| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
ARTICLES |
Department of Physiology and Pharmacology, Oregon Health Sciences University, Portland, Oregon 97201
Address all correspondence and requests for reprints to: Dr. Charles E. Roselli, Department of Physiology and Pharmacology L334, Oregon Health Sciences University, 3181 S.W. Sam Jackson Park Road, Portland, Oregon 97201-3098. E-mail: rosellic{at}OHSU.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Behavioral gender differences in androgen responsiveness are presumably related to specific morphological and/or biochemical brain characteristics that differ between males and females. Several brain regions involved in the regulation of sexual behavior are known to be sexually dimorphic (10). There is evidence that perinatal androgen exposure masculinizes the brain by altering the number of neurons in specific brain areas, as well as their connectivity and neurotransmitter content (10, 11, 12, 13). Moreover, region-specific sex differences in steroid receptor binding [protein and messenger RNA (mRNA)] have been demonstrated in the rat brain. Males have greater androgen receptor concentrations and/or greater numbers of androgen receptor-containing neurons within brain areas mediating the hormonal activation of sexual behavior by androgen (2, 14, 15, 16). As with brain differentiation, the conversion of T to estradiol by aromatase is an important part of the pathway mediating copulatory behaviors in adult rats (17). Thus, it is also possible that the capacity for conversion of T to active estrogenic metabolites in the adult mammalian brain is influenced by perinatal exposure to gonadal steroids.
Aromatase activity (AA) is highest within the medial amygdala (MA), bed nucleus of the stria terminalis (BNST), medial preoptic nucleus (MPN), periventricular preoptic area (PVPOA), anterior hypothalamus (AH), and ventromedial nucleus of the hypothalamus (VMN) of adult rats (18, 19). These nuclei in large part comprise the brain circuitry that regulates masculine sexual behavior (20). AA in these nuclei is stimulated by androgens and has been referred to as androgen dependent (19). The induction of aromatase by T is mediated through an androgen receptor mechanism that regulates aromatase mRNA levels (21, 22). Expression of androgen-dependent AA in the adult brain is greater in males than in females because of normal sex differences in circulating androgen levels (19). However, the mechanism of enzyme induction is also sexually dimorphic because equivalent physiological doses of T stimulate aromatase to a greater extent in males than in females (2). Dose-response studies indicate that the sex difference is apparent over a range of circulating T concentrations and constitutes a gender difference in the efficacy of T stimulation (23). Measurements of aromatase mRNA in androgen-treated gonadectomized (Gdx) rats demonstrate that the sex difference in regulation is exerted pretranslationally (24). Taken together, these results suggest a sexually dimorphic mechanism that could potentially limit the action of T in females and may in part account for the enhanced expression of T-stimulated sexual behaviors in males.
A major issue that needs to be resolved is whether the adult gender difference in the expression and regulation of brain aromatase is developmentally determined. Therefore, the present study tested the hypothesis that exposure to T during the critical perinatal period for sexual differentiation accounts for the sex difference in the neural expression of androgen-regulated aromatase in adults. In addition, as the aromatase hypothesis predicts that androgen-derived estrogens are responsible for sexual differentiation in rats (17), we examined whether perinatal estrogen exposure is also capable of altering the phenotypic expression of aromatase in the adult. Finally, because it has been suggested that puberty is associated with an increase in neural responsiveness to T (25, 26, 27), we examined what effect, if any, prepubertal Gdx exerts on enzyme regulation in adults. The present study has addressed these issues by using a micropunch technique to measure basal and androgen-stimulated AA in a group of limbic and basal forebrain nuclei that are known to be involved in the modulation of androgen-dependent behaviors and neuroendocrine functions. Comparisons are made among control males, control females, and females treated perinatally with androgen or estrogen.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Experimental protocol
The experimental protocol is diagramed in Fig. 1
. Rat fetuses were treated prenatally
with daily sc injections of testosterone propionate (TP; 2 mg/100 µl
oil), diethylstilbestrol (DES; 10 µg/100 µl oil), or 100 µl
sesame oil vehicle to pregnant dams from GD16 to GD19. At birth, all
pups were weighed, and their anogenital distances were measured with a
micrometer. When possible, litters were adjusted to no more than eight
pups. After birth, the pups were given sc injections of TP (0.1 mg/50
µl oil), DES (1 µg/50 µl oil), or 50 µl sesame oil vehicle each
day beginning on the day of delivery and continuing through the fifth
postnatal day (PD). All pups in a litter were given the same treatment
and were kept separate from all other litters. The five perinatal
treatment groups were: 1) control 
, males treated pre- and
postnatally with the sesame oil vehicle; 2) control 
, females
treated pre- and postnatally with the sesame oil vehicle; 3) perinatal
TP , females treated pre- and postnatally with TP; 4) postnatal TP
, females treated prenatally with vehicle and postnatally with TP;
and 5) perinatal DES , females treated pre- and postnatally with
DES. These treatments were chosen because they were shown previously to
masculinize brain anatomy and adult sexual behaviors (7, 8, 28).
