| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Other Original Articles |
Prince Henry’s Institute of Medical Research and Monash University, Departments of Obstetrics and Gynecology and Medicine, Monash Medical Center, Clayton, Victoria 3168, Australia
Address all correspondence and requests for reprints to: Dr. Peter J. Fuller, Prince Henry’s Institute of Medical Research, P.O. Box 5152, Clayton, Victoria 3168, Australia. E-mail: . peter.fuller{at}med.monash.edu.au
Abstract
Granulosa cell tumors of the ovary (GCT) express the inhibin subunit genes and secrete dimeric inhibin. Transgenic mice null for the 
-inhibin gene develop GCT. It has been suggested that this apparent contradiction may be reconciled if the human GCT are resistant to the tumor-suppressive effects of inhibin. Inhibin receptors have recently been characterized as consisting of either betaglycan or p120 in association with the type II or type I activin receptor subunits (ActR), respectively. To test the hypothesis that GCT may exhibit loss of inhibin receptor expression we have examined the expression of the receptor subunits in a cohort of GCT and in mucinous and serous cystadenocarcinomas and normal ovary. Expression was determined by RT-PCR using gene-specific primers and probes combined with Southern blot analysis of the PCR products. The ActRI subunits and ActRIIA exhibited widespread albeit variable expression across tissues, with the highest levels in the serous tumors. ActRIIB expression was relatively low in the mucinous tumors and high in the GCT. Betaglycan expression was abundant, widespread, and variable across all tissues; highest mean levels occurred in the GCT and normal ovary. p120 expression was low or absent in all tissues except the GCT. Within the GCT there was parallel expression of the ActR subunits, betaglycan and p120; the levels, however, varied considerably between tumors. Expression of betaglycan and p120 in most GCT argues against the hypothesis, but does not exclude the possibility that low or absent expression of p120 might be significant in a subset of these tumors.
INHIBIN AND ACTIVIN are related peptides, that share subunits derived from separate homologous genes of the TGFß superfamily of peptides (1). Inhibin is formed by heterodimerization of the -subunit with one of two possible ß-subunits, where :ßA gives inhibin A and :ßB gives inhibin B. Activins are homo- or heterodimers of the inhibin ß-subunits. The activins have a diverse range of essentially paracrine actions, whereas inhibin appears to function through inhibition of activin (2). Inhibin synthesis and secretion are prominent features of many ovarian tumors, particularly granulosa cell tumors (GCT) and mucinous cystadenocarcinomas (MC) (3, 4). In the GCT this is not unexpected, as inhibin is a normal product of proliferating granulosa cells (5). In patients with GCT, FSH levels are generally suppressed, suggesting that the inhibin secreted by the tumor is bioactive (4, 6). Several studies have now demonstrated the utility of inhibin as a tumor marker, not only in GCT (7) but also in some epithelial tumors (4, 8, 9).
In transgenic mice, deletion of the inhibin -subunit gene results in the development of stromal tumors of the ovary, leading the authors to suggest that -inhibin is a tumor suppressor gene (10). This result is inconsistent with the situation in the human, where expression of the inhibin -subunit gene clearly occurs, and biologically active inhibin is indeed secreted. Moreover, Watson et al. (11) failed to find any evidence of LOH at the inhibin -subunit gene locus in a cohort of granulosa cell tumors. To explore the apparent contradiction between the murine and human tumors, Matzuk et al. (2, 12) suggested that in the human GCT, there may be loss of a tumor suppressor gene that is downstream in the inhibin signal transduction pathway, such as the inhibin receptor.
Identification of inhibin receptors has proven remarkably elusive. Other TGFß and superfamily ligands interact with type I and type II receptors, which are transmembrane serine-threonine kinase receptors (13, 14, 15). The ligands bind to the type II receptors, which subsequently recruit a type I receptor. Two cell surface molecules that bind inhibin with high affinity and antagonize the action of activin have recently been identified (2). The first, betaglycan, acts with the activin type II receptor (ActRII) to bind inhibin with high affinity (16), and the second, p120, is a cell adhesion molecule that binds inhibin with high affinity in association with the activin type I receptor (ActRI) (17).
