Organizational Role for Testosterone and Estrogen on Adult Hypothalamic-Pituitary-Adrenal Axis Activity in the Male Rat
http://www.100md.com
内分泌学杂志 2005年第4期
Laboratories for Integrative Neuroscience and Endocrinology, University of Bristol, Bristol BS1 3NY, United Kingdom
Address all correspondence and requests for reprints to: Dr. S. Lightman, Laboratories for Integrative Neuroscience and Endocrinology, Dorothy Hodgkin Building, Whitson Street, University of Bristol, Bristol BS1 3NY, United Kingdom. E-mail: stafford.lightman@bristol.ac.uk.
Abstract
Organizational effects of testosterone during a critical period of neonatal life have major irreversible effects on adult sexual behavior. We have investigated whether perinatal androgen changes also affect another major sexually differentiated system, the hypothalamo-pituitary-adrenal axis. This was assessed in male rats who had been exposed to perinatal flutamide or 1,4,6-androstatriene-3,17-dione (ATD). Once the animals reached adulthood, an automated sampling system was used to collect blood from freely moving animals at 10-min intervals over 24 h, followed by a noise stress and then the administration of lipopolysaccharide (LPS). Perinatal flutamide- and ATD-treated rats not only had higher mean corticosterone levels and increased frequency and amplitude of corticosterone pulses over the 24 h compared with vehicle-injected controls, but they also showed markedly increased corticosterone responses to both noise and LPS. All parameters of increased hypothalamo-pituitary-adrenal activity resembled the normal physiological state of the intact adult female rather than that of the intact adult male rat. Furthermore, 3 h after LPS administration, both flutamide- and ATD-treated animals had markedly higher levels of corticotropin-releasing factor mRNA in the parvocellular paraventricular nucleus (PVN) and proopiomelanocortin mRNA in the adenohypophysis. Flutamide-treated rats also had a greater level of PVN arginine vasopressin mRNA. PVN glucocorticoid receptor mRNA levels were significantly lower in both the flutamide- and the ATD-treated male rats. These data highlight the importance of perinatal exposure to both testosterone and estrogen(s) on the development of a masculinized circadian corticosterone profile and stress-induced hypothalamo-pituitary-adrenal axis activity in the adult male rat.
Introduction
AN INFLUENCE OF GONADAL steroids on the development of sexually dimorphic brain regions has been known since the 1930s. The organizational effects of androgens were first described by Arai and Gorski in the late 1960s (1), and the sexual dimorphism of brain structure was described by Raisman and Field in 1971 (2). Around the same time, Levine et al. (3) investigated the effect of neonatal stimulation on adult hypothalamo-pituitary-adrenal (HPA) activity and clearly showed that this could be programmed by neonatal stress.
The presence of sexual dimorphism in corticosterone secretion was originally reported by Kitay (4). On the basis of this research, we and others have demonstrated an enhanced basal and stress-induced HPA axis and subsequent corticosterone response for adult female compared with male rats (5, 6, 7, 8). The different levels of circulating gonadal steroids in the two genders are considered integral in producing this divergent HPA axis activity in the adult. However, the question arises of whether this sexually dimorphic HPA axis is the product of a gonadal steroid influence at the perinatal stage.
Male rats undergo two testosterone surges, one between gestational d 17 and 18 (9) and another 1–3 h post parturition (10). These surges are vital for both the anatomical (11) and behavioral (12) masculinization of male rats. The effects of these surges can be antagonized (13, 14, 15, 16) by antiandrogens such as flutamide, a potent nonsteroidal antiandrogen that inhibits androgen uptake and nuclear binding in target tissues (17). The influence of perinatal testosterone may not, however, be solely mediated through the androgen receptor (AR), because testosterone can also be aromatized to estradiol, which can, in turn, exert effects through the estrogen receptor (ER) (18). Indeed, there are prevalent aromatizing enzymes (19), high levels of aromatization (20), and functional hypothalamic ERs (21) in the newborn male rat brain. The ability of postnatal injections of estrogen, but not nonaromatizable dihydrotestosterone, to suppress lordotic and female gonadotropin cyclic activities in neonatally castrated male rats (22) supports this aromatization hypothesis. Furthermore, prenatal and postnatal treatments with a specific aromatization inhibitor, 1,4,6-androstatriene-3,17-dione (ATD) (23), reduce male sexual behavior in the adult rat (24).
Current findings implicate an important role for testosterone in shaping masculine behavior, although it is unclear whether such an effect is taking place via AR or aromatization and subsequent ER activity. Additionally, the majority of studies examining the effect of early gonadal action on the adult male rat have focused on behavioral or anatomical aspects. In contrast, the effects of early life manipulation of gonadal steroids on HPA axis activity have received comparatively little attention. The ability of castration to enhance basal and stress-induced corticosterone release in adult male rats significantly above those of intact or gonadally replaced male rats (8, 25, 26) demonstrates a mediatory effect of androgens on HPA activity in the adult. However, whether the behavioral effects of early exposure to gonadal steroids are reflected in organizational changes in adult male corticosterone release and HPA axis activity remains to be determined. Demonstration of such an influence may help us to understand the mechanisms underlying the sexually differentiated corticosterone profiles found for adult male and female rats.
An automated sampling system that allows the collection of blood samples over a prolonged period was used to examine the 24-h corticosterone profile of feminized male rats compared with controls under basal conditions. The release pattern of basal corticosterone influences corticosterone secretion in response to stressful stimuli. Accordingly, the effects of feminization on corticosterone secretion after an acute (noise) and an immune-mediated (LPS) stress were also studied using the automated sampling system. The unique ability of this system to collect blood samples in a remote manner over a considerable number of hours allows for a very detailed analysis of corticosterone release while minimizing any confounding factors associated with conventional manual sampling techniques. To elucidate the potential mechanisms through which neonatal testosterone might result in altered adult HPA axis activity, we used two treatment paradigms. The first administered flutamide pre- and postnatally to assess the effect of androgen uptake into target cells on subsequent adult corticosterone secretion. The second administered ATD both pre- and postnatally to examine the effects of aromatization on adult corticosterone release profiles. Changes in circulating levels of corticosteroid-binding globulin (CBG) could be confounding influences on the interpretation of plasma corticosterone levels, with greater levels in the female animal buffering the effects of higher circulating corticosterone levels (4, 5). We therefore measured both basal and stress-induced CBG levels in our experimental groups.
Materials and Methods
Flutamide experimental animals
Each week, two Sprague Dawley female rats were time-mated with a male Sprague Dawley rat in-house, so that two litters were born every week over a 1-month period. Four dams and the subsequent litters (10–14 rats/litter) were used for the flutamide group, and four dams and litters were used for the control group. Flutamide was dissolved in a vehicle of 5% ethanol in sunflower oil. All pregnant dams were individually housed and injected (sc) every day in the dorsum of the neck with flutamide (100 mg/kg body weight) from d 13 of gestation to parturition. Upon birth, male pups were injected sc with flutamide (50 mg/kg body weight) every day until postnatal d 20. This protocol has been shown to effectively feminize male pups (13). Pups remained with the dam until weaned (21 d post parturition). During the early postnatal days, both female and male pups were administered flutamide, because similarities in the anogenital distances of the males and females made accurate sexing impossible. Females were culled once their sex could be determined (3–4 wk post parturition). Control pregnant dams and resultant pups were injected with vehicle only, using the same time regimen as the flutamide group. All animals from both groups were handled for a similar amount of time during flutamide or vehicle administration.
ATD experimental animals
Each week two Sprague Dawley female rats were time-mated with a male Sprague Dawley rat, so that two litters were born per week in-house over the period of 1 month. Four dams and the subsequent litters (10–14 rats/litter) were used for the ATD group, and four dams and litters were used for the control group. For the ATD group, individually housed, pregnant, Sprague Dawley dams received two sc implants containing ATD (2.5 cm; outside diameter, 2.1 mm; inside diameter, 1.5 mm) on d 13 of gestation. Upon birth, male pups were injected sc with ATD (dissolved in a vehicle of 5% ethanol in sunflower oil) at a dosage of 1 mg/pup every other day from birth (d 0) to d 12. This method is effective in preventing aromatization of testosterone surges in the male rat (27). Using the same time protocol as that in the ATD group, control pregnant dams received two sc implants containing cholesterol, and subsequent male pups were injected every other day until d 12 with vehicle only. Female pups were culled at the time of birth in both ATD and control groups. All animals from both groups were handled for a similar amount of time during ATD or vehicle administration.
Automated sampling system surgery
Once treatment was finished, male pups from each group were left to grow up in cages of four under a 14-h light, 10 -h dark schedule with ad libitum access to food and water. At 9–10 wk of age (275–300 g), all male rats were anesthetized using Hypnorm (0.32 mg/kg fentanyl citrate and 10 mg/kg fluanisone; Janssen Pharmaceuticals, Buckinghamshire, UK) and diazepam (2.6 mg/kg; Phoenix Pharmaceuticals, St. Joseph, MO). A SILASTIC brand (Dow Corning, Midland, MI)-tipped polythene cannula (inside diameter, 0.58 mm; Portex, Hythe, UK) filled with heparinized saline (10 U/ml heparin; CP Pharmaceuticals Ltd., Wrexham, Wales, UK) was inserted into the right jugular vein of each rat. The cannula was exteriorized at the crown of the rat’s head and protected by a spring attached to a 360° mechanical swivel. Animals were connected to an automated blood sampling system, allowing the collection of blood samples at preset time points (28). All animal procedures were carried out in accordance with the Animal (Scientific Procedures) Act of 1986. Rats from both studies underwent the same sampling protocol and the same assays on their blood and tissues. Hematocrits were calculated after the sampling procedures and never fell below 23%. This fall in hematocrit was not associated with any change in HPA axis activity or any alteration in behavior as recorded by remote video recording.
Sampling procedure
Sampling began at 0700 h on d 5 post surgery. Blood samples were collected every 10 min from each rat over a period of 24 h. Blood (37.7 μl) was removed for each sample and replaced with heparinized saline. At 0700 h (lights on at 0500 h) on d 6 post surgery, a noise generator was used to expose rats to a 10-min white noise stress (110 db). Blood samples were taken every 10 min for 2 h after noise onset. At 0900 h (d 6; lights on at 0500 h), 100 μl LPS (Escherichia coli; 055:B5, 250 μg/ml; Sigma-Aldrich Corp., Poole, UK) was administered to each rat through the iv cannula. Samples were collected every 15 min for an additional 3 h after LPS administration. At 1200 h, rats were overdosed with pentobarbitone (iv) and decapitated. Trunk blood, brains, and pituitaries were collected. Previous work reported LPS-induced increases in corticotropin-releasing factor (CRF) mRNA using this serotype 3 h after LPS administration (29).
Corticosterone RIA
The sampling protocol produced 169 blood samples/rat. During sampling, each 37.7-μl blood sample was diluted 1:5 in heparinized saline. Fifty microliters of each blood sample were further diluted into 50 μl citrate buffer (pH 3.0) and incubated overnight at 4 C with 50 μl [125I]corticosterone tracer (ICN Biomedicals, Aurora, OH) and 50 μl rabbit antirat corticosterone primary antibody (donated by G. Makara, Institute of Experimental Medicine, Budapest, Hungary). All samples were processed in duplicate. On d 2, a charcoal/dextran solution was added to all samples, which were then centrifuged (15 min, 4000 rpm, 4 C) and aspirated before being loaded onto a -counter. Intra- and interassay coefficients of variation for the corticosterone assays were 12.4% and 16%, respectively.
