ORIGINAL
RESEARCH
Stress Levels of Glucocorticoids Inhibit LH␤-Subunit
Gene Expression in Gonadotrope Cells
Kellie M. Breen, Varykina G. Thackray, Tracy Hsu, Rachel A. Mak-McCully,
Djurdjica Coss, and Pamela L. Mellon
Department of Reproductive Medicine and Center for Reproductive Science and Medicine, University of
California, San Diego, La Jolla, California 92093-0674
Increased glucocorticoid secretion is a common response to stress and has been implicated as a
mediator of reproductive suppression upon the pituitary gland. We utilized complementary in
vitro and in vivo approaches in the mouse to investigate the role of glucocorticoids as a stressinduced intermediate capable of gonadotrope suppression. Repeated daily restraint stress lengthened the ovulatory cycle of female mice and acutely reduced GnRH-induced LH secretion and
synthesis of LH ␤-subunit (LH␤) mRNA, coincident with increased circulating glucocorticoids. Administration of a stress level of glucocorticoid, in the absence of stress, blunted LH secretion in
ovariectomized female mice, demonstrating direct impairment of reproductive function by glucocorticoids. Supporting a pituitary action, glucocorticoid receptor (GR) is expressed in mouse
gonadotropes and treatment with glucocorticoids reduces GnRH-induced LH␤ expression in immortalized mouse gonadotrope cells. Analyses revealed that glucocorticoid repression localizes to
a region of the LH␤ proximal promoter, which contains early growth response factor 1 (Egr1) and
steroidogenic factor 1 sites critical for GnRH induction. GR is recruited to this promoter region in
the presence of GnRH, but not by dexamethasone alone, confirming the necessity of the GnRH
response for GR repression. In lieu of GnRH, Egr1 induction is sufficient for glucocorticoid repression of LH␤ expression, which occurs via GR acting in a DNA- and dimerization-independent
manner. Collectively, these results expose the gonadotrope as an important neuroendocrine site
impaired during stress, by revealing a molecular mechanism involving Egr1 as a critical integrator
of complex formation on the LH␤ promoter during GnRH induction and GR repression. (Molecular
Endocrinology 26: 1716 –1731, 2012)
NURSA Molecule Pages†: Ligands: Corticosterone.
S
tress profoundly disrupts reproductive function.
Whether the nature of the stressor is physical (e.g.
foot-shock, exercise), immunological (e.g. infection, administration of cytokines or endotoxins), or psychological (e.g. isolation, mental performance tasks), each has
been shown to decrease circulating levels of gonadotropins in mammals (1–5). Associated with this reproductive
disturbance is an activation of the hypothalamic-pituitary-adrenal axis and an elevation in circulating glucocorticoids from the adrenal cortex, the final hormonal
effectors of the hypothalamic-pituitary-adrenal axis. Ev-
idence that the glucocorticoid receptor (GR) antagonist,
RU486, attenuates the inhibitory effect of immobilization
stress on LH secretion in male rats or psychosocial stress
on pituitary responsiveness to GnRH in ovariectomized
ewes implies a physiological role for glucocorticoids in
mediating the inhibitory effects of stress on LH secretion,
although RU486 can also block the effects of progesterone (6 – 8).
Although there is little doubt that glucocorticoids suppress gonadotropin secretion, the neuroendocrine mechanism underlying this effect is not well understood. Inhi-
ISSN Print 0888-8809 ISSN Online 1944-9917
Printed in U.S.A.
Copyright © 2012 by The Endocrine Society
doi: 10.1210/me.2011-1327 Received November 18, 2011. Accepted July 2, 2012.
First Published Online July 31, 2012
†
1716
mend.endojournals.org
Annotations provided by Nuclear Receptor Signaling Atlas (NURSA) Bioinformatics Resource.
Molecule Pages can be accessed on the NURSA website at www.nursa.org.
Abbreviations: ChIP, Chromatin immunoprecipitation; DBD, DNA-binding domain; Egr1, early
growth response factor 1; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; GFP, green
fluorescent protein; GR, glucocorticoid receptor; GST, glutathione-S-transferase; ␣GSU, glycoprotien hormone alpha-subunit; SF1, steroidogenic factor 1; TK, thymidine kinase.
Mol Endocrinol, October 2012, 26(10):1716 –1731
Mol Endocrinol, October 2012, 26(10):1716 –1731
bition at the hypothalamic level is supported by evidence
that glucocorticoids reduce the frequency of LH pulses in
ovary-intact female sheep, ovariectomized female rats,
and women during the follicular phase of the ovulatory
cycle (9 –11). Because LH pulse frequency is generally
modulated by the GnRH neurosecretory system, these
findings suggest an action of glucocorticoids to suppress
the frequency of GnRH pulses. A recent study in follicular
phase sheep provides the first definitive evidence that glucocorticoids inhibit GnRH pulses in pituitary portal
blood (12). GR is expressed within hypothalamic neurons
implicated in GnRH regulation (13), and such neurons
provide a potential indirect target by which glucocorticoids may inhibit GnRH secretion or GnRH synthesis (14);
however, a direct action within the GnRH neuron itself is
supported by evidence that glucocorticoids blunt GnRH
synthesis and release from immortalized GnRH neurons,
GT1–7 cells (15, 16). Thus, the mechanism whereby glucocorticoids suppress GnRH and LH remains unclear and
may involve direct actions upon the GnRH neuron itself,
indirect actions via another neuronal cell type, or actions
upon an extrahypothalamic site.
With regard to a site outside of the central nervous
system, the most obvious possibility is that glucocorticoids act via GR located within gonadotrope cells of the
anterior pituitary gland. Evidence that glucocorticoids reduce the amplitude of the LH response to a GnRH challenge in rodents, pigs, cows, and women is consistent with
this possibility (17–20). Further, suppression of responsiveness to GnRH in vitro has been observed in rodent,
porcine, and bovine pituitary cell cultures, indicating that
glucocorticoids can act directly upon the gonadotrope cell
to inhibit responsiveness to GnRH (19 –21). Consistent
with an action upon the gonadotrope cell, GR has been
identified in rat gonadotropes (22), and studies in rat and
pig primary cells suggest that glucocorticoids inhibit signaling mechanisms downstream of the GnRH receptor,
including protein kinase C and cAMP (20, 23). It is not
known, however, whether these nongenomic actions of
glucocorticoids that inhibit intracellular signaling pathways ultimately lead to a reduction in LH release. Alternatively, evidence suggests that glucocorticoids can act
genomically to suppress gonadotrope responsiveness by
regulating transcription and translation of the GnRH receptor gene (24, 25).
Another potential mechanism whereby glucocorticoids could diminish GnRH responsiveness of the gonadotrope is via regulation of gonadotropin synthesis. At the
molecular level, LH and FSH are glycoprotein hormones
that exist as heterodimers, consisting of a common and
abundantly expressed glycoprotein hormone alpha-subunit (␣GSU) complexed with a unique ␤-subunit that con-
mend.endojournals.org
1717
fers biological specificity (26). Synthesis of the ␤-subunit
gene of each hormone is the rate-limiting step in the overall production of LH and FSH (26, 27). Because expression of each ␤-subunit is tightly controlled by endocrine,
paracrine, and autocrine actions, including hypothalamic
GnRH, the activin-inhibin-follistatin system, and steroid
hormones of gonadal origin (27, 28), it is possible that
GR regulation of transcriptional activity underlies the inhibitory effects of stress on the gonadotrope.
Transcriptional effects of steroid hormones within the
gonadotrope have been shown for the gonadotropin
genes, including Cga, Fshb, and Lhb (29, 30). With regard
to Lhb, androgen repression of Lhb involves protein-protein interactions between the androgen receptor and steroidogenic factor 1 (SF1) and is localized to a bipartite
SF1 element within the LH␤ proximal promoter, critical
for mediating GnRH responsiveness (31, 32). Progesterone repression also involves indirect receptor binding but
differs from androgen repression of LH␤ gene expression
in that, rather than SF1 elements, progesterone repression
involves two novel promoter regions located upstream of
the SF1 sites. Similar to progestins and androgens, glucocorticoids have been shown to inhibit LH␤ gene expression (29, 33), although the mechanism is unclear, raising
the possibility that stress impairs fertility by way of disruption of gene expression within the gonadotrope cell.
We initiated two lines of investigation in the mouse to
tease apart the mechanisms whereby elevated glucocorticoids inhibit gonadotrope responsiveness during episodes
of stress. First, we tested the hypothesis that restraint
stress, and/or an elevation in glucocorticoids mimicking
the level induced by restraint stress, can disrupt reproductive function and inhibit gonadotrope production of LH
in female mice. Second, we conducted a series of studies to
examine the molecular mechanisms underlying glucocorticoid regulation of the LH␤ promoter utilizing the immortalized L␤T2 gonadotrope cell line.
Materials and Methods
Animals
Female C57Bl/6 mice (6 wk of age) were purchased from The
Jackson Laboratory (Bar Harbor, ME), and housed in a UCSD
vivarium animal facility under standard conditions. All animals
were housed under a 12-h light, 12-h dark cycle (lights on at
0700 h) and provided with food and water ad libitum. Mice
were group housed (four females per cage) for 2 wk of acclimatization. All experimentation was performed between 0900 and
1300 h in a room within the vivarium. Care was taken to minimize pain and discomfort for the animals. Mouse colonies were
maintained in agreement with protocols approved by the Institutional Animal Care and Use Committee at the University of
1718
Breen et al.
Glucocorticoid Repression of LH␤ Gene Expression
California, San Diego. All procedures were approved by the
University of California, San Diego IACUC.
Determination of phase of estrous cycle
At 8 wk of age, vaginal lavage was performed daily (at
0900 h) by flushing the vagina with distilled H2O. Collected
smears were mounted on glass slides and examined microscopically for cell type (34). The smears were classified into one of
three phases of estrus: diestrus, proestrus, or estrus. Female mice
exhibiting two consecutive 4- to 6-d estrous cycles, including
positive classification of diestrus, proestrus, and estrus, were
used in animal experiments. Estrous cycle length was calculated
as the length of time between two successive occurrences of
estrus. The time spent in each cycle stage was calculated as the
proportion of time classified in each cycle phase during the
observation period, and values were analyzed by two-way repeated measures ANOVA, group (no stress vs. stress) ⫻ time (d
1– d 18 vs. d 19 – d 36). All statistics were performed using JMP
7.0 (SAS Institute, Cary, NC), and significance was established
as P ⬍ 0.05.
Restraint stress protocol
After vaginal lavage and estrus classification, mice were either returned to their group home cage (no stress) or placed
individually into clear plastic restraint tubes (stress). The ventilated tubes (Harvard Apparatus, Holliston, MA) are designed to
be small enough to restrain a mouse so that it is able to breathe
but unable to move freely. The restraint devices were cleaned
between uses with soap, water, and ethanol (70%). Mice are
continually observed by experienced personnel during the 180min restraint period. After the restraint period, stress mice are
returned to individual home cages or killed by decapitation, and
trunk blood or pituitary tissue was collected from individual
animals. Hormone values were analyzed by one-way ANOVA
followed by Tukey’s post hoc test or two-way ANOVA, group
(no stress vs. stress) ⫻ time (vehicle vs. GnRH).
Corticosterone response protocol
Female mice (8 wk of age) were ovariectomized and allowed
to recover for 2 wk before experimentation. To test the LH
response to a stress level of corticosterone, animals received a
bolus injection of corticosterone (200 ng/kg, sc) or vehicle. After
90 min, animals were killed by decapitation, and trunk blood
was collected from individual animals. Hormone values were
analyzed by one-way ANOVA followed by Tukey’s post hoc
test.
