| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
-Subunit of Human Chorionic Gonadotropin by the Embryonic Transcription Factor Oct-3/4
Departments of Biological Sciences (L.L.) and Animal Sciences and Biochemistry (D.L., M.V., R.M.R.), University of Missouri, Columbia, Missouri 65211
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
-subunit (hCG)
expression in JAr cells. Oct-3/4 reduced human GH reporter expression
from the -170 hCG promoter in either the presence or absence of
cAMP by about 70% in transient cotransfection assays, but had no
effect on expression from either the -148 hCG or the -99 hCG
promoter. Unexpectedly, no Oct-3/4-binding site was identified within
the -170 to -148 region of the hCG promoter, although one was
found around position -115 by both methylation interference
footprinting and electrophoretic mobility shift assays. Site-directed
mutagenesis of this binding site destroyed the affinity of the promoter
for Oct-3/4, but did not affect repression of the promoter. Therefore,
inhibition of hCG gene transcription by Oct-3/4 appears not to
involve direct binding of this factor to the site responsible for
silencing. When stably transfected into JAr cells, Oct-3/4 reduced the
amounts of both endogenous hCG mRNA and protein by 70–80%. Oct-3/4
is therefore capable of silencing both hCG and hCGß gene
expression. We suggest that as the trophoblast begins to form,
reduction of Oct-3/4 expression permits the coordinated onset of
transcription from the hCG and hCGß genes. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
-subunit (hCG), common to all the glycoprotein hormones,
and a distinct ß-subunit (hCGß) responsible for biological
specificity. To produce biologically active hCG, the two subunit
molecules must be produced in a coordinated manner. Not unexpectedly,
therefore, synthesis of hCG- and ß-subunits is reasonably well
balanced in the first trimester of human pregnancy (3, 4), although
little is understood about how this coordinated synthesis is
achieved.
Transcriptional regulation appears to be paramount in the control of
hCG production (1, 2). The hCG subunit is encoded by only a single
gene (5, 6), whereas there is a cluster of six hCGß subunit genes (1, 7). Multiple regulatory elements in the upstream promoter region of the
hCG gene have been found to be important for transcriptional
activation of the gene in choriocarcinoma cells (8, 9, 10, 11, 12, 13, 14, 15). They include
two tandem repeats of a cAMP response element (CRE), a complex upstream
regulatory element (URE), the -activator (-ACT) element, the
junctional regulatory element (JRE), and the CCAAT region. The
transacting factors that bind to these elements, with the exception of
the CRE-binding protein (CREB), remain to be identified and cloned.
Although much less studied, the hCGß gene also contains multiple
regulatory regions that contribute toward expression in choriocarcinoma
cells (16, 17, 18, 19). Curiously, the promoter regions of hCG- and
ß-subunit genes share little similarity, and no transcription factor
important for regulation of both genes in developing trophectoderm has
been identified.
Oct-3/4, characterized by its conserved POU DNA-binding domain, is expressed in totipotent/pluripotent embryonic cells and is a strong candidate for a regulator of early embryogenesis (20, 21, 22, 23, 24, 25). In mouse, Oct-3/4 mRNA and the protein itself have been detected in early stage trophectoderm but not in trophoblast cells after the blastocyst has hatched from the zona pellucida (20, 23, 24). By analogy with the mouse, down-regulation of Oct-3/4 in the human embryo probably coincides with the first expression of hCG in trophectoderm.