ACTH RIA
Plasma derived from trunk blood taken 3 h after LPS administration was analyzed for ACTH using a rabbit antirat ACTH primary antibody (donated by G. Makara) and [125I]ACTH (Amersham Biosciences, Little Chalfont, UK). A polyethylene glycol solution, sheep antirabbit secondary antibody (1:50 in standard assay buffer with 0.4% normal rabbit serum), and centrifugation were used to separate the bound from the unbound hormone fraction into a pellet. The resultant pellets were counted on a -counter.
CBG assay
Plasma samples from 0800–0850 h before LPS administration (d 6 post surgery) were pooled and assayed to provide pre-LPS CBG levels. Plasma collected 3 h post LPS administration were used to assess post-LPS levels. Plasma samples were stripped of endogenous corticosterone, eluted, and incubated overnight in a total binding or nonspecific binding [3H]corticosterone solution as previously described (28). On d 2, incubates were processed in triplicate through LH20 columns to separate bound and unbound steroids before being eluted and counted on a ?-scintillation counter (28). The Bradford method was used to determine protein content (30). Resultant values are reported as picomoles of [3H]corticosterone bound per milligram of protein.
Oligonucleotide in situ hybridization
Paraventricular nucleus (PVN) and pituitary sections (12 μm) were cut and mounted onto gelatin-coated slides. Sections were fixed in 4% formaldehyde (5 min) and taken through a prehybridization washing procedure as previously described (31). Slides were incubated overnight at 37 C in hybridization buffer (31) containing 1 M dithiothreitol and the required [35S]deoxy-ATP oligonucleotide probe. Approximately 100,000–200,000 cpm were applied to each slide in 45 μl hybridization buffer. The specificity of the oligonucleotide probes have been previously determined for arginine vasopressin (AVP) (32), CRF (33), and proopiomelanocortin (POMC) (34). The specific activities of the probes were 1.82 x 1018 dpm/mol (CRF), 2.03 x 1018 dpm/mol (AVP), and 1.75 x 1018 dpm/mol (POMC). For each study, all control and experimental sections were hybridized in the same reaction. On d 2, slides were taken through four changes of 1x saline sodium citrate (SSC) and washed in 1x SSC at 55 C (four times, 15 min each time) and in 1x SSC at room temperature (2 x 30 min), dipped in distilled water, and dried. Slides were exposed (AVP, 3 d; POMC, 5 d; CRF, 14 d) to Hyperfilm MP (Amersham Biosciences). Analysis of mRNA levels in comparison to 14C-labeled standards was achieved using image analysis software (Image 1.6.2, W. Rasband, NIH, Bethesda, MD). POMC mRNA was determined by measuring gray levels over the anterior pituitary, and CRF mRNA was determined using a threshold method to highlight probe bound over the nucleus (31, 35). To differentiate parvocellular from magnocellular AVP mRNA content, a threshold method was used to threshold out the magnocellular cells and therefore measure signal over the parvocellular cells only (36).
Riboprobe in situ hybridization
Antisense 35S-labeled glucocorticoid receptor (GR) and mineralocorticoid receptor (MR) riboprobes (donated by J. Seckl, University of Edinburgh, Edinburgh, Scotland) were used to detect GR mRNA expression in PVN and hippocampal sections (12 μm) and MR mRNA in hippocampal sections only. GR and MR were reverse transcribed as previously described (36). Sections were fixed in 4% paraformaldehyde, processed through prehybridization washes, and incubated overnight at 55 C in hybridization buffer containing the required 35S-radiolabeled riboprobe (40). Approximately 1,000,000 cpm were applied to each slide in 45 μl hybridization buffer. The specific activities of the probes were 1.68 x 1019 dpm/mol (GR) and 1.40 x 1019 dpm/mol (MR). On d 2, sections were taken through three changes of 1x SSC before being washed in 50% formamide in 1x SSC three times for 15 min each time at 55 C. Sections were briefly dipped in 1x SSC at 37 C and incubated in ribonuclease A (25 mg/ml; 37 C) for 30 min. Slides were dipped in 1x SSC (37 C), washed in 50% formamide in 1x SSC (three times, 15 min each time, 55 C), 1x SSC (twice, 5 min each time, room temperature), and dipped in distilled water. Slides were left to air-dry before exposure to Hyperfilm for 2 wk. Image analysis software was used to analyze resultant mRNA levels in subregions of the hippocampus or in the PVN. Sense 35S-labeled GR and MR riboprobes were included as negative controls and displayed no radioactive signal on the treated hippocampal and PVN sections.
Statistical analysis
For each study, data are presented as mean plasma corticosterone blood levels within each group over the 24-h basal period. A Pulsar analysis program (37) analyzed pulse number, frequency, amplitude and height. Unpaired t tests were used to detect any differences in these parameters between the two groups for each study.
For statistical analysis of the postnoise stress response (noise onset occurred at 0700 h on d 6 post surgery), the mean corticosterone level 20 min after noise onset was averaged for each group. The point 20 min after noise onset was chosen due to previous findings demonstrating maximal corticosterone secretion at this time (38). A baseline control value was obtained using the mean of values from 0600–0650 h (d 6 post surgery) according to the recommendation of Festing et al. (39). For the two studies, an unpaired t test compared the stress responses of the group. To clarify any statistical differences in corticosterone secretion occurring at these discrete time points between the treatment groups, an area under the curve measurement from the time of noise onset (0700 h on d 6 post surgery) to the time at which values had returned to baseline (0750 h on d 6) was calculated for each rat. Unpaired t tests compared area under the curve measurements between the flutamide and control groups and between the ATD and control groups.
A similar statistical procedure analyzed corticosterone levels after LPS administration (LPS injected at 0900 h on d 6 post surgery). In this situation, the mean of values from 0800–0850 h on d 6 post surgery was used to calculate a baseline control value. Mean corticosterone levels 90 min after LPS administration were used as an indication of the LPS-induced stress response. Previous findings have shown maximal corticosterone secretion after LPS exposure to occur between 60–120 min post administration (40). For additional statistical analysis of the LPS stress response, an area under the curve measurement, using the trapezoid rule, from LPS administration (0900 h) to experimental termination (1200 h on d 6) was calculated for each rat. Unpaired t tests were used to compare area under the curve measurements between flutamide and control and between ATD and control groups.
Unpaired t tests were used to compare post-LPS ACTH levels for both studies. ANOVA was used to compare pre- and post-LPS CBG levels within and between the two groups in each study.
For mRNA analysis, a minimum of four PVN or hippocampal or six pituitary sections were analyzed for each rat. A mean value of the sections per rat was then calculated, so that each rat contributed a single mean value, which was used to calculate a group mean. Statistical analysis was performed on the dataset before calculation of percent change in values used for comparative purposes only.
Results
Flutamide study
Previous research has reported disrupted testes descent and/or epididymus development (13, 14) and enhanced testosterone levels (41) in flutamide-treated compared with control males. In the present study, all flutamide-treated rats demonstrated disrupted testes descent and/or epididymus development, whereas no such disruption was found in control males. Significantly enhanced (P < 0.01) levels of testosterone in the plasma of flutamide-treated (mean ± SE, 1.88 ± 0.09 ng/ml) compared with control (mean SE, 1.58 ± 0.1 ng/ml) males was also found.
Blood samples taken every 10 min over a period of 24 h were assayed for corticosterone from seven male adults pre- and postnatally exposed to flutamide (flutamide group) and seven male adults pre- and postnatally exposed to sunflower oil (control group). A mean basal 24-h corticosterone profile for each group was calculated from the mean corticosterone level for each 10-min point (Fig. 1). The mean group profiles demonstrated a higher corticosterone secretion pattern for flutamide compared with the control group. Flutamide males had a significantly greater number of corticosterone pulses over the 24-h period compared with controls (P < 0.001). A significantly increased pulse frequency per hour (P < 0.001), mean pulse height (P < 0.001), and mean pulse amplitude (P < 0.001) were also found for flutamide compared with control rats (Table 1).
FIG. 1. Mean blood corticosterone levels for flutamide-treated and control male rats over a 24-h period. Each 10-min data point represents the mean for each group (n = 7 rats/group). , Dark phase (1900–0500 h).
TABLE 1. Mean ± SEM of pulsar parameter measurements for flutamide and control male rats (n = 7 rats per group)
After the 24-h blood collection, rats received a 10-min noise stress. Blood samples were collected for 2 h after noise onset, and mean values for each rat group were calculated (Fig. 2A). Flutamide rats had significantly increased corticosterone levels at the 20 min postnoise time point chosen for comparison purposes (P < 0.05). The area under the curve from 0700–0750 h after noise stress was significantly greater for the flutamide (mean, 4767 ± 141) compared with the control (mean, 2333 ± 138) rats (P < 0.01).
FIG. 2. A, Mean (±SEM; n = 7 rats/group) blood corticosterone levels of flutamide-treated and control male rats 30 min before and 90 min after noise stress administration (, noise stress from 0700–0710 h; lights on at 0500 h). *, P < 0.05 compared with controls. Statistical comparison was made between the groups at the 20 min point according to the recommendations of Festing et al. (39 ). B, Mean (±SEM; n = 7 rats/group) corticosterone levels of flutamide-treated and control male rats 30 min before and 180 min after LPS administration at 0900 h (indicated by arrow; lights on at 0500 h). *, P < 0.001 compared with controls. Statistical comparison was made between the groups 90 min after LPS administration.
LPS administration at 0900 h was followed by blood collection every 15 min for an additional 3 h. The corticosterone profile after LPS for each treatment group was determined (Fig. 2B). Flutamide males had significantly higher corticosterone levels than controls (P < 0.001) at the chosen comparative time point (90 min after LPS). A significantly higher area under the curve measurement after LPS administration was found for the flutamide (mean, 36605 ± 1960) compared with the control (mean, 20268 ± 1216) males (P < 0.001).
Plasma collected 3 h post LPS administration was assayed for ACTH levels. Flutamide males (mean, 414 ± 37 pg/ml; n = 7) had significantly (P < 0.001) higher ACTH plasma levels post LPS compared with controls (mean, 175 ± 28 pg/ml; n = 7). Post-LPS CBG levels were significantly reduced in comparison to pre-LPS CBG levels for both groups (P < 0.001). There were no significant differences in CBG levels between the two groups at analogous time points (Fig. 3).
FIG. 3. Mean (±SEM; n = 7 rats/group) plasma CBG (picomoles per mg protein) levels for flutamide and control male rats pre- and postLPS administration. *, P < 0.001 compared with post-LPS condition.
CRF, AVP, and GR mRNAs in the PVN and POMC mRNA in the anterior pituitary at the termination of the experiment are shown in Table 2. Significantly increased levels of parvocellular PVN AVP, PVN CRF, and anterior pituitary POMC mRNA were found for flutamide males compared with controls (P < 0.001). GR PVN mRNA levels were significantly lower in flutamide compared with control males (P < 0.001). No significant differences between the two groups were found in MR and GR mRNA levels for the CA1, CA2, CA3, and dentate gyrus hippocampal regions (Fig. 4).
TABLE 2. Mean percentage change in PVN CRF, AVP, and GR mRNA and anterior pituitary POMC mRNA levels compared with control males (n = 8 rats per group)
FIG. 4. Mean MR and GR mRNA levels in the CA1, CA2, CA3, and dentate gyrus hippocampal regions of flutamide rats as a percentage of the control value (n = 8 rats/group). No significant differences in GR or MR mRNA levels in the different hippocampal regions were found between the treatment groups.
ATD study
Mean blood corticosterone levels over a basal 24-h period demonstrated an enhanced corticosterone secretion profile for ATD-treated males compared with oil-treated males (controls; Fig. 5). Pulsar analysis demonstrated a significantly greater number of pulses over 24 h for ATD compared with control male rats (P < 0.01). ATD males also demonstrated significantly higher levels of pulse amplitude (P < 0.001), height (P < 0.05), and frequency (P < 0.05) compared with controls (Table 3).