Hormone analysis
Trunk blood was collected and serum separated by centrifugation and stored frozen at ⫺20 C before analysis at the Center
for Research in Reproduction Ligand Assay and Analysis Core
at the University of Virginia (Charlottesville, VA). Corticosterone concentrations were determined by RIA in single 25- to
50-␮l aliquots of serum. Assay sensitivity averaged 20.0 ng/ml.
LH concentrations were determined by two-site sandwich immunoassay in duplicate 50-␮l aliquots of serum. Assay sensitivity averaged 0.07 ng/ml.
Mol Endocrinol, October 2012, 26(10):1716 –1731
Plasmids
The ⫺1800 rat LH␤-luc in pGL3 was kindly provided by Dr.
Mark Lawson. The 5⬘-truncations of the ⫺1800-bp LH␤-luc
reporter plasmid were created by inserting fragments between
KpnI and HindIII in pGL3 have been partially described by
Thackray et al. (35): ⫺400, ⫺300, ⫺200, ⫺150, ⫺87. The 5⬘and 3⬘-mutations of either Egr1 or SF1 binding sites in the ⫺200
rat LH␤-luc have been reported previously (35). The reporter
plasmids with the multimerized consensus SF1 site (TGACCTTGA) or consensus Egr1 site (CGCCCCCGC) were created by
ligating an oligonucleotide, containing four copies of the indicated site, between KpnI and NheI in pGL3, upstream of the
81-bp thymidine kinase (TK) promoter. The sequences of all
promoter fragments were confirmed by dideoxynucleotide
sequencing.
The mouse Ptx1 pcDNA3 expression vector has been previously described (36). The rat Egr1 and mouse SF1 cDNA were
kindly provided by Dr. Jacques Drouin and Dr. Bon-chu Chung,
respectively. The cDNA were cloned into the pCMV expression
vector using the ClaI/XbaI restriction sites of both plasmids. The
human Egr1 cDNA was provided by Dr. Hermann Pavenstadt
and was cut using XhoI/EcoRI and inserted into pGEX-5X expression vector using the SmaI restriction site by blunt end cloning. Dr. Douglass Forbes provided the green fluorescent protein
(GFP) expression plasmid. The wild-type rat GR pSG5 plasmid
was provided by Dr. Keith Yamamoto. Both GR mutants are in
pSG5 and have been previously described; the GRdim4 mutant
contains four point mutations that prevent dimerization and
DNA binding of the receptor (29), and the GR DNA-binding
domain (DBD) mutant contains a mutation in the DNA-binding
domain (33).
Cell culture and transient transfection
L␤T2 cells, cultured as previously described (29), were
seeded into 12-well plates at 3 ⫻ 105 cells per well and incubated
overnight at 37 C. Each well was transfected with 400 ng of the
luciferase-reporter plasmid or control pGL3 vector and 100 ng
of a ␤-galactosidase reporter gene regulated by the TK promoter
(TK-␤gal) as a control for transfection efficiency using FuGENE
6 transfection reagent (Roche Applied Science, Indianapolis,
IN). In experiments utilizing Egr1, SF1, or Ptx1 to induce promoter activity, cells were also transfected with 100 ng Egr1
(unless indicated otherwise) or empty pCMV vector, 100 ng SF1
or empty pCMV vector, 50 ng Ptx1 or empty pcDNA3 vector.
Eighteen hours after transfection, cells were transferred to serum-free DMEM (supplemented with 0.1% BSA, 5 mg/liter
transferrin, and 50 nM sodium selenite) containing either the
natural glucocorticoid, corticosterone (100 nM, Sigma Aldrich,
St. Louis, MO), synthetic glucocorticoid, dexamethasone (100
nM, Sigma Aldrich), or vehicle (0.1% ethanol). When cells were
treated with GnRH (10 nM; Sigma Aldrich), treatment with
GnRH or vehicle (0.1% BSA) began 6 h before harvest. Cells
were harvested and extracts were prepared for assay of luciferase and ␤-galactosidase activity as previously described
(33).
CV-1 cells, a monkey kidney cell line that does not express
detectable endogenous GR (37), were seeded into 12-well plates
at 1.5 ⫻ 105 cells per well as previously described (38) and
treated as indicated above with the following addition. Each
Mol Endocrinol, October 2012, 26(10):1716 –1731
well was transfected with 50 ng of the GR expression vector
(wild type or mutant) or empty pGS5 vector.
Luciferase reporter experiments were performed in triplicate
and were repeated at least three times. To normalize for transfection efficiency, all luciferase values were divided by ␤-galactosidase, and the triplicate values were averaged. To control for
interexperimental variation, the control pGL3 reporter plasmid
was transfected with TK-␤gal and any relevant expression vectors, and the average pGL3/␤gal value was calculated. Average
luc/␤gal values were divided by the corresponding pGL3/␤gal
value. Individual values obtained from each independent experiment were averaged and analyzed by Student’s t test or one-way
ANOVA followed by Tukey’s post hoc test.
Quantitative real-time PCR
Preparation of cDNA from mouse pituitary or L␤T2 cells
was performed as previously described (39). Briefly, RNA was
extracted with Trizol reagent (Invitrogen/GIBCO, Carlsbad,
CA) according to the manufacturer’s instructions, treated to
remove contaminating DNA (DNA-free; Ambion, Austin, TX),
and reverse transcribed using Superscript III First-Strand Synthesis System (Invitrogen). Quantitative real-time PCR was performed in an iQ5 Real-Time PCR instrument (Bio-Rad Laboratories, Inc., Hercules, CA) and used iQ SYBR Green Supermix
(Bio-Rad Laboratories) with specific primers for glyceraldehyde-3-phosphate dehydrogenase (GAPDH), LH␤, or Egr1
cDNA.
LH␤ forward: CTGTCAACGCAACTCTGG
LH␤ reverse: ACAGGAGGCAAAGCAGC
Egr1 forward: ATTTTTCCTGAGCCCCAAAGC
Egr1 reverse: ATGGGAACCTGGAAACCACC
GAPDH forward: TGCACCACCAACTGCTTAG
GAPDH reverse: GGATGCAGGGATGATGGTTC
The iQ5 real-time PCR program was as follows: 95 C for 15
min, followed by 40 cycles at 95 C for 15 sec, 55 C for 30 sec,
and 72 C for 30 sec. Within each experiment, the amount of
LH␤ or Egr1 and GAPDH mRNA was calculated by comparing
a threshold cycle obtained for each sample with the standard
curve generated from serial dilutions of a plasmid containing
GAPDH, ranging from 1 ng to 1 fg. All samples were assayed (in
triplicate) within the same run, and the experiment was conducted three times. Values were analyzed by one-way ANOVA
followed by Tukey’s post hoc test or two-way ANOVA, group
(no stress vs. stress) ⫻ time (vehicle vs. GnRH).
Dual immunofluorescence
Adult mouse pituitary paraffin tissue sections (Zyagen, San
Diego, CA) were dewaxed in xylene, rehydrated through a series
of graded ethanol baths, and washed in H20. Sections were
immersed in 10 mM sodium citrate buffer (pH 6.0), heated in a
standard microwave twice for 5 min, and allowed to stand for
20 min at room temperature. After a brief wash in PBS, nonspecific binding was blocked with 5% goat serum/0.3% Triton
X-100 for 60 min at room temperature. Dual fluorescence labeling was tested on the same section with a guinea pig antirat
LH␤ primary antibody (anti-r␤ LH-IC-2, NIDDK NHPP; 1:200
dilution in 5% goat serum/0.3% Triton X-100) plus a rabbit
antimouse GR primary antibody (sc-1004, Santa Cruz Biotechnology, Inc., Santa Cruz, CA; 1:500 dilution) for 48 h at 4 C.
LH␤- and GR-containing cells were revealed using a goat rho-
mend.endojournals.org
1719
damine antiguinea pig IgG secondary antibody (ab6905, 1:300
dilution; Abcam, Cambridge, MA) plus a goat fluorescein antirabbit IgG secondary antibody (FI-1000; Vector Laboratories,
Burlingame, CA; 1:300 dilution) for 60 min at room temperature. After rinsing with PBS, sections were coverslipped with
Vectashield HardSet mounting medium with 4⬘,6-diamidino-2phenylindole (Vector Laboratories). Exclusion of each primary
antibody was run as a negative control, and each antibody was
run separately to confirm that each immunostaining pattern was
similar to published reports (40, 41). Specificity of the GR antibody was confirmed by immunoblot analysis of L␤T2 and
CV-1 cell lysates, cell lines previously shown to contain and lack
GR (29, 37), respectively, which detected a single band of the
expected size.
Sections were viewed using an inverted fluorescence microscope (Nikon Eclipse TE 2000-U; Nikon, Tokyo, Japan), and
digital images were collected using a Sony CoolSNAP EZ cooled
charge-coupled device camera (Roper Scientific, Trenton, NJ)
and analyzed with Nikon Imaging System—Elements image
analysis system (version 2.3; Nikon).
Western blot analysis
Nuclear extracts were prepared as previously described (36)
from L␤T2 cells treated with dexamethasone (100 nM, 18 h),
GnRH (10 nM, 45 min), GnRH ⫹ dexamethasone or vehicle
(0.1% BSA/0.1% ethanol). Nuclear extract (20 ␮g) was boiled
for 5 min in 5⫻ Western-loading buffer, fractionated on a 10%
SDS-PAGE gel, and electroblotted for 90 min at 300 mA onto
polyvinylidene difluoride (Millipore Corp., Billerica, MA) in 1⫻
Tris-glycine-sodium dodecyl sulfate/20% methanol. Blots were
blocked overnight at 4 C in 3% BSA and then probed for 1 h at
room temperature with rabbit antihuman Egr1 antibody (sc110, Santa Cruz Biotechnology) diluted 1:750 in blocking buffer. Blots were then incubated with a horseradish peroxidaselinked secondary antibody (Santa Cruz Biotechnology), and
bands were visualized using the SuperSignal West Pico chemiluminescent substrate (Pierce Biotechnology, Inc., Rockford,
IL). Bio-Rad Pre-stained Protein Ladder Plus serves as a size
marker.
Chromatin immunoprecipitation (ChIP)
ChIP assays were performed as previously described (29, 35).