Recently, we found that hCGß expression in JAr human choriocarcinoma
cells was almost completely silenced by Oct-3/4 (25). Reporter
expression from an hCGß gene promoter was reduced about 90% by
Oct-3/4 in transient transfection assays. Oct-3/4 specifically bound to
a sequence (CAATAATCA; -276/-268) in the hCGß gene promoter as
measured in electrophoretic mobility shift assays (EMSAs) and
methylation interference footprinting analyses. Mutation of this
binding site abolished Oct-3/4 repression (25). Here we report that
Oct-3/4 is also a potent inhibitor of hCG expression in JAr cells
and suggest that a loss of Oct-3/4 expression in developing
trophectoderm may permit the coordinated production of both CG subunits
in the early human embryo.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
-GH Expression by Oct-3/4 promoter is sufficient to provide full expression
and cAMP responsiveness to a reporter gene in human choriocarcinoma
cells (26, 27, 28). Initial experiments with the -170 hCG-CAT construct
showed that it was inhibited in JAr cells in a dose-dependent manner by
pcDNA3-Oct-3/4 cotransfection and that 8-bromo-cAMP (1 mM),
despite enhancing promoter activity approximately 10-fold in controls,
did not reverse the Oct-3/4 inhibition (data not shown).
To define where Oct-3/4 effects in the promoter were manifested,
various truncated promoter constructs were tested in the cotransfection
assay (Fig. 1
). Whereas human GH (hGH)
expression from -170 hCG-GH was inhibited approximately 70% by
pcDNA3-Oct-3/4 transfection, there was no inhibition of the -99
construct. Similarly, Oct-3/4 failed to inhibit expression from the
-148 hCG promoter, which lacks the URE (11, 26) but contains intact
CRE elements (8, 9, 10). Therefore, a site within the -170 to -149 URE
region is likely responsible for the Oct-3/4 effects. When -170
hCG-GH was cotransfected with an Oct-1 expression plasmid, hGH
expression was not affected (112% ± 17 of the control value).
|
Enhancer Activity by Oct-3/4 promoter transcriptionally
activated a variety of heterologous promoters in human trophoblast cell
lines (8, 10, 29). Whether Oct-3/4 could inhibit these enhancer effects
was studied by cotransfection of pcDNA3-Oct-3/4 with a series of
hCG-TKGH constructs in which hCG fragments (-170/-53,
-172/-100, and -153/-129) had been fused to a TKGH hybrid gene
(Fig. 2). Human GH reporter expression
from the thymidine kinase (TK) promoter was increased 9- and 6-fold by
the -170/-53 and -170/-100 hCG fragments, respectively, in the
absence of Oct-3/4 (Fig. 2). When pcDNA3-Oct-3/4 was cotransfected, hGH
expression from -170/-53-TKGH and -170/-100-TKGH was in each
case reduced to about 30% of control values. Expression from the
-153/-129 construct, which lacked the URE but contained an intact
CRE, was not affected.
|
might be through the TK promoter, which has a canonical ATTTGCAT
octamer motif (30). This sequence, therefore, was removed. Expression
from the modified construct (-170/-53-2.3TKGH) was inhibited by
Oct-3/4 about as effectively as the original -170/-53-TKGH
construct (Fig. 2).
Binding of Oct-3/4 to the hCG Promoter in Vitro
Oct-3/4 synthesized by coupled in vitro transcription
and translation formed a complex with a 32P-labeled
-170/-53 hCG promoter fragment (Fig. 3A, lane 2). Such a labeled complex was
absent when the reticulocyte lysate alone was used (lane 1). It was not
detected when the OCT consensus oligonucleotide was added as competitor
(lane 3).
|
). The -117 to -110
sequence within the hCG gene promoter is identical at six positions
to the consensus octamer motif (Fig. 3E). Methylation at other
positions in the sense strand of -170/-53 hCG (Fig. 3B) or at
positions in the antisense strand of -170/-100 hCG (data not
shown) did not appear to interfere with Oct-3/4 binding.