FIG. 5. Mean corticosterone levels for ATD and control male rats over a 24-h period. Each 10-min data point represents the mean for each group (n = 7 rats/group). , Dark phase (1900–0500 h).
TABLE 3. Mean ± SEM of pulsar parameter measurements for ATD and control male rats (n = 7 rats per group)
Mean corticosterone levels after noise onset are shown in Fig. 6A. At the comparative 20 min point, ATD-treated males showed significantly increased levels of corticosterone secretion compared with controls (P < 0.001). The postnoise area under the curve was significantly greater for ATD (mean, 4273 ± 153) compared with control (mean, 2073 ± 152) rats (P < 0.01).
FIG. 6. A, Mean (±SEM; n = 7 rats/group) corticosterone levels of ATD and control male rats 30 min before and 90 min after noise stress administration (, noise stress from 0700–010 h; lights on at 0500 h). *, P < 0.001 compared with controls Statistical comparison was made between the groups at the 20 min point. B, Mean (±SEM; n = 7 rats/group) corticosterone levels of ATD and control male rats 30 min before and 180 min after LPS administration at 0900 h (indicated by arrow; lights on at 0500 h). *, P < 0.001 compared with controls. Statistical comparison was made between the groups 90 min after LPS administration.
Mean corticosterone secretion every 15 min for 3 h post LPS administration was calculated for each group (Fig. 6B). Significantly elevated corticosterone levels were found for ATD compared with control males at the 90 min point used for comparative purposes (P < 0.001). A significantly higher post-LPS area under the curve was demonstrated for ATD (mean, 29,057 ± 1,119) compared with control (mean, 16,694 ± 1,349) males (P < 0.01).
Plasma ACTH levels 3 h post LPS exposure demonstrated significantly (P < 0.001) higher levels for the ATD (mean, 317 ± 29 pg/ml; n = 7) compared with control rats (mean, 168 ± 23 pg/ml; n = 7). Pre-LPS CBG levels were significantly higher compared with post-LPS CBG levels for both ATD and control groups (P < 0.001). Treatment group had no effect on CBG levels (Fig. 7).
FIG. 7. Mean (±SEM; n = 7 rats/group) plasma CBG (picomoles per mg protein) levels for ATD and control male rats before and after LPS administration. *, P < 0.001 compared with post-LPS condition.
Mean percent changes in PVN CRF, AVP, and GR and anterior pituitary POMC mRNA at the end of the experiment were calculated for the ATD group compared with the control group (Table 4). ATD males demonstrated significantly greater levels of PVN CRF mRNA (P < 0.05). No significant differences in PVN AVP mRNA were found for ATD males compared with control males. PVN GR mRNA was significantly lower in the ATD males compared with controls (P < 0.001). Anterior pituitary POMC mRNA levels were significantly higher in ATD males compared with controls (P < 0.05). GR and MR mRNA levels in the hippocampal CA1, CA2, CA3, and dentate gyrus regions were not significantly different between the two treatment groups (Fig. 8).
TABLE 4. Mean percentage change in PVN CRF, AVP, and GR mRNA and anterior pituitary POMC mRNA levels compared with control males (n = 8 rats per group)
FIG. 8. Mean MR and GR mRNA levels in the CA1, CA2, CA3, and dentate gyrus hippocampal regions of ATD rats as a percentage of the control value (n = 7 rats/group). No significant differences in GR or MR mRNA levels in the different hippocampal regions were found between the treatment groups.
Discussion
These novel data demonstrate an increase in the basal corticosterone secretion of male adult rats experimentally deprived of pre- and postnatal testosterone activity by exposure to the antiandrogen flutamide or the aromatase inhibitor ATD. The enhanced secretion of corticosterone was characterized by an increase in the number, frequency, height, and amplitude of corticosterone pulses. These data demonstrate that feminization not only altered the amount of corticosterone secreted in adulthood, but also the underlying characteristics of such release. Additionally, flutamide- and ATD-treated rats had significantly elevated corticosterone secretion after noise stress and LPS administration compared with their controls. The endocrine differences between the experimental groups were also mirrored by significant changes in hypothalamic and pituitary transcripts at the end of the experimental period, particularly an increase in PVN CRF mRNA (and parvocellular AVP mRNA in the flutamide group) and adenohypophyseal POMC mRNA as well as decreased PVN GR mRNA in both the flutamide- and ATD-treated groups. Therefore, we have good evidence for an organizational effect of endogenous androgens, probably acting through both ARs and ERs to increase the activity of the HPA axis in the adult animal.
The perinatal period is critically important for the maturation and programming of neuroendocrine functions within the central nervous system. As early as 1957, Levine demonstrated a link between neonatal experience and the response to stress of the same animal when it reached adulthood (42). More recently, Meaney and others (43, 44, 45) have demonstrated how variations in handling procedures or neonatal immunological challenges can program different patterns of HPA responsiveness in the adult. In addition to its ability to respond to environmental conditions, the perinatal brain is also very sensitive to changes in circulating androgens. Phoenix and colleagues (46) were the first to show that testosterone treatment of pregnant guinea pigs resulted in masculinized sexual activity of the female offspring, and Gorski et al. (47) demonstrated that gender differences in the sexually dimorphic nucleus were due to variations in neonatal sex hormone levels as opposed to levels circulating during adulthood. We have previously reported that under basal conditions, the number, frequency, height, and amplitude of corticosterone pulses over 24 h are significantly enhanced in adult female and castrated male rats compared with intact males (8). What has not been investigated, however, is the effect of neonatal androgenization on the developmental programming of the HPA axis. Because rats show a major sexual dimorphism in HPA axis activity with females exhibiting greater basal activity and a considerably larger response to stress (4, 5, 6, 7, 8, 48), it is clearly possible that both activational and organizational effects of sex hormones may be important. Our research has examined this possibility.
In the present study flutamide-treated rats had a mean of 18 pulses over the basal 24-h period at a frequency of 0.76 pulses/h. These values are closer to the enhanced values previously reported (8) for both intact females (mean number, 24; mean frequency, 1) and castrated males (mean number, 22; mean frequency, 0.95) compared with those of intact males (mean number, 11; mean frequency, 0.49). The amplitude and height of the corticosterone pulses of flutamide-treated rats over the basal period also approximate those previously found for intact females and castrated males (8). Additionally, an excitatory influence of flutamide exposure was found on stress-induced corticosterone release and the ACTH and HPA axis response to an immune-mediated challenge (LPS). These findings support an organizational role for testosterone in producing a characteristically masculine corticosterone profile and HPA axis stress response in the adult male rat.
The fact that flutamide results in reduced testes descent and/or underdevelopment of the epididymus (13, 14) suggests that testosterone production is reduced in these rats. Given the inhibitory effects of testosterone on corticosterone secretion, it may be surmised that such reduced testosterone levels are responsible for the enhanced corticosterone and HPA axis activity evidenced in flutamide-treated rats. However, the plasma testosterone levels of the flutamide group (1.88 ng/dl) were significantly greater than those in the controls (1.58 ng/dl) supporting previous reports of elevated plasma testosterone levels after neonatal flutamide (41). These findings suggest that early flutamide exposure may disrupt androgen-mediated negative feedback. The increased levels of testosterone in the flutamide-treated rats could be associated with changes in levels of other gonadal steroids, although we are not aware of any effect of neonatal androgen blockade on testicular estrogen production, adrenal steroidogenesis, or aromatase activity. We cannot, however, exclude the possibility of an alteration in circulating levels of estrogens.
There are two potential mechanisms through which flutamide could be exerting an effect on corticosterone secretion. It could be due to the prevention of androgen uptake and subsequent activation of ARs in the target cell, or it could be a secondary phenomenon arising from blocked androgen uptake removing the substrate for neuronal aromatase and thus inhibiting the formation of estradiol and the activation of ERs. The latter possibility was directly tested using ATD, which prevented the production of estrogens during the proposed critical stage of gonadal steroid organizational activity in the male rat. This treatment resulted in an enhanced number, frequency, height, and amplitude of corticosterone pulses over the basal 24-h period compared with control males. Corticosterone release in response to a noise stress and an immune-mediated stress was also elevated for ATD-treated males compared with controls and implies a role for the aromatization of testosterone to estrogens in early development in producing a masculinized corticosterone secretion pattern. The presence of nuclear ERs (21) and high levels of aromatizing activity in the hypothalamus of male neonatal and fetal rat brains (20) also suggests that early ER activation in this region may be involved in producing the reduced corticosterone levels found under normal conditions in untreated males. The occurrence of increased ACTH levels and anterior pituitary POMC mRNA after an immune-mediated stress highlights the adenohypophysis as one area in which perinatal aromatization may exert an organizational influence. Additionally, in the present study, hypothalamic CRF, but not AVP, mRNA after LPS administration was enhanced, perhaps indicating an inhibitory effect of aromatized estrogen on neuronal CRF development in the early stages of life in the male rat. The occurrence of an estrogen hormone response element on the CRF gene (49) provides a means by which estrogen may directly act to mediate CRF activity.
No significant differences were found in CBG levels for the flutamide and ATD experimental conditions. As with the control groups, CBG levels decreased significantly 3 h after LPS administration, concurring with previous studies with both acute (50) and chronic (51) stressors. The enhanced corticosterone secretion found after flutamide or ATD treatment thus cannot be accounted for by an elevation in CBG activity as has been proposed to explain some of the differences in corticosterone secretion seen between males and females (52).
Our data demonstrate that prevention of both nonaromatized and aromatized androgen activities disrupts the development of a characteristically masculine corticosterone release profile and stress response in the male rat. However, although significant, the differences in basal pulsatile measurements in ATD compared with controls were not as dramatic as those in the flutamide group compared with their controls. For example, ATD rats had a mean of 14 pulses over the 24 h compared with a control value of 10, whereas a mean number of 18 pulses were found for the flutamide group compared with 11 in their controls. The greater effect in the flutamide-treated than in the ATD-treated animals suggests that both AR and ER activation may be necessary to ensure complete masculinization of the HPA axis and corticosterone secretion profile. The greater effect of flutamide than ATD on PVN GR mRNA may similarly suggest a role for AR in addition to ER. The demonstration that estradiol produces the same masculine behavior in adult castrated rats as testosterone replacement only if administered with dihydrotestosterone (53) further reinforces this dual hormone idea. Accordingly, it may be surmised that testosterone acts via AR activity to mediate AVP expression in the PVN, but acts via ER activity to mediate PVN CRF expression. The suggestion that testosterone acts primarily via PVN AVP inhibition to mediate corticosterone secretion in the adult rat supports this theory (54).
The site(s) at which this perinatal programming occurs is unclear. The presence of both ARs (55) and ERs (56) in the hippocampus of the male rat suggests that this could be an area where androgens might modify the negative feedback properties of hippocampal corticosteroid receptors on the HPA axis. The present study, however, failed to demonstrate any differences in MR or GR mRNA in the subfields of the hippocampus, making it very unlikely that the differential corticosterone secretion arising from our perinatal treatment was caused by changes in corticosteroid receptor negative feedback at this level. In contrast, significantly reduced levels of PVN GR mRNA were found in both flutamide- and ATD-treated rats compared with their respective controls. It would appear, therefore, that hypothalamic GR are sensitive to early life exposure to androgens and may represent an important locus for the perinatal effects of androgens on regulation of the HPA activity by glucocorticoid feedback in the adult.
Our data clearly show a strong influence of pre- and postnatal testosterone activities on the development of a characteristically masculine corticosterone secretion profile and stress response. It appears that testosterone exerts its organizational effects via a combination of both AR and ER activities in the developing neonate. The occurrence of gender-based differences in the susceptibility to a number of disease models in the rodent (57, 58, 59) has been attributed in part to the differential HPA axis activity of male and female rats. Modification of basal and stress-induced corticosterone secretion after manipulation of perinatal androgen activity may have important implications for the susceptibility of the adult rat to these disease models.
