| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Departments of Physiology (F.H., R.F.) and Obstetrics and Gynecology (R.F.), McGill University, Montréal, Québec, Canada H3A 1A1
Address all correspondence and requests for reprints to: Dr. Riaz Farookhi, F3.44 Women’s Pavilion, Royal Victoria Hospital, 687 Pine Avenue West, Montréal, Québec, Canada H3A 1A1. E-mail: rfarookh{at}rvhob2.lan.mcgill.ca
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
GCs were obtained from the ovaries of diethylstilbestrol-primed immature rats. Cells were prepared as aggregate or dispersed populations by isolating GCs in either the presence or absence of Ca2+. Nonviable cells were removed by a brief exposure to trypsin. We have shown previously that trypsin treatment in the absence of Ca2+ removes a class of cell adhesion molecules, termed cadherins, from the plasma membranes of GCs. Hence, the dispersed GCs are incapable of reaggregating. Dispersed or aggregate GC preparations were incubated with different doses of human FSH (0–1 µg) for 0–60 min in the presence of isobutylmethylxanthine, a phosphodiesterase inhibitor. Incubations were terminated, and the cAMP accumulated was measured using a specific RIA. As desensitization to hormonal stimuli is a characteristic property of many G protein-coupled response systems, cAMP production of cell aggregates and dispersed cells in response to a repeated stimulation with FSH was assessed.
Our results indicate that aggregate GCs have a significantly attenuated cAMP response to all doses of FSH compared with dispersed GC preparations. Changing cell densities did not alter the nature of these responses, indicating that nonspecific cell interactions were not responsible for this difference. The number of FSH receptors and their affinity were unaltered in the two cell preparations. Cholera toxin- and forskolin-stimulated cAMP production were similar in the two preparations, demonstrating that the changes in responsiveness did not arise from alterations in G protein activation or adenylate cyclase activity. Only the aggregate GCs could be desensitized. The dispersed cells displayed undiminished cAMP responsiveness to a second FSH stimulation. Finally, culture of the GC preparations with cholera toxin induced LH receptors in GC aggregates only. LH receptor induction in dispersed cell cultures required the addition of estradiol.
These results indicate that contact-dependent cell interactions can modulate GC cAMP production in response to FSH. cAMP responses, however, were not the sole determinant of cell differentiation, as assessed by LH receptor induction. We speculate that cell-cell interactions within the follicular epithelium are important determinants for cell differentiation leading to follicle selection for ovulation or atresia.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
GCs also interact with one another through gap junctions. Gap junctions are aggregates of transmembrane channels through which small molecules (<1000–2000 daltons) can pass from the cytoplasm of one cell to a neighboring cell (7). The basic structural unit of the gap junction is termed a connexon and consists of an oligomer of gap junction proteins termed connexins. The apposition of two connexons, each contributed by an adjacent cell, form the gap junction channel. The oocyte, cumulus cells, and mural GCs in ovarian follicles are interconnected to one another by an array of gap junctional complexes. The role of these gap junctions in follicular function is not well understood. These junctions are involved, however, in the metabolic cooperation between the oocyte and the surrounding cumulus cells and in the maintenance of meiotic arrest of the oocyte within the follicle (8). Whether gap junctions play a role in GC differentiation is presently not clear. Intercellular coupling has been implicated in growth regulation and differentiation in cell lines and in a number of developmental systems (9–12; reviewed in Ref.13). A role for gap junction-dependent cell communication in tissue patterning has been suggested (14, 15, 16).
FSH exerts its action on GCs by stimulating increased levels of intracellular cAMP (17, 18). The FSH receptor is a member of the G protein-coupled serpentine (seven-transmembrane domain) receptor superfamily (19). GCs in all growing follicles in the rat ovary have been shown to express FSH receptors (20).