Effect of Mutating the 0ct-3/4 Binding Site in the hCG
Promoter
The -170/-100 hCG fragment formed a single complex with
Oct-3/4 in the EMSA (Fig. 3C), but after the nucleotides A (-111) and
T (-110) within the putative Oct-3/4 binding region (Fig. 3E) were
changed to C and G, respectively, the resulting mutant µ-170/-100
hCG fragment was unable to bind to Oct-3/4 (Fig. 3D). Whether this
Oct-3/4 site defined by mobility shift assays and by methylation
interference was necessary for Oct-3/4 inhibition of the hCG
promoter was studied by fusing the µ-170/-100 hCG fragment to
the -99 hCG promoter to form the µ-170 hCG-GH and
-chloramphenicol acetyl transferase (CAT) constructs. As expected (13),
reporter expression from the mutant promoter was only about 30%
(±10%) of that from the wild type -170 hCG promoter in the
absence of Oct-3/4 cotransfection. However, both hGH and CAT expression
from the mutant promoter were repressed as strongly as from the wild
type promoter by Oct-3/4 (Fig. 1; data not shown for CAT
constructs).
It would appear that even though there is a well defined
Oct-3/4-binding site on the hCG promoter, it is not required for
Oct-3/4 inhibitory effects.
Inhibition of Endogenous hCG-Subunit Production in JAr Cells by
0ct-3/4 Stable Transfection
To study the effect of Oct-3/4 on endogenous hCG gene
expression, JAr cells were stably transfected with pcDNA3-Oct-3/4 (25).
Total RNA was isolated from both stable Oct-3/4 clones and stable
control clones and subjected to a ribonuclease protection assay in the
presence of an antisense hCG RNA probe expected to hybridize to a
440-bp fragment of the hCG transcript. As shown in Fig. 4A, the quantity of hCG mRNA in both
of the stable Oct-3/4 clones tested (S1 and S4) was clearly much lower
than in the two stable control clones (C1 and C2) and in the normal JAr
cells (J). In contrast, the content of ß-actin mRNA was comparable
among all clones, whether they expressed Oct-3/4 or not. When
quantitated by densitometry and normalized to ß-actin mRNA, the
hCG mRNA content of the clones expressing Oct-3/4 was about 23% of
that in the controls (Fig. 4B).
|
subunit measured by a RIA was significantly
reduced (P < 0.001) in the clones expressing Oct-3/4
(Fig. 5). For clones S1 and S4, the
amount of hCG was 33% and 19%, respectively, of the average
production of two control lines C1 and C2.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
gene expression in JAr choriocarcinoma
cells. Stable expression of Oct-3/4 reduced the amounts of both
endogenous hCG mRNA and hCG protein by 70 to 80% in these
cells. Oct-3/4 also strongly inhibited reporter expression from the
hCG gene promoter in transient transfection assays. Nevertheless,
the mechanism whereby hCG transcription is repressed by Oct-3/4
appears distinct from the manner in which hCGß transcription is
silenced. Whereas the Oct-3/4-binding site in the hCGß promoter is
critical for its repression, the binding site in the hCG promoter
appears dispensable.
Transient transfection experiments with the hCG promoter-reporter
gene constructs revealed selective inhibition of the promoter by
Oct-3/4 relative to controls (Figs. 1 and 2). The responsive region was
located between -170 and -149 (Fig. 1), but no direct binding of
Oct-3/4 to this region could be demonstrated (Fig. 3, B and D),
although the existence of a cryptic site available in vivo
cannot be ruled out. Expression from constructs lacking the -170/-149
region were not affected by Oct-3/4 expression. The inhibitory effects,
therefore, were not due to sequence-independent squelching of some
general transcription factor as has been observed for the
transcriptional repressor Dr1/Dc2 (31, 32). Preliminary experiments,
less complete than those shown in Figs. 1 and 2 and performed with a
CAT rather than a GH reporter gene, have shown that a deletion of the
-160/-155 region on the promoter (the µVIII hCG-CAT construct of
26 abolished the responsiveness of the promoter to Oct-3/4.
Curiously, however, as assessed by EMSA and by methylation interference
assays (Fig. 3, A–C), Oct-3/4 was found to associate with the promoter
not within the URE, but at another site (-117/-110). However,
mutation of this sequence had no influence on the ability of Oct-3/4 to
repress the hCG promoter (Fig. 1). The significance, if any, of
Oct-3/4 binding remains unclear.