References
Arai Y, Gorski RA 1968 The critical exposure time for androgenisation of the developing hypothalamus in the female rat. Endocrinology 82:1010–1014
Raisman G, Field PM 1971 Sexual dimorphism in the preoptic area. Science 173:731–733
Levine S, Haltmeyer GC, Karas GG, Denenberg VH 1967 Physiological and behavioural effects of infantile stimulation. Physiol Behav 2:55–63
Kitay JI 1961 Sex differences in adrenal cortical secretion in the rat. Endocrinology 68:818–824
Critchlow V, Liebelt A, Barsela M, Mountcastle W, Lipscomb HS 1963 Sex differences in resting pituitary-adrenal function in the rat. Am J Physiol 205:807–815
Le Mevel JC, Abitbol S, Beraud G, Maniey J 1979 Temporal changes in plasma adrenocorticotropin concentration after repeated neurotropic stress in male and female rats. Endocrinology 105:812–817
Patchev VK, Hayashi S, Orikasa C, Almeida OFX 1995 Implications of estrogen dependent brain organisation for gender differences in hypothalamo-pituitary-adrenal regulation. FASEB J 9:419–423
Seale JV, Wood SA, Atkinson HC, Bate E, Lightman SL, Ingram CD, Jessop DS, Harbuz MS 2004 Gonadectomy reverses the sexually diergic patterns of circadian and stress-induced hypothalamo-pituitary-adrenal axis activity in male and female rats. J Neuroendocrinol 16:516–524
Weisz JW, Ward IL 1980 Plasma testosterone and progesterone titers of pregnant rats, the male and female fetuses, and neonatal offspring. Endocrinology 106:306–316
Slob AK, Ooms MP, Vreeberg JMT 1980 Prenatal and early postnatal sex differences in plasma and gonadal testosterone and plasma luteinising hormone in female and male rats. J Endocrinol 87:81–87
Grady KL, Phoenix CH, Young WC 1965 Role of the developing rats testis in differentation of the neural tissues mediating mating behaviour. J Comp Physiol Psychol 59:176–182
Pfaff DW, Zigmond RE 1971 Neonatal androgen effects on sexual and nonsexual behaviour of adult rats tested under various hormone regimes. Neuroendocrinology 7:129–145
Husmann DA, McPhaul MJ 1991 Time-specific androgen blockade with flutamide inhibits testicular descent in the rat. Endocrinology 129:1409–1416
Kassim NM, McDonald SW, Reid O, Bennett NK, Gilmore DP, Payne AP 1997 The effects of pre- and postnatal exposure to the nonsteroidal antiandrogen flutamide on testis descent and morphology in the albino Swiss rat. J Anat 190:577–588
Meaney MJ, Stewart J, Poulin P, McEwen BS 1983 Sexual differentiation of social play in rat pups is mediated by the neonatal androgen-receptor system. Neuroendocrinology 37:85–90
Axelson JF, Smith M, Duarte M 1999 Prenatal flutamide treatment eliminates the adult male rat’s dependency upon vasopressin when forming social-olfactory memories. Horm Behav 36:109–118
Husmann DA, Wilson CM., McPhaul MJ, Tilley WO, Wilson JD 1990 Antipeptide antibodies to two distinct regions of the androgen receptor protein localize the receptor protein to the nuclei of cells in the rat and human prostate. Endocrinology 126:2359–2368
Maclusky NJ, Hurlburt PC, Naftolin F 1985 Estrogen formation in the developing rat brain: sex differences in aromatase activity during early post-natal life. Psychoneuroendocrinology 10:355–361
Reddy VVR, Naftolin F, Ryan KJ 1974 Conversion of androstenedione to estrone by neural tissues from fetal and neonatal rats. Endocrinology 94:117–121
George FW, Ojeda SR 1982 Changes in aromatase activity in the rat brain during embryonic, neonatal and infantile development. Endocrinology 111:522–528
Westly BR, Salaman DF 1976 Role of oestrogen receptor in androgen-induced sexual differentiation of the brain. Nature 264:407–408
Booth JE 1977 Sexual behaviour of neonatally castrated rats injected during infancy with oestrogen and dihydrotestosterone. J Endocrinol 72:135–141
Schwarzel WC, Kruggel WG, Brodie HJ 1973 Studies on the mechanism of estrogen biosynthesis. VIII. The development of inhibitors of the enzyme system in human placenta. Endocrinology 92:866–880
Houtsmuller EJ, Brand T, De Jonge FH, Joosten RJNMA, Van de Poll NE, Slob AK 1994 SDN-POA volume, sexual behaviour and partner preference of male rats affected by perinatal treatment with ATD. Physiol Behav 56:535–541
Handa RJ, Nunley KM, Lorens SA, Louie JP, McGivern RF, Bollnow MR 1994 Androgen regulation of adrenocorticotropin and corticosterone secretion in the male rat following novelty and foot shock stressors. Physiol Behav 55:117–124
Viau V, Meaney MJ 1996 The inhibitory effect of testosterone on hypothalamic-pituitary-adrenal responses to stress is mediated by the medial preoptic area. J Neurosci 16:1866–1876
Domínguez-Salazar, E, Portillo W, Baum MJ, Bakker, J, Paredes RG 2002 Effect of prenatal androgen receptor antagonist or aromatase inhibitor on sexual behavior, partner preference and neuronal Fos responses to estrous female odors in the rat accessory olfactory system. Physiol Behav 75:337–346
Windle RJ, Wood SA, Shanks N, Lightman SL, Ingram CD 1998 Ultradian rhythm of basal corticosterone release in the female rat: dynamic interaction with the response to acute stress. Endocrinology 139:443–450
Conde GL, Renshaw D, Zubelewicz B, Lightman SL, Harbuz MS 1999 Central LPS-induced c-fos expression in the PVN and the A1/A2 brainstem noradrenergic cell groups is altered by adrenalectomy. Neuroendocrinology 70:175–185
Bradford MA 1976 A rapid and sensitive method for quantitation of microgram quantities of protein utilising the principle of protein-dye binding. Anal Biochem 72:248–254
Harbuz MS, Lightman SL 1989 Responses of hypothalamic and pituitary mRNA to physical and psychological stress in the rat. J Endocrinol 122:705–711
Young III WS, Mezey E, Siegel RE 1986 Vasopressin and oxytocin mRNAs in adrenalectomised and Brattleboro rats: analysis by quantitative in situ hybridisation histochemistry. Mol Brain Res 1:231–241
Young III WS, Mezey E, Siegel RE 1986 Quantitative in situ hybridisation histochemistry reveals increased levels of corticotrophin-releasing factor mRNA after adrenalectomy in rats. Neurosci Lett 70:198–203
Takahashi H, Hakamata Y, Wantabe Y, Kikuno R, Miyata T, Numa S1983 POMC probe. Nucleic Acids Res 11:6647–6858
Kinoshita H, Jessop DS, Finn DP, Harbuz MS 2000 Cyanamide-induced activation of the hypothalamo-pituitary-adrenal axis. J Neuroendocrinol 12:255–262
Seckl JR, Dickson KL, Yates C, Fink G 1991 Distribution of glucocorticoid and mineralocorticoid receptor messenger RNA expression on human postmortem hippocampus. Brain Res 562:332–337
Merriam GR, Wachter KW 1982 Algorhythms for the study of episodic hormone secretion. Am J Physiol 243:E310–E318
Windle RJ, Wood SA, Lightman SL, Ingram CD 1998 The pulsatile characteristics of the hypothalamo-pituitary-adrenal activity in female Lewis and Fischer 344 rats and its relationship to differential stress responses. Endocrinology 139:4044–4052
Festing MFW, Overend P, Gaines Das R, Borja MC, Berdoy M 2002 The design of animal experiments. London: Royal Society of Medicine Press
Harbuz MS, Windle RJ, Jessop DS, Renshaw D, Ingram C.D, Lightman SL 1999 Differential effects of psychological and immunological challenge on the hypothalamo-pituitary-adrenal axis function in adjuvant-induced arthritis. Ann NY Acad Sci 876:43–52
McCormick CM, Mahoney E 1999 Persistent effects of prenatal, neonatal, or adult treatment with flutamide on the hypothalamic-pituitary-adrenal stress response of adult male rats. Horm Behav 35:902–101
Levine S 1957 Infantile experience and resistance to physiological stress. Science 126:406–406
Meaney MJ, Aitkin DH, Viau V, Sharma S, Sarrieau A 1989 Neonatal handling alters adrenocortical negative feedback sensitivity and hippocampal type II glucocorticoid receptor binding in the rat. Neuroendocrinology 50:597–604
Viau V, Sharma S, Plotsky PM, Meaney MJ 1993 Increased plasma ACTH responses to stress in nonhandled compared with handled rats require levels of corticosterone and are associated with increased levels of ACTH secretagogues in the median eminence. J Neurosci 13:1097–1105
Shanks N, Windle RJ, Harbuz MS, Jessop DJ, Ingram CD, Lightman SL 2000 Early-life exposure to endotoxin alters hypothalamic-pituitary-adrenal function and predisposition to inflammation. Proc Natl Acad Sci USA 97:5645–5650
Phoenix CH, Goy RW, Gerall AA, Young WC 1959 Organising action of prenatally administered testosterone propionate on the tissues mediating mating behaviour in the female guinea pig. Endocrinology 65:369–382
Gorski RA, Gordon JH, Shryne JE, Southam AM 1978 Evidence for a morphological sex difference within the medial preoptic area of the rat brain. Brain Res 148:333–346
Viau V, Meaney J 1991 Basal and stress hypothalamic-adrenal activity in cycling and ovariectomised-steroid treated rats. Endocrinology 129:2503–2511
Vamvakopolous NC, Chrousos GP 1993 Structural organization of the 5' flanking region of the human corticotropin releasing hormone gene. DNA Sequencing Mapping 4:197–206
Tannenbaum B, Rowe W, Sharma S, Diorio J, Steverman A, Walker M, Meaney MJ 1997 Dynamic variations in plasma corticosteroid-binding globulin and basal activity following acute stress in adult rats. J Neuroendocrinol 9:163–168
Neufeld JH, Breen L, Hauger R 1994 Extreme posture elevate corticosterone in a forced ambulation model of chronic stress in rats. Pharmacol Biochem Behav 47:223–240
Gala RR, Westphal U 1965 Corticosteroid-binding globulin in the rat: studies on the sex difference. Endocrinology 77:841–851
Baum MJ 1979 Differentation of coital behaviour in mammals: a comparative analysis. Neurosci Biobehav Rev 3:265–284
Viau V, Chu A, Soriano L, Dallman M 1999 Independent and overlapping effects of corticosterone and testosterone on corticotropin-releasing hormone and arginine vasopressin mRNA expression in the paraventricular nucleus of the hypothalamus and stress-induced adrenocorticotropic hormone release. J Neurosci 19:6684–6693
Kerr JE, Allore RJ, Beck, SG, Handa RJ 1995 Distribution and hormonal regulation of androgen receptor (AR) and AR messenger ribonucleic acid in the rat hippocampus. Endocrinology 136:3213–3221
O’Keefe, JA, Handa RJ 1990 Transient increases in estrogen receptor in the neonatal rat hippocampus. Brain Res Dev Brain Res 57:119–127
Schuurs AH, Verheul HA 1990 Effects of gender and sex steroids on the immune response. J Steroid Biochem 35:157–172
Homo-Delarche F, Fitzpatrick F, Christeff N, Nunez EA, Bach JF, Dardenne M 1991 Sex steroids, glucocorticoids stress and autoimmunity. J Steroid Biochem 40:619–637
Whitacre CC, Dowdell K, Griffin AC 1998 Neuroendocrine influences on experimental autoimmune encephalomyelitis. Ann NY Acad Sci 840:705–716(J. V. Seale, S. A. Wood, )
Address all correspondence and requests for reprints to: Dr. S. Lightman, Laboratories for Integrative Neuroscience and Endocrinology, Dorothy Hodgkin Building, Whitson Street, University of Bristol, Bristol BS1 3NY, United Kingdom. E-mail: stafford.lightman@bristol.ac.uk.
