| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Department of Human Physiology, University of California School of Medicine, Davis, California 95616
Address all correspondence and requests for reprints to: Dr. Judith L. Turgeon, Department of Human Physiology, University of California School of Medicine, Davis, California 95616. E-mail: jlturgeon{at}ucdavis.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-dihydrotestosterone (DHT), generally is found to be effective. The mechanisms involved in the direct inhibitory action of androgens on GnRH-stimulated LH secretion from female gonadotropes have not been established, although T has been reported to inhibit LH release induced by phorbol ester-activated protein kinase C (9) and to either suppress (10) or potentiate (11) release induced by cAMP. There also is a report that DHT can decrease GnRH-binding sites in female pituitary cells (12). Most of the early studies that uncovered the suppressive effect of androgen on LH secretion were carried out with prolonged, nonpulsatile exposure of pituitary cells to GnRH, thus making it difficult to dissect out the complex actions. Given that androgens can have both negative and positive actions on the female gonadotrope and that the sites of action could involve steps in the GnRH signaling pathway, LH gene expression and protein synthesis, or the exocytotic machinery, it is not surprising that kinetic aspects of the stimulus and response as well as end points other than the aggregate LH secretory response must be considered.
Much of the recent work on androgen action in the gonadotrope has focused on cells from males and within the framework of its primary physiological role as a negative feedback regulator of LH secretion in the male (13, 14, 15). For the female, the physiological context is less obvious. During the rat estrous cycle, serum T and DHT levels are lowest during estrus and highest on proestrus after the onset of the preovulatory gonadotropin surges, and these fluctuations are independent of contributions from the adrenal glands (16, 17). The proestrous increase in serum T concentration appears to precede the onset of the LH surge by a few hours (16). Thus, there is a context in which androgens reasonably could be considered as playing a physiological role. Timing is paramount in the reproductive cycle. Most studies of androgen action on the gonadotrope have used extended exposure times (24 h or more), and the minimum exposure requirements or, in fact, whether there are different temporal requirements for the negative and positive actions of androgens have not been determined.
Prompted by the intriguing observations suggesting that androgens have a positive as well as a negative role in the regulation of LH, we asked whether androgens could modulate the LH secretory events in an in vitro model of preovulatory gonadotropes. Specifically, we examined the effect of androgens on GnRH self priming and on acute progesterone augmentation of stimulated LH secretion. Part of this study has been presented in preliminary form (18).
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Pituitary cell culture
Rats were ovariectomized and maintained for 2 weeks, after which
pituitaries were removed following CO2 narcosis and
decapitation. Anterior pituitary cells, obtained by trypsin dispersion
(day 0), were cultured in Eagle’s MEM containing D-valine
(MEM), 0.2 mM kanamycin sulfate, and 10% charcoal-treated
FBS (FBS-CT) with or without proestrous levels of E2 (0.2
nM) (21) in a humidified atmosphere of 5% CO2
in air as previously described (19). Residual steroid concentrations in
FBS-CT were 10 pM for progesterone and less than 1
pM for E2 as determined by RIA. Short term
ovariectomized rats were used to reduce the variability in steroid
background for donor pituitaries and to allow the cells to be
synchronized in vitro by the presence or absence of
E2 in the culture medium, thus providing for an in
vitro model of preovulatory gonadotropes. For secretion
experiments, cells were plated at 3 x 105 on
Matrigel-coated coverslips inserted into 22-mm wells. For
transfection experiments, cells were plated at 6 x
105 in 35-mm dishes.
Secretion studies
The medium was changed on day 2; at that time, exposure to
either T (1.7 nM) or DHT (0.7 nM) was initiated
for some groups. Androgen concentrations were based on levels occurring
during the preovulatory LH surge (16, 17) and correspond to the levels
reported to result in facilitation of GnRH-stimulated LHß in female
rats (7). Experiments were begun 16–19 h after the initiation of
androgen treatment unless noted otherwise and were carried out in
medium containing 1 mg/ml BSA without serum. For all experiments,
successive 15-min incubations were collected before, during, and after
the pulses to monitor LH secretion. Samples were stored at -70 C until
assayed by RIA as described previously (19).