Oct-3/4 probably exerts its effects within the URE region (-177/-141)
of the hCG promoter (Fig. 1). The narrower -170/-149 region
responsive to Oct-3/4 is believed to bind at least three different
nuclear proteins from choriocarcinoma cells in vitro (11, 15, 18, 26). The URE2/TSE binding factor that requires the sequence
from -177 to -156 (12, 15) is probably unable to bind the shorter
-170 hCG promoter used primarily in the present study and is
unlikely, therefore, to be the target of Oct-3/4. The UREl-binding
protein (12, 26) and the ACT-binding protein (15), on the other
hand, are known to bind the -170 promoter directly. A possible
mechanism of Oct-3/4 inhibition is that it interacts directly with one
or more of these URE-associated transcriptional activators and
influences their abilities to associate with either the promoter or
CREB, the CRE-binding transcription activator (33), in the downstream
CRE elements upon which the URE depends exclusively for its enhancing
activity (10, 11, 12). An alternative hypothesis is that Oct-3/4 sequesters
(squelches), or represses by some other means, transcriptional
coactivators necessary for proper functioning of the URE-associated
factors. Some POU-containing proteins, including Oct-1, regulate
transcription through their cooperative interactions with other
transcription cofactors (34, 35, 36, 37).
In summary, the gene for hCG-subunit, like those for the
ß-subunit, has been shown to be a target of the embryonic
transcription factor Oct-3/4. Both subunit genes are coordinately
repressed in the presence of Oct-3/4, a phenomenon that may explain why
these genes are silent in the inner cell mass of the human embryo but
expressed in trophectoderm. Surprisingly, the mechanisms by which
silencing of the two genes is mediated appear to differ.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
(SanI-BamHI),
-99/+44 hCG (RsaI-BamHI) and -148/+29 hCG
(isolated by SacI and HindIII digestion of
pµIXhCG-CAT; 26 promoter fragments (8) were fused to the hGH
gene of the p0GH plasmid (Nichols Institute Diagnostics, San Juan
Capistrano, CA) to form p-170 hCG-GH, p-99 hCG-GH, and p-148
hCG-GH, respectively. The mutant µ-170/-100 hCG fragment was
produced by PCR with oligonucleotide primer µOctR (see Table 1) and was cloned upstream of the -99
hCG-GH fusion gene to form pµ-170 hCG-GH, in which the
µ-170/-100 and -99/+44 hCG fragments were in the same
orientation (confirmed by DNA sequencing). The hCG-CAT plasmids,
p-170 hCG-CAT, p-99 hCG-CAT, and pµ-170 hCG-CAT, were
constructed in a way similar to that described for hCG-GH
construction.
|
fragments were
cloned upstream of the TKGH fusion gene through the HindIII
site in the pTKGH plasmid (Nichols Institute Diagnostics) to form the
plasmid p-170/-53-TKGH and p-170/-100-TKGH, respectively. The
p-170/-53-2.3TKGH plasmid was constructed by fusing the -170/-53
hCG fragment to a 2.3-kb EcoRI fragment of the TKGH
fusion gene, which was devoid of the -198/-79 TK promoter fragment
that contains a consensus octamer motif (ATGCAAAT at -131/-138; 30 . Oligonucleotide CREf and CREr (Table 1) were annealed to
each other to form the double-stranded CRE fragment. Two tandem
repeats of the CRE fragment were then cloned into the
HindIII site of pTKGH to form p-153/-129-TKGH. The
hCG-GH, hCG-CAT, and hCG-TKGH constructs were confirmed by DNA
sequencing.
A 440-bp PstI hCG cDNA fragment that covers exon 1, 2,
and part of exon 3 of the hCG gene (5) was isolated and cloned into
the PstI site of the pGEM-4Z vector (Promega, Madison, WI).
The orientation of the hCG fragment relative to the T7
promoter in the pGEM-hCG plasmid was determined by DNA
sequencing.