Abstract
Organizational effects of testosterone during a critical period of neonatal life have major irreversible effects on adult sexual behavior. We have investigated whether perinatal androgen changes also affect another major sexually differentiated system, the hypothalamo-pituitary-adrenal axis. This was assessed in male rats who had been exposed to perinatal flutamide or 1,4,6-androstatriene-3,17-dione (ATD). Once the animals reached adulthood, an automated sampling system was used to collect blood from freely moving animals at 10-min intervals over 24 h, followed by a noise stress and then the administration of lipopolysaccharide (LPS). Perinatal flutamide- and ATD-treated rats not only had higher mean corticosterone levels and increased frequency and amplitude of corticosterone pulses over the 24 h compared with vehicle-injected controls, but they also showed markedly increased corticosterone responses to both noise and LPS. All parameters of increased hypothalamo-pituitary-adrenal activity resembled the normal physiological state of the intact adult female rather than that of the intact adult male rat. Furthermore, 3 h after LPS administration, both flutamide- and ATD-treated animals had markedly higher levels of corticotropin-releasing factor mRNA in the parvocellular paraventricular nucleus (PVN) and proopiomelanocortin mRNA in the adenohypophysis. Flutamide-treated rats also had a greater level of PVN arginine vasopressin mRNA. PVN glucocorticoid receptor mRNA levels were significantly lower in both the flutamide- and the ATD-treated male rats. These data highlight the importance of perinatal exposure to both testosterone and estrogen(s) on the development of a masculinized circadian corticosterone profile and stress-induced hypothalamo-pituitary-adrenal axis activity in the adult male rat.
Introduction
AN INFLUENCE OF GONADAL steroids on the development of sexually dimorphic brain regions has been known since the 1930s. The organizational effects of androgens were first described by Arai and Gorski in the late 1960s (1), and the sexual dimorphism of brain structure was described by Raisman and Field in 1971 (2). Around the same time, Levine et al. (3) investigated the effect of neonatal stimulation on adult hypothalamo-pituitary-adrenal (HPA) activity and clearly showed that this could be programmed by neonatal stress.
The presence of sexual dimorphism in corticosterone secretion was originally reported by Kitay (4). On the basis of this research, we and others have demonstrated an enhanced basal and stress-induced HPA axis and subsequent corticosterone response for adult female compared with male rats (5, 6, 7, 8). The different levels of circulating gonadal steroids in the two genders are considered integral in producing this divergent HPA axis activity in the adult. However, the question arises of whether this sexually dimorphic HPA axis is the product of a gonadal steroid influence at the perinatal stage.
Male rats undergo two testosterone surges, one between gestational d 17 and 18 (9) and another 1–3 h post parturition (10). These surges are vital for both the anatomical (11) and behavioral (12) masculinization of male rats. The effects of these surges can be antagonized (13, 14, 15, 16) by antiandrogens such as flutamide, a potent nonsteroidal antiandrogen that inhibits androgen uptake and nuclear binding in target tissues (17). The influence of perinatal testosterone may not, however, be solely mediated through the androgen receptor (AR), because testosterone can also be aromatized to estradiol, which can, in turn, exert effects through the estrogen receptor (ER) (18). Indeed, there are prevalent aromatizing enzymes (19), high levels of aromatization (20), and functional hypothalamic ERs (21) in the newborn male rat brain. The ability of postnatal injections of estrogen, but not nonaromatizable dihydrotestosterone, to suppress lordotic and female gonadotropin cyclic activities in neonatally castrated male rats (22) supports this aromatization hypothesis. Furthermore, prenatal and postnatal treatments with a specific aromatization inhibitor, 1,4,6-androstatriene-3,17-dione (ATD) (23), reduce male sexual behavior in the adult rat (24).
Current findings implicate an important role for testosterone in shaping masculine behavior, although it is unclear whether such an effect is taking place via AR or aromatization and subsequent ER activity. Additionally, the majority of studies examining the effect of early gonadal action on the adult male rat have focused on behavioral or anatomical aspects. In contrast, the effects of early life manipulation of gonadal steroids on HPA axis activity have received comparatively little attention. The ability of castration to enhance basal and stress-induced corticosterone release in adult male rats significantly above those of intact or gonadally replaced male rats (8, 25, 26) demonstrates a mediatory effect of androgens on HPA activity in the adult. However, whether the behavioral effects of early exposure to gonadal steroids are reflected in organizational changes in adult male corticosterone release and HPA axis activity remains to be determined. Demonstration of such an influence may help us to understand the mechanisms underlying the sexually differentiated corticosterone profiles found for adult male and female rats.
An automated sampling system that allows the collection of blood samples over a prolonged period was used to examine the 24-h corticosterone profile of feminized male rats compared with controls under basal conditions. The release pattern of basal corticosterone influences corticosterone secretion in response to stressful stimuli. Accordingly, the effects of feminization on corticosterone secretion after an acute (noise) and an immune-mediated (LPS) stress were also studied using the automated sampling system. The unique ability of this system to collect blood samples in a remote manner over a considerable number of hours allows for a very detailed analysis of corticosterone release while minimizing any confounding factors associated with conventional manual sampling techniques. To elucidate the potential mechanisms through which neonatal testosterone might result in altered adult HPA axis activity, we used two treatment paradigms. The first administered flutamide pre- and postnatally to assess the effect of androgen uptake into target cells on subsequent adult corticosterone secretion. The second administered ATD both pre- and postnatally to examine the effects of aromatization on adult corticosterone release profiles. Changes in circulating levels of corticosteroid-binding globulin (CBG) could be confounding influences on the interpretation of plasma corticosterone levels, with greater levels in the female animal buffering the effects of higher circulating corticosterone levels (4, 5). We therefore measured both basal and stress-induced CBG levels in our experimental groups.
Materials and Methods
Flutamide experimental animals
Each week, two Sprague Dawley female rats were time-mated with a male Sprague Dawley rat in-house, so that two litters were born every week over a 1-month period. Four dams and the subsequent litters (10–14 rats/litter) were used for the flutamide group, and four dams and litters were used for the control group. Flutamide was dissolved in a vehicle of 5% ethanol in sunflower oil. All pregnant dams were individually housed and injected (sc) every day in the dorsum of the neck with flutamide (100 mg/kg body weight) from d 13 of gestation to parturition. Upon birth, male pups were injected sc with flutamide (50 mg/kg body weight) every day until postnatal d 20. This protocol has been shown to effectively feminize male pups (13). Pups remained with the dam until weaned (21 d post parturition). During the early postnatal days, both female and male pups were administered flutamide, because similarities in the anogenital distances of the males and females made accurate sexing impossible. Females were culled once their sex could be determined (3–4 wk post parturition). Control pregnant dams and resultant pups were injected with vehicle only, using the same time regimen as the flutamide group. All animals from both groups were handled for a similar amount of time during flutamide or vehicle administration.
ATD experimental animals
Each week two Sprague Dawley female rats were time-mated with a male Sprague Dawley rat, so that two litters were born per week in-house over the period of 1 month. Four dams and the subsequent litters (10–14 rats/litter) were used for the ATD group, and four dams and litters were used for the control group. For the ATD group, individually housed, pregnant, Sprague Dawley dams received two sc implants containing ATD (2.5 cm; outside diameter, 2.1 mm; inside diameter, 1.5 mm) on d 13 of gestation. Upon birth, male pups were injected sc with ATD (dissolved in a vehicle of 5% ethanol in sunflower oil) at a dosage of 1 mg/pup every other day from birth (d 0) to d 12. This method is effective in preventing aromatization of testosterone surges in the male rat (27). Using the same time protocol as that in the ATD group, control pregnant dams received two sc implants containing cholesterol, and subsequent male pups were injected every other day until d 12 with vehicle only. Female pups were culled at the time of birth in both ATD and control groups. All animals from both groups were handled for a similar amount of time during ATD or vehicle administration.
Automated sampling system surgery
Once treatment was finished, male pups from each group were left to grow up in cages of four under a 14-h light, 10 -h dark schedule with ad libitum access to food and water. At 9–10 wk of age (275–300 g), all male rats were anesthetized using Hypnorm (0.32 mg/kg fentanyl citrate and 10 mg/kg fluanisone; Janssen Pharmaceuticals, Buckinghamshire, UK) and diazepam (2.6 mg/kg; Phoenix Pharmaceuticals, St. Joseph, MO). A SILASTIC brand (Dow Corning, Midland, MI)-tipped polythene cannula (inside diameter, 0.58 mm; Portex, Hythe, UK) filled with heparinized saline (10 U/ml heparin; CP Pharmaceuticals Ltd., Wrexham, Wales, UK) was inserted into the right jugular vein of each rat. The cannula was exteriorized at the crown of the rat’s head and protected by a spring attached to a 360° mechanical swivel. Animals were connected to an automated blood sampling system, allowing the collection of blood samples at preset time points (28). All animal procedures were carried out in accordance with the Animal (Scientific Procedures) Act of 1986. Rats from both studies underwent the same sampling protocol and the same assays on their blood and tissues. Hematocrits were calculated after the sampling procedures and never fell below 23%. This fall in hematocrit was not associated with any change in HPA axis activity or any alteration in behavior as recorded by remote video recording.
Sampling procedure
Sampling began at 0700 h on d 5 post surgery. Blood samples were collected every 10 min from each rat over a period of 24 h. Blood (37.7 μl) was removed for each sample and replaced with heparinized saline. At 0700 h (lights on at 0500 h) on d 6 post surgery, a noise generator was used to expose rats to a 10-min white noise stress (110 db). Blood samples were taken every 10 min for 2 h after noise onset. At 0900 h (d 6; lights on at 0500 h), 100 μl LPS (Escherichia coli; 055:B5, 250 μg/ml; Sigma-Aldrich Corp., Poole, UK) was administered to each rat through the iv cannula. Samples were collected every 15 min for an additional 3 h after LPS administration. At 1200 h, rats were overdosed with pentobarbitone (iv) and decapitated. Trunk blood, brains, and pituitaries were collected. Previous work reported LPS-induced increases in corticotropin-releasing factor (CRF) mRNA using this serotype 3 h after LPS administration (29).
Corticosterone RIA
The sampling protocol produced 169 blood samples/rat. During sampling, each 37.7-μl blood sample was diluted 1:5 in heparinized saline. Fifty microliters of each blood sample were further diluted into 50 μl citrate buffer (pH 3.0) and incubated overnight at 4 C with 50 μl [125I]corticosterone tracer (ICN Biomedicals, Aurora, OH) and 50 μl rabbit antirat corticosterone primary antibody (donated by G. Makara, Institute of Experimental Medicine, Budapest, Hungary). All samples were processed in duplicate. On d 2, a charcoal/dextran solution was added to all samples, which were then centrifuged (15 min, 4000 rpm, 4 C) and aspirated before being loaded onto a -counter. Intra- and interassay coefficients of variation for the corticosterone assays were 12.4% and 16%, respectively.
ACTH RIA
Plasma derived from trunk blood taken 3 h after LPS administration was analyzed for ACTH using a rabbit antirat ACTH primary antibody (donated by G. Makara) and [125I]ACTH (Amersham Biosciences, Little Chalfont, UK). A polyethylene glycol solution, sheep antirabbit secondary antibody (1:50 in standard assay buffer with 0.4% normal rabbit serum), and centrifugation were used to separate the bound from the unbound hormone fraction into a pellet. The resultant pellets were counted on a -counter.