|
Tissue dissections
Frozen brains were sectioned coronally at 300-µm intervals
beginning where the anterior commissure crosses the midline and
extending caudally 3.6 mm as depicted in the brain map of Palkovits and
Brownstein (30). The tissue sections were thaw mounted onto glass
microscope slides and stored overnight at -80 C. Selected brain nuclei
and regions were dissected bilaterally using 500- and 1000-µm
calibrated stainless steel cannulas, as described previously (19). The
tissue punches were expelled into 500-µl propylene microtubes and
chilled on ice until homogenized.
AA assay
Tissue punch dissections collected from individual animals were
homogenized using a tissue sonicator (E/MC Corp., Hauppauge, NY) in 125
µl phosphate buffer (10 mM
KH2PO4, 100 mM KCl, 1
mM EDTA, and 1 mM dithiothreitol, pH 7.4).
Aliquots of these homogenates (100 µl) were then incubated for 1
h at 37 C with [1ß-3H]androstenedione (New England
Nuclear-Dupont, Boston, MA; SA, 24.1 Ci/mmol). AA was estimated by
quantifying the amount of 3H2O generated by the
stereospecific loss of the C-1ß tritium, which is proportional to the
amount of estrogen formed (31). This assay has been validated in our
laboratory, and the procedural details have been described previously
(31). Protein concentrations were determined using the Lowry method
(32). AA was expressed as femtomoles of 3H2O
produced per h/mg protein.
Testosterone assay
Trunk blood was collected after decapitation and allowed to clot
at 4 C overnight. They were then centrifuged (1500 x g
for 30 min), and sera were harvested. T levels were measured in
individual samples by RIA after extraction and chromatography on
Sephadex LH-20 using previously described methodologies (33). All
samples were assayed in duplicate in a single RIA. The percent
recovery, water blanks, and intraassay coefficient of variation were
79%, 0.7 pg/tube, and 7.3%, respectively.
Statistical analysis
The AA data for each brain nucleus were analyzed by parametric
3 x 3 ANOVA (perinatal treatment x age at castration
x adult treatment). When a significant main effect of perinatal
treatment was found in a tissue (P < 0.05), multiple
one-way ANOVAs were performed across treatment groups.
Post-hoc comparisons were then made between perinatal
treatments with post-hoc Newman-Keuls tests and were
considered significant at P < 0.05 (34).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-5,
respectively. In each figure, A shows the results from rats Gdx on
PD30, and B shows the results for rats Gdx on PD56. To simplify the
data presentation, only comparisons of perinatal treatment groups with
control males and control females are noted in the figures. As
expected, adult T treatments significantly stimulated AA in these
tissues regardless of prior perinatal treatment (compare
gray vs. black bars) or age at
castration (compare A, Gdx on PD 30, with B, Gdx on PD 56). The
induction of AA by T was significantly greater in control males than in
control females. In the prepubertal Gdx group, the percentages by which
AA levels in males was greater than those in females were 174% (BNST),
154% (MPN), 212% (PVPOA), and VMN (122%). In the adult Gdx group,
the percentages by which AA levels in males were greater than those in
females were 134% (BNST), 212% (MPN), 184% (PVPOA), and 143% (VMN).