In this study we addressed the hypothesis that loss of expression of the inhibin receptor may be part of the molecular pathogenesis of ovarian granulosa cell tumors (2, 12). The expression profiles of the subunits that constitute the inhibin receptor in a cohort of GCT were compared with the expression in the normal ovary as well as in epithelial tumors of the ovary. Specifically we determined the expression of the betaglycan, p120, and activin receptor subunit genes.
Materials and Methods
Isolation of RNA from tissue specimens
Ovarian GCT (n = 7), MC (n = 8), and serous cystadenocarcinomas (SC; n = 9) were obtained in a study of serum inhibin levels in ovarian tumors (4, 7). The tumors were consecutive tumors for which adequate tissue was available for RNA extraction. Some of these tissues have been examined in previous studies for inhibin subunit expression (18) and ERß gene expression (19). Normal ovarian tissue was obtained from eight premenopausal women who had undergone elective hysterectomy with oophorectomy for a range of conditions not associated with ovarian malignancy. Clinical details for many of the tumors (Table 1
) have been presented previously (18, 19). The RNA was isolated using the guanidine thiocyanate/cesium chloride method as described previously (18).
|
One microgram of total RNA was reverse transcribed for 120 min at 42 C in a total volume of 20 µl using AMV reverse transcriptase (Roche Molecular Biochemicals, Mannheim, Germany). First strand synthesis for ActRIA, ActRIB, ActRIIA, ActRIIB, betaglycan, p120, and ß2-microglobulin was performed using 30 pmol oligo(deoxythymidine). The oligonucleotide primers for the ActRIIA and ß2-microglobulin genes have previously been described (18). The primers for the other activin-inhibin receptor subunit genes (Table 2) were designed from published sequences (20, 21, 22, 23) with OLIGO Primer Analysis software version 5.0 (Natural Biosciences, North Plymouth, MN).
|
, and extension at 72 C for 2 min with a final incubation at 72 C for 10 min. The number of cycles is also shown in Table 2; these were chosen to ensure that amplification was in the linear phase. The products were visualized on a 2.0% agarose gel, stained with ethidium bromide, and photographed under UV transillumination. Controls for the RT-PCR were the reaction mixtures described above, but with reverse transcriptase omitted. The identity of the amplicons was confirmed by automated sequencing in the Wellcome Trust Joint Sequencing Facility at Monash Medical Center.
Southern blot analysis
For Southern blot analysis using gene-specific 32P-labeled probes (Table 2), the PCR products described above were transferred to Hybond N+ membranes (Amersham Pharmacia Biotech, Aylesbury, UK) as described previously (18, 19, 24). Other amplicons were included in each transfer as hybridization controls. Radiolabeled membranes were exposed to a storage phosphor screen, which was scanned using a Storm PhosphorImager (Molecular Dynamics, Inc., Sunnyvale, CA). Semiquantitative analysis using the Storm PhosphorImager was performed by measuring densitometric values for each band and correcting for ß2-microglobulin, and data were plotted as a scattergram. Statistical analysis for each dataset was performed using PRISM software version 2.1 (GraphPad Software, Inc., San Diego, CA). They were analyzed by one-way ANOVA, followed by Dunnett’s test comparing each dataset with the normal ovary.
Results
Each of the PCR reactions produced an amplicon of the predicted size, which could be visualized on ethidium-stained gels (data not shown). The identity of the product was confirmed by both direct sequencing and Southern blot analysis using internal gene-specific oligonucleotide probes. Each of the samples was analyzed on at least three occasions; in all cases the relative levels of expression were reproduced across experiments. The Southern blot analyses were used to assess the pattern of expression for each of the genes across the various tissue samples (Fig. 1). Relative levels of expression were compared using a PhosphorImager with correction using ß2-microglobulin levels (Fig. 2).
|
|
). Considerable variability of expression was seen for the activin type IIA receptor in all tissues examined, although the mean levels for the mucinous tumors, GCT, and normal ovary were relatively similar. The levels in the serous tumors were significantly higher, with several tumors having extremely high expression. The profile of expression of the type IIB receptor at an individual tumor level did not parallel that of the type IIA receptor; indeed, in some tumor types there appeared to be almost an inverse correlation. Particularly striking is the relatively low level in the mucinous tumors and the relatively higher levels in the GCT.