CBG assay
Plasma samples from 0800–0850 h before LPS administration (d 6 post surgery) were pooled and assayed to provide pre-LPS CBG levels. Plasma collected 3 h post LPS administration were used to assess post-LPS levels. Plasma samples were stripped of endogenous corticosterone, eluted, and incubated overnight in a total binding or nonspecific binding [3H]corticosterone solution as previously described (28). On d 2, incubates were processed in triplicate through LH20 columns to separate bound and unbound steroids before being eluted and counted on a ?-scintillation counter (28). The Bradford method was used to determine protein content (30). Resultant values are reported as picomoles of [3H]corticosterone bound per milligram of protein.
Oligonucleotide in situ hybridization
Paraventricular nucleus (PVN) and pituitary sections (12 μm) were cut and mounted onto gelatin-coated slides. Sections were fixed in 4% formaldehyde (5 min) and taken through a prehybridization washing procedure as previously described (31). Slides were incubated overnight at 37 C in hybridization buffer (31) containing 1 M dithiothreitol and the required [35S]deoxy-ATP oligonucleotide probe. Approximately 100,000–200,000 cpm were applied to each slide in 45 μl hybridization buffer. The specificity of the oligonucleotide probes have been previously determined for arginine vasopressin (AVP) (32), CRF (33), and proopiomelanocortin (POMC) (34). The specific activities of the probes were 1.82 x 1018 dpm/mol (CRF), 2.03 x 1018 dpm/mol (AVP), and 1.75 x 1018 dpm/mol (POMC). For each study, all control and experimental sections were hybridized in the same reaction. On d 2, slides were taken through four changes of 1x saline sodium citrate (SSC) and washed in 1x SSC at 55 C (four times, 15 min each time) and in 1x SSC at room temperature (2 x 30 min), dipped in distilled water, and dried. Slides were exposed (AVP, 3 d; POMC, 5 d; CRF, 14 d) to Hyperfilm MP (Amersham Biosciences). Analysis of mRNA levels in comparison to 14C-labeled standards was achieved using image analysis software (Image 1.6.2, W. Rasband, NIH, Bethesda, MD). POMC mRNA was determined by measuring gray levels over the anterior pituitary, and CRF mRNA was determined using a threshold method to highlight probe bound over the nucleus (31, 35). To differentiate parvocellular from magnocellular AVP mRNA content, a threshold method was used to threshold out the magnocellular cells and therefore measure signal over the parvocellular cells only (36).
Riboprobe in situ hybridization
Antisense 35S-labeled glucocorticoid receptor (GR) and mineralocorticoid receptor (MR) riboprobes (donated by J. Seckl, University of Edinburgh, Edinburgh, Scotland) were used to detect GR mRNA expression in PVN and hippocampal sections (12 μm) and MR mRNA in hippocampal sections only. GR and MR were reverse transcribed as previously described (36). Sections were fixed in 4% paraformaldehyde, processed through prehybridization washes, and incubated overnight at 55 C in hybridization buffer containing the required 35S-radiolabeled riboprobe (40). Approximately 1,000,000 cpm were applied to each slide in 45 μl hybridization buffer. The specific activities of the probes were 1.68 x 1019 dpm/mol (GR) and 1.40 x 1019 dpm/mol (MR). On d 2, sections were taken through three changes of 1x SSC before being washed in 50% formamide in 1x SSC three times for 15 min each time at 55 C. Sections were briefly dipped in 1x SSC at 37 C and incubated in ribonuclease A (25 mg/ml; 37 C) for 30 min. Slides were dipped in 1x SSC (37 C), washed in 50% formamide in 1x SSC (three times, 15 min each time, 55 C), 1x SSC (twice, 5 min each time, room temperature), and dipped in distilled water. Slides were left to air-dry before exposure to Hyperfilm for 2 wk. Image analysis software was used to analyze resultant mRNA levels in subregions of the hippocampus or in the PVN. Sense 35S-labeled GR and MR riboprobes were included as negative controls and displayed no radioactive signal on the treated hippocampal and PVN sections.
Statistical analysis
For each study, data are presented as mean plasma corticosterone blood levels within each group over the 24-h basal period. A Pulsar analysis program (37) analyzed pulse number, frequency, amplitude and height. Unpaired t tests were used to detect any differences in these parameters between the two groups for each study.
For statistical analysis of the postnoise stress response (noise onset occurred at 0700 h on d 6 post surgery), the mean corticosterone level 20 min after noise onset was averaged for each group. The point 20 min after noise onset was chosen due to previous findings demonstrating maximal corticosterone secretion at this time (38). A baseline control value was obtained using the mean of values from 0600–0650 h (d 6 post surgery) according to the recommendation of Festing et al. (39). For the two studies, an unpaired t test compared the stress responses of the group. To clarify any statistical differences in corticosterone secretion occurring at these discrete time points between the treatment groups, an area under the curve measurement from the time of noise onset (0700 h on d 6 post surgery) to the time at which values had returned to baseline (0750 h on d 6) was calculated for each rat. Unpaired t tests compared area under the curve measurements between the flutamide and control groups and between the ATD and control groups.
A similar statistical procedure analyzed corticosterone levels after LPS administration (LPS injected at 0900 h on d 6 post surgery). In this situation, the mean of values from 0800–0850 h on d 6 post surgery was used to calculate a baseline control value. Mean corticosterone levels 90 min after LPS administration were used as an indication of the LPS-induced stress response. Previous findings have shown maximal corticosterone secretion after LPS exposure to occur between 60–120 min post administration (40). For additional statistical analysis of the LPS stress response, an area under the curve measurement, using the trapezoid rule, from LPS administration (0900 h) to experimental termination (1200 h on d 6) was calculated for each rat. Unpaired t tests were used to compare area under the curve measurements between flutamide and control and between ATD and control groups.
Unpaired t tests were used to compare post-LPS ACTH levels for both studies. ANOVA was used to compare pre- and post-LPS CBG levels within and between the two groups in each study.
For mRNA analysis, a minimum of four PVN or hippocampal or six pituitary sections were analyzed for each rat. A mean value of the sections per rat was then calculated, so that each rat contributed a single mean value, which was used to calculate a group mean. Statistical analysis was performed on the dataset before calculation of percent change in values used for comparative purposes only.
Results
Flutamide study
Previous research has reported disrupted testes descent and/or epididymus development (13, 14) and enhanced testosterone levels (41) in flutamide-treated compared with control males. In the present study, all flutamide-treated rats demonstrated disrupted testes descent and/or epididymus development, whereas no such disruption was found in control males. Significantly enhanced (P < 0.01) levels of testosterone in the plasma of flutamide-treated (mean ± SE, 1.88 ± 0.09 ng/ml) compared with control (mean SE, 1.58 ± 0.1 ng/ml) males was also found.
Blood samples taken every 10 min over a period of 24 h were assayed for corticosterone from seven male adults pre- and postnatally exposed to flutamide (flutamide group) and seven male adults pre- and postnatally exposed to sunflower oil (control group). A mean basal 24-h corticosterone profile for each group was calculated from the mean corticosterone level for each 10-min point (Fig. 1). The mean group profiles demonstrated a higher corticosterone secretion pattern for flutamide compared with the control group. Flutamide males had a significantly greater number of corticosterone pulses over the 24-h period compared with controls (P < 0.001). A significantly increased pulse frequency per hour (P < 0.001), mean pulse height (P < 0.001), and mean pulse amplitude (P < 0.001) were also found for flutamide compared with control rats (Table 1).
FIG. 1. Mean blood corticosterone levels for flutamide-treated and control male rats over a 24-h period. Each 10-min data point represents the mean for each group (n = 7 rats/group). , Dark phase (1900–0500 h).
TABLE 1. Mean ± SEM of pulsar parameter measurements for flutamide and control male rats (n = 7 rats per group)
After the 24-h blood collection, rats received a 10-min noise stress. Blood samples were collected for 2 h after noise onset, and mean values for each rat group were calculated (Fig. 2A). Flutamide rats had significantly increased corticosterone levels at the 20 min postnoise time point chosen for comparison purposes (P < 0.05). The area under the curve from 0700–0750 h after noise stress was significantly greater for the flutamide (mean, 4767 ± 141) compared with the control (mean, 2333 ± 138) rats (P < 0.01).
FIG. 2. A, Mean (±SEM; n = 7 rats/group) blood corticosterone levels of flutamide-treated and control male rats 30 min before and 90 min after noise stress administration (, noise stress from 0700–0710 h; lights on at 0500 h). *, P < 0.05 compared with controls. Statistical comparison was made between the groups at the 20 min point according to the recommendations of Festing et al. (39 ). B, Mean (±SEM; n = 7 rats/group) corticosterone levels of flutamide-treated and control male rats 30 min before and 180 min after LPS administration at 0900 h (indicated by arrow; lights on at 0500 h). *, P < 0.001 compared with controls. Statistical comparison was made between the groups 90 min after LPS administration.
LPS administration at 0900 h was followed by blood collection every 15 min for an additional 3 h. The corticosterone profile after LPS for each treatment group was determined (Fig. 2B). Flutamide males had significantly higher corticosterone levels than controls (P < 0.001) at the chosen comparative time point (90 min after LPS). A significantly higher area under the curve measurement after LPS administration was found for the flutamide (mean, 36605 ± 1960) compared with the control (mean, 20268 ± 1216) males (P < 0.001).
Plasma collected 3 h post LPS administration was assayed for ACTH levels. Flutamide males (mean, 414 ± 37 pg/ml; n = 7) had significantly (P < 0.001) higher ACTH plasma levels post LPS compared with controls (mean, 175 ± 28 pg/ml; n = 7). Post-LPS CBG levels were significantly reduced in comparison to pre-LPS CBG levels for both groups (P < 0.001). There were no significant differences in CBG levels between the two groups at analogous time points (Fig. 3).
FIG. 3. Mean (±SEM; n = 7 rats/group) plasma CBG (picomoles per mg protein) levels for flutamide and control male rats pre- and postLPS administration. *, P < 0.001 compared with post-LPS condition.
CRF, AVP, and GR mRNAs in the PVN and POMC mRNA in the anterior pituitary at the termination of the experiment are shown in Table 2. Significantly increased levels of parvocellular PVN AVP, PVN CRF, and anterior pituitary POMC mRNA were found for flutamide males compared with controls (P < 0.001). GR PVN mRNA levels were significantly lower in flutamide compared with control males (P < 0.001). No significant differences between the two groups were found in MR and GR mRNA levels for the CA1, CA2, CA3, and dentate gyrus hippocampal regions (Fig. 4).
TABLE 2. Mean percentage change in PVN CRF, AVP, and GR mRNA and anterior pituitary POMC mRNA levels compared with control males (n = 8 rats per group)
FIG. 4. Mean MR and GR mRNA levels in the CA1, CA2, CA3, and dentate gyrus hippocampal regions of flutamide rats as a percentage of the control value (n = 8 rats/group). No significant differences in GR or MR mRNA levels in the different hippocampal regions were found between the treatment groups.
ATD study
Mean blood corticosterone levels over a basal 24-h period demonstrated an enhanced corticosterone secretion profile for ATD-treated males compared with oil-treated males (controls; Fig. 5). Pulsar analysis demonstrated a significantly greater number of pulses over 24 h for ATD compared with control male rats (P < 0.01). ATD males also demonstrated significantly higher levels of pulse amplitude (P < 0.001), height (P < 0.05), and frequency (P < 0.05) compared with controls (Table 3).
FIG. 5. Mean corticosterone levels for ATD and control male rats over a 24-h period. Each 10-min data point represents the mean for each group (n = 7 rats/group). , Dark phase (1900–0500 h).