In BNST, MPN, and PVPOA, females that were treated perinatally with TP
or postnatally only with TP, exhibited significantly greater levels of
T-stimulated AA than control females, whereas T-stimulated AA levels in
these groups were not significantly different from levels in control
males. This effect was observed regardless of whether the rats were Gdx
on PD 30 or as adults. In VMN, a significant effect of perinatal and
postnatal TP treatments was observed only in rats that were Gdx
as adults. Perinatal DES treatment masculinized the T-stimulated
induction of AA in BNST, MPN, and PVPOA, but not in VMN. Gonadectomized
(i.e. unstimulated) levels of AA did not differ between
perinatal treatment groups in the MPN and VMN. However, in the PVPOA
and BNST, AA levels in control males, perinatal TP-treated females, and
postnatal TP-treated females were generally higher than the levels in
control females and perinatal DES-treated females. In general, both
Gdx and T-stimulated levels of AA in all four tissues were higher
in rats Gdx at 56 days of age (Figs. 2B–5B) than in rats Gdx at 30
days of age (Figs. 2A–5A), which accounts for the significant main
effect due to age at Gdx.
|
and 7, respectively. Three-way ANOVA revealed
a significant main effect of adult T treatment, but no effect of
perinatal treatment or age of castration in these tissues. In general,
T treatment stimulated AA in adults regardless of any prior perinatal
treatment (compare gray vs. black
bars) or pubertal treatments (compare A, Gdx on PD 30, with B, Gdx
on PD 56).
|
|
. There was a significant effect of
perinatal treatment on anogenital distances, but no effect of age at
castration or adult treatment. In general, anogenital distance was
greatest in control males, intermediate in females treated peri- or
postnatally with TP, and lowest in control females and females treated
perinatally with DES. There was a significant effect of perinatal
treatment group, age at castration, and adult treatment on body weight,
with a significant interaction between perinatal treatment and age at
castration (Fig. 9). In general, control
males weighed more than any other treatment group. However, females
that were treated postnatally with TP or perinatally with DES and
castrated as adults were significantly heavier than control females,
but were not as heavy as control males (Fig. 9B). Females that were
gonadectomized prepubertally weighed more than their counterparts that
were gonadectomized as adults.
|
|
Serum T levels
The levels of serum T are summarized in Fig. 10. Equivalent levels of T were
achieved in all groups 1 week after Gdx and androgen treatment.
However, females treated with TP perinatally or postnatally exhibited a
slight, but significant, elevation in serum T levels in the sham group
Gdx at 30 days of age.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We demonstrated that the organizational effects of androgen and estrogen are brain region specific. An effect on the expression of androgen-stimulated AA was evident in the BNST, MPN, PVPOA, and VMN, but not in the AH and MA. These results generally agree with our previous studies that identified gender differences in the induction of brain aromatase by T within these same nuclei (2, 23). Our results confirm and extend the work of Steimer and Hutchison (35), who demonstrated that the sex difference in the induction of aromatase in the preoptic area of rats is abolished by neonatal Gdx and exposure to TP from PD 2 to 70. However, in this early study, neonatal Gdx and extended treatment with TP appeared to substantially decrease the responses of both sexes to TP compared with that of normal adult males. In contrast, the perinatal treatments used in the present experiment were sufficient to completely masculinize the expression of AA, so that the enzyme activity in hormone-treated genetic females was equivalent to that in control males.