FSH stimulates LH receptor expression in rat GCs in vivo (21) and in vitro (22). Sanders and Midgley (23) have shown that cAMP can stimulate LH receptor expression in primary cultures of rat GCs. Our observation that FSH-stimulated LH receptor induction requires direct cell interactions raises the question of whether cAMP production is altered in aggregate vs. dispersed GCs. Hence, we examined whether the aggregation state of freshly isolated rat GCs affects FSH-stimulated cAMP production in vitro. Aggregate and dispersed GCs were prepared from the ovaries of diethylstilbestrol (DES)-primed immature rats and were stimulated with various doses of FSH in vitro. As receptor desensitization is an important aspect of G protein-coupled receptor responses in general (24) and of GC responsiveness to gonadotropins in particular (25), we examined cAMP production by the two types of cell preparations in response to repeated stimuli with FSH. Finally, to determine whether differences in cAMP production are important for GC differentiation, we examined whether FSH-independent elevation of cAMP allows LH receptor induction in these cell preparations.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
FSH receptor level and affinity characteristics
Receptor levels in preparations of dispersed and aggregate
(nondispersed) GCs were determined as described previously (2, 26). FSH
binding affinity was assessed by competition of [125I]FSH
binding to the cells with increasing amounts of unlabeled FSH (27). FSH
was labeled with 125I by the lactoperoxidase method (21).
The highly purified human (h) FSH (AFP-4822B) used in the binding
assays was a gift from the National Hormone and Pituitary Program
(NHPP), NIDDK (Bethesda, MD).
Cell incubation and cAMP stimulation
cAMP stimulation by gonadotropins and other agents (forskolin or
cholera toxin) in the two preparations of GCs was assessed as described
previously (2, 26). Briefly, 0.1 ml cell suspension (3 x
106 cells/ml medium) was aliquoted into 12 x 75-mm
polypropylene tubes (Sarstedt, Montreal, Canada). Then, in order, 0.1
ml 1.5 mM isobutylmethylxanthine (IBMX; Sigma Chemical Co.,
St. Louis, MO) and 0.1 ml of various doses of FSH [0–1000 ng ovine
(o) FSH-17, NHPP, NIDDK], hCG (0–10 IU CR-121; 13,450 IU/mg; NHPP,
NIDDK), cholera toxin (0–1000 ng/ml; Calbiochem, San Diego, CA), or
forskolin (0–100 µM; ICN, Cleveland OH) were added,
giving a final incubation volume of 0.3 ml. All additions were in
Ham’s F-12-DMEM. IBMX was added to inhibit endogenous
phosphodiesterase activity and prevent cAMP metabolism. The oFSH
preparation used has a biological potency of 20 U/mg and negligible
(<0.04 x NIH LH-S1) LH bioactivity. Tubes were incubated for
0–120 min at 37 C in a Dubnoff shaker (10 oscillations/min) in 95%
air-5% CO2 (28). At the end of the incubation periods, 0.7
ml boiling distilled water was added to each tube, and the tubes were
incubated for a further 10 min in a boiling water bath. The tubes were
allowed to cool to room temperature, after which they were centrifuged
(3000 x g for 30 min). The supernatants were removed
and frozen for cAMP analyses. The pellets were analyzed for DNA content
using the fluorescence assay of LaBarca and Paigen (29).
To assess the partitioning of the cAMP produced between the extracellular and intracellular compartments, the incubations were terminated by rapid centrifugation (200 x g for 5 min). The supernatant media were removed, and immediately boiled and stored. The cell pellets were washed with medium and then lysed with boiling distilled water. The lysates were centrifuged, and the supernatants were saved for cAMP determinations. The pellets were used for DNA determination.
Desensitization of GCs
Cells (3 x 106/ml) were aliquoted (0.1
ml/tube) to polypropylene tubes as described above. After the addition
of IBMX, cells were incubated with various doses of FSH as described
previously. After a 60- to 240-min incubation period, the cells were
pelleted by centrifugation (200 x g for 5 min), and
the supernatants were removed. The supernatants were heat inactivated
and stored at -40 C for later assay of cAMP. The pelleted cells were
washed twice with 1 ml Ham’s F-12-DMEM, after which they were
resuspended in 0.3 ml medium containing 0.5 mM IBMX and 500
ng oFSH and incubated for an additional 30 min. At the end of the
incubation period, the tubes were processed as described above.
cAMP assay
cAMP was measured using an RIA as described previously (2, 26).