Briefly, confluent L␤T2 cells in 10-cm plates were treated with
dexamethasone (100 nM, 1 h), GnRH (10 nM, 1 h), GnRH ⫹
dexamethasone (cotreatment 1 h), or vehicle (0.1% BSA/0.1%
ethanol) and cross-linked with 1% formaldehyde. The nuclear
fraction was obtained, and chromatin was sonicated to an average length of 300 –500 bp using a Branson Sonifier 250 (Branson Ultrasonics Corp., Danbury, CT). Protein-DNA complexes
were incubated overnight with nonspecific rabbit IgG (sc-2027,
Santa Cruz Biotechnology) or rabbit antihuman GR (ab3579,
Abcam) and precipitated with protein A/G beads. Immunoprecipitated DNA and DNA from input chromatin were analyzed
by quantitative PCR using primers specific for a 220-bp sequence of the mouse LH␤ proximal promoter (⫺180 LH␤/⫹40
LH␤). Primers specific to the mouse FSH␤ promoter (⫺223
FSH␤/⫹57 FSH␤) and FSH␤ coding region were used as positive and negative controls, respectively. DNA from immunoprecipitated samples was quantified relative to a standard curve
representing percent of input chromatin. For ChIP assays com-
Glucocorticoid Repression of LH␤ Gene Expression
A
prestress
E
P
D
E
P
D
1
35
S-labeled proteins were produced using the TnT Coupled
Reticulolysate System (Promega Corp., Madison, WI). Bacteria
transformed with the GST plasmids were grown to an OD of 0.6
and then induced with isopropyl-␤-d-thiogalactoside overnight
at 30 C. The bacterial pellets were sonicated in 0.1% Triton
X-100 and 5 mm EDTA in 1⫻ PBS and centrifuged, and the
supernatant was bound to glutathione Sepharose 4B resin (Amersham Pharmacia Biotech, Piscataway, NJ). The beads were
washed four times in PBS and then in HEPES/Nonidet P-40/
dithiothreitol (HND) buffer [10 mg/ml BSA, 20 mm HEPES (pH
7.8), 50 mm NaCl, 5 mm dithiothreitol, and 0.1% Nonidet
P-40]. For the interaction assay, 35S-labeled in vitro-transcribed
and -translated GR (20 ␮l), SF1 (5 ␮l), or GFP (5 ␮l) was added
to the beads with 400 ␮l HND buffer. The beads were incubated
for 1 h at 4 C and then washed twice with HND buffer and twice
with 0.1% Nonidet P-40 in PBS. Thirty microliters of 2⫻ Laemmli load buffer were added, and the samples were boiled and
then electrophoresed on a 10% sodium dodecyl sulfate-polyacrylamide gel. One tenth of the 35S-labeled in vitro-transcribed
and -translated product was loaded onto the gel as input. The gel
was dried, and the proteins were visualized by autoradiography.
Results
Chronic restraint stress compromises estrous
cyclicity
To evaluate the mechanism whereby elevated glucocorticoids impair reproduction, we developed a model to
assess whether daily restraint stress disrupts estrous cyclicity. Vaginal cytology was examined daily by vaginal
lavage in a cohort of female mice during an 18-d control
period (d 1– d 18). After this period of observation, mice
were randomly assigned to either the stress or no stress
9
18
27
Day of assessment
36
C
B
8
7
*
6
5
4
3
Glutathione-S-transferase (GST) interaction assay
daily restraint stress
E
P
D
1-18 19-36 1-18 19-36
No stress
Stress
Time in cycle stage (days)
paring chromatin from hormone-treated L␤T2 cells, the fold
enrichment of antibody signal over IgG was calculated for each
primer set, and data from each independent experiment were
normalized to the indicated negative control gene. ChIP and
input samples were quantified using a standard curve made
from ChIP input DNA. ChIP samples were normalized to their
appropriate input samples and are expressed as fold enhancement over nonspecific IgG.
⫺180 LH␤-forward: CGAGTGTGAGGCCAATTCACTGG
⫹40 LH␤-reverse: GGCCCTACCCATCTTACCTGGAGC
⫺223 FSH␤-forward: GGTGTGCTGCCATATCAGATTCGG
⫹57 FSH␤-reverse: GCATCAAGTGCTGCTACTCACCTGTG
FSH␤-coding forward: GCCGTTTCTGCATAAGC
FSH␤-coding reverse: CAATCTTACGGTCTCGTATACC
The iQ5 real-time PCR program was as follows: 95 C for 15
min, followed by 40 cycles at 95 C for 15 sec, 55 C for 30 sec,
and 72 C for 30 sec. All samples were assayed within the same
run, and the experiment was conducted three times. Individual
values obtained from each independent experiment were averaged and analyzed by one-way ANOVA followed by Tukey’s
post hoc test.
Mol Endocrinol, October 2012, 26(10):1716 –1731
Vaginal histological classification
Breen et al.
Cycle length (days)
1720
5
4
*
3
2
1
0 E PD E PD
1-18 19-36
Stress
FIG. 1. Daily stress disrupts estrous cyclicity in the mouse. A,
Representative profiles depicting estrous cyclicity, as measured by
vaginal cytology, during the prestress period (d 1– d 18) and
subsequent daily restraint stress period (d 19 – d 36) in three female
mice subjected to 180 min of daily restraint stress. E, Estrus; P,
proestrus; D, diestrus. Shading indicates period of exposure to daily
restraint stress. B, Average estrous cycle length in the no stress group
and stress group (n ⫽ 9/group) during the two periods of assessment,
d 1– d 18 and d 19 – d 36. C, Time spent in each stage of the cycle in
stress mice during the prestress period, d 1– d 18, and stress period, d
19 – d 36. *, Significant (P ⬍ 0.05) group (no stress vs. stress) ⫻ time (d
1– d 18 vs. d 19 – d 36) interaction.
group (n ⫽ 9/group), and estrous cyclicity was monitored
for an additional 18 d (d 19 – d 36). Figure 1A illustrates
profiles of vaginal histological classification during the
prestress and stress period of three female mice exposed to
180 min of daily restraint stress. In mice not subjected to
stress, cycle length was not significantly different between
the two periods of assessment (P ⬎ 0.05; d 1– d 18 vs. d
19 – d 36, 5.33 ⫾ 0.31 vs. 4.81 ⫾ 0.45 d, Fig. 1B). In
contrast, stressed mice exhibited a significant increase in
average cycle length in the stress period compared with
the prestress period (P ⬍ 0.05; d 1– d 18 vs. d 19 – d 36,
5.25 ⫾ 0.35 vs. 6.75 ⫾ 0.55 d). Specifically, exposure to
daily restraint stress significantly increased the time spent
in diestrus during the stress period, d 19 – d 36, without
altering the time spent in estrus or proestrus as compared
with the prestress period, d 1– d 18 (P ⬎ 0.05; Fig. 1C).
Acute restraint stress disrupts gonadotrope
function
Having found that repeated exposure to stress compromises reproductive neuroendocrine activity as evi-
Mol Endocrinol, October 2012, 26(10):1716 –1731
A
Blood
(n=7/grp)
Blood
(n=7/grp)
1721
GnRH Blood
or Veh (n=7/grp/trt)
No stress
Stress
30
0
*
400
*
200
*
2.0
*
1.5
1.0
0.5
0
0 30
180
Time (min)
No stress
D
GnRH
or Veh
Ve
G h
nR
H
0
180
#
2.5
Ve
G h
nR
H
600
C
No stress
Stress
LH (ng/ml)
Corticosterone (ng/ml)
800
170
Time (min)
B
Stress
Pit
(n=7/grp/trt)
No stress
Stress
Time (min)
0
180
E
25
LHβ/GAPDH mRNA
denced by a disruption in estrous cyclicity in female mice,
the next step was to focus on the role of the pituitary
gland and test whether gonadotrope responsiveness, as
assessed by GnRH-induced LH synthesis or secretion, is
diminished by acute restraint stress. Separate cohorts of
female mice, in which estrus cyclicity was confirmed and
diestrus females selected, were used to assess gonadotrope
responsiveness to GnRH in the absence or presence of
stress. In the first study, GnRH-induced LH secretion was
monitored in groups of control mice (no stress) or mice
exposed to 180 min of restraint stress (Fig. 2A). Blood
was collected before stress, as well as 30 and 180 min
after the initiation of the control or stress period for measurement of corticosterone (n ⫽ 7 per time point per
group). To assess LH release in response to GnRH, female
diestrus mice were administered GnRH (200 ng/kg, sc;
n ⫽ 7 per group per treatment) or saline vehicle and killed
10 min after injection. Blood was collected and processed
for measurement of LH at 180 min after the initiation of
stress. This GnRH dose was selected because it has been
shown to produce a significant, yet submaximal, LH response, that peaks 10 min after administration in a mouse
model of diestrus in which female mice are ovariectomized and estrogen primed (42, 43). Serum levels of corticosterone remained low in no stress control animals
(Fig. 2B, open circles); corticosterone levels were significantly increased in stressed mice at 30 min (P ⬍ 0.05;
stress vs. no stress, 588.6 ⫾ 26.4 vs. 108.8 ⫾ 35.5 ng/ml)
and remained significantly elevated 180 min after initiation of restraint (Fig. 2B, gray circles). Stress did not significantly induce serum levels of progesterone at either 30
min or 180 min after the initiation of restraint (data not
shown). Mean LH in the no stress diestrus females receiving vehicle was not significantly different from values in
the vehicle-treated stress animals, indicating that stress
does not significantly impair responsiveness to endogenous GnRH in this animal model (P ⬎ 0.05; stress Veh vs.
no stress Veh, Fig. 2C). In the no stress group, exogenous
GnRH caused a robust increase in circulating LH levels as
compared with vehicle-treated animals (P ⬍ 0.05; 1.83 ⫾
0.23 vs. 0.56 ⫾ 0.11 ng/ml, Fig. 2C). GnRH also significantly increased LH in stressed animals (P ⬍ 0.05, Fig.
2C). However, the LH response to exogenous GnRH was
significantly blunted in restraint-stressed animals compared with the response in no stress controls (P ⬍ 0.05;
stress GnRH vs. no stress GnRH, 1.19 ⫾ 0.18 vs. 1.83 ⫾
0.23 ng/ml; Fig. 2C), suggesting that stress diminishes the
ability of the pituitary to respond to GnRH. Taken together, these experiments reveal an interplay between
GnRH and stress and implicate responsiveness of the pituitary gonadotrope as a potential neuroendocrine site of
LH suppression.
mend.endojournals.org
20
Pituitary
#
*
15
*
10
5
0
Veh GnRH
No stress
Veh GnRH
Stress
FIG. 2. Acute stress disrupts pituitary responsiveness to GnRH. A,
Schematic depicting events during the 180-min observation period in
which animals were maintained in no stress conditions (white bar) or
subjected to restraint stress (gray bar) for measurement of circulating
corticosterone and GnRH-induced LH. Time of euthanasia and blood
collection (Blood coll’n) are indicated: 0, 30, and 180 min. At 170 min,
no stress and stressed animals (group) are divided into two treatments
(n ⫽ 7/group per treatment) receiving either GnRH (200 ng/kg, sc) or
vehicle (Veh). B, Serum corticosterone (ng/ml) was measured in no
stress (white circles) and stress animals (gray circles). *, Significant
(P ⬍ 0.05) effect of stress. C, Serum LH (ng/ml) was measured in no stress
(white bars) and stressed (gray bars) animals that received vehicle or
GnRH, respectively, 10 min before euthanasia. *, Significant (P ⬍ 0.05)
effect of GnRH; #, difference between no stress and stress. D, Schematic
depicting events during 180-min observation period in which animals are
maintained in no stress conditions (white bar) or subjected to restraint
stress (gray bar) for measurement of GnRH-induced LH␤ mRNA. No stress
and stressed animals are divided into two groups (n ⫽ 7/group) receiving
either GnRH (200 ng/kg, sc) or vehicle (Veh) at 0 min of observation. Time
of euthanasia and blood collection (Blood coll’n) occurred at 180 min. E,
Quantitative RT-PCR analysis of LH␤ mRNA was performed on individual
mouse pituitary glands, and the amount of LH␤ mRNA was compared
with the amount of GAPDH mRNA and expressed as relative transcript
level. *, Significant (P ⬍ 0.05) effect of GnRH; #, difference between no
stress and stress. grp, Group; trt, treatment.