The induction of AA by T is transcriptionally mediated through a specific androgen receptor mechanism (36). Therefore, one important inference that can be drawn from the current study is that the perinatal exposure to androgen permanently alters the adult responsiveness to T. Moreover, the cellular basis for the effect of early androgen and estrogen exposure on androgen-dependent aromatase expression may relate in part to the differentiation of androgen receptor-positive neurons or their connections. Consistent with this idea, the distribution of androgen receptors and androgen-dependent aromatase exhibit an extensive regional overlap in the rat brain (37, 38, 39, 40), and sex differences in androgen receptor concentrations are found largely within the same brain regions as sex differences in aromatase (2).
In the current study, significant sex differences in the expression of basal levels of AA were found in the BNST and PVPOA of Gdx adults. Basal levels of aromatase were significantly higher in control males than in control females, but were not significantly different from those in females treated perinatally with TP. Sex differences in AA that are present after Gdx could be due to gender differences in androgens produced by the adrenals, but no sex-related differences in plasma T were detected in the Gdx sera in this study. Moreover, in a preliminary unpublished study, we did not detect differences between Gdx and Gdx-adrenalectomized rats in the levels of AA measured in brain punch samples. Thus, the sex differences in the Gdx levels of AA most likely reflect actual differences in the number of aromatase neurons independent of their capacity for hormonal induction and suggest that greater numbers of aromatase-containing cells are present within certain regions of the masculinized adult brain. Consistent with this hypothesis, Wagner and Morell (38) reported that adult male rats tended to have higher numbers of aromatase mRNA-expressing cells than females, but not more aromatase mRNA per cell. Likewise, a sex difference in the number of aromatase-immunoreactive cells has been found in the preoptic area of adult Japanese quail (41) and shrew (42). Finally, Beyer and Hutchison (43) have shown that numbers of aromatase-immunoreactive neurons are higher in hypothalamic cultures originating from male compared with female mice.
Interestingly, perinatal exposure to TP, but not DES, masculinized the basal levels of AA in the BNST and PVPOA. This observation contrasts with our finding that both TP and DES exposure masculinized the capacity for androgen induction of AA in these nuclei as well as in PVPOA and MPN. The fact that these two end points (i.e. basal and androgen-induced aromatase) can be differentially affected by perinatal steroid exposure suggests that they may represent independent processes that have different hormonal requirements for differentiation. Both end points appear to be sexually differentiated by androgen, but aromatization may be required to determine the capacity for induction of aromatase by androgen, i.e. androgen responsiveness, whereas androgen per se may determine the absolute number of aromatase neurons that develop or survive. There are precedents for considering that both local estrogen synthesis and unmetabolized androgens contribute to male-specific developmental processes in the perinatal brain. Locally synthesized estrogens are thought to irreversibly masculinize the size and connectivity of several brain nuclei, including the sexually dimorphic nucleus of the preoptic area (44). On the other hand, T, not its aromatized metabolites, affects the survival of neurons in the spinal nucleus of the bulbocavernosis (10). Moreover, it was recently demonstrated that T treatment increases the absolute number of aromatase-immunoreactive neurons in hypothalamic cultures and that, unlike estrogen, T stimulated their morphological differentiation (45). Taken together, these studies suggest that androgen and estrogen exert independent, perhaps complementary, effects on the developing brain that, in turn, affect the basal levels and androgen responsiveness of aromatase in the adult.
Rats that were Gdx on PD 56 exhibited significantly higher levels of basal and T-stimulated aromatase in BNST, MPN, and PVPOA than rats that were Gdx on PD30. This effect was observed regardless of sex or perinatal treatment. In VMN, the effects of perinatal T treatments reached significance only in the group that was Gdx on PD 56. These data suggest that peripubertal exposure to gonadal secretions can enhance the expression of AA in certain areas of the adult brain, which may, in turn, enhance adult responsiveness to T. There was no effect of prepubertal Gdx on the expression of aromatase in AH or MA, indicating that this response was exerted in a regionally specific manner. The nature of the effect we observed was subtle and appeared to be secondary to the effect exerted by perinatal androgen exposure. However, if confirmed, this observation may help explain the results of earlier behavioral studies which demonstrated that pubertal exposure to testosterone sensitizes the neural circuits mediating male sexual behavior to the activational effects of T in adulthood (25, 26, 27, 46).