Single GnRH pulse protocol. Starting at time zero, cells were incubated in control medium with or without 20 or 200 nM progesterone and challenged with a single 1-nM GnRH pulse of 15-min duration, initiated at 90 min. In certain experiments, DHT exposure was either terminated or initiated at 30 min before progesterone stimulation was initiated.
Triple GnRH pulse protocol. Cells were challenged with three 15-min pulses of 1 nM GnRH at 60-min intervals.
Multiple K+ pulses. Cells were challenged with multiple pulses of 54 mM K+ of 15-min duration at 60-min intervals. In some groups, 20 nM progesterone was included in the medium beginning 90 min before the first K+ pulse. For the K+ pulse experiments, MEM/BSA medium was replaced with medium of similar composition, pH 7.4, that contained 1 mg BSA/ml, 15 mg phenol red/liter, 110 mM NaCl, 5.4 mM KCl, 1.8 mM CaCl2, 1.6 mM MgSO4, 24 mM NaHCO3, 0.8 mM Na2HPO4, and 25 mM glucose; 54 mM K+ medium was prepared by equimolar replacement of NaCl with KCl.
Transfection studies
Cells were incubated in MEM/FBS-CT medium plus E2.
On day 2 of culture, cells were transfected with 2 µg
TAT3LUC plasmid by the CaPO4 precipitation
method and were subjected to glycerol shock 4 h later as
previously described (22). After glycerol shock, incubation was
continued in fresh medium with or without 0.7 nM DHT.
Beginning 16 h later, the transfected cells were rinsed in MEM/BSA
and challenged for 6 h in MEM/BSA with medium only (control),
progesterone (20 or 200 nM), DHT (continued), DHT
(continued) plus progesterone, or acute DHT plus progesterone.
Extract preparation
Cells were rinsed in cold PBS, incubated in lysis buffer at 4 C
for 15 min, scraped and lysed in a final volume of 180 µl, and stored
at -70 C until assayed at 50 µl in duplicate in a luminometer.
Data analysis
Data are presented as the mean ± SEM. Each
experiment represents a separate pool of dispersed pituitary cells; n
refers to the number of times an experiment was repeated. For LH
secretion studies in response to pulsatile secretagogue administration,
the integrated secretory response was calculated as the total amount of
LH secreted during the 15-min exposure to a secretagogue plus that
secreted in the subsequent 15 min. In some cases, the t
distribution was used to test the hypothesis that the response was
significantly different from 100%. All other statistical analyses were
performed using SigmaStat (version 2.0, SPSS, Inc.,
Chicago, IL). For multiple comparisons, differences between groups were
determined by ANOVA followed by the Tukey test; where differences are
indicated as significant, P < 0.05. Where appropriate,
differences between two groups were determined using Student’s
t test, with the level of significance noted in the report
of the results. In simple linear regression we tested the hypothesis
that the slope of the regression line was significantly different from
zero.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, the LH
secretory response is significantly suppressed in the presence of
either androgen to about 40% of the control value (compared with
control, P < 0.01 for either T or DHT). Also shown in
Fig. 1 is the acute augmentation by progesterone of GnRH-stimulated LH
secretion; exposure to 20 nM progesterone for 90 min before
a pulse of GnRH led to a doubling of the LH secretory response. When
the experiment was repeated for cells incubated in the presence of T or
DHT for 16 h, the progesterone-induced augmentation was completely
suppressed; no significant difference was found regardless of whether
progesterone was present in either androgen-treated group.
|
Effect on response to multiple GnRH pulses
To address whether E2 is required for the suppressive
action of androgen on GnRH-stimulated LH secretion, pituitary cells
cultured in the absence of E2 were subjected to hourly GnRH
pulses, which led to repetitive, small amplitude secretory responses.