Glucocorticoid Repression of LH␤ Gene Expression
We further investigated the response of the gonadotrope by analyzing GnRH-induced LH␤ mRNA transcript level in mice exposed to restraint stress. Initially,
diestrus female mice received saline vehicle or GnRH (200
ng/kg, sc; n ⫽ 7 per group per treatment) and were subsequently sorted into no stress or stress treatment groups
(Fig. 2D). Animals were killed and the pituitary glands
were collected for mRNA analysis by quantitative RTPCR. Stress did not alter basal LH␤ transcript levels compared with expression in no stress vehicle controls (P ⬍
0.05; stress Veh vs. no stress Veh; Fig. 2E). In the absence
of stress, exogenous GnRH resulted in a 4-fold increase in
LH␤ mRNA compared with values in vehicle-treated animals (P ⬍ 0.05; Fig. 2E). In stressed mice, the LH␤ transcript level in response to GnRH was significantly reduced by approximately 45% (P ⬍ 0.05; no stress GnRH
vs. stress GnRH, 16.5 ⫾ 2.3 vs. 8.9 ⫾ 1.1; Fig. 2E).
Collectively, these observations raise the possibility that
stress can interfere with ovarian cyclicity by disrupting
the synthesis and secretion of LH at the level of the anterior pituitary gonadotrope cell.
Glucocorticoids impair LH secretion in vivo
potentially via receptors expressed in mouse
gonadotrope cells
Our studies thus far demonstrate that circulating glucocorticoids are increased within 30 min and remain significantly elevated for the 180-min stress paradigm (Fig.
2). Because stress likely induces a host of inhibitory intermediates, any of which could alter reproductive activity,
we directly assessed the role of glucocorticoids by testing
the hypothesis that a stress-like level of glucocorticoid in
female mice reduces LH secretion. Pilot studies were conducted to identify a dose of glucocorticoid that approximated a stress level (⬃750 ng/ml) and an animal model
that eliminated confounding effects of ovarian steroids.
Blood was collected 90 min after a bolus administration
of vehicle or corticosterone (200 ng/kg, sc; n ⫽ 6 per
treatment) to ovariectomized female mice (Fig. 3A). Corticosterone remained low in mice treated with vehicle, yet
values were markedly elevated after administration of
corticosterone (P ⬍ 0.05; Fig. 3B). Treatment with corticosterone significantly reduced mean LH as compared
with the value in mice treated with vehicle (P ⬍ 0.05; Cort
vs. Veh, 3.4 ⫾ 0.3 vs. 1.9 ⫾ 0.6 ng/ml, Fig. 3C), demonstrating sufficiency of glucocorticoids to disrupt reproductive neuroendocrine function and relevancy as an inhibitory factor induced during stress.
Evidence that GR is expressed in gonadotrope cells in
the rat (22) supports our hypothesis of a direct action of
glucocorticoids via GR within mouse gonadotrope cells.
To confirm the presence of this receptor in mouse go-
Mol Endocrinol, October 2012, 26(10):1716 –1731
A
Cort
or Veh
Blood
(n=6/trt)
0
B
D
Time (min)
* C
1000
750
500
250
0
90
4
LH (ng/ml)
Breen et al.
Corticosterone (ng/ml)
1722
3
*
2
1
0
Veh
E
Veh
Cort
Cort
F
*
*
*
FIG. 3. Glucocorticoids inhibit LH and GR are expressed in mouse
gonadotrope cells. A, Schematic depicting experimental events in
which animals were administered corticosterone (200 ng/kg, sc; gray
bar) or vehicle (white bar), 90 min before euthanasia and blood
collection (Blood coll’n). B and C, Serum corticosterone (ng/ml, panel
B) or LH (ng/ml, panel C) measured in animals administered vehicle
(white bars) or corticosterone (gray bars). *, Significant (P ⬍ 0.05)
effect of treatment. D–F, Photomicrographs of representative mouse
pituitary sections subjected to two-color immunofluorescence staining
using a fluorescein isothiocyanate-conjugated anti-GR (green
fluorescent signal), followed by rhodamine-conjugated anti-LH␤ (red
fluorescent signal). Red (panel D), green (panel E), and merge (panel F)
immunofluorescence images were taken of the same microscopic field
using appropriate filters. White stars, GR-positive/LH␤-positive cells.
Scale bar, 20 ␮m. Cort, Corticosterone; trt, treatment; Veh, vehicle.
nadotrope, we used dual-label immunofluorescence of
adult mouse anterior pituitary sections for LH␤ and GR
(rhodamine- and fluorescein isothiocyanate-conjugated
secondary antibodies, respectively; Fig. 3, D–F). LH␤ immunostaining identified gonadotropes that accounted for a
small proportion of labeled anterior pituitary cells (Fig. 3D),
whereas GR immunostaining occurred in an extensive population of pituitary cells (Fig. 3E). Of interest, numerous
LH␤-containing gonadotropes showed GR staining, confirming the presence of GR in this anterior pituitary cell type
in the mouse (merge, white stars, Fig. 3F).
Glucocorticoids regulate LH␤ gene expression in
gonadotrope cells
Having confirmed the presence of GR in adult mouse
pituitary gland, we next tested the hypothesis that the
stress-induced decrease in LH␤ mRNA expression could
be recapitulated in cultured gonadotrope cells. As expected, immortalized pituitary gonadotrope cells, L␤T2,
responded to GnRH with a 2-fold increase in endogenous
LH␤ mRNA as measured by quantitative RT-PCR (Fig.
Mol Endocrinol, October 2012, 26(10):1716 –1731
mend.endojournals.org
A
LHβ/GAPDH mRNA
4A). Treatment of L␤T2 cells with a physiological, stress
level of corticosterone decreased GnRH-induced LH␤
mRNA expression (P ⬍ 0.05; GnRH vs. GnRH ⫹ Cort,
2.1 ⫾ 0.3 vs. 1.6 ⫾ 0.1; Fig. 4A); however, corticosterone
alone did not reduce LH␤ expression. These findings indicate that repression by glucocorticoids occurs in the presence of GnRH and confirm the L␤T2 gonadotrope cell is an
LHβ-Luc/βgal
Fold induction
B
LHβ-Luc/βgal
Fold induction
C
6
5
4
3
2
1
0
#
3.0
2.5
*
2.0
*
1.5
1.0
0
Veh
Cort GnRH GnRH
+Cort
*
*
*
GnRH
GnRH + Dex
4
2
#
GnRH GnRH GnRH
+Cort +Dex
Veh Cort Dex
5
3
#
#
#
#
#
#
1
0
-1800 -400 -300 -200 -150
-87
Truncation from transcription start site
FIG. 4. Glucocorticoid repression localizes to the LH␤ proximal
promoter. A, Quantitative RT-PCR analysis of LH␤ mRNA extracted
from L␤T2 cells cultured in the presence of GnRH (10 nM, 6 h),
corticosterone (Cort; 100 nM, 24 h), GnRH ⫹ corticosterone [Cort (100
nM, entire 24 h); GnRH (10 nM, final 6 h)], or vehicle (Veh; 0.1% BSA/
0.1% ethanol). Results are expressed as LH␤ mRNA levels normalized
to GAPDH mRNA levels and are the mean of three separate
experiments performed in triplicate. Results shown are average ⫾ SEM
relative to the vehicle treatment. *, Significant (P ⬍ 0.05) effect of
GnRH; #, significant repression by corticosterone. B, The ⫺1800-bp rat
LH␤-luc reporter gene was transfected into L␤T2 cells, and cells were
subsequently treated with corticosterone (Cort; 100 nM, 24 h),
dexamethasone (Dex; 100 nM, 24 h), or vehicle alone (Veh; 0.1% BSA/
0.1% ethanol) or either glucocorticoid (100 nM, entire 24 h) in
combination with GnRH (10 nM, final 6 h), and harvested for luciferase
as a measure of LH␤ promoter activity. Results are depicted as fold
induction by hormone treatment relative to vehicle (dashed line) as
indicated. *, Significant induction by GnRH vs. vehicle control; #,
significant repression by glucocorticoid treatment on GnRH-induced
LH␤ expression. C, L␤T2 cells were transfected with a series of 5⬘truncated LH␤-luc reporter plasmids and treated with GnRH (10 nM,
final 6 h) or GnRH ⫹ dexamethasone [Dex (100 nM, entire 24 h); GnRH
(10 nM, final 6 h)] to determine regions of the LH␤ promoter that are
responsive to glucocorticoids. Results for each truncation are depicted
as LH␤ fold induction relative to vehicle of that truncation (dashed
line). #, Significant repression by glucocorticoid treatment on GnRHinduced LH␤ expression.
1723
appropriate model for investigating the role of glucocorticoid-induced suppression of LH␤ gene expression.
We began to dissect out the mechanism whereby glucocorticoids reduce LH␤-subunit induction in L␤T2 cells,
by investigating the regulation of an LH␤-luciferase reporter after incubation with either natural or synthetic
glucocorticoids. For this purpose, L␤T2 cells were transiently transfected with ⫺1800 bp of the rat LH␤ regulatory region fused upstream of a luciferase reporter gene
(LH␤-luc) or pGL3 control plasmid and treated with corticosterone (100 nM), dexamethasone (100 nM), or vehicle
(0.1% ethanol) for 24 h before harvest. Neither the natural glucocorticoid, corticosterone, nor the synthetic glucocorticoid, dexamethasone, significantly altered basal
expression of the ⫺1800-bp LH␤ promoter (P ⬎ 0.05;
Veh vs. Cort or Dex; Fig. 4B). In contrast, both glucocorticoids significantly blunted the robust increase in promoter activity induced by GnRH (10 nM, final 6 h of
glucocorticoid treatment; P ⬍ 0.05; Fig. 4B). Specifically,
GnRH resulted in a 4.6-fold induction of LH␤-luc, which
was reduced 35% by corticosterone, or 45% by dexamethasone. Although both glucocorticoids are capable of
transcriptional repression of GnRH induction of the LH␤
promoter, we focused on the synthetic glucocorticoid,
dexamethasone, based on the intensity of its effect and
evidence for its potent interaction with the endogenous
steroid receptor, GR, expressed in L␤T2 cells (29) and
identified in mouse gonadotrope cells (Fig. 3).
Using a promoter truncation approach, we identified
regions of the LH␤ gene that are functionally involved in
glucocorticoid regulation. L␤T2 cells were transiently
transfected with a series of truncated LH␤ reporter plasmids, ranging in length from ⫺1800 to ⫺87 bp of the
5⬘-regulatory sequence. Figure 4C illustrates the effect of
treatment with GnRH in the presence or absence of dexamethasone on progressive 5⬘-LH␤ promoter truncations.
As observed previously, GnRH induction of the LH␤ gene
declined incrementally as the promoter was progressively
truncated from ⫺1800 to ⫺87 (Fig. 4C and Ref. 44).
Dexamethasone repressed LH␤ promoter activity by approximately 40% when the region contained at least
⫺150 bp of the proximal promoter. Interestingly, further
truncation of the region from ⫺150 to ⫺87, which removed the proximal GnRH responsive elements [such as
early growth response factor 1 (Egr1) and SF1], resulted
in a loss of GnRH induction and elimination of dexamethasone repression, indicating that GnRH responsiveness of the LH␤ gene is required for glucocorticoid repression. Collectively, these data suggest that GR exerts its
inhibitory effects upon the LH␤ proximal promoter,
likely via interactions with GnRH-responsive factors,
such as Egr1 and SF1.
1724
Breen et al.