Expression of the betaglycan gene was widespread and abundant; relative levels varied considerably within tumor types (Fig. 1). Between-tumor expression was higher in GCT and normal ovary (Fig. 2). In view of the putative relationship of the betaglycan with the ActRII in inhibin signaling (16), the ratio of betaglycan to the type II receptor subunit mRNA levels in each tissue was determined. This ratio was 2- to 3-fold higher in the serous tumors than in the other tissues, which had relatively similar ratios.
Expression of the p120 gene was particularly interesting, with little or no expression seen in tissues other than the GCT where, again, considerable variability in the levels of expression was observed (Fig. 1). As with betaglycan, the ratio of p120 expression to that of the relevant activin receptor subunits, i.e. the ActRIs (17), was determined for each tissue. This further emphasized the predominance of p120 expression in the GCT with a ratio that was approximately 10-fold higher in the GCT than in the other tissues.
GCT were the primary focus of this study; as a group they showed a markedly heterogeneous pattern of expression. This is in contrast to our other studies of gene expression in these tumors (18, 19), where there has been considerable homogeneity of expression within the GCT. Of particular note is the observation that the expression of all six genes appeared to be parallel in the GCT, with only one or two exceptions (Fig. 1).
Discussion
Ovarian granulosa cells normally secrete inhibin under the control of FSH (5). In 1989, Lappohn et al. (3) reported that serum inhibin levels were elevated in patients with granulosa cell tumors of the ovary. This finding has subsequently been confirmed in a number of studies (4, 7, 25). The major form of inhibin secreted by the tumors is inhibin B (26, 27), and on the basis of the suppressed FSH levels before the removal of these tumors, it is biologically active (4, 6). In addition, a number of epithelial tumors of the ovary, particularly MC (4, 6, 8), also secrete immunoreactive inhibin. We have previously demonstrated expression of the inhibin gene in both GCT and epithelial tumors (18). Although it is clear that inhibin may be a very useful tumor marker in ovarian cancer, a possible pathological role had not been suspected until the landmark study of Matzuk et al. (10). They found that female mice null for the inhibin gene developed granulosa cell tumors. The mice have cachexia-like symptoms (28) that appear to result from excessive secretion of activin. Deletion of the ActRIIA gene ameliorates the cachexia, but not the tumor development (29). Gonadotropins are essential for the development of these tumors, in that when the inhibin null mice are crossed with gonadotropin-deficient mice (hpg mice), the tumors fail to develop in the double null homozygotes (30). Conversely, when FSHß-deficient mice were crossed with the inhibin null mice, the tumors still developed, but at a later stage (31). The major difference between the two double knockouts is the retention of LH secretion in the latter line (31). The importance to the tumor phenotype of the high levels of activin that are produced in the inhibin null mice remains unresolved. Although the studies with ActRIIA knockout mice (29) argue against a role, it is possible that the signaling may be occurring through ActRIIB. Activin stimulates growth of GCT cell lines derived from these mice (32), and Di Simone et al. (33) reported that activin stimulates the growth of epithelial ovarian cancer cell lines. Overexpression of follistatin under the control of a metallothionine promoter in the inhibin null mice also blocks the activin-mediated cancer cachexia syndrome; however, although gonadal tumors still develop, survival is prolonged (34). The mechanism of the tumorigenesis in this mouse model is not clear, but it does not appear to be explicable by either activin (29, 33) or FSH (30, 31) hypersecretion alone; rather, the loss of inhibin signaling appears to be critical (2, 12).
Watson et al. (11) sought evidence to implicate the inhibin gene as a tumor suppressor gene in human malignancy, but they were unable to find any evidence of LOH at the inhibin locus in GCT. The relevance of this mouse model to human tumors is not apparent given the clear and abundant expression of the inhibin -subunit gene in human GCT (18). To reconcile the murine and human situations, Matzuk (2, 10) suggested that in the human GCT, loss of expression of some component of the inhibin signaling pathway may be pathogenetic.