As shown in Fig. 2A, exposure to DHT for
16–19 h in the absence of E2 significantly suppressed the
response to the first GnRH pulse (P < 0.001) just as
was found in the presence of E2 (Fig. 1
). However, although
the response for the control cells remained constant over the three
GnRH pulses, the response for the androgen-treated cells increased with
each pulse, such that by the third pulse, LH secretion was doubled
compared with the response to the first (P < 0.002).
The integrated secretory data are presented in Table 1.
|
|
); the response to the third GnRH
pulse was 145 ± 2% of that to the first pulse (significantly
greater at P < 0.01). When E2-treated
cells were exposed to DHT or T (16–19 h), the initial secretory
response was significantly suppressed, as demonstrated for the
DHT-treated group in Fig. 1, but the responses to subsequent GnRH
pulses increased (P < 0.002, response to third pulse
vs. response to first pulse) similar to that shown in the
absence of E2 (Fig. 2B and Table 1). Although this is
reminiscent of a self-priming response, the result showing that
estrogen is not required for the apparent facilitatory effect of
androgen on GnRH-stimulated LH secretion (Fig. 2A) suggested that the
androgen action was distinct from GnRH self priming, which is
absolutely dependent on a background of E2 for its
expression. To further distinguish between the positive action of
androgen in the presence of multiple GnRH pulses and GnRH self priming,
we repeated the experiments in the presence of RU486, which we have
shown to eliminate GnRH self priming even in the absence of
progesterone (23). As indicated in Table 1, the LH secretory response
in DHT-treated cells was unaffected by RU486, whereas the same
treatment in the absence of androgen abolished self priming (responses
not significantly different from each other within the control plus
RU486 group). This result provides further evidence that the
facilitatory action of androgen on LH secretion in response to multiple
GnRH pulses is unrelated to the self-priming action of GnRH.
Effect on response to pulses of elevated extracellular
K+
In the next series, the GnRH receptor was bypassed and LH
secretion was elicited with hourly pulses of 54 mM
K+ resulting in depolarization and repetitive secretory
episodes (Fig. 3). The aim was to test
whether GnRH receptor activation is required for either the inhibitory
or facilitatory actions of androgen on LH secretion. As shown in Fig. 3A, exposure to DHT for 16–19 h resulted in a significant and
unambiguous suppression of depolarization-stimulated LH secretion to
less than 40% of the control value for the first pulse of
K+ (P < 0.001). Although the responses in
the DHT-treated cells continued to be suppressed compared with those in
control cells over the next four pulses of K+, unlike the
responses in the controls there was a gradual increase in the LH
secretory response for the DHT-treated cells. The rise was significant,
as shown by the significantly positive slope of the regression line
drawn through the peak secretory responses to the five K+
pulses (P < 0.005). However, the increase seen after
multiple depolarization pulses over 5 h was not as dramatic as the
2- to 3-fold increases seen during stimulation with multiple GnRH
pulses over 3 h (Fig. 2 and Table 1).
|
shows that
incubation in the presence of 20 nM progesterone led to an
augmentation of the response to depolarizing K+ pulses
(note differences in scale for the ordinate in B compared
with that in A). This augmentation by progesterone was unambiguously
inhibited by preexposure to androgen. As an example for direct
comparison, the integrated secretory data for the second K+
pulse for all groups are shown in Fig. 4;
by 150 min of exposure to progesterone, the depolarization-induced LH
secretory response was doubled, but this augmentation was significantly
suppressed for cells incubated in DHT before exposure to
progesterone.
