help button home button Endocrine Society Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology, doi:10.1210/en.2007-0155
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Farnworth, P. G.
Right arrow Articles by Findlay, J. K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Farnworth, P. G.
Right arrow Articles by Findlay, J. K.
Endocrinology Vol. 148, No. 11 5355-5368
Copyright © 2007 by The Endocrine Society

Transforming Growth Factor-β Blocks Inhibin Binding to Different Target Cell Types in a Context-Dependent Manner through Dual Mechanisms Involving Betaglycan

Paul G. Farnworth, Yao Wang, Ruth Escalona, Pauline Leembruggen, Guck T. Ooi and Jock K. Findlay

Prince Henry’s Institute of Medical Research (P.G.F., Y.W., R.E., P.L., G.T.O., J.K.F.) and Department of Physiology (P.G.F., G.T.O.), Monash University, Clayton, Victoria 3168, Australia; and Sun Biomedical Technologies Inc. (G.T.O.), Ridgecrest, California 93555

Address all correspondence and requests for reprints to: Paul Farnworth, Prince Henry’s Institute of Medical Research, PO Box 5152, Clayton, Victoria 3168, Australia. E-mail: paul.farnworth{at}princehenrys.org.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Inhibin antagonizes activin and bone morphogenetic protein actions by sequestering their type II receptors in high-affinity complexes with betaglycan, a coreceptor that inhibin shares with TGF-β. To clarify the nature and extent of interactions between inhibin and TGF-β, we therefore examined 1) the mutual competition between these ligands for binding, 2) the regulation of endogenous betaglycan expression by inhibin and TGF-β isoforms, and 3) the consequences of such betaglycan regulation for subsequent inhibin binding in mouse Leydig (TM3), Sertoli (TM4), adrenocortical cancer (AC), and gonadotroph (LβT2) cell lines, chosen to model cellular targets for local and endocrine actions of inhibin. Recognized inhibin, activin, and TGF-β binding proteins and TGF-β/activin signaling components were expressed by all four cell types, but AC and LβT2 cells notably lacked the type II receptor for TGF-β, TβRII. Overnight treatment of TM3 and TM4 cells with TGF-β1 suppressed the levels of betaglycan mRNA by 73 and 46% of control and subsequent [125I]inhibin A binding by 64 and 41% of control (IC50 of 54 and 92 pM), respectively. TGF-β2 acted similarly. TGF-β pretreatments commensurately decreased the [125I]inhibin A affinity labeling of betaglycan on TM3 and TM4 cells. TGF-β isoforms as direct competitors blocked up to 60% of specific inhibin A binding sites on TM3 and TM4 cells but with 9- to 17-fold lower potency than when acting indirectly via regulation of betaglycan. Only the competitive action of TGF-β was observed with TβRII-deficient AC and LβT2 cells. Neither inhibin A nor inhibin B regulated betaglycan mRNA or competed for binding of [125I]TGF-β1 or -β2. Thus, inhibin binding to its target cell types is controlled by TGF-β through dual mechanisms of antagonism, the operation of which vary with cell context and display different sensitivities to TGF-β. In contrast, TGF-β binding is relatively insensitive to the presence of either inhibin A or inhibin B.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
INHIBINS AND ACTIVINS are members of the TGF-β superfamily that also includes the growth/differentiation factor (GDF)/bone morphogenetic protein (BMP) family of secreted ligands (1, 2, 3). Each superfamily member is locally produced within multiple tissues and can display a wide range of autocrine/paracrine actions, the specific nature of which depends on the ontogenetic and cellular context (4). Inhibins are expressed in the gonads and adrenal glands, where they have local actions in processes such as steroidogenesis (4, 5, 6). Inhibin secreted from the gonads additionally acts as an endocrine feedback inhibitor to suppress FSH production by gonadotropes of the anterior pituitary gland (1).

Activin, TGF-β, and BMP/GDF ligands signal by bringing together distinct combinations of constitutively active serine/threonine kinase type II receptors and dormant type I activin receptor-like kinase (ALK) receptors. Once transactivated, each ALK receptor initiates downstream signaling by receptor-specific subsets of intracellular Smads, in combination with the common signaling Smad4 and cell-specific transcription factors, to modify gene expression. Secreted and membrane-associated binding proteins selectively modify the actions of TGF-β superfamily agonists. Notably, the transmembrane proteoglycan, betaglycan, acts as a coreceptor for TGF-β, enhancing its interaction with TGF-β type II receptor (TβRII) and type I receptor (ALK5) (7). In most circumstances, the TGF-β2 isoform in fact requires betaglycan for signaling to occur (7, 8). More recently, betaglycan was found to additionally function as a coreceptor for inhibin (9, 10).