The cAMP antibody was purchased from Biomedical Technologies
(Stoughton, MA). The tyrosine methyl ester derivative of cAMP (Sigma
Chemical Co.) was iodinated and purified as described by Brooker
et al. (30). The assay sensitivity was 8 fmol cAMP. Inter-
and intraassay coefficients of variation were less than 10%.
LH receptor induction
LH receptor induction in aggregate or dispersed GC cultures was
assessed using 125I-labeled hCG as the ligand. These
procedures were described previously (2).
Statistical analysis
All experiments were repeated at least twice, with at least
three replicates per treatment. Data are presented as the arithmetic
mean with associated SEM and were analyzed by ANOVA.
Multiple comparisons were made using the least significance difference
test. Differences were considered significant for P 
0.05.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
) and their
affinity (Fig. 1B) appeared to be unaffected by the procedures involved
in cell preparation.
|
illustrates the temporal
response of dispersed and aggregate GCs to a single stimulation with
100 ng FSH. Both cell preparations displayed a rapid increase in cAMP
levels by the first time interval examined (15 min), with the dispersed
cell preparations showing a significantly higher cAMP level than the
aggregate cells. The difference in cAMP levels persisted over the
entire time interval examined (120 min). Both preparations appeared to
have attained steady state levels of cAMP by 30–60 min, although the
dispersed cell preparations seemed to maintain a steady increase in
cAMP levels over the 30- to 120-min interval examined. This
observation, although not appreciated at the time, was an indication
that desensitization had not occurred in these cell preparations (see
below). To ensure that the differences in responsiveness were not a
dose-dependent phenomenon, cAMP production in response to different FSH
doses was evaluated at 60 min (Fig. 2B). cAMP levels were significantly
higher (
2-fold) in the dispersed cell preparations compared with
those in the cell aggregates at all of the FSH doses considered.
Maximal levels of cAMP were attained in the dispersed cell preparations
with FSH doses of 30 ng or more. In contrast, maximal levels of cAMP
were observed for FSH doses of 500 ng or more in the aggregate cells.
Nonetheless, even at the higher FSH doses (
500 ng), cAMP accumulation
was significantly higher in the dispersed cells than in the cell
aggregates. No response, in terms of cAMP elevation, was seen when the
cells were stimulated with hCG, indicating that this response was
specific for FSH and resulted from interaction with the FSH receptor
(data not shown). Furthermore, the latter observation indicates that
thecal cell contamination in the GC preparations was minimal. In the
absence of FSH stimulation, a small, but similar, amount of cAMP was
accumulated in both preparations of cells (0.1 pmol/µg DNA) by 60
min.
|
illustrates the linear
regression lines obtained for the cAMP levels attained in the cell
preparations as a function of DNA measured in the cell pellet. Both the
dispersed and aggregate preparations showed a highly linear correlation
(r2 = 0.98 and 0.99, respectively) of cAMP production with
cell number, as reflected by DNA. The slopes of the two regression
lines reflect the increased responsiveness of the dispersed cell
preparations compared with that of the aggregate cells (30
vs. 19 pmol/µg DNA, respectively) and agree with the
levels obtained in the dose-response studies (Fig. 2A
).
|
). Both cholera
toxin (Fig. 4A), which activates Gs by ADP ribosylation
(31), or the diterpene forskolin (Fig. 4B), which activates adenylate
cyclase directly (32), were fully capable of stimulating comparable
levels of cAMP in both cell preparations. As cholera toxin requires a
prolonged incubation period of 120 min to exert its maximal effect on
rat GCs (data not shown), the incubations with both of these agents
were conducted for 120 min. Forskolin produced a more robust response
than cholera toxin. A similar difference in cholera toxin
vs. forskolin responses has been reported for pig GCs
(33).