|
, exposure
to 20 nM progesterone for 6 h resulted in a 3-fold
increase in luciferase activity. When transfected cells were pretreated
with DHT for 16 h before and during stimulation with progesterone,
the progesterone-stimulated luciferase response was significantly
dampened by about 30% (Fig. 5). DHT by itself had no effect on
luciferase expression. When the progesterone concentration was
increased to 200 nM, or exposure to DHT was limited to the
acute period during which the cells were exposed to progesterone, there
was still a tendency for the progesterone-stimulated luciferase
response to be dampened by DHT, but the differences were not
significant (data not shown).
|
for control pituitary cells that have a
chronic (16-h) background of DHT exposure, the LH secretory response to
a GnRH pulse was suppressed by about 40% regardless of whether the DHT
was continued or removed before the GnRH pulse; however, DHT was
without significant effect when control cells were exposed to the
steroid for only 2 h. For cells responding to the acute
augmentation action of progesterone, the temporal pattern of DHT
suppression differed, in that 1) removal of DHT 30 min before
initiation of the acute 20-nM progesterone
treatment resulted in a slight, but significant, abatement in the
suppression compared with the response when DHT was continuously
present (P < 0.02); and 2) acute addition of DHT just
before progesterone exposure is initiated results in a significant
decrease in the response to GnRH (75% of that found when DHT is not
present; P < 0.001; Table 2). Although the maximal
effective concentration of progesterone in these cells was in the range
of 10-7 M based on augmentation of
GnRH-stimulated LH secretion (19), we tested whether increasing the
concentration of progesterone would diminish the ability of DHT to
interfere with progesterone. As shown in Table 2, although there is a
slight, but significant, lessening of the chronic DHT-induced
suppression in the 200-nM progesterone group
(P < 0.005 for 200-nM compared with
20-nM progesterone groups), the GnRH response remained
suppressed to about 40%; for acute DHT treatment there were no
differences between the two concentrations of progesterone. In
conjunction with the transfection studies these results suggest that in
addition to the posttranscriptional consequences of DHT’s action,
another component of the suppressive effect of DHT may involve
interference with progesterone’s action at the transcriptional
level.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Inhibitory androgen action
The temporal requirements for the suppression of
secretagogue-induced LH secretion by androgens provide some clue to its
possible role in the female reproductive cycle. In humans, ovarian
androgen secretion varies throughout the cycle, with the peak occurring
during the periovulatory period (28, 29, 30). In the rat, the serum T
concentration appears to exhibit three phases. The lowest levels are
found on estrus; by the first day of diestrus the levels have doubled
and stay in that range through the second day of diestrus and the
morning of proestrus; from there the levels increase 2- to 3-fold on
the afternoon of proestrus beginning about 4 h before the onset of
the preovulatory LH surge (16). Using an androgen concentration similar
to that observed during the proestrous gonadotropin surge, we found
that 8 h, but not 4 h, of exposure in vitro led to
only a slight reduction in stimulus-induced LH secretion, but that
16 h of androgen resulted in a 60% decrease in the response.
Given this latency, it is unlikely that the negative action of androgen
is operable or at least has any physiological consequences during the
approximately 4-h duration of the rat LH surge. However, it is not
unreasonable to speculate that the timing of the onset of the
inhibitory action of androgen would be consistent with a role for the
steroid in the termination of the surge. Several candidate mechanisms
have been suggested to participate in the termination of the surge
(31, 32, 33). Indicative of the potential for redundancy in the system,
which would be expected, is a report from 1974 suggesting that
antiserum to T administered on the evening before the day of the
preovulatory surge was without effect on the LH surge (34).