Glucocorticoid Repression of LH␤ Gene Expression
Proximal binding elements are important for
glucocorticoid repression
The proximal 150 bp of the rat LH␤ promoter contain
multiple binding elements that are critically important for
GnRH regulation of LH␤ transcription (Fig. 5A), including tandem elements for both Egr1 and SF1 that are arranged on either side of a homeobox element, previously
shown to bind pituitary homeobox factor 1 (Ptx1). Ptx1 is
A
-87
SF1 Egr1 Ptx1 SF1 Egr1
*
Veh
20
10
0
6.0
Veh
Dex GnRH GnRH
+Dex
Ab: αEgr1
ChIP: GR/IgG
*
4.0
GnRH + Dex
*
30
D
Fold enhancement
C
n.s.
Dex
Egr1/GAPDH mRNA
B
rat LHβ
GnRH
-150
*
2.0
0
Veh
Dex
GnRH
GnRH
+Dex
FIG. 5. GnRH-responsive factor necessary for GR recruitment. A,
Schematic of the proximal 150 bp of the rat LH␤ 5⬘-regulatory region
illustrating the known promoter elements involved in expression of the
LH␤ gene. Proteins binding each site are indicated. B, Quantitative RTPCR analysis of Egr1 mRNA extracted from L␤T2 cells cultured in the
presence of dexamethasone (Dex; 100 nM, 18 h), GnRH (10 nM, 45
min), GnRH ⫹ dexamethasone [Dex (100 nM, entire 18 h); GnRH (10
nM, final 45 min)], or vehicle (Veh; 0.1% BSA/0.1% ethanol). Results
are expressed as Egr1 mRNA levels normalized to GAPDH mRNA levels
and are the mean of three separate experiments performed in
triplicate. Results shown are average ⫾ SEM. *, Significant induction by
GnRH vs. vehicle control. C, Western blotting analysis of nuclear
extracts from L␤T2 cells treated with dexamethasone (Dex; 100 nM,
18 h), GnRH (10 nM, 2 h), GnRH ⫹ dexamethasone [Dex (100 nM,
entire 18 h); GnRH (10 nM, final 2 h)], or vehicle (Veh; 0.1% BSA/0.1%
ethanol) was performed using an antibody for Egr1. A protein band
was detected at the expected size of 82 kDa for Egr1. The experiment
was repeated three times with similar results, and a representative gel
is shown. D, ChIP was performed using cross-linked chromatin from
L␤T2 cells treated with dexamethasone (Dex; 100 nM, 1 h), GnRH (10
nM, 1 h), GnRH ⫹ dexamethasone [Dex (100 nM, 1 h); GnRH (10 nM,
1 h)], or vehicle (Veh; 0.1% BSA/0.1% ethanol) and antibodies
directed against GR or nonspecific IgG as a negative control. ChIP and
input DNA were analyzed by quantitative PCR using primers
encompassing the proximal promoter of Lhb to determine the amount
of immunoprecipitated DNA. ChIP samples were normalized to the
respective input samples and then expressed as fold enrichment
relative to the nonspecific IgG ChIP samples.
Mol Endocrinol, October 2012, 26(10):1716 –1731
expressed throughout pituitary development and plays a
critical role in activation of a number pituitary genes,
including Lhb (45). SF1 is specifically expressed in the
gonadotrope of the anterior pituitary gland, and its importance for reproduction is underscored by a lack of
gonads in SF1-null mice (46). Although highly important
for LH␤ transcription, SF1 and Ptx1 are not regarded as
factors induced or regulated by GnRH (47). On the other
hand, Egr1 is rapidly induced by GnRH and considered a
critical regulator of LH␤ gene expression (47), because
Egr1 null mice lack LH␤ expression in the pituitary gonadotropes (48). Therefore, we focused on the role of
Egr1 in glucocorticoid repression of LH␤ gene expression
and tested the hypothesis that glucocorticoids interfere
with GnRH induction of LH␤ by blunting the GnRHstimulated increase in Egr1 mRNA and protein. To investigate hormone regulation of Egr1 mRNA, L␤T2 cells
were treated with hormone, and after 45 min, RNA was
harvested for Egr1 mRNA analysis by quantitative RTPCR. Egr1 mRNA was low in vehicle-treated control
cells, and expression was unchanged by treatment with
dexamethasone alone (Fig. 5B). As expected, GnRH
caused a dramatic 27.2-fold induction in Egr1 mRNA
(P ⬍ 0.05 vs. Veh). This increase in GnRH-induced Egr1
transcript, however, was not significantly decreased by
dexamethasone (P ⬎ 0.05; GnRH vs. GnRH ⫹ Dex,
27.2⫾ 4.6 vs. 23.1 ⫾ 4.5; Fig. 5B). Because we found that
treatment with glucocorticoids does not suppress levels of
GnRH-induced Egr1 transcript, we next tested the hypothesis that dexamethasone regulates translation or degradation of Egr1 in L␤T2 cells using Western blot analysis. Protein expression of Egr1 was undetectable in
nuclear extracts of L␤T2 cells after treatment with vehicle
or dexamethasone (Fig. 5C). After treatment with GnRH,
Egr1 protein was readily detected, yet protein expression
did not significantly change after cotreatment with GnRH
and dexamethasone, indicating that GR does not interfere
with GnRH induction of Egr1 protein.
We further examined how glucocorticoids might influence GnRH-induced LH␤ expression by performing ChIP
assays on the endogenous mouse LH␤ promoter in L␤T2
cells. Cells were treated with dexamethasone or GnRH
alone or in combination for 60 min before cross-linking.
Sonicated chromatin was immunoprecipitated using either anti-GR or nonspecific IgG. Cross-linking was reversed and precipitates analyzed by quantitative PCR for
220 bp of the LH␤ promoter (⫺180 to ⫹40). Dexamethasone treatment alone did not increase GR binding to the
proximal promoter over vehicle treatment (P ⬍ 0.05; Veh
vs. Dex; Fig. 5D). However, we can only rule out a change
in GR occupancy of the LH␤ promoter at 60 min, and
that later (or earlier) changes in the response to glucocor-
Mol Endocrinol, October 2012, 26(10):1716 –1731
LHβ-Luc/βgal
60
1725
#
50
Empty Vec
Egr1
*
40
30
*
20
10
0
Veh
Dex
25
20
GnRH
GnRH+Dex
15
10
5
*#
*#
* # *#
0
3.0
2.0
**
25
50
75
100
EV
Egr1
GnRH
2.5
1.5
0
Egr1
EV
D
Relief of repression (%)
C
B
LHβ-Luc/βgal
Fold induction
Glucocorticoids interfere with Egr1 actions at the
level of the promoter
Having determined that GR recruitment is dependent
on GnRH (i.e. a responsive factor such as Egr1), yet the
inhibitory effect of glucocorticoids is downstream of
GnRH-induced Egr1 mRNA or protein, we investigated
the role of Egr1 at the level of the LH␤ promoter. We
began by testing the hypothesis that glucocorticoids reduce activity of the LH␤ promoter when induced by Egr1
itself, in the absence of GnRH. Egr1 is a potent activator
of LH␤ transcription, and transfection of an Egr1 expression plasmid in L␤T2 cells treated with vehicle caused a
robust 35-fold increase in LH␤ activity (P ⬍ 0.05; Veh:
Empty Vec vs. Egr1; Fig. 6A). Treatment with dexamethasone significantly blunted the increase in LH␤ activity induced by Egr1 compared with the response in cells treated
with vehicle (P ⬍ 0.05; Egr1 (black bars): Dex vs. Veh; Fig.
6A), indicating that glucocorticoids can interfere with activation by Egr1 at the level of the LH␤ promoter in gonadotrope cells and that GR suppression does not require factors
involved in GnRH signaling upstream of Egr1.
We next attempted to rescue the glucocorticoid repression of GnRH-induced LH␤ activity in L␤T2 cells. We
hypothesized that if Egr1 were the sole factor affected by
glucocorticoids, then titrating increasing amounts of Egr1
into L␤T2 cells would restore full induction and prevent
diminishment by glucocorticoids. LH␤-luc/␤gal values
are represented as fold induction of GnRH or GnRH ⫹
Dex relative to the vehicle control condition containing
the same amount of Egr1. In the absence of exogenous
Egr1, dexamethasone significantly blunted GnRH-induced LH␤ activity (P ⬍ 0.05; empty vector: GnRH vs.
GnRH ⫹ Dex; Fig. 6B). In the presence of increasing
amounts of overexpressed Egr1 (50 –200 ng), the percent
repression significantly decreased (P ⬍ 0.05; empty vector vs. 100 or 200 ng Egr1; Fig. 6C), indicating that repression occurs, in part, via interfering with Egr1 function
on the promoter.
We focused our attention on the action of glucocorticoids at the level of the LH␤ promoter and analyzed the
A
-200 LHβ-Luc/βgal
Fold induction
ticoids are still possible. In contrast, GnRH treatment
increased GR binding 3.8-fold compared with vehicle
(P ⬍ 0.05; Veh vs. GnRH), indicating that recruitment of
GR to the LH␤ promoter is not dependent on dexamethasone binding, but rather is dependent on a GnRH-responsive factor. No difference in GR binding was observed when the cells were concomitantly stimulated with
dexamethasone and GnRH as compared with GnRH
alone, suggesting that a change in GR conformation or
recruitment due to ligand does not underlie the mechanism of glucocorticoid repression of LH␤ transcription.
mend.endojournals.org
*
#
GnRH+Dex
*#
* * # * # n.s.
1.0
0.5
0
WT
5'SF1 5'Egr1 3'SF1 3'Egr1
Mutations
FIG. 6. Glucocorticoids interfere with Egr1 actions at the level of the
proximal promoter. A, To test whether glucocorticoids can interfere
with Egr1-induced LH␤ expression, Egr1 (black bars) or empty vector
(Empty Vec, open bars) was transfected with the ⫺1800-bp LH␤-luc
reporter plasmid into L␤T2 cells and subsequently treated with
dexamethasone (Dex; 100 nM, 24 h) or vehicle (Veh; 0.1% BSA/0.1%
ethanol). Data are shown relative to vehicle-treated in the absence of
Egr1. *, Significant induction by Egr1 vs. vector control; #, significant
repression by glucocorticoid treatment on Egr1-induced LH␤
expression. B, To determine whether titrating in increasing levels of
Egr1 can rescue glucocorticoid repression of ⫺1800-bp LH␤-luc
activity, L␤T2 cells were transfected with empty vector (EV, 200 ng) or
increasing amounts of Egr1 [50 ng (plus 150 ng empty vector), 100 ng
(plus 100 ng empty vector), 200 ng] and treated with GnRH (white
hatched bars, 10 nM, final 6 h) or GnRH ⫹ dexamethasone [gray
hatched bars, Dex (100 nM, entire 24 h); GnRH (10 nM, final 6 h)].
Results are depicted as LH␤-fold induction relative to the vehicletreated condition in the presence of the same amount of Egr1 (dashed
line) as indicated. *, Significant induction by GnRH; #, significant
repression by glucocorticoid treatment on GnRH-induced LH␤
expression. C, The effect of increasing Egr1 on the ratio of LH␤-fold
induction in cells treated with GnRH alone vs. GnRH with
dexamethasone was calculated and expressed as percent repression. *,
Significant difference in the ratio as compared with EV determined by
Student’s t test. D, L␤T2 cells were transfected with either the ⫺200bp LH␤-luc reporter plasmid (WT) or reporter plasmids containing the
⫺200-bp LH␤ region with mutations in SF1 or Egr1 binding elements
as indicated and cultured in the presence of GnRH (white hatched
bars) or GnRH ⫹ Dex (gray hatched bars). Data are shown for each
mutant promoter, relative to its own vehicle treatment. See Fig. 6B
legend for more details. n.s., Not significant P ⫽ 0.08; WT, wild type.