Construction of pcDNA3-Oct4 from pcMV-Oct4 (22) was described previously (25). Expression plasmid pCGOct-1 was provided by Dr. W. Herr (38). Plasmids p0GH, pTKGH, and pXGH5 were purchased from Nichols Institute Diagnostics.
Transient Transfection, hGH RIA, and CAT Assay
The phCG-GH plasmid (2 µg) was cotransfected with either
pcDNA3-Oct-3/4 (2 µg) or pcDNA3 (2 µg) into JAr cells in six-well
tissue culture plates (25). The amount of hGH secreted was measured by
a specific RIA (25). Expression from the phCG-TKGH construct was
analyzed in a similar manner. Either pcDNA3-Oct-3/4 (2 µg) or pcDNA3
(2 µg) was cotransfected with phCG- CAT (2 µg) and a pTKGH
internal control plasmid (0.2 µg) into JAr cells. CAT activity was
then normalized in relation to hGH production (25). ANOVA and a
standard Student’s t test were used for statistical
analyses (39).
EMSA and Methylation Interference Analysis
Oct-3/4 protein, produced by coupled in vitro
transcription and translation in a reticulocyte lysate (25), was used
with the 32P-labeled -170/-53 hCG, -170/-100 hCG,
and µ-170/-100 hCG fragments in electrophoretic mobility shift
assays (25). The double-stranded OCT oligonucleotide, which had been
produced by annealing oligonucleotide OCTf and OCTr (Table 1), was
included as a competitor in some reactions. Single-ended labeled
-170/-53 hCG fragments were prepared and subjected to methylation
interference analyses (25).
Stable Transfection of Oct-3/4, Ribonuclease Protection Assay,
and RIA of hCG
JAr cells were stably transfected with pcDNA3-Oct-3/4 or pcDNA3
(25). Ribonuclease protection assays of the total RNA (10 µg)
isolated from several stable clones (-2 x 108 cells)
were performed by standard procedures (25). Antisense hCG probe was
synthesized from pGEM-hCG by using T7 bacteriophage RNA
polymerase in the presence of -[32P]CTP (800 Ci/mmol,
DuPont NEN, Weymouth, MA). Antisense human ß-actin RNA probe was
synthesized to an 80-fold lower specific activity (by diluting the
original 800 Ci/mmol -[32P]CTP with a concentrated
solution of unlabeled CTP) from the pTRI-ß-actin-125-human antisense
control template (Ambion Inc, Austin, TX). The human ß-actin probe
(10,000 cpm) was used together with the hCG probe (100,000 cpm) for
each protection assay. The relative amounts of radiolabel in the
protected ß-actin and hCG fragments were measured by densitometry.
The radiolabel in hCG mRNA was then normalized relative to the label
in ß-actin mRNA.
The amount of hCG (either as the free subunit or as the subunit of
intact hCG heterodimer) secreted by cultured JAr cells (
2 x
106 cells per culture) over a 24-h period was measured by a
double-antibody RIA (40). The procedure employed rabbit anti-hCG
antiserum (AFP-310784) and hCG subunit standard (CR-119). Both
AFP-310784 and CR-119 were supplied by the National Hormone and
Pituitary Program (NIDDK, Rockville, MD). Mean intra- and interassay
coefficients of variation were 3.5% and 6.4%, respectively.
|
ACKNOWLEDGMENTS |
|---|
cDNA clone, Dr. J. H. Nilson
for pµIXhCG-CAT, Dr. D. H. Keisler for reagents and help in
the RIA of hCG, and the National Hormone and Pituitary Program, NIH,
for anti-hCG antiserum and hCG subunit standard. We thank Ellen
Swanson for help in preparing the manuscript. |
FOOTNOTES |
|---|
This work was supported by NIH Grants HD-21896 and HD-29843. This paper is a contribution from the Missouri Agricultural Experiment Station, Journal Series Number 12,488.
1 Present address: Duke University Medical Center, Department of
Medicine, Durham, NC 27710. 