Facilitatory androgen action
A positive action of androgens on LH in female rats has been shown
as a facilitation of GnRH pulse-induced gonadotropin subunit mRNA
expression (7, 8). An up-regulation of LHß message was seen after
6 h of GnRH pulses in vivo or 24 h in
vitro for pituitary glands or cells that had been exposed to T for
1 day; shorter exposure times to either the steroid or to the GnRH
pulse regimens were not examined in these studies (7, 8). In our work
we found that after the initial suppressed response in cells incubated
in an androgen background, there was a gradual increase in the LH
secretory response to GnRH pulses that reached significance by 2 h
after the first pulse, and this occurred whether the cells had been
cultured in the presence or absence of E2. Whether this
correlates with an increase in LH available for secretion remains to be
established, but it is interesting that once the androgen background is
in place, the relatively short time course for a putative up-regulation
of LHß in response to GnRH pulses is consistent with the rapid events
occurring during the preovulatory LH surge in rats. The lowest androgen
concentration required for the facilitatory effect was not determined
in the in vivo work of Yasin et al. (7), but they
do report that a pre-LH surge level of T was more effective than higher
concentrations. The requirement for pulsatile GnRH delivery to observe
a positive modulatory role for androgens could explain why many of the
previous studies that used long duration GnRH exposures failed to
detect positive androgen actions in female pituitary cells (2, 3, 4, 5, 27).
Androgen and depolarization-induced secretion
In the current study, depolarization of gonadotropes with hourly
pulses of elevated extracellular K+ resulted in LH
secretion episodes of constant dimensions. On a background of DHT that
continued through the K+ challenges, the secretory episodes
were reduced by more than half, but by the fourth hour of pulsing there
was evidence for some abatement of the suppression. This increase in
responsiveness to depolarization-induced calcium influx even in the
presence of the inhibitory action of androgen on the secretion process
correlates with the observation by Haisenleder et al. that
in the presence of T, pulsatile calcium influx stimulated an increase
in gonadotropin subunit mRNAs after 24 h of pulsing, the only time
point examined (8). Whether the increase in secretion reported in our
studies is a reflection of increased available LH remains to be
established. It is of interest, however, that, for LH secretion
stimulated by either depolarization or GnRH, androgen treatment led
to an initial suppression that was gradually overcome with subsequent
secretagogue pulses. This indicates that neither the suppressive effect
nor the facilitatory action of androgen has an absolute requirement for
a change in the GnRH receptor. However, the more rapid time course for
the positive response to GnRH pulses compared with depolarization is
probably due to activation of pathways in addition to calcium-related
signals and perhaps to an increase in GnRH receptor as well. It would
be important in future studies to test for a possible effect of
androgen on GnRH receptor expression and binding.
Androgen and the acute action of progesterone
Acute progesterone exposure (<6 h) of pituitary cells in culture
results in a severalfold increase in the LH secretory response to GnRH
(19, 35, 36, 37), and this acute action of progesterone shares common
characteristics with the GnRH self-priming effect (22, 23). In the
female reproductive cycle, the augmentation by progesterone of
stimulated secretion probably plays a role in GnRH signal amplification
during the preovulatory gonadotropin surge; therefore, our studies
showing that an extended background of elevated androgen levels can
interfere with the acute effect of progesterone may have some relevance
to the actions of hyperandrogenism in the disruption of cyclicity.
The sites through which androgen achieves this dampening of the acute actions of progesterone on secretagogue-stimulated LH secretion are not known. Based on these studies, however, we can speculate that the primary transcriptional targets of androgen for both the suppression of LH secretion in general as well as the inhibition of progesterone’s ability to augment LH secretion are likely to overlap; these consequences of androgen’s genomic actions could be proteins involved in the proximal secretagogue signaling pathway and/or the LH exocytotic pathways. We suggest that an additional target specific to the acute action of progesterone is interference by androgen with the transcriptional activity of the progesterone receptor.