Glucocorticoid Repression of LH␤ Gene Expression
Interaction and involvement of Lhb promoter
proximal binding factors
Egr1 conveys GnRH induction of the LH␤ promoter
by interaction with other regulators of gene expression in
the gonadotrope (47, 49 –51). To assess the complex and
cooperative roles of Egr1, SF1, and Ptx1, we used heterologous CV-1 cells. Unlike gonadotrope cells, CV-1 cells
lack GnRH receptors, an Egr1 response to GnRH, and are
devoid of endogenous SF1, Ptx1, and GR, which allowed
us to reconstitute these factors and determine the necessity and sufficiency of proteins involved in suppression by
dexamethasone. In addition to GR, CV-1 cells were
cotransfected with 1800-bp LH␤-luc reporter plasmid or
pGL3 control plasmid and Egr1, SF1, or Ptx1 alone, or in
combination, and tested for activation of LH␤ transcription in the presence of dexamethasone or vehicle. As
shown in Fig. 7A (white bars), overexpression of Egr1,
SF1, or Ptx1 alone induced small increases in LH␤ transcription, with induction by Egr1 reaching significance.
Dexamethasone significantly diminished induction by
Egr1 (P ⬍ 0.05; Egr1: Dex vs. Veh). Interestingly, dexamethasone also significantly repressed LH␤ induction
when Egr1 was cotransfected with SF1 alone or SF1 plus
Ptx1 (P ⬍ 0.05; Egr1: Dex vs. Veh, Egr1/SF1: Dex vs.
Veh; Fig. 7A).
To address whether the Egr1-binding site is sufficient
for glucocorticoid repression of LH␤ transcription, reporter constructs containing four copies of the Egr1 consensus site ligated into pGL3 upstream of a minimal TK
promoter were created (Egr1 multimer, Fig. 7B). CV-1
A
25 CV-1 cells
20
15
10
5
0
B
Veh
Dex
*#
Egr1
SF1
*
4
#
3
*#
*
1
0
Ptx1
C
5
2
*# **
*
Egr1 Egr1
SF1
Ptx1
Egr1 Egr1
SF1 Ptx1
SF1 multimer
Luc/βgal fold induction
necessity of known GnRH-responsive elements within the
proximal 150 bp of the LH␤ gene for glucocorticoid repression by creating cis mutations in the 5⬘-SF1, 3⬘-SF1,
5⬘-Egr1, and 3⬘-Egr1 binding elements in the context of a
minimal ⫺200 bp LH␤ promoter (WT). We used this
minimal LH␤ promoter because it is sufficient for responsiveness to GnRH and glucocorticoids. cis mutation of
either the 5⬘- or 3⬘-SF1 or the 5⬘-Egr1 site preserved significant GnRH induction and was sufficient for glucocorticoid repression, suggesting that these elements are not
required for either GnRH induction or GR repression
(P ⬍ 0.05; *, significant induction by GnRH; #, significant repression of GnRH induction; Fig. 6D). In contrast,
the 3⬘-Egr1 cis mutation was the only mutation to abrogate glucocorticoid repression (Fig. 6D, 3⬘-Egr1 mutation). Of interest to our study, this Egr1 binding site has
been shown to be critical for GnRH induction as well
(49), and cis mutation of this element eliminated significant induction by GnRH in our hands, implying that
GnRH responsiveness is necessary for repression by
glucocorticoids.
Mol Endocrinol, October 2012, 26(10):1716 –1731
LHβ-Luc/βgal
Fold induction
Breen et al.
Egr1 multimer
Luc/βgal fold induction
1726
5
4
3
2
*#
*
Egr1
SF1
Ptx1
Veh
Dex
*#
1
0
SF1 SF1
Egr1
Ptx1
FIG. 7. Interruption of LH␤ transcriptional complex formation by
glucocorticoids. A, To investigate glucocorticoid-mediated interference
of Egr1, SF1, and/or Ptx1 induction of LH␤ promoter activity, CV-1 cells
were transfected with a GR expression vector and the ⫺1800-bp LH␤luc reporter plasmid, along with Egr1, SF1, or Ptx1 alone or in
combination, as indicated, and treated with dexamethasone (gray bars,
Dex; 100 nM, 24 h) or vehicle (white bars, Veh; 0.1% BSA/0.1%
ethanol). *, Significant induction of LH␤ promoter activity vs. vector
control (dashed line); #, significant repression by glucocorticoid
treatment. B and C, Induction of the Egr1 multimer (B) or SF1 multimer
(C) by the indicated Egr1 or SF1 alone, respectively, or in combination
with Ptx1, was assessed after treatment with dexamethasone (gray
bars, Dex; 100 nM, 24 h) or vehicle (white bars, Veh; 0.1% BSA/0.1%
ethanol) and depicted as fold induction relative to induction of the
control luciferase reporter driven by the TK promoter. *, Significant
induction of multimer activity vs. vector control (dashed line); #,
significant repression by glucocorticoid treatment.
cells were cotransfected with the Egr1 multimer or control TKluc pGL3 plasmid, GR, and either Egr1 alone, or
Egr1 in combination with SF1 and Ptx1, to test for sufficiency to activate transcription in the presence or absence
of dexamethasone. Egr1 was sufficient to induce transcriptional activation of the Egr1 multimer, and this effect
was blunted by glucocorticoids (P ⬍ 0.05; #, significant
repression by Dex; Fig. 7B). Of interest, this site was
sufficient to allow for enhanced activation by the combination of Egr1, SF1, and Ptx1, an affect that was also
diminished by glucocorticoids. In contrast to the sufficiency of the Egr1 site, a consensus SF1-binding site multimer does not convey responsiveness to dexamethasone
when induced by SF1 alone, but activation of the SF1 site
by the combination of Egr1, SF1, and Ptx1 is disrupted by
dexamethasone. Taken together, these findings indicate
that the Egr1 site activated by Egr1 alone is sufficient for
mediating repression. In contrast, the SF1 site activated
Mol Endocrinol, October 2012, 26(10):1716 –1731
mend.endojournals.org
by SF1 alone is not sufficient although a complex with
SF1, Egr1, and Ptx1 on the SF1 site can be repressed.
Role of the GR at the level of the Lhb promoter
Because CV-1 cells lack GR, this cell model allowed us
to determine the necessity of DNA binding or dimerization by GR for repression in gonadotrope cells. In the
absence of transfected GR in CV-1 cells, dexamethasone
does not inhibit LH␤ promoter activity (P ⬎ 0.05; no GR:
Veh vs. Dex; Fig. 8A), confirming that GR is necessary for
repression of LH␤ by glucocorticoids. To determine
whether direct DNA binding by GR plays a critical role in
the repression of LH␤ by glucocorticoids, we transfected
CV-1 cells with two different GR mutants that are incapable of binding DNA. The first mutant, GR Dim mut,
contains four point mutations in the dimerization domain
of GR. These mutations prevent homodimerization of
GR, but do not prevent indirect DNA binding through
A
LHβ-Luc/βgal
Fold induction by
Egr1/SF1/Ptx1
25
CV-1 cells
20
#
15
#
#
10
5
0
No GR
B
10% Input
WT
GR
GST
Dim
mut
DBD
mut
1727
interactions with other transcription factors (52, 53). The
second GR mutant, GR DBD mut, prevents direct DNA
binding by the mutant. Dexamethasone elicited suppression in the presence of either transfected GR mutant, indicating that GR dimerization and DNA binding are not
necessary for repression of GnRH-induced LH␤ promoter activity (P ⬍ 0.05; Dim mut or DBD mut: Veh vs.
Dex; Fig. 8A).
After demonstrating that GR does not require DNA
binding to elicit repression upon the LH␤ promoter, we
investigated the ability of GR to be tethered to DNA by a
GnRH-induced factor, by testing the hypothesis that GR
is capable of binding Egr1. We asked whether in vitrotranscribed and -translated GR could bind bacterially expressed GST-Egr1 in pull-down experiments. Figure 8B
demonstrates precipitation of 35S-labeled GR with GSTEgr1, as compared with GST alone, illustrating a physical
interaction between GR and Egr1. As a positive control,
we show that 35S-labeled SF1 binds GST-Egr1, a strong
interaction that has been previously reported (54). In contrast, the interaction between 35S-labeled GFP with either
GST construct was undetectable. Together with the ChIP
results in Fig. 5D, these findings indicate that GR does not
directly bind the LH␤ promoter; rather it physically interacts with Egr1 and thus is recruited to the LH␤ promoter as a complex with Egr1, identifying Egr1 as a critical factor mediating GnRH induction and GR repression
of LH␤ gene expression.
GST-Egr1
GR
SF1
GFP
FIG. 8. GR does not require dimerization or DNA binding, but is
capable of physically interacting with Egr1. A, The necessity of
dimerization or DNA binding by GR, or by GR itself, was assessed in
CV-1 cells. The ⫺1800-bp LH␤-luc reporter plasmid was maximally
induced with a combination of Egr1, SF1, and Ptx1 in CV-1 cells
cotransfected with empty vector (No GR), wild-type GR (WT GR), GR
dimerization mutant (Dim mut), or GR DBD mutant (DBD mut) and
subsequently treated with dexamethasone (gray bars, Dex; 100 nM,
24 h) or vehicle (open bars, Veh; 0.1% BSA/0.1% ethanol). Data are
presented relative to the ⫺1800-bp LH␤-luc reporter cotransfected
with the empty vectors for Egr1, SF1, and Ptx1 for each GR expression
vector. #, Significant repression by glucocorticoid treatment. B, GST
interaction assays were performed using bacterially expressed Egr1GST fusion protein and 35S-labeled in vitro translated GR, SF1, or GFP
as a control. One tenth of the protein input (10% Input) and the GST
tag-alone (GST, negative control) are shown (note that the SF1
complex with GST-Egr1 has spilled over into the intermediate empty
lane between the GST and the GST-Egr1 lanes).
Discussion
Utilizing a restraint stress paradigm that robustly stimulates corticosterone to investigate stress-induced suppression of gonadotrope function in female mice, we demonstrate that chronic exposure to stress impairs estrous
cyclicity and reduces GnRH-induced LH synthesis and
secretion in diestrus female mice. Whether the increase in
glucocorticoids is induced by stress or exogenously administered, both conditions lead to repression of gonadotrope function, confirming the inhibitory role of glucocorticoids within the reproductive axis and
demonstrating the capacity for suppression during stress.
Although we acknowledge that stress likely increases a
variety of mediators, including other hormones of the
adrenal stress axis that have been shown to alter reproductive function during stress (2, 55, 56), we conclude,
based on our current investigation, that glucocorticoids
are sufficient to disrupt reproductive neuroendocrine
function.
Seminal work by Dr. Neena Schwartz and colleagues
(57) clearly identified inhibitory effects of glucocorticoids
1728
Breen et al.
Glucocorticoid Repression of LH␤ Gene Expression
within the reproductive neuroendocrine axis and postulated that glucocorticoids may alter hypothalamic and
pituitary function. Indeed, in vivo analyses demonstrated
that glucocorticoids could prevent the postcastration rise
in LH in rats but also blunt the response to GnRH in
anterior pituitary fragments in culture (20, 57). Our present investigation expands upon these early studies by
demonstrating that glucocorticoids can act directly upon
the anterior pituitary gonadotrope cell to suppress GnRH
induction of LH␤ gene expression. Intriguingly, we observe a similar suppression in LH␤ gene expression after
stress, but it remains to be determined whether this is a
direct effect of glucocorticoids within the pituitary gland.