Inhibin is a member of the TGFß superfamily of secreted growth factors. Receptors for these ligands have been recently characterized and reviewed in detail (13, 14, 15). The ligand brings together members from two families of receptor serine/threonine kinases The type II receptors activate the type I receptor, which, in turn, phosphorylates the SMAD signal transduction proteins, which translocate into the nucleus where they have direct transcriptional effects (14). In the case of inhibin, most, if not all, of this cellular action can be explained in terms of antagonism of activin (2). Activin binds to specific type II receptors (ActRIIA and -B) to promote recruitment and phosphorylation of the type I serine kinase (ActRIA and -B). Although inhibin-binding sites had been shown in a number of tissues (2, 16, 17), their biochemical characterization has proven difficult. Recently, Lewis et al. (16) found that the so-called type III TGFß receptor, betaglycan, binds inhibin with a high affinity, particularly in association with ActRII. This interaction facilitates inhibition of activin action, presumably by sequestering ActRII. This is in contrast to the situation for TGFß, where betaglycan enhances binding of TGFß to the type II receptors. These researchers reported betaglycan immunoreactivity in follicular granulosa and thecal cells as well as the oocyte. It is also expressed in the gonadotropes of the pituitary, the classic inhibin target tissue. Chong et al. (17) used a biochemical approach to isolate an inhibin-binding protein from bovine pituitary, which they designated p120. The peptide sequence obtained was homologous to a known Ig superfamily/cell adhesion protein. As with betaglycan, it has a large extracellular domain, a single transmembrane domain, and a short cytoplasmic domain. p120 was immunolocalized to the pituitary gonadotrophs and the Leydig cells of the testes; the ovary was not examined (17). Mazzarella et al. (23) screened a range of human tissues for p120 expression by Northern blot analysis; they found high levels of expression in adult testis, with moderate expression in several other tissues and low, but detectable, levels in ovary among other tissues. Similar relative abundance was observed in the rat by Bernard and Woodruff (35), who also found high levels of expression in the pituitary.
Chapman and Woodruff (36) analyzed the interaction of p120 with the activin receptor subunits. Central to inhibin action is an interaction between p120 and the ActRI; in the absence of inhibin and in the presence of activin, this interaction stabilizes the ActRI/ActRII complex. The recruitment of p120 to the activin receptor complex allows inhibin B (but not inhibin A) to bind with much higher affinity to the complex, which consequently destabilizes the complex and abrogates activin signal transduction (36). This study (36) suggests ligand specificity with inhibin B preferentially interacting with p120; given that inhibin B is the product of proliferating preovulatory granulosa cells and indeed the predominant form of inhibin produced by GCT (18, 26, 27), it may be that p120 is relatively more important in this context than betaglycan. Conversely, the low levels of p120 expression observed in the normal ovary (23, 36) might argue against a role for p120, although it is entirely plausible that a subset of granulosa or inhibin-sensitive cells expresses p120.
In GCT our studies reveal the expression of both isoforms of the type I and type II receptors as well as expression of betaglycan and p120, which would argue against a role for the absence of the receptors in the pathogenesis of the tumors. The apparent importance in inhibin signaling of the interactions between betaglycan and p120 with ActRII and ActRI, respectively, suggests that the ratios of the interacting partners may be relevant. For betaglycan and the type II receptors this analysis reveals relatively higher betaglycan expression in the serous tumors. The ratio of p120 expression to that of either of the type I receptors is approximately 10-fold higher in the GCT than in the other tissues, reinforcing the potential importance of p120 and perhaps inhibin B signaling in GCT and/or granulosa cells per se. The analysis is semiquantitative, and it is difficult to relate the expression levels to those of normal granulosa cells; the normal ovaries examined are both collectively and individually heterogeneous. Although granulosa/lutein cells can be obtained during oocyte retrieval for in vitro fertilization, these cells probably do not equate to the cells from which the GCT arise; our studies (Chu, S., manuscript in preparation) suggest that the GCT most closely resemble preovulatory granulosa cells, which are very difficult to obtain. Of more interest is the variability of expression of these genes between GCT; previous studies of this cohort of tumors for the expression of other genes including the ERß (19), FSH receptor (37), and inhibin subunit (18) genes have shown relatively uniform expression within the GCT. Indeed, ß2-microglobulin levels are remarkably uniform across these six GCT. The trend appears to be either for expression of all six components of the activin-inhibin receptor (e.g. GCT 9, 11, 12, and 14) or for there to be low to absent expression of most, if not all, genes (GCT 13). This pattern does not appear to correlate with any other clinical, biochemical, histological, or molecular parameters determined to date.