Androgen receptor
The androgen receptor has been shown to have both
trans-activation and trans-repression modes of
regulation of gene expression and to use a variety of strategies
involving regions outside of its DNA-binding domain to achieve some of
its multiple functions (38, 39, 40, 41, 42, 43, 44). In relation to our work showing DHT
suppression of progesterone’s augmentation action on stimulated LH
secretion, Yen et al. (41) demonstrated in cotransfection
studies in CV-1 cells that wild-type androgen receptor can have
striking dominant negative activity on glucocorticoid receptor- and
progesterone receptor-mediated transcription. For rat pituitary cells,
we found that DHT was able to only partially repress progesterone
receptor-mediated transcription and that this could be overcome with
increased progesterone concentration. The reporter plasmid used for our
transfection studies contained PRE sequences similar to that used by
Yen et al. (41), but differences in promoter context,
e.g. the use of three tandem PREs in the construct we used,
as well as cell-specific factors, agonist concentration, and receptor
abundance could explain the lack of an effect of androgen alone on
trans-activation as well as the relatively modest cross-talk
between the androgen and progesterone receptors in our pituitary cell
study (38, 45). However, as only one construct was tested, we cannot
exclude the possibility of significant androgen receptor interference
with the trans-activation function of the progesterone
receptor at target genes in gonadotropes. In fact, our secretion
results with acute androgen addition or removal in the presence of
progesterone suggest that at least part of the DHT-induced abolition of
the augmentation effect of progesterone on stimulated LH secretion
could be due to a direct inhibitory action at the level of the
progesterone receptor as well as to downstream targets, e.g.
a short-lived protein specific to the progesterone augmentation
pathway.
In summary, the facilitatory and inhibitory functions suggest that androgens have complex modulatory roles in the cyclic regulation of LH secretion that are context dependent, particularly with regard to temporal requirements, GnRH pulse patterns, and the presence of progesterone. Based on the results reported here, a possible scenario during the estrous cycle would include an androgen background-dependent facilitation of GnRH pulse-induced increase in gonadotropin subunit expression that becomes manifest during the preovulatory period as GnRH pulse frequency and/or amplitude increases. The latency of the inhibitory action attributable to the increase in androgen occurring before the onset of the gonadotropin surges is consistent with a role for androgen in the termination of the LH surge. These studies provide the framework for further investigation of the potential multiple targets and interplay of steroids within the gonadotrope for the complex androgen action.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received September 14, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
- and
lutropin (LH) ß-subunit but not release of stored LH in cultured rat
pituitary cells. Biochim Biophys Acta 1310:348–354[Medline]
-subunit gene by androgen receptor
occurs independently of DNA binding but requires the DNA-binding and
ligand-binding domains of the receptor. Mol Endocrinol 11:1497–1506This article has been cited by other articles:
 |
|
 |
|
  M. G. Ropelato, M. C. Garcia Rudaz, M. E. Escobar, S. V. Bengolea, M. L. Calcagno, J. D. Veldhuis, and M. Barontini Acute Effects of Testosterone Infusion on the Serum Luteinizing Hormone Profile in Eumenorrheic and Polycystic Ovary Syndrome Adolescents J. Clin. Endocrinol. Metab., September 1, 2009; 94(9): 3602 - 3610. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M Weiss, H. Huller, S. Polack, M. Friedrich, K. Diedrich, O. Treeck, G. Pfeiler, and O. Ortmann Estradiol differentially modulates the exocytotic proteins SNAP-25 and munc-18 in pituitary gonadotrophs J. Mol. Endocrinol., February 1, 2007; 38(2): 305 - 314. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. L Turgeon and D. W Waring Differential expression and regulation of progesterone receptor isoforms in rat and mouse pituitary cells and L{beta}T2 gonadotropes. J. Endocrinol., September 1, 2006; 190(3): 837 - 846. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. S. Clark, M. C. Kelton, and A. C. Whitney Chronic Administration of Anabolic Steroids Disrupts Pubertal Onset and Estrous Cyclicity in Rats Biol Reprod, February 1, 2003; 68(2): 465 - 471. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. L. Turgeon and D. W. Waring Luteinizing Hormone Secretion from Wild-Type and Progesterone Receptor Knockout Mouse Anterior Pituitary Cells Endocrinology, July 1, 2001; 142(7): 3108 - 3115. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