No agonist signaling pathway has yet been identified for inhibin. Inhibin directly binds to betaglycan, thereby increasing the affinity of inhibin binding to activin and BMP type II receptors [activin receptor type II (ActRII), ActRIIB, and BMP receptor type II (BMPRII)] (9, 10, 11). This sequestration of type II receptors reduces the capacity of activin and BMP to access their respective type I receptors, thereby disrupting their signaling. Primary cultures of rat and sheep anterior pituitary cells and mouse LβT2 gonadotropes serve as models for endocrine targets of inhibin action (12, 13, 14, 15), and adrenocortical cancer (AC), Leydig-like TM3, and Sertoli-like TM4 cell lines model target cells for paracrine/autocrine actions of inhibin (12, 16, 17). These cell types bind radiolabeled inhibin A with high affinity via not only betaglycan and proteins corresponding in size to activin/BMP type II receptors, consistent with the simple betaglycan model of inhibin action, but also multiple additional inhibin binding proteins (12, 14, 16, 17).

Although inhibin and TGF-β share betaglycan as a coreceptor, little is known of interactions between these factors under physiological conditions. First, interaction can occur directly at the level of binding. In some circumstances, especially where cells are made to constitutively overexpress exogenous betaglycan, TGF-β competes for inhibin binding to the more membrane-proximal of two binding sites for TGF-β on betaglycan (9, 18). However, we hypothesized that the endogenous expression of multiple additional inhibin binding proteins alongside betaglycan could lead to mutual competition between inhibin and TGF-β for binding to inhibin target cell types. Second, inhibin-TGF-β interaction under physiological conditions might also occur through the control of endogenous betaglycan expression. Betaglycan mRNA is down-regulated by TGF-β1 in small-cell lung cancer cell lines and C2C12 myoblast cells (19, 20), whereas TGF-β1 up-regulates betaglycan expression in cultured human periodontal ligament cells (21). Inhibin regulation of betaglycan expression has not been reported. Because inhibin and TGF-β are coexpressed in several tissues and cell types, our second hypothesis was that TGF-β regulation of betaglycan expression in inhibin target cells indirectly controls the binding and action of inhibin. If inhibin regulates betaglycan, then the converse might also be true.

To investigate the interactions between inhibin and TGF-β, we examined the direct effects of TGF-β isoforms as competitors for the binding of radiolabeled inhibin A to inhibin target cell lines, each of which endogenously expresses betaglycan. The direct effects of inhibin on radiolabeled TGF-β binding were also examined. Two of the tested cell lines were found not to express TβRII, so we compared the effects arising from direct competition between inhibin and TGF-β for binding sites during concurrent exposure with the effects resulting from prior treatment of the cells with TGF-β. The studies showed that TGF-β isoforms directly compete for a major portion of inhibin A binding sites on each of the tested cell lines. In the cell lines that express TβRII, much lower concentrations of TGF-β isoforms, particularly TGF-β1, suppress the endogenous expression of betaglycan and thereby indirectly reduce inhibin binding. It is concluded that inhibin binding to, and anti-activin/BMP/GDF action on, its target cells can be differentially regulated in a context-dependent manner by TGF-β through dual mechanisms that reflect both binding to betaglycan and regulation of betaglycan expression.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Materials
The mouse TM3 and TM4 cell lines were from the American Type Culture Collection (Rockville, MD), the AC cell line was derived from the C{alpha}-1 cell line provided by Prof. Ilpo Huhtaniemi and Dr. Nafis Rahman (University of Turku, Turku, Finland) (17, 22), and LβT2 cells were provided by Dr. Pam Mellon (University of California, San Diego, CA) (23). Cell culture media and supplements, including fetal bovine serum (FBS), were obtained from Thermo Trace Biosciences (Melbourne, Victoria, Australia), antibiotics (final concentrations of 100 U/ml penicillin, 100 µg/ml streptomycin, 250 ng/ml fungizone) were from Commonwealth Serum Laboratories (Parkville, Victoria, Australia), and insulin (Actrapid, 100 IU/ml) was from Novo Nordisk Pharmaceuticals (North Rocks, New South Wales, Australia). BSA (fatty acid content 0.004%), dog serum albumin, human transferrin, and inhibin B for early experiments were obtained from Sigma-Aldrich Co. (St. Louis, MO). Recombinant human 31-kDa inhibin A and additional inhibin B (obtained in crude form from Biotech Australia, Sydney, Australia) were purified and then stored at –70 C in 0.1% trifluoroacetic acid/acetonitrile (24). Recombinant human activin A and activin B and antiserum to human betaglycan were obtained from R&D Systems (Minneapolis, MN), TGF-β1 and TGF-β2 were obtained from PeproTech Inc. (Rocky Hill, NJ). Reagents and materials for RNA extraction, RT-PCR amplification, ligand radioiodination, and protein cross-linking were as previously described (6, 17). Primers for gene amplification were made commercially (Sigma-Genosys, Castle Hill, New South Wales, Australia).

Culture and treatment of mouse cell lines
Each cell line was cultured in a 1:1 (vol/vol) mixture of DMEM:F12 media buffered with bicarbonate and containing antibiotics, nonessential amino acids for MEM, and supplementary glutamine (2 mM final). Medium was supplemented with 10% FBS for passaging and preincubations and, for AC cells, also included HEPES buffer (2 mM). Cells were maintained at 37 C in a humidified atmosphere of 5% CO2 in air.