2 Present address: Department of Molecular Biology, Research Institute,
Cleveland Clinic Foundation, Cleveland, Ohio 44195-5210.
3 Present address: Servicio de Investigaci 92 n Agraria Diputaci
92 n General de Aragon, Apartado 727, 50.080-Zaragoza, Spain.
Received for publication February 21, 1997. Revision received July 3, 1997. Accepted for publication July 17, 1997.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
hCG
and ßhCG as measured by specific monoclonal immunoradiometric assays.
Endocrinology 120:549–558-subunit gene in normal tissue and
tumor-derived cell lines. J Biol Chem 256:5121–5127
subunit of the four human glycoprotein hormones. J Mol Appl Genet 1:3–18[Medline]
gene transcription is directed
by a repeated 18-base pair enhancer. J Biol Chem 262:12169–12174-subunit gene: dependence
on cyclic AMP-inducible elements. Mol Cell Biol 7:3994–4002-gene. Mol Endocrinol 3:763–772 subunit
glycoprotein hormone gene utilizes a unique CCAAT binding factor.
J Biol Chem 265:6279–6285-subunit genes. Mol Endocrinol 5:243–255- and ß-subunit genes of human chorionic
gonadotropin by trophoblast-specific element-binding protein. Mol
Endocrinol 7:1579–1588 and ß genes through distinct types of CR Es. J
Biol Chem 269:31090–31096
subunit gene. J Biol Chem 269:19360–19368-gene contains multiple protein binding domains
that interact to modulate basal and cAMP-responsive transcription.
J Biol Chem 263:9879–9886This article has been cited by other articles:
 |
|
 |
|
  T. Ezashi, P. Das, R. Gupta, A. Walker, and R. M. Roberts The Role of Homeobox Protein Distal-Less 3 and Its Interaction with ETS2 in Regulating Bovine Interferon-Tau Gene Expression-Synergistic Transcriptional Activation with ETS2 Biol Reprod, July 1, 2008; 79(1): 115 - 124. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Lee, H. K. Kim, J.-Y. Rho, Y.-M. Han, and J. Kim The Human OCT-4 Isoforms Differ in Their Ability to Confer Self-renewal J. Biol. Chem., November 3, 2006; 281(44): 33554 - 33565. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Ghosh, S. Sachdev, M. Hannink, and R. M. Roberts Coordinate Regulation of Basal and Cyclic 5'-Adenosine Monophosphate (cAMP)-Activated Expression of Human Chorionic Gonadotropin-{alpha} by Ets-2 and cAMP-Responsive Element Binding Protein Mol. Endocrinol., April 1, 2005; 19(4): 1049 - 1066. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. J. Soares and M. W. Wolfe Human Embryonic Stem Cells Assemble and Fulfill Their Developmental Destiny Endocrinology, April 1, 2004; 145(4): 1514 - 1516. [Full Text] [PDF]
|
|
|
|
|
|
D. Ghosh, T. Ezashi, M. C. Ostrowski, and R. M. Roberts A Central Role for Ets-2 in the Transcriptional Regulation and Cyclic Adenosine 5'-Monophosphate Responsiveness of the Human Chorionic Gonadotropin-{beta} Subunit Gene Mol. Endocrinol., January 1, 2003; 17(1): 11 - 26. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Ezashi, D. Ghosh, and R. M. Roberts Repression of Ets-2-Induced Transactivation of the Tau Interferon Promoter by Oct-4 Mol. Cell. Biol., December 1, 2001; 21(23): 7883 - 7891. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. L. Strauss and I. Boime Cellular Localization of the Human Chorionic Gonadotropin {beta}-Subunit in Transgenic Mouse Placenta Endocrinology, January 1, 2000; 141(1): 430 - 437. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Knofler What factors regulate HCG production in Down's syndrome pregnancies?: Regulation of HCG during normal gestation and in pregnancies affected by Down's syndrome Mol. Hum. Reprod., October 1, 1999; 5(10): 895 - 897. [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 |
