses of P makes it difficult to assess the effects of endogenous P in the modulation of male sexual behavior, we sought to examine the effects of physiological P concentrations. Our results, that exogenous administration of low doses of P had no effect on behavior, are consistent with other published reports. Intact male mice administered a physiological dose of P exhibited no significant change in mounting and intromission behavior (34). Simultaneous injections of P and testosterone propionate failed to produce statistically significant changes in sexual behavior (36). Tonic levels of PR activation through either endogenous P or ligand-independent activation in the intact male may be sufficiently high to preclude the effects of exogenous P administration. It is possible that the circulating levels of P achieved in our studies were too low to augment endogenous levels and thus failed to affect behavior. Changes in route of administration and/or concentration could result in different levels of P available for receptor activation. Further studies in castrated and/or adrenalectomized animals would determine whether endogenous P influences normal male sexual behavior. Despite the failure of exogenous P to influence sexual behavior in the male mouse, we have demonstrated both pharmacologically and genetically that PR activation is involved in the inhibitory mechanisms regulating male sexual behavior.
The enhancement of sex behavior in the male PRKO mouse does not appear to be secondary to any alterations in serum T levels because we found no such changes in the PRKO animals vs. WT controls in a previous study (24). The absence of any elevations in LH secretion in the present study is also consistent with this previous finding and reinforces the idea that PR deletion or antagonism more directly influences the neural circuitries that facilitate male sex behaviors. It is possible, nevertheless, that the inhibitory effects of PR activation on sexual behavior may be manifest only in the context of physiological T concentrations. A previous study using castrate WT and PRKO mice (37) demonstrated that sexually experienced PRKO males exhibited a greater reduction of sexual behavior after castration than WT. Although that study focused on the castrate animal and is thus not directly comparable with our studies in testes-intact animals, it nevertheless suggests a facilitatory, rather than inhibitory role for PR in T-induced sexual behavior. Thus, PR activation may produce differing effects in the presence and absence of T. The time of day in which tests were conducted may also explain the difference in the results of the two studies; Phelps et al. (37) conducted their sex behavior testing shortly (2–3 h) after light onset, whereas tests in the present study were conducted shortly after dark onset. It is possible that regulation of the anticipatory component of sex behavior in the male mouse occurs only against a background of heightened arousal that occurs during the early portion of the dark period.
Deletion of the PR gene is accompanied by minor changes in reproductive physiology that include reduced FSH and increased inhibin levels in serum, and delayed pubertal appearance of mature spermatozoa in the vas deferens. The decrease in circulating FSH appears to be a consequence of increased inhibin levels. The finding of both increased α-inhibin subunit and decreased FSH suggests that activation of PR contributes to the development and/or function of testicular cells, specifically Sertoli cells, which express inhibin. Thus, a PR-sensitive inhibitory mechanism may normally operate to regulate inhibin expression, although it is not clear whether PR-mediated effects are exerted during Sertoli cell development or whether PR may exert an inhibitory influence on homeostatic mechanisms in mature testicular cells. We are not aware of any clear association of P and PR activation and inhibition of inhibin production in the male. In the adult female, blockade of PR with RU486 has been shown to affect FSH secretion and thus, inhibin secretion, depending on when it is given during the estrous cycle (38, 39). Additionally, a circumstantial relationship between circulating FSH and inhibin may be inferred from studies in the female that show a drop in inhibin during the luteal phase of the ovulatory cycle (40–43). This putative inhibitory mechanism may be important in the homeostatic regulation of inhibin production, but it does not seem to have any lasting effects on regulation of the hypothalamic-pituitary-gonadal axis or testicular development and function
The moderate, but significant reduction in FSH in PRKO mice is not associated with any overt changes in testis morphology or function. The effects of greater disruption of FSH expression in males are well documented and include altered testicular morphology, decreased testis weight, and decreased spermatogenesis (44–47). It is possible that PRs may normally subserve endocrine functions that are maintained in the PRKO mouse via the emergence of compensatory mechanisms during development. The most straightforward interpretation of the endocrine observations in these animals is that PR activation does not play a critically important role in the basic operation of the male reproductive axis.