With regard to direct actions within the gonadotrope, we
demonstrate a novel mechanism whereby GR is recruited
to the 5⬘-region of the mouse LH␤ gene in live L␤T2 cells
and blunts GnRH induction of LH␤ by interfering
with the genomic effects of Egr1 on the proximal LH␤
promoter.
Glucocorticoid repression via GR maps to a highly
active region of the LH␤ promoter, which contains elements critical for GnRH induction. Promoter analyses
revealed that glucocorticoid-mediated repression is lost
upon 5⬘-truncation of the bipartite SF1 and Egr1 elements
in the LH␤ proximal promoter. Truncation of this region,
however, also eliminated GnRH induction of the promoter (Fig. 4C, ⫺87 LH␤-luc), suggesting that the ability
of glucocorticoids to repress is closely tied to the highly
coordinated and complex mechanism of GnRH action on
Lhb. Not surprisingly, the only cis mutation that relieved
suppression by glucocorticoids (Fig. 6D, the 3⬘-Egr site)
also prevented significant induction of the LH␤ promoter
by GnRH, supporting the conclusion that glucocorticoidinduced repression is dependent on the response to GnRH
and highlights the importance of Egr1 for repression by
glucocorticoids.
Glucocorticoids have the potential to interfere with
GnRH induction of LH␤ via interference of GnRH receptor signaling upstream and downstream of Egr1 induction. For example, studies performed using rat and pig
pituitary cell cultures suggest that chronic exposure to
glucocorticoids can disrupt GnRH receptor-signaling
mechanisms involved in gonadotropin release, including
activation of protein kinase C and cAMP (20, 23). Further, GR has been shown to interact rapidly with c-Jun
N-terminal kinase, a kinase implicated in mediating effects of GnRH in gonadotrope cells (58). In contrast to
these actions of glucocorticoids on GnRH signaling, our
findings support an action of glucocorticoids downstream of Egr1 induction by GnRH via GR acting within
the LH␤ chromatin. Glucocorticoids do not alter GnRHinduced Egr1 mRNA or protein in L␤T2 cells, demon-
Mol Endocrinol, October 2012, 26(10):1716 –1731
strating that the pathway between GnRH binding its receptor and induction of the immediate early gene, Egr1, is
intact. Furthermore, glucocorticoids blunt the ability of
overexpressed Egr1 to induce LH␤ activity, providing evidence that GR inhibition does not require elements upstream of Egr1 and focus our attention on a mechanism
involving Egr1 at the promoter level. We show that titrating in increasing amounts of Egr1 partially overcomes
glucocorticoid repression of Lh␤ in L␤T2 cells, solidifying a mediatory role of Egr1.
With regard to the mechanism of repression, similar to
other steroid hormone receptors, GR mediates transcriptional regulation of target genes via a host of direct and
indirect mechanisms. For example, within the gonadotrope cell, induction of either the murine FSH␤ or human
␣GSU gene occurs via direct GR binding to DNA at conserved glucocorticoid response elements (29, 30). In contrast, using cell models either devoid of or expressing GR
(CV-1 vs. L␤T2 cells, respectively), we find that neither an
intact DBD nor dimerization domain is necessary for GR
repression of the rat LH␤ gene, implicating a genomic
action that occurs via indirect GR binding. In addition,
GR is recruited to the LH␤ promoter chromatin by GnRH
induction in the presence or absence of glucocorticoids.
Coupled with our findings that GR physically interacts
with Egr1 in GST-pulldown experiments, this provides
evidence in support of our hypothesis that GR is tethered
to the LH␤ promoter via a GnRH-induced factor, and in
particular by Egr1.
The requirement for a GnRH-induced factor may also
contribute to the differential effect of glucocorticoids on
transcription within the gonadotrope cell. On the one
hand, genes encoding FSH␤, ␣GSU, and GnRH receptor
are each induced by glucocorticoids alone (30, 33, 59).
On the other hand, our data reveal that GR repression
requires GnRH, likely due to the deficiency of GR recruitment to the proximal promoter in the absence of GnRH.
On the GnRH receptor gene in gonadotrope cells, GnRHinduced GR recruitment has been shown to be involved in
glucocorticoid induction of transcription (59). In that
case, however, the recruitment of GR to the GnRH receptor promoter is dependent on rapid phosphorylation of
GR by GnRH-signaling pathways, which differs from our
finding that GR repression of LH␤ transcription does not
require GnRH signaling upstream of Egr1 activation.
Rather, we conclude that GR is recruited by a factor induced by GnRH and hypothesize that Egr1 itself mediates
the balance between GnRH induction and GR repression
of the LH␤ promoter.
In summary, the present study shows that restraint
stress potently activates the adrenal stress axis in mice and
interferes with gonadotropin synthesis and secretion. We
Mol Endocrinol, October 2012, 26(10):1716 –1731
expand upon this finding by demonstrating that administration of a stress level of glucocorticoids inhibits LH
secretion in female mice and detailing a mechanism
whereby glucocorticoids repress activity of the anterior
pituitary gland via regulation of LH␤ gene expression in
gonadotrope cells. We identify GR in native gonadotrope
cells in the mouse and demonstrate that the recruitment of
GR to the mouse LH␤ promoter via Egr1 interferes with
the tripartite transcriptional complex necessary for mediating responsiveness of this gonadotropin gene to GnRH.
Utilizing dual in vivo and in vitro approaches, collectively, this work reveals new insights regarding the interaction between the adrenal stress and reproductive neuroendocrine axes by identifying a mechanism whereby
LH␤ expression is dampened after a stress elevation in
glucocorticoid.
Acknowledgments
We thank Dr. Alexander (Sasha) Kauffman (University of California, San Diego, La Jolla, CA) and Dr. Catherine Rivier (Salk
Institute, La Jolla, CA) for insightful discussions regarding the in
vivo experiments. We are grateful to Dr. Jacques Drouin (University of Montreal, Quebec, Canada) for generously providing
the mouse Ptx1 and rat Egr1 cDNAs; to Dr. Hermann Pavenstadt (University of Munster, Munster, Germany) for providing
the human Egr1 cDNA; and to Dr. Douglass Forbes (University
of California, San Diego, La Jolla, CA) for providing the GFP
expression plasmid. The mouse SF1 pCMV expression plasmid
was a kind gift of Dr. Bon-chu Chung (Academia Sinica, Taipei,
Taiwan), and the rat LH␤-luciferase plasmid was kindly provided by Dr. Mark Lawson (University of California, San Diego,
La Jolla, CA). We thank Dr. Keith Yamamoto (University of
California, San Francisco, San Francisco, CA) for providing the
rat GR plasmid and Dr. Al Parlow of the National Hormone and
Peptide Program for providing the NIDDK-anti-r␤ LH-IC-2 antibody. We also thank Jason Meadows, Emily Witham, Chuqing (Carol) Yao, Courtney Benson, and Dr. Suzanne Rosenberg
(University of California, San Diego, La Jolla, CA) for technical
assistance and helpful discussions throughout this work.
Address all correspondence and requests for reprints to: Pamela L. Mellon Ph.D., Department of Reproductive Medicine/
Neuroscience, University of California, San Diego, 9500 Gilman Drive, La Jolla, California 92093-0674. E-mail:
pmellon@ucsd.edu.
This work was supported by National Institutes of Health
(NIH) grants R01 HD020377, R01 HD072754, and R01
DK044838 (to P.L.M.) and by the Eunice Kennedy Shriver National Institute of Child Health and Human Development/NIH
through cooperative agreement (U54 HD012303) as part of the
Specialized Cooperative Centers Program in Reproduction and
Infertility Research (to P.L.M.). P.L.M. was also partially supported by P30 CA023100, P30 DK063491, and P42 ES010337.
K.M.B. was partially supported by NIH Grant K99 HD060947.
mend.endojournals.org
1729
V.G.T. was partially supported by NIH Grants K01 DK080467,
and R01 HD067448. D.C. was partially supported by NIH
Grants R01 HD057549, R21 HD058752, and R03 HD054595.
The LH␤ antibody was provided by Dr. A. F. Parlow from the
National Hormone and Pituitary Program, Harbor-UCLA
Medical Center. DNA sequencing was performed by the DNAsequencing shared resource, University of California, San Diego,
Cancer Center, which is funded in part by National Cancer
Institute Cancer Support Grant P30 CA023100. Serum hormone assays were performed by The University of Virginia Ligand Assay Core Laboratory, which is supported through National Institute of Child Health and Human Development Grant
U54 HD028934.
Disclosure Summary: The authors have nothing to disclose.
References
1. Ferin M 1999 Clinical review 105: Stress and the reproductive cycle.
J Clin Endocrinol Metab 84:1768 –1774
2. Rivier C, Rivest S 1991 Effect of stress on the activity of the hypothalamic-pituitary-gonadal axis: peripheral and central mechanisms. Biol Reprod 45:523–532
3. Tilbrook AJ, Turner AI, Clarke IJ 2002 Stress and reproduction:
central mechanisms and sex differences in non-rodent species. Stress
5:83–100
4. Tilbrook AJ, Turner AI, Clarke IJ 2000 Effects of stress on reproduction in non-rodent mammals: the role of glucocorticoids and sex
differences. Rev Reprod 5:105–113
5. Dobson H, Smith RF 2000 What is stress, and how does it affect
reproduction? Anim Reprod Sci 60 – 61:743–752
6. Briski KP, Vogel KL, McIntyre AR 1995 The antiglucocorticoid,
RU486, attenuates stress-induced decreases in plasma-luteinizing
hormone concentrations in male rats. Neuroendocrinology 61:
638 – 645
7. Dong Q, Salva A, Sottas CM, Niu E, Holmes M, Hardy MP 2004
Rapid glucocorticoid mediation of suppressed testosterone biosynthesis in male mice subjected to immobilization stress. J Androl
25:973–981
8. Breen KM, Oakley AE, Pytiak AV, Tilbrook AJ, Wagenmaker ER,
Karsch FJ 2007 Does cortisol acting via the type II glucocorticoid
receptor mediate suppression of pulsatile luteinizing hormone secretion in response to psychosocial stress? Endocrinology 148:
1882–1890
9. Saketos M, Sharma N, Santoro NF 1993 Suppression of the hypothalamic-pituitary-ovarian axis in normal women by glucocorticoids. Biol Reprod 49:1270 –1276
10. Breen KM, Billings HJ, Wagenmaker ER, Wessinger EW, Karsch FJ
2005 Endocrine basis for disruptive effects of cortisol on preovulatory events. Endocrinology 146:2107–2115
11. Li XF, Edward J, Mitchell JC, Shao B, Bowes JE, Coen CW, Lightman SL, O’Byrne KT 2004 Differential effects of repeated restraint
stress on pulsatile lutenizing hormone secretion in female Fischer,
Lewis and Wistar rats. J Neuroendocrinol 16:620 – 627
12. Oakley AE, Breen KM, Clarke IJ, Karsch FJ, Wagenmaker ER,
Tilbrook AJ 2009 Cortisol reduces gonadotropin-releasing hormone pulse frequency in follicular phase ewes: influence of ovarian
steroids. Endocrinology 150:341–349
13. Dufourny L, Skinner DC 2002 Progesterone receptor, estrogen receptor ␣, and the type II glucocorticoid receptor are coexpressed in
the same neurons of the ovine preoptic area and arcuate nucleus: a
triple immunolabeling study. Biol Reprod 67:1605–1612
14. Gore AC, Attardi B, DeFranco DB 2006 Glucocorticoid repression
of the reproductive axis: effects on GnRH and gonadotropin subunit mRNA levels. Mol Cell Endocrinol 256:40 – 48
1730
Breen et al.