TABLE 3. Mean ± SEM of pulsar parameter measurements for ATD and control male rats (n = 7 rats per group)
Mean corticosterone levels after noise onset are shown in Fig. 6A. At the comparative 20 min point, ATD-treated males showed significantly increased levels of corticosterone secretion compared with controls (P < 0.001). The postnoise area under the curve was significantly greater for ATD (mean, 4273 ± 153) compared with control (mean, 2073 ± 152) rats (P < 0.01).
FIG. 6. A, Mean (±SEM; n = 7 rats/group) corticosterone levels of ATD and control male rats 30 min before and 90 min after noise stress administration (, noise stress from 0700–010 h; lights on at 0500 h). *, P < 0.001 compared with controls Statistical comparison was made between the groups at the 20 min point. B, Mean (±SEM; n = 7 rats/group) corticosterone levels of ATD and control male rats 30 min before and 180 min after LPS administration at 0900 h (indicated by arrow; lights on at 0500 h). *, P < 0.001 compared with controls. Statistical comparison was made between the groups 90 min after LPS administration.
Mean corticosterone secretion every 15 min for 3 h post LPS administration was calculated for each group (Fig. 6B). Significantly elevated corticosterone levels were found for ATD compared with control males at the 90 min point used for comparative purposes (P < 0.001). A significantly higher post-LPS area under the curve was demonstrated for ATD (mean, 29,057 ± 1,119) compared with control (mean, 16,694 ± 1,349) males (P < 0.01).
Plasma ACTH levels 3 h post LPS exposure demonstrated significantly (P < 0.001) higher levels for the ATD (mean, 317 ± 29 pg/ml; n = 7) compared with control rats (mean, 168 ± 23 pg/ml; n = 7). Pre-LPS CBG levels were significantly higher compared with post-LPS CBG levels for both ATD and control groups (P < 0.001). Treatment group had no effect on CBG levels (Fig. 7).
FIG. 7. Mean (±SEM; n = 7 rats/group) plasma CBG (picomoles per mg protein) levels for ATD and control male rats before and after LPS administration. *, P < 0.001 compared with post-LPS condition.
Mean percent changes in PVN CRF, AVP, and GR and anterior pituitary POMC mRNA at the end of the experiment were calculated for the ATD group compared with the control group (Table 4). ATD males demonstrated significantly greater levels of PVN CRF mRNA (P < 0.05). No significant differences in PVN AVP mRNA were found for ATD males compared with control males. PVN GR mRNA was significantly lower in the ATD males compared with controls (P < 0.001). Anterior pituitary POMC mRNA levels were significantly higher in ATD males compared with controls (P < 0.05). GR and MR mRNA levels in the hippocampal CA1, CA2, CA3, and dentate gyrus regions were not significantly different between the two treatment groups (Fig. 8).
TABLE 4. Mean percentage change in PVN CRF, AVP, and GR mRNA and anterior pituitary POMC mRNA levels compared with control males (n = 8 rats per group)
FIG. 8. Mean MR and GR mRNA levels in the CA1, CA2, CA3, and dentate gyrus hippocampal regions of ATD rats as a percentage of the control value (n = 7 rats/group). No significant differences in GR or MR mRNA levels in the different hippocampal regions were found between the treatment groups.
Discussion
These novel data demonstrate an increase in the basal corticosterone secretion of male adult rats experimentally deprived of pre- and postnatal testosterone activity by exposure to the antiandrogen flutamide or the aromatase inhibitor ATD. The enhanced secretion of corticosterone was characterized by an increase in the number, frequency, height, and amplitude of corticosterone pulses. These data demonstrate that feminization not only altered the amount of corticosterone secreted in adulthood, but also the underlying characteristics of such release. Additionally, flutamide- and ATD-treated rats had significantly elevated corticosterone secretion after noise stress and LPS administration compared with their controls. The endocrine differences between the experimental groups were also mirrored by significant changes in hypothalamic and pituitary transcripts at the end of the experimental period, particularly an increase in PVN CRF mRNA (and parvocellular AVP mRNA in the flutamide group) and adenohypophyseal POMC mRNA as well as decreased PVN GR mRNA in both the flutamide- and ATD-treated groups. Therefore, we have good evidence for an organizational effect of endogenous androgens, probably acting through both ARs and ERs to increase the activity of the HPA axis in the adult animal.
The perinatal period is critically important for the maturation and programming of neuroendocrine functions within the central nervous system. As early as 1957, Levine demonstrated a link between neonatal experience and the response to stress of the same animal when it reached adulthood (42). More recently, Meaney and others (43, 44, 45) have demonstrated how variations in handling procedures or neonatal immunological challenges can program different patterns of HPA responsiveness in the adult. In addition to its ability to respond to environmental conditions, the perinatal brain is also very sensitive to changes in circulating androgens. Phoenix and colleagues (46) were the first to show that testosterone treatment of pregnant guinea pigs resulted in masculinized sexual activity of the female offspring, and Gorski et al. (47) demonstrated that gender differences in the sexually dimorphic nucleus were due to variations in neonatal sex hormone levels as opposed to levels circulating during adulthood. We have previously reported that under basal conditions, the number, frequency, height, and amplitude of corticosterone pulses over 24 h are significantly enhanced in adult female and castrated male rats compared with intact males (8). What has not been investigated, however, is the effect of neonatal androgenization on the developmental programming of the HPA axis. Because rats show a major sexual dimorphism in HPA axis activity with females exhibiting greater basal activity and a considerably larger response to stress (4, 5, 6, 7, 8, 48), it is clearly possible that both activational and organizational effects of sex hormones may be important. Our research has examined this possibility.
In the present study flutamide-treated rats had a mean of 18 pulses over the basal 24-h period at a frequency of 0.76 pulses/h. These values are closer to the enhanced values previously reported (8) for both intact females (mean number, 24; mean frequency, 1) and castrated males (mean number, 22; mean frequency, 0.95) compared with those of intact males (mean number, 11; mean frequency, 0.49). The amplitude and height of the corticosterone pulses of flutamide-treated rats over the basal period also approximate those previously found for intact females and castrated males (8). Additionally, an excitatory influence of flutamide exposure was found on stress-induced corticosterone release and the ACTH and HPA axis response to an immune-mediated challenge (LPS). These findings support an organizational role for testosterone in producing a characteristically masculine corticosterone profile and HPA axis stress response in the adult male rat.
The fact that flutamide results in reduced testes descent and/or underdevelopment of the epididymus (13, 14) suggests that testosterone production is reduced in these rats. Given the inhibitory effects of testosterone on corticosterone secretion, it may be surmised that such reduced testosterone levels are responsible for the enhanced corticosterone and HPA axis activity evidenced in flutamide-treated rats. However, the plasma testosterone levels of the flutamide group (1.88 ng/dl) were significantly greater than those in the controls (1.58 ng/dl) supporting previous reports of elevated plasma testosterone levels after neonatal flutamide (41). These findings suggest that early flutamide exposure may disrupt androgen-mediated negative feedback. The increased levels of testosterone in the flutamide-treated rats could be associated with changes in levels of other gonadal steroids, although we are not aware of any effect of neonatal androgen blockade on testicular estrogen production, adrenal steroidogenesis, or aromatase activity. We cannot, however, exclude the possibility of an alteration in circulating levels of estrogens.
There are two potential mechanisms through which flutamide could be exerting an effect on corticosterone secretion. It could be due to the prevention of androgen uptake and subsequent activation of ARs in the target cell, or it could be a secondary phenomenon arising from blocked androgen uptake removing the substrate for neuronal aromatase and thus inhibiting the formation of estradiol and the activation of ERs. The latter possibility was directly tested using ATD, which prevented the production of estrogens during the proposed critical stage of gonadal steroid organizational activity in the male rat. This treatment resulted in an enhanced number, frequency, height, and amplitude of corticosterone pulses over the basal 24-h period compared with control males. Corticosterone release in response to a noise stress and an immune-mediated stress was also elevated for ATD-treated males compared with controls and implies a role for the aromatization of testosterone to estrogens in early development in producing a masculinized corticosterone secretion pattern. The presence of nuclear ERs (21) and high levels of aromatizing activity in the hypothalamus of male neonatal and fetal rat brains (20) also suggests that early ER activation in this region may be involved in producing the reduced corticosterone levels found under normal conditions in untreated males. The occurrence of increased ACTH levels and anterior pituitary POMC mRNA after an immune-mediated stress highlights the adenohypophysis as one area in which perinatal aromatization may exert an organizational influence. Additionally, in the present study, hypothalamic CRF, but not AVP, mRNA after LPS administration was enhanced, perhaps indicating an inhibitory effect of aromatized estrogen on neuronal CRF development in the early stages of life in the male rat. The occurrence of an estrogen hormone response element on the CRF gene (49) provides a means by which estrogen may directly act to mediate CRF activity.
No significant differences were found in CBG levels for the flutamide and ATD experimental conditions. As with the control groups, CBG levels decreased significantly 3 h after LPS administration, concurring with previous studies with both acute (50) and chronic (51) stressors. The enhanced corticosterone secretion found after flutamide or ATD treatment thus cannot be accounted for by an elevation in CBG activity as has been proposed to explain some of the differences in corticosterone secretion seen between males and females (52).
Our data demonstrate that prevention of both nonaromatized and aromatized androgen activities disrupts the development of a characteristically masculine corticosterone release profile and stress response in the male rat. However, although significant, the differences in basal pulsatile measurements in ATD compared with controls were not as dramatic as those in the flutamide group compared with their controls. For example, ATD rats had a mean of 14 pulses over the 24 h compared with a control value of 10, whereas a mean number of 18 pulses were found for the flutamide group compared with 11 in their controls. The greater effect in the flutamide-treated than in the ATD-treated animals suggests that both AR and ER activation may be necessary to ensure complete masculinization of the HPA axis and corticosterone secretion profile. The greater effect of flutamide than ATD on PVN GR mRNA may similarly suggest a role for AR in addition to ER. The demonstration that estradiol produces the same masculine behavior in adult castrated rats as testosterone replacement only if administered with dihydrotestosterone (53) further reinforces this dual hormone idea. Accordingly, it may be surmised that testosterone acts via AR activity to mediate AVP expression in the PVN, but acts via ER activity to mediate PVN CRF expression. The suggestion that testosterone acts primarily via PVN AVP inhibition to mediate corticosterone secretion in the adult rat supports this theory (54).
The site(s) at which this perinatal programming occurs is unclear. The presence of both ARs (55) and ERs (56) in the hippocampus of the male rat suggests that this could be an area where androgens might modify the negative feedback properties of hippocampal corticosteroid receptors on the HPA axis. The present study, however, failed to demonstrate any differences in MR or GR mRNA in the subfields of the hippocampus, making it very unlikely that the differential corticosterone secretion arising from our perinatal treatment was caused by changes in corticosteroid receptor negative feedback at this level. In contrast, significantly reduced levels of PVN GR mRNA were found in both flutamide- and ATD-treated rats compared with their respective controls. It would appear, therefore, that hypothalamic GR are sensitive to early life exposure to androgens and may represent an important locus for the perinatal effects of androgens on regulation of the HPA activity by glucocorticoid feedback in the adult.
Our data clearly show a strong influence of pre- and postnatal testosterone activities on the development of a characteristically masculine corticosterone secretion profile and stress response. It appears that testosterone exerts its organizational effects via a combination of both AR and ER activities in the developing neonate. The occurrence of gender-based differences in the susceptibility to a number of disease models in the rodent (57, 58, 59) has been attributed in part to the differential HPA axis activity of male and female rats. Modification of basal and stress-induced corticosterone secretion after manipulation of perinatal androgen activity may have important implications for the susceptibility of the adult rat to these disease models.