In contrast to the minimal effects of PR ablation on the hypothalamic-pituitary-gonadal axis, PR gene deletion does produce substantial effects on reproductive behavior. The molecular signals that regulate the display of masculine sex behavior are not well understood, and hence the mechanisms by which PRs may normally influence male sex behavior are not clear. In most male rodents, the ability of T to facilitate sex behavior has been attributed in part to the effects of E derived by aromatization of T (reviewed in Ref.48). As a metabolite of T, estrogen (E) activates intracellular estrogen receptors (ERs), principally ERα, to elicit increase motivational and consummatory components of sexual behavior (25, 49–52). The requisite involvement of ERα in this regard has been demonstrated by the observation that mice in which ERα has been deleted exhibit profound deficits in several components of male sexual behavior (50, 52, 54). We have noted in the present study that deletion of the PR gene is accompanied by an enhancement of some of the same behaviors (e.g. latency to mount) that are reduced or absent in mice in which ERα has been deleted. It is thus possible that the effects of PR on male behavior may be due to its ability to oppose the action of E, particularly by down-regulating ERα-mediated transcriptional responses to estrogen.
In several cell contexts, molecular studies have revealed that transcriptional responses to ER activation can be potently modulated by PR activation. Liganded PRA and PRB can each suppress E-stimulated ER activity in a cell-, ligand-, and isoform-specific manner (55). P can also mediate a reduction in ER protein in vitro, most likely resulting from decreased cellular ER mRNA levels (56–58). Increased sexual behavior in PRKO males may therefore be due to loss of this negative regulation of ER. Because abolition of ERα in mice results in infertility and completely disrupts ejaculatory behavior (50), it is possible that loss of PR regulation in PRKO males leads to an increase in ER-stimulated behaviors, specifically the increase in sexual behavior observed in PRKO males. Inhibitory effects of P on E-induced sexual behavior are well known in female rodents, having been demonstrated in rats (59) and hamsters (60) as concurrent inhibition (61, 62) or sequential inhibition (63). It is possible that similar molecular mechanisms may mediate the inhibitory effects of PR activation on ER-dependent sex behavior in both males and females.
In the present studies, we have documented that AR expression is significantly increased in the MPN and BST of PRKO vs. WT mice. Our finding of increased AR expression in these structures provides a second possible explanation for the enhancement of sexual behaviors in the PRKO mice; that is, absence or antagonism of PRs leads to increased AR expression and hence, more robust effects of androgens via AR activation on some components of male sexual behavior. There is now considerable evidence that AR-dependent androgen signaling, particularly in the MPN (64–66) and BST (67, 68), contributes to the expression of sexual behavior in male rodents. Recent studies in AR knockout mice have confirmed that AR gene inactivation in intact males results in decreased male sexual behavior (69), an effect that cannot be reversed with dihydrotestosterone treatment. Interestingly, treatment of AR knockout mice with estrogen can restore sexual behavior (53), suggesting that AR and ER influences on sexual behavior in male mice are convergent, perhaps via AR regulation of aromatase. It remains to be determined whether PR gene deletion or blockade leads to enhancement of male sexual behaviors via disinhibition of AR expression, ER signaling, or a combination of these processes.
In summary, ablation of PR in males increases the intensity of some components of sexual behavior. Moreover, a similar increase, rather than a decrease, in male sexual behavior can be induced by pharmacological blockade of PRs in WT mice. The results of this study, therefore, suggest an inhibitory role for PR in male reproductive behaviors.
Acknowledgments
The authors gratefully acknowledge Brigitte Mann for her expert technical assistance with hormone measurements.
Abbreviations
- AR,
Androgen receptor;
- BST,
bed nucleus of the stria terminalis;
- E,
estrogen;
- ER,
estrogen receptor;
- EPM,
elevated plus maze;
- ISO,
isogenic;
- MPA,
medroxyprogesterone acetate;
- MPN,
medial preoptic nucleus;
- OF,
open field;
- P,
progesterone;
- PR,
progesterone receptor;
- PRKO,
progesterone receptor knockout;
- WT,
wild type.