Glucocorticoid Repression of LH␤ Gene Expression
15. Attardi B, Tsujii T, Friedman R, Zeng Z, Roberts JL, Dellovade T,
Pfaff DW, Chandran UR, Sullivan MW, DeFranco DB 1997 Glucocorticoid repression of gonadotropin-releasing hormone gene expression and secretion in morphologically distinct subpopulations
of GT1–7 cells. Mol Cell Endocrinol 131:241–255
16. DeFranco DB, Attardi B, Chandran UR 1994 Glucocorticoid receptor-mediated repression of GnRH gene expression in a hypothalamic GnRH-secreting neuronal cell line. Ann NY Acad Sci 746:
473– 475
17. Melis GB, Mais V, Gambacciani M, Paoletti AM, Antinori D,
Fioretti P 1987 Dexamethasone reduces the postcastration gonadotropin rise in women. J Clin Endocrinol Metab 65:237–241
18. Pearce GP, Paterson AM, Hughes PE 1988 Effect of short-term
elevations in plasma cortisol concentration on LH secretion in prepubertal gilts. J Reprod Fertil 83:413– 418
19. Li PS, Wagner WC 1983 In vivo and in vitro studies on the effect of
adrenocorticotropic hormone or cortisol on the pituitary response
to gonadotropin releasing hormone. Biol Reprod 29:25–37
20. Suter DE, Schwartz NB, Ringstrom SJ 1988 Dual role of glucocorticoids in regulation of pituitary content and secretion of gonadotropins. Am J Physiol 254:E595–E600
21. Suter DE, Orosz G 1989 Effect of treatment with cortisol in vivo on
secretion of gonadotropins in vitro. Biol Reprod 41:1091–1096
22. Kononen J, Honkaniemi J, Gustafsson JA, Pelto-Huikko M 1993
Glucocorticoid receptor colocalization with pituitary hormones in
the rat pituitary gland. Mol Cell Endocrinol 93:97–103
23. Li PS 1994 Modulation by cortisol of luteinizing hormone secretion
from cultured porcine anterior pituitary cells: effects on secretion
induced by phospholipase C, phorbol ester and cAMP. Naunyn
Schmiedebergs Arch Pharmacol 349:107–112
24. Maya-Núñez G, Conn PM 2003 Transcriptional regulation of the
GnRH receptor gene by glucocorticoids. Mol Cell Endocrinol 200:
89 –98
25. Adams TE, Sakurai H, Adams BM 1999 Effect of stress-like concentrations of cortisol on estradiol-dependent expression of gonadotropin-releasing hormone receptor in orchidectomized sheep. Biol
Reprod 60:164 –168
26. Pierce JG, Parsons TF 1981 Glycoprotein hormones: structure and
function. Annu Rev Biochem 50:465– 495
27. Kaiser UB, Conn PM, Chin WW 1997 Studies of gonadotropinreleasing hormone (GnRH) action using GnRH receptor-expressing pituitary cell lines. Endocr Rev 18:46 –70
28. Vale W, Rivier C, Brown M 1977 Regulatory peptides of the hypothalamus. Annu Rev Physiol 39:473–527
29. Thackray VG, McGillivray SM, Mellon PL 2006 Androgens, progestins and glucocorticoids induce follicle-stimulating hormone
␤-subunit gene expression at the level of the gonadotrope. Mol
Endocrinol 20:2062–2079
30. Sasson R, Luu SH, Thackray VG, Mellon PL 2008 Glucocorticoids
induce human glycoprotein hormone ␣-subunit gene expression in
the gonadotrope. Endocrinology 149:3643–3655
31. Curtin D, Jenkins S, Farmer N, Anderson AC, Haisenleder DJ,
Rissman E, Wilson EM, Shupnik MA 2001 Androgen suppression
of GnRH-stimulated rat LH␤ gene transcription occurs through
Sp1 sites in the distal GnRH-responsive promoter region. Mol Endocrinol 15:1906 –1917
32. Jorgensen JS, Nilson JH 2001 AR suppresses transcription of the
LH␤ subunit by interacting with steroidogenic factor-1. Mol Endocrinol 15:1505–1516
33. McGillivray SM, Thackray VG, Coss D, Mellon PL 2007 Activin
and glucocorticoids synergistically activate follicle-stimulating hormone ␤-subunit gene expression in the immortalized L␤T2 gonadotrope cell line. Endocrinology 148:762–773
34. Becker KL 1995 Principles and practice of endocrinology and metabolism. 2nd ed. Philadelphia: Lippincott
35. Thackray VG, Hunnicutt JL, Memon AK, Ghochani Y, Mellon PL
2009 Progesterone inhibits basal and GnRH induction of luteiniz-
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
Mol Endocrinol, October 2012, 26(10):1716 –1731
ing hormone ␤-subunit gene expression. Endocrinology 150:2395–
2403
Rosenberg SB, Mellon PL 2002 An Otx-related homeodomain protein binds an LH␤ promoter element important for activation during gonadotrope maturation. Mol Endocrinol 16:1280 –1298
Hollenberg SM, Evans RM 1988 Multiple and cooperative transactivation domains of the human glucocorticoid receptor. Cell 55:
899 –906
Corpuz PS, Lindaman LL, Mellon PL, Coss D 2010 FoxL2 is required for activin induction of the mouse and human follicle-stimulating hormone ␤-subunit genes. Mol Endocrinol 24:1037–1051
Coss D, Hand CM, Yaphockun KK, Ely HA, Mellon PL 2007 p38
mitogen-activated kinase is critical for synergistic induction of the
FSH ␤ gene by gonadotropin-releasing hormone and activin
through augmentation of c-Fos induction and Smad phosphorylation. Mol Endocrinol 21:3071–3086
Larder R, Clark DD, Miller NL, Mellon PL 2011 Hypothalamic
dysregulation and infertility in mice lacking the homeodomain protein Six6. J Neurosci 31:426 – 438
Miyamoto J, Matsumoto T, Shiina H, Inoue K, Takada I, Ito S, Itoh
J, Minematsu T, Sato T, Yanase T, Nawata H, Osamura YR, Kato
S 2007 The pituitary function of androgen receptor constitutes a
glucocorticoid production circuit. Mol Cell Biol 27:4807– 4814
Chappell PE, Schneider JS, Kim P, Xu M, Lydon JP, O’Malley BW,
Levine JE 1999 Absence of gonadotropin surges and gonadotropinreleasing hormone self-priming in ovariectomized (OVX), estrogen
(E2)-treated, progesterone receptor knockout (PRKO) mice. Endocrinology 140:3653–3658
Bronson FH, Vom Saal FS 1979 Control of the preovulatory release
of luteinizing hormone by steroids in the mouse. Endocrinology
104:1247–1255
Coss D, Thackray VG, Deng CX, Mellon PL 2005 Activin regulates
luteinizing hormone ␤-subunit gene expression through Smadbinding and homeobox elements. Mol Endocrinol 19:2610 –2623
Tremblay JJ, Lanctôt C, Drouin J 1998 The pan-pituitary activator
of transcription, Ptx1 (pituitary homeobox 1), acts in synergy with
SF-1 and Pit1 and is an upstream regulator of the Lim-homeodomain gene Lim3/Lhx3. Mol Endocrinol 12:428 – 441
Zhao L, Bakke M, Parker KL 2001 Pituitary-specific knockout of
steroidogenic factor 1. Mol Cell Endocrinol 185:27–32
Tremblay JJ, Drouin J 1999 Egr-1 is a downstream effector of
GnRH and synergizes by direct interaction with Ptx1 and SF-1 to
enhance luteinizing hormone ␤ gene transcription. Mol Cell Biol
19:2567–2576
Topilko P, Schneider-Maunoury S, Levi G, Trembleau A, Gourdji
D, Driancourt MA, Rao CV, Charnay P 1998 Multiple pituitary
and ovarian defects in Krox-24 (NGFI-A, Egr-1)-targeted mice.
Mol Endocrinol 12:107–122
Weck J, Anderson AC, Jenkins S, Fallest PC, Shupnik MA 2000
Divergent and composite gonadotropin-releasing hormone-responsive elements in the rat luteinizing hormone subunit genes. Mol
Endocrinol 14:472– 485
Dorn C, Ou Q, Svaren J, Crawford PA, Sadovsky Y 1999 Activation of luteinizing hormone ␤ gene by gonadotropin-releasing hormone requires the synergy of early growth response-1 and steroidogenic factor-1. J Biol Chem 274:13870 –13876
Kaiser UB, Halvorson LM, Chen MT 2000 Sp1, steroidogenic factor 1 (SF-1), and early growth response protein 1 (egr-1) binding
sites form a tripartite gonadotropin-releasing hormone response
element in the rat luteinizing hormone-␤ gene promoter: an integral
role for SF-1. Mol Endocrinol 14:1235–1245
Rogers SL, Nabozny G, McFarland M, Pantages-Torok L, Archer J,
Kalkbrenner F, Zuvela-Jelaska L, Haynes N, Jiang H, Four mutations in the GR dimerization domain result in perinatal lethal mice.
In: Nuclear Receptors: Steroid Sisters, Keystone, CO 2004, p 194
(Abstract 316)
Mol Endocrinol, October 2012, 26(10):1716 –1731
53. Heck S, Kullmann M, Gast A, Ponta H, Rahmsdorf HJ, Herrlich P,
Cato AC 1994 A distinct modulating domain in glucocorticoid
receptor monomers in the repression of activity of the transcription
factor AP-1. EMBO J 13:4087– 4095
54. Halvorson LM, Ito M, Jameson JL, Chin WW 1998 Steroidogenic
factor-1 and early growth response protein 1 act through two composite DNA binding sites to regulate luteinizing hormone ␤-subunit
gene expression. J Biol Chem 273:14712–14720
55. Cates PS, Li XF, O’Byrne KT 2004 The influence of 17beta-oestradiol on corticotrophin-releasing hormone induced suppression of
luteinising hormone pulses and the role of CRH in hypoglycaemic
stress-induced suppression of pulsatile LH secretion in the female
rat. Stress 7:113–118
56. Rivier C, Rivier J, Vale W 1986 Stress-induced inhibition of repro-
mend.endojournals.org
1731
ductive functions: role of endogenous corticotropin-releasing factor. Science 231:607– 609
57. D’Agostino J, Valadka RJ, Schwartz NB 1990 Differential effects of
in vitro glucocorticoids on luteinizing hormone and follicle-stimulating hormone secretion: dependence on sex of pituitary donor.
Endocrinology 127:891– 899
58. Yokoi T, Ohmichi M, Tasaka K, Kimura A, Kanda Y, Hayakawa J,
Tahara M, Hisamoto K, Kurachi H, Murata Y 2000 Activation of
the luteinizing hormone ␤ promoter by gonadotropin-releasing
hormone requires c-Jun NH2-terminal protein kinase. J Biol Chem
275:21639 –21647
59. Kotitschke A, Sadie-Van Gijsen H, Avenant C, Fernandes S, Hapgood JP 2009 Genomic and nongenomic cross talk between the
gonadotropin-releasing hormone receptor and glucocorticoid receptor signaling pathways. Mol Endocrinol 23:1726 –1745