Both epithelial tumor types express the activin receptor subunits, which, given the widespread expression of the inhibin ß genes (18), is perhaps not surprising. Betaglycan is also variably expressed in these tumors, although its pattern does not correlate with the activin receptor subunit genes, which would be consistent with its primary role being in TGFß signaling. In contrast, p120 expression is low in the mucinous tumors and, with two exceptions, is absent in the serous tumors. The human p120 gene maps to chromosome Xq25 (22, 23). LOH at Xq25–26 has been reported in advanced human epithelial ovarian carcinoma; the absence of p120 expression in all but two serous tumors could be consistent with the loss and/or mutation of genes at this chromosomal locus (38). The very low p120 expression previously reported in the ovary (23, 36) and observed in this study is consistent with p120 not normally being expressed in ovarian epithelial cells.
A conceptual difficulty with the model of inhibin as a molecular antagonist is that if inhibin is a tumor suppressor in GCT, then it implies that activin must be a tumor promoter. It therefore follows that any interruption to this signaling pathway at the receptor or beyond will inhibit tumor promotion without the need to invoke a role for inhibin unless the effect is inhibin specific, i.e. only affects betaglycan or p120. This would appear to be true for GCT 15, where expression of the activin receptor subunit genes is present, but both betaglycan and p120 expressions are very low, or perhaps GCT 10, where p120 expression is relatively very low. Conversely, in GCT 13 the lack of expression of all components would seem to be neutral. A recent study from Ala-Fossi et al. (39) provides some support for the inhibin hypothesis; in a retrospective study of 30 patients who had previous surgery for GCT they found that 4 patients whose tumors were negative for inhibin immunoreactivity had a far worse outcome than those that were inhibin positive. Unfortunately there was no information on serum inhibin levels (39). The inhibin-negative tumors were associated with a more advanced stage. Although loss of inhibin synthesis may in this context simply be an epiphenomenon of the malignant process, one could also argue, particularly in the context of the mouse models, that the absence of inhibin has pathogenetic significance.
The related ligand, TGFß, also has a role in the malignancy, where it has a biphasic action. During malignant progression, the tumor cells may acquire resistance to the antimitogenic effect of TGFß while retaining a tumor-promoting function through stimulation of angiogenesis, immunosuppression, and synthesis of extracellular matrix (13). Inactivation of the type II receptor has been reported in some colon cancer cells, and homozygous deletion of SMAD4 in pancreatic cancers led to its discovery as the DPC4 tumor-suppressor gene. It has been estimated that nearly all pancreatic and other cancers have mutations in some component of the TGFß-signaling pathway (15). Wang et al. (40) examined 32 ovarian cancers for mutations in the TGFß signal transduction pathway. The tumors included a wide range of histological types, but no stromal tumors. Eleven of the tumors did not express TGFßR1 due to a range of genetic mutations. Recently, mutations of the activin type 1B receptor genes have been reported in pancreatic carcinoma, suggesting that activin may be involved in tumor suppression in this epithelial tissue (41).
GCT of the ovary represent a distinct relatively uncommon subgroup of ovarian cancers with a distinct endocrine phenotype. Previous studies suggest a relatively homogenous pattern of gene expression (18, 19) and therefore perhaps a relatively homogenous etiology. The phenotype has many features in common with the FSH- and estrogen-dependent proliferation phase of granulosa cell development (5), suggesting that activation of the FSH signal transduction pathway may contribute to the pathogenesis. However, activating mutations of either the FSH receptor or its coupled G proteins (37, 42, 43) have not been found. Failure of leutinization/terminal differentiation or atresia-apoptosis, the two usual fates of granulosa cells (5), might also be important. The inhibin null transgenic mouse model has been interpreted as showing that failure of a growth inhibitory signal may also be a pathogenetic mechanism. These various possibilities are of course not mutually exclusive, and it is likely that more than one pathway will be compromised. The findings presented demonstrate an unexpected degree of heterogeneity in activin-inhibin receptor subunit gene expression, which superficially argues against a specific role in the pathogenesis at least as a key etiological factor. In the case of p120 expression the differences are quite striking, and in some tumors are certainly consistent with an absence of p120 and, hence, inhibin action contributing to the pathogenesis. Clearly the mechanism of this decreased expression needs to be determined, as do the levels of p120 protein on the surface of the tumor cells where p120 mRNA was detected. If some tumors do indeed lack functional p120, then ideally a correlation with inhibin binding needs to be sought; it is conceivable that those GCT expressing p120 are expressing a mutant nonfunctioning molecule.