|
, with the exception of the 30-ng dose,
both cell preparations showed a similar distribution of extra-
vs. intracellular cAMP. Interestingly, the distribution
showed dose dependency. Maximal amounts of intracellular cAMP were seen
for the 10-ng FSH dose, whereas at higher doses, cAMP appeared to be
partitioned equally between the two compartments.
|
illustrates the result of one such
experiment in which GCs were stimulated initially with various doses of
FSH (shown on the abscissa in Fig. 6), followed 60 min later
by stimulation with 500 ng FSH for 30 min. As shown in Fig. 6, an
initial stimulation with 100-1000 ng FSH almost completely abolished
the response to the second stimulation with 500 ng FSH. If the initial
stimulation period were prolonged (to 240 min), lower initial doses
(10–30 ng FSH) also desensitized the cells (data not shown). Prolonged
incubations, however, can also lead to receptor down-regulation (24).
Hence, we examined the responses of the aggregate and dispersed cell
preparations to this desensitization paradigm using 500 ng FSH and a
1-h initial stimulation period followed by restimulation with 500 ng
FSH. As shown in Fig. 7, only the
aggregate GCs display desensitization. The dispersed GCs, in contrast,
maintain a robust response to the second stimulus, with levels of cAMP
equivalent to those seen in the primary response. The desensitization
is exerted at the level of the FSH receptor, as responses to forskolin
were not affected in either the FSH-treated aggregate or dispersed cell
preparations (Fig. 8, A and B; second
stimulation with forskolin, open bars). The effect of
initial stimulation with forskolin followed by acute (30-min)
stimulation with FSH was also assessed (Fig. 8, A and B, second
stimulation with FSH, solid bars). In this case, responses
to FSH were maintained, demonstrating that desensitization did not
result solely from an elevation in cAMP levels, but also required FSH
occupation of its receptor.
|
|
|
shows that although cholera toxin can
induce LH receptors in cultures of GC aggregates, it fails to induce
these receptors in dispersed cell cultures. Including estradiol in
dispersed cell cultures, however, permits cholera toxin induction of LH
receptors.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Previous studies from our laboratory have demonstrated that cell cooperativity is required for GC differentiation. We have shown that FSH-stimulated induction of LH receptors and aromatase activity requires cell adhesion (2, 3). Other investigators have shown that the addition of cAMP analogs or adenylate cyclase activators to GC cultures can mimic the actions of FSH (23). Thus, the hormone-mediated elevation in intracellular cAMP levels in GCs is an important aspect of follicular cell differentiation. It should be noted, however, that the expression of LH receptors and aromatase is not uniform throughout the follicle. Expression of both the receptors and aromatase is most pronounced in mural GCs and appears to display a gradient within the follicle, where the highest expression levels are in the cells closest to the basement membrane and the lowest levels are in the cumulus cells (37). How this patterning is produced is not known. We suspect, however, that the cell adhesion and/or connectivity between GCs may play a role in this process. Cadherins and gap junctions have been shown to be important in tissue morphogenesis and patterning (15, 16, 38).
Christ et al. (39) suggested that gap junctions act to amplify the effects of local receptor activation by spreading second messengers to adjacent cells that are not directly activated by agonist. It is equally possible that inhibitory influences can be transmitted to or activated in adjacent cells. Our finding that desensitization only occurs in cell aggregates would be compatible with this idea.
Our expectation was that disaggregated GCs would show diminished cAMP production. Vander Molen et al. (40) have shown that uncoupled osteoblastic cells have reduced cAMP responses to PTH. These uncoupled cells were developed by stable transfection with complementary DNA antisense to connexin-43, the predominant connexin isoform expressed by these cells. In GCs, however, the disruption of coupling and cell associations resulted in an elevation in cAMP production, suggesting that cellular aggregation had an attenuating, rather than an amplifying, influence. Our experimental design, however, does not allow us to distinguish whether these effects are due to the loss of uncoupling or the loss of cadherin-mediated adhesion. Cadherin expression and gap junction coupling have been shown to be related (41, 42).