References
Arai Y, Gorski RA 1968 The critical exposure time for androgenisation of the developing hypothalamus in the female rat. Endocrinology 82:1010–1014
Raisman G, Field PM 1971 Sexual dimorphism in the preoptic area. Science 173:731–733
Levine S, Haltmeyer GC, Karas GG, Denenberg VH 1967 Physiological and behavioural effects of infantile stimulation. Physiol Behav 2:55–63
Kitay JI 1961 Sex differences in adrenal cortical secretion in the rat. Endocrinology 68:818–824
Critchlow V, Liebelt A, Barsela M, Mountcastle W, Lipscomb HS 1963 Sex differences in resting pituitary-adrenal function in the rat. Am J Physiol 205:807–815
Le Mevel JC, Abitbol S, Beraud G, Maniey J 1979 Temporal changes in plasma adrenocorticotropin concentration after repeated neurotropic stress in male and female rats. Endocrinology 105:812–817
Patchev VK, Hayashi S, Orikasa C, Almeida OFX 1995 Implications of estrogen dependent brain organisation for gender differences in hypothalamo-pituitary-adrenal regulation. FASEB J 9:419–423
Seale JV, Wood SA, Atkinson HC, Bate E, Lightman SL, Ingram CD, Jessop DS, Harbuz MS 2004 Gonadectomy reverses the sexually diergic patterns of circadian and stress-induced hypothalamo-pituitary-adrenal axis activity in male and female rats. J Neuroendocrinol 16:516–524
Weisz JW, Ward IL 1980 Plasma testosterone and progesterone titers of pregnant rats, the male and female fetuses, and neonatal offspring. Endocrinology 106:306–316
Slob AK, Ooms MP, Vreeberg JMT 1980 Prenatal and early postnatal sex differences in plasma and gonadal testosterone and plasma luteinising hormone in female and male rats. J Endocrinol 87:81–87
Grady KL, Phoenix CH, Young WC 1965 Role of the developing rats testis in differentation of the neural tissues mediating mating behaviour. J Comp Physiol Psychol 59:176–182
Pfaff DW, Zigmond RE 1971 Neonatal androgen effects on sexual and nonsexual behaviour of adult rats tested under various hormone regimes. Neuroendocrinology 7:129–145
Husmann DA, McPhaul MJ 1991 Time-specific androgen blockade with flutamide inhibits testicular descent in the rat. Endocrinology 129:1409–1416
Kassim NM, McDonald SW, Reid O, Bennett NK, Gilmore DP, Payne AP 1997 The effects of pre- and postnatal exposure to the nonsteroidal antiandrogen flutamide on testis descent and morphology in the albino Swiss rat. J Anat 190:577–588
Meaney MJ, Stewart J, Poulin P, McEwen BS 1983 Sexual differentiation of social play in rat pups is mediated by the neonatal androgen-receptor system. Neuroendocrinology 37:85–90
Axelson JF, Smith M, Duarte M 1999 Prenatal flutamide treatment eliminates the adult male rat’s dependency upon vasopressin when forming social-olfactory memories. Horm Behav 36:109–118
Husmann DA, Wilson CM., McPhaul MJ, Tilley WO, Wilson JD 1990 Antipeptide antibodies to two distinct regions of the androgen receptor protein localize the receptor protein to the nuclei of cells in the rat and human prostate. Endocrinology 126:2359–2368
Maclusky NJ, Hurlburt PC, Naftolin F 1985 Estrogen formation in the developing rat brain: sex differences in aromatase activity during early post-natal life. Psychoneuroendocrinology 10:355–361
Reddy VVR, Naftolin F, Ryan KJ 1974 Conversion of androstenedione to estrone by neural tissues from fetal and neonatal rats. Endocrinology 94:117–121
George FW, Ojeda SR 1982 Changes in aromatase activity in the rat brain during embryonic, neonatal and infantile development. Endocrinology 111:522–528
Westly BR, Salaman DF 1976 Role of oestrogen receptor in androgen-induced sexual differentiation of the brain. Nature 264:407–408
Booth JE 1977 Sexual behaviour of neonatally castrated rats injected during infancy with oestrogen and dihydrotestosterone. J Endocrinol 72:135–141
Schwarzel WC, Kruggel WG, Brodie HJ 1973 Studies on the mechanism of estrogen biosynthesis. VIII. The development of inhibitors of the enzyme system in human placenta. Endocrinology 92:866–880
Houtsmuller EJ, Brand T, De Jonge FH, Joosten RJNMA, Van de Poll NE, Slob AK 1994 SDN-POA volume, sexual behaviour and partner preference of male rats affected by perinatal treatment with ATD. Physiol Behav 56:535–541
Handa RJ, Nunley KM, Lorens SA, Louie JP, McGivern RF, Bollnow MR 1994 Androgen regulation of adrenocorticotropin and corticosterone secretion in the male rat following novelty and foot shock stressors. Physiol Behav 55:117–124
Viau V, Meaney MJ 1996 The inhibitory effect of testosterone on hypothalamic-pituitary-adrenal responses to stress is mediated by the medial preoptic area. J Neurosci 16:1866–1876
Domínguez-Salazar, E, Portillo W, Baum MJ, Bakker, J, Paredes RG 2002 Effect of prenatal androgen receptor antagonist or aromatase inhibitor on sexual behavior, partner preference and neuronal Fos responses to estrous female odors in the rat accessory olfactory system. Physiol Behav 75:337–346
Windle RJ, Wood SA, Shanks N, Lightman SL, Ingram CD 1998 Ultradian rhythm of basal corticosterone release in the female rat: dynamic interaction with the response to acute stress. Endocrinology 139:443–450
Conde GL, Renshaw D, Zubelewicz B, Lightman SL, Harbuz MS 1999 Central LPS-induced c-fos expression in the PVN and the A1/A2 brainstem noradrenergic cell groups is altered by adrenalectomy. Neuroendocrinology 70:175–185
Bradford MA 1976 A rapid and sensitive method for quantitation of microgram quantities of protein utilising the principle of protein-dye binding. Anal Biochem 72:248–254
Harbuz MS, Lightman SL 1989 Responses of hypothalamic and pituitary mRNA to physical and psychological stress in the rat. J Endocrinol 122:705–711
Young III WS, Mezey E, Siegel RE 1986 Vasopressin and oxytocin mRNAs in adrenalectomised and Brattleboro rats: analysis by quantitative in situ hybridisation histochemistry. Mol Brain Res 1:231–241
Young III WS, Mezey E, Siegel RE 1986 Quantitative in situ hybridisation histochemistry reveals increased levels of corticotrophin-releasing factor mRNA after adrenalectomy in rats. Neurosci Lett 70:198–203
Takahashi H, Hakamata Y, Wantabe Y, Kikuno R, Miyata T, Numa S1983 POMC probe. Nucleic Acids Res 11:6647–6858
Kinoshita H, Jessop DS, Finn DP, Harbuz MS 2000 Cyanamide-induced activation of the hypothalamo-pituitary-adrenal axis. J Neuroendocrinol 12:255–262
Seckl JR, Dickson KL, Yates C, Fink G 1991 Distribution of glucocorticoid and mineralocorticoid receptor messenger RNA expression on human postmortem hippocampus. Brain Res 562:332–337
Merriam GR, Wachter KW 1982 Algorhythms for the study of episodic hormone secretion. Am J Physiol 243:E310–E318
Windle RJ, Wood SA, Lightman SL, Ingram CD 1998 The pulsatile characteristics of the hypothalamo-pituitary-adrenal activity in female Lewis and Fischer 344 rats and its relationship to differential stress responses. Endocrinology 139:4044–4052
Festing MFW, Overend P, Gaines Das R, Borja MC, Berdoy M 2002 The design of animal experiments. London: Royal Society of Medicine Press
Harbuz MS, Windle RJ, Jessop DS, Renshaw D, Ingram C.D, Lightman SL 1999 Differential effects of psychological and immunological challenge on the hypothalamo-pituitary-adrenal axis function in adjuvant-induced arthritis. Ann NY Acad Sci 876:43–52
McCormick CM, Mahoney E 1999 Persistent effects of prenatal, neonatal, or adult treatment with flutamide on the hypothalamic-pituitary-adrenal stress response of adult male rats. Horm Behav 35:902–101
Levine S 1957 Infantile experience and resistance to physiological stress. Science 126:406–406
Meaney MJ, Aitkin DH, Viau V, Sharma S, Sarrieau A 1989 Neonatal handling alters adrenocortical negative feedback sensitivity and hippocampal type II glucocorticoid receptor binding in the rat. Neuroendocrinology 50:597–604
Viau V, Sharma S, Plotsky PM, Meaney MJ 1993 Increased plasma ACTH responses to stress in nonhandled compared with handled rats require levels of corticosterone and are associated with increased levels of ACTH secretagogues in the median eminence. J Neurosci 13:1097–1105
Shanks N, Windle RJ, Harbuz MS, Jessop DJ, Ingram CD, Lightman SL 2000 Early-life exposure to endotoxin alters hypothalamic-pituitary-adrenal function and predisposition to inflammation. Proc Natl Acad Sci USA 97:5645–5650
Phoenix CH, Goy RW, Gerall AA, Young WC 1959 Organising action of prenatally administered testosterone propionate on the tissues mediating mating behaviour in the female guinea pig. Endocrinology 65:369–382
Gorski RA, Gordon JH, Shryne JE, Southam AM 1978 Evidence for a morphological sex difference within the medial preoptic area of the rat brain. Brain Res 148:333–346
Viau V, Meaney J 1991 Basal and stress hypothalamic-adrenal activity in cycling and ovariectomised-steroid treated rats. Endocrinology 129:2503–2511
Vamvakopolous NC, Chrousos GP 1993 Structural organization of the 5' flanking region of the human corticotropin releasing hormone gene. DNA Sequencing Mapping 4:197–206
Tannenbaum B, Rowe W, Sharma S, Diorio J, Steverman A, Walker M, Meaney MJ 1997 Dynamic variations in plasma corticosteroid-binding globulin and basal activity following acute stress in adult rats. J Neuroendocrinol 9:163–168
Neufeld JH, Breen L, Hauger R 1994 Extreme posture elevate corticosterone in a forced ambulation model of chronic stress in rats. Pharmacol Biochem Behav 47:223–240
Gala RR, Westphal U 1965 Corticosteroid-binding globulin in the rat: studies on the sex difference. Endocrinology 77:841–851
Baum MJ 1979 Differentation of coital behaviour in mammals: a comparative analysis. Neurosci Biobehav Rev 3:265–284
Viau V, Chu A, Soriano L, Dallman M 1999 Independent and overlapping effects of corticosterone and testosterone on corticotropin-releasing hormone and arginine vasopressin mRNA expression in the paraventricular nucleus of the hypothalamus and stress-induced adrenocorticotropic hormone release. J Neurosci 19:6684–6693
Kerr JE, Allore RJ, Beck, SG, Handa RJ 1995 Distribution and hormonal regulation of androgen receptor (AR) and AR messenger ribonucleic acid in the rat hippocampus. Endocrinology 136:3213–3221
O’Keefe, JA, Handa RJ 1990 Transient increases in estrogen receptor in the neonatal rat hippocampus. Brain Res Dev Brain Res 57:119–127
Schuurs AH, Verheul HA 1990 Effects of gender and sex steroids on the immune response. J Steroid Biochem 35:157–172
Homo-Delarche F, Fitzpatrick F, Christeff N, Nunez EA, Bach JF, Dardenne M 1991 Sex steroids, glucocorticoids stress and autoimmunity. J Steroid Biochem 40:619–637
Whitacre CC, Dowdell K, Griffin AC 1998 Neuroendocrine influences on experimental autoimmune encephalomyelitis. Ann NY Acad Sci 840:705–716(J. V. Seale, S. A. Wood, )