In conclusion, the present experiment has demonstrated that AA in the neural centers mediating masculine sexual behavior is sexually differentiated in the rat. We have shown that manipulations of the early hormonal milieu, which are known to differentiate sexual behavior, also irreversibly determine the capacity for aromatization in the adult. The process that differentiates aromatase appears to depend on androgen exposure and in part local estrogen synthesis and is brain region specific. As the activation of copulation by androgen requires aromatization of T to estradiol (47, 48, 49), this quantitative enzymatic difference between sexes could contribute in part to the greater behavioral responsiveness displayed by males. Finally, our data suggest that although exposure of the brain to steroid hormones during perinatal life appears to be necessary for sexual differentiation of aromatase expression to occur, gonadal hormones also appear to exert additional effects during puberty.
|
|
|
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received January 26, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-reduction of androgens in discrete
hypothalamic and limbic regions of the male and female rat.
Endocrinology 101:841–848[Abstract]
This article has been cited by other articles:
 |
|
 |
|
  C. Bodo and E. F. Rissman The Androgen Receptor Is Selectively Involved in Organization of Sexually Dimorphic Social Behaviors in Mice Endocrinology, August 1, 2008; 149(8): 4142 - 4150. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J Bancroft The endocrinology of sexual arousal J. Endocrinol., September 1, 2005; 186(3): 411 - 427. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. M. Foecking, M. Szabo, N. B. Schwartz, and J. E. Levine Neuroendocrine Consequences of Prenatal Androgen Exposure in the Female Rat: Absence of Luteinizing Hormone Surges, Suppression of Progesterone Receptor Gene Expression, and Acceleration of the Gonadotropin-Releasing Hormone Pulse Generator Biol Reprod, June 1, 2005; 72(6): 1475 - 1483. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Omoto, R. Lathe, M. Warner, and J.-A. Gustafsson Early onset of puberty and early ovarian failure in CYP7B1 knockout mice PNAS, February 22, 2005; 102(8): 2814 - 2819. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Matagne, G. Rasier, M.-C. Lebrethon, A. Gerard, and J.-P. Bourguignon Estradiol Stimulation of Pulsatile Gonadotropin-Releasing Hormone Secretion in Vitro: Correlation with Perinatal Exposure to Sex Steroids and Induction of Sexual Precocity in Vivo Endocrinology, June 1, 2004; 145(6): 2775 - 2783. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. E. Roselli, K. Larkin, J. A. Resko, J. N. Stellflug, and F. Stormshak The Volume of a Sexually Dimorphic Nucleus in the Ovine Medial Preoptic Area/Anterior Hypothalamus Varies with Sexual Partner Preference Endocrinology, February 1, 2004; 145(2): 478 - 483. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Kishida and G. V. Callard Distinct Cytochrome P450 Aromatase Isoforms in Zebrafish (Danio rerio) Brain and Ovary Are Differentially Programmed and Estrogen Regulated during Early Development Endocrinology, February 1, 2001; 142(2): 740 - 750. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Gomez and M. F. Dallman Manipulation of androgens causes different energetic responses to cold in 60- and 40-day-old male rats Am J Physiol Regulatory Integrative Comp Physiol, January 1, 2001; 280(1): R262 - R273. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. D. Romeo, J. Wade, J. E. Venier, and C. L. Sisk Androgenic Regulation of Hypothalamic Aromatase Activity in Prepubertal and Postpubertal Male Golden Hamsters Endocrinology, January 1, 1999; 140(1): 112 - 117. [Abstract] [Full Text]
|
|
,
P < 0.05 vs. sham control 
, P < 0.05 vs. T-treated control