A possible explanation for the enhanced cAMP production in disaggregated GCs could be an alteration in FSH receptor number or affinity. Disruption of cell aggregates may disclose cryptic receptors or allow access to previously inaccessible receptors. It is unlikely that new receptor synthesis occurs over the brief time intervals used in our studies. Our findings that both types of cell preparations had similar receptor levels with similar affinity characteristics demonstrate that receptor changes are not responsible for the increased responsiveness. We did note, however, the small, but significant, increases in cAMP response to the higher doses of FSH in cell aggregates. This finding shows that responsiveness of the receptors in cell aggregates, although lower than that in dispersed cells, can be increased if the stimulus is increased. One possible explanation for this, which would also explain the lower responsiveness of aggregates, is if the cAMP generated affects the responsiveness of the FSH receptor. Evidence for phosphorylation-dependent desensitization of peptide receptors, including FSH and LH receptors, has been reported (43). Furthermore, if the cAMP generated in one cell is transmitted rapidly to the adjoining coupled cells via gap junctions, then desensitization of FSH receptors before their activation may occur. This scenario would be consistent with the FSH responses seen in the two cell preparations, as the disaggregated cells would respond as individual units and not as coupled cells. The elevation of cAMP, however, cannot be the only determinant of this receptor nonresponsiveness, because both cell preparations respond similarly if the receptor is bypassed by stimulation with either cholera toxin or forskolin. It has been shown that phosphorylation of peptide receptors requires occupation of the ligand-binding site (19). Hence, a scenario involving transferred cAMP may be operative if the kinetics of phosphorylation are more rapid than the ligand-induced activation of adenylate cyclase. The transmission of cAMP between cells has been demonstrated (7). In fact, Lawrence et al. (44) have shown that GCs can couple to cardiac cells and transmit cAMP, as assessed by changes in the beat rate of the cardiac cells in response to FSH and changes in progesterone secretion in response to isoproterenol in cocultures of these cells.
Cadherin-mediated adhesion may be responsible for the changes in responsiveness. There is increasing evidence that cadherins can act as signaling molecules (45). Cadherins have been shown to associate with the actin cytoskeleton through linker proteins termed catenins (46). Changes in the actin cytoskeleton have been shown to affect GC function (47, 48). Cote et al. (49) have shown that disruption of the cytoskeleton attenuates ACTH-stimulated cAMP production in rat adrenal cells. Modulation of N-cadherin-mediated adhesion in rat cardiomyocytes, using dominant negative mutants of N-cadherin, has been shown to cause gap junction retraction (50). Similar observations were reported for endothelial cells when VE-cadherin was altered (51).
The similarity in responses of the two cell preparations to cholera toxin and forskolin indicate that neither G protein coupling nor adenylate cyclase activity are affected by the cell preparation procedures. Furthermore, as cell density had no effect on the nature of these responses, nonspecific cell contacts do not provide an explanation. We also confirmed that these differences were not the result of changes in the distribution of cAMP between the intracellular and extracellular compartments. With respect to cAMP distribution, it is noteworthy that at low FSH doses, the major proportion of cAMP generated is retained within the cell. At higher doses and correspondingly higher levels of cAMP generated, approximately half of the cAMP is found in the incubation medium. This suggests that mechanisms for cAMP removal by these cells are sensitive to the intracellular level of cAMP. It should be noted, however, that phosphodiesterase activity was inhibited in these cells, and this may be responsible for the profiles of cAMP secretion observed. In a limited series of studies conducted in the absence of IBMX, however, we noted that the differences in FSH-stimulated cAMP production were maintained (data not shown). If phosphodiesterases play a role in regulating intracellular cAMP levels in GCs, this is not apparent over the short time periods of our studies.
It is possible that the cell preparation procedures lead to the selection of different populations of GCs. We have shown that GCs from DES-primed rats are heterogeneous with respect to size and lectin-binding characteristics (36). If these populations also have differences in their FSH responsiveness, but not in their FSH-binding capabilities, the observed differences in cAMP production could occur. Truncated receptors, lacking the ability to stimulate cAMP but retaining binding capability, have been reported (20). Although we cannot discount the possibility that we may be dealing with different cell populations, the fact that dispersed cells are capable of expressing LH receptors in response to cholera toxin, albeit in the presence of estradiol, makes population differences unlikely. This possibility, however, needs to be investigated before it can be ruled out conclusively.