In conclusion, although the relatively specific expression of p120 in the GCT emphasizes its potential importance, the unexpected heterogeneity of its expression within the GCT neither clearly reflutes nor establishes its role in pathogenesis. Functional characterization of p120 in those GCT in which expression is observed will be needed to resolve these issues.
Acknowledgments
We thank Sue Panckridge and Claudette Thiedeman for their contributions to preparation of the manuscript. We thank Guck Ooi for the p120 primers. We also acknowledge the contribution of our clinical colleagues, Tom Jobling and David Healy and Bruce Ward.
Footnotes
This work was supported by a project grant from the National Health and Medical Research Council of Australia (no. 122201) and was presented in part at the 83rd Annual Meeting of The Endocrine Society, Denver, Colorado, June 2001.
1 Recipient of a Principal Research Fellowship from the National Health and Medical Research Council of Australia. 
2 Recipient of a Dora Lush Postgraduate Scholarship from the National Health and Medical Research Council of Australia.
Abbreviations: ActR, Activin receptor; GCT, granulosa cell tumors; MC, mucinous cystadenocarcinomas; SC, serous cystadenocarcinomas.
Received April 9, 2001.
Accepted November 30, 2001.
References
-Inhibin is a tumor-suppressor gene with gonadal specificity in mice. Nature 360:313–319[CrossRef][Medline]
-inhibin locus on chromosome 2q is not a feature of human granulosa cell tumors. Gynecol Oncol 65:387–390[CrossRef][Medline]
and p53-deficient mice: the role of activin as an autocrine growth factor. Mol Endocrinol 8:983–995This article has been cited by other articles:
 |
|
 |
|
  M. Bilandzic, S. Chu, P. G. Farnworth, C. Harrison, P. Nicholls, Y. Wang, R. M. Escalona, P. J. Fuller, J. K. Findlay, and K. L. Stenvers Loss of Betaglycan Contributes to the Malignant Properties of Human Granulosa Tumor Cells Mol. Endocrinol., April 1, 2009; 23(4): 539 - 548. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.-H. Choi, A. S. T. Wong, H.-F. Huang, and P. C. K. Leung Gonadotropins and Ovarian Cancer Endocr. Rev., June 1, 2007; 28(4): 440 - 461. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Hempel, T. How, M. Dong, S. K. Murphy, T. A. Fields, and G. C. Blobe Loss of Betaglycan Expression in Ovarian Cancer: Role in Motility and Invasion Cancer Res., June 1, 2007; 67(11): 5231 - 5238. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. C.K. Leung and J.-H. Choi Endocrine signaling in ovarian surface epithelium and cancer Hum. Reprod. Update, March 1, 2007; 13(2): 143 - 162. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. D. Steller, T. J. Shaw, B. C. Vanderhyden, and J.-F. Ethier Inhibin Resistance Is Associated with Aggressive Tumorigenicity of Ovarian Cancer Cells Mol. Cancer Res., January 1, 2005; 3(1): 50 - 61. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Xu, Y. Zhong, S. Munir, B. B. Yang, B. K. Tsang, and C. Peng Nodal Induces Apoptosis and Inhibits Proliferation in Human Epithelial Ovarian Cancer Cells via Activin Receptor-Like Kinase 7 J. Clin. Endocrinol. Metab., November 1, 2004; 89(11): 5523 - 5534. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Liu, T. Kuulasmaa, V.-M. Kosma, R. Butzow, T. Vanttinen, C. Hyden-Granskog, and R. Voutilainen Expression of Betaglycan, an Inhibin Coreceptor, in Normal Human Ovaries and Ovarian Sex Cord-Stromal Tumors and Its Regulation in Cultured Human Granulosa-Luteal Cells J. Clin. Endocrinol. Metab., October 1, 2003; 88(10): 5002 - 5008. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Chu, S. Rushdi, E.T. Zumpe, P. Mamers, D.L. Healy, T. Jobling, H.G. Burger, and P.J. Fuller FSH-regulated gene expression profiles in ovarian tumours and normal ovaries Mol. Hum. Reprod., May 1, 2002; 8(5): 426 - 433. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