The physiological relevance of the different behaviors of cell aggregates and dispersed cells is not apparent. Aharoni et al. (52) have shown that the maintenance of elevated levels of cAMP in GCs from preovulatory follicles is detrimental and directs these cells toward apoptosis. These authors have suggested that the desensitization mechanism in these cells may act as a protective preservation response. Peluso et al. (6) have shown that single GCs show an increased probability for apoptosis compared with aggregate cells. Studies in our laboratory agree with these findings. We have shown that dispersed GCs are more susceptible to a serum-mediated cell deletion effect than are estrogen-treated cells, where aggregation is promoted (53). Wiesen and Midgley (54) have reported the loss of connexin-43 in follicles undergoing atresia. The loss of cell-cell and cell-matrix interactions has been implicated in tissue apoptosis. As FSH secretion in mammals is pulsatile, and a large discharge of FSH is seen coincident with the ovulatory surge of LH, changes in cell-cell connectivity may be an important determinant of whether particular follicles continue to develop or undergo atresia.
In summary, we have shown that cell contact is an important determinant of hormone responsiveness in GCs. Our experimental design does not allow us to identify whether cell adhesion, gap junction connectivity, or both are the reasons for the altered responses to FSH. Studies using blocking antibodies and peptides that would allow us to make these distinctions are in progress.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received August 11, 1997.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-1 adrenergic agonist efficacy in isolated rat aorta. J
Pharmcol Exp Ther 266:1054–1065-catenin with the cadherin-catenin cell-cell adhesion
complex via -catenin. J Cell Biol 130:67–77s-subunit of the Gs protein with
microfilaments and microtubules: implication during ACTH stimulation in
rat adrenal glomerulosa cells. Endocrinology 138:69–78This article has been cited by other articles:
 |
|
 |
|
  Y. L. Pon and A. S. T. Wong Gonadotropin-Induced Apoptosis in Human Ovarian Surface Epithelial Cells Is Associated with Cyclooxygenase-2 Up-Regulation via the {beta}-Catenin/T-Cell Factor Signaling Pathway Mol. Endocrinol., December 1, 2006; 20(12): 3336 - 3350. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Zygmunt, F. Herr, S. Keller-Schoenwetter, K. Kunzi-Rapp, K. Munstedt, C. V. Rao, U. Lang, and K. T. Preissner Characterization of Human Chorionic Gonadotropin as a Novel Angiogenic Factor J. Clin. Endocrinol. Metab., November 1, 2002; 87(11): 5290 - 5296. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Ricken*, P. Lochhead, M. Kontogiannea, and R. Farookhi Wnt Signaling in the Ovary: Identification and Compartmentalized Expression of wnt-2, wnt-2b, and Frizzled-4 mRNAs Endocrinology, July 1, 2002; 143(7): 2741 - 2749. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.J. Peluso, A. Pappalardo, and G. Fernandez E-Cadherin-Mediated Cell Contact Prevents Apoptosis of Spontaneously Immortalized Granulosa Cells by Regulating Akt Kinase Activity Biol Reprod, April 1, 2001; 64(4): 1183 - 1190. [Abstract] [Full Text]
|
|
|
|
|
|
N. A. Grieshaber, S. Boitano, I. Ji, J. P. Mather, and T. H. Ji Differentiation of Granulosa Cell Line: Follicle-Stimulating Hormone Induces Formation of Lamellipodia and Filopodia via the Adenylyl Cyclase/Cyclic Adenosine Monophosphate Signal Endocrinology, September 1, 2000; 141(9): 3461 - 3470. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. H. Machell, O. W. Blaschuk, and R. Farookhi Developmental Expression and Distribution of N- and E-Cadherin in the Rat Ovary Biol Reprod, September 1, 2000; 63(3): 797 - 804. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