After plating, cells were preincubated for 1 d in FBS-containing medium, and then culture medium was changed to a chemically defined medium containing 10% artificial serum (AS), prepared as previously described (6, 17). Unless otherwise indicated, treatments were applied to cells overnight (18 h) under standard culture conditions in this chemically defined AS-containing medium before the analyses.

mRNA purification and analysis
Total RNA was extracted from cell monolayers using UltraSpec RNA reagent, purified using standard procedures, and digested with DNase, and then cDNAs were synthesized from 2 µg RNA by oligo(dT) priming using 50 U Expand reverse transcriptase. Primers for amplification of specific cDNA products were as follows: mouse ALK1 (forward primer, 5'-ATGACCTTGGGGAGCTTCAGA-3'; reverse primer, 5'-GATGGGCATCAACTTCTGGCT-3'), yielding a 352-bp product, and mouse ALK5 (forward primer, 5'-GTCCGCAGCTCCTCATCGT-3'; reverse primer, 5'-GACAGTGCGGTTATGGCAGAT-3'), yielding a 431-bp product (25). Primers and conditions for amplifications of mouse betaglycan, ALK2, -3, -4, -6, and -7, ActRII/IIB, BMPRII, TβRII, Smad2, and Smad3 were as previously described (6, 17, 26).

Real-time PCR amplification assays were performed in a LightCycler (Roche Diagnostics Australasia, Castle Hill, New South Wales, Australia) (17). The LightCycler amplification and assessment run for ALK1 (or ALK5) consisted of 1) denaturation at 95 C for 6 (or 8) min to activate the Taq polymerase, 2) amplification and quantification for 40 cycles of 15 sec at 95 C for denaturation, 5 sec at 66 C for annealing, 16 (or 20) sec at 72 C for extension, and a single fluorescence measurement at 72 C for quantitation, 3) melting curve assessment between 57 and 95 C at a temperature transition rate of 0.2 C/sec with a continuous fluorescence measurement; and 4) cooling to 40 C. Under these conditions, only single ALK1 and ALK5 transcripts were amplified. Amplification specificity was assessed by consistent melting point analyses and the occurrence of a single product on agarose gel electrophoresis. The identity of each cell product was confirmed by nucleotide sequencing of representative samples. Products obtained from 40 cycles of PCR amplification were run on 1.7% agarose gels, stained with ethidium bromide, and photographed for qualitative evidence of the presence of each mRNA species. The procedures for quantitatively comparing the levels of mRNA encoding binding and signaling proteins for TGF-β superfamily members in the cell lines have been described previously (17). Samples from each experiment for real-time RT-PCR were analyzed in duplicate and corrected for the corresponding glyceraldehyde-3-phosphate dehydrogenase (GAPDH) cDNA level determined by RT-PCR (17, 26) and then normalized by comparison with the value determined for matching control samples from the same cell culture assayed in the same run on the same day. Yields were converted to femtograms of amplicon per microgram total RNA after correction against the respective 28S RNA content of the total RNA sample determined densitometrically from an ethidium bromide-stained gel, arbitrarily setting the average level in the four cell lines as 1.

Radioiodination of inhibin and TGF-β and assays of binding to cells
Inhibin A (5-µg aliquots) was radioiodinated by a lactoperoxidase method (13), whereas TGF-β1 and -β2 were radiolabeled using chloramine T (27, 28). [125I]Inhibin A binding assays and subsequent affinity labeling and immunoprecipitations with betaglycan antiserum were conducted on d 2 of culture, as previously described (14, 17). Similar procedures were used for [125I]TGF-β isoform binding, affinity labeling, and immunoprecipitation, except that the cell monolayers were incubated at room temperature (24 C) with [125I]TGF-β [~200,000 cpm/ml in 48-well plates (0.125 ml/well) and six-well plates (0.60 ml/well), corresponding to a final concentration of 80–150 pM] for 2 h in the presence or absence of unlabeled TGF-β or inhibin competitors (0–50 nM final concentration). Where appropriate, binding data were corrected for nonspecific binding, which was determined in the presence of excess unlabeled ligand (see legends to figures and tables for details). The proteins to which [125I]inhibin A and [125I]TGF-β isoforms bind on the surfaces of TM3, TM4, AC, and LβT2 cells were compared by affinity labeling with the bifunctional cross-linker, BS3, as previously described (17). After the separation of labeled complexes by nonreducing 7.5% SDS-PAGE, Kodak MS films were exposed to dried gels for 3–21 d using BioMax TranScreen-HE intensifying screen (Eastman Kodak, Rochester, NY). Resulting autoradiographs for some [125I]inhibin A tracer preparations showed species of higher molecular mass than 31 kDa, particularly material that migrated as a 50-kDa complex, and also traces around 100 kDa, especially where long film exposure times were required (e.g. Fig. 2AGo), but these complexes were absent in other tracer preparations (e.g. Fig. 9Go). Apart from the 50-kDa complex, these tracer species migrated at different sizes from the affinity-labeled complexes in cell lysates and immunoprecipitates. In addition, the intensities of radioactivity associated with each cell-derived complex differed greatly between cell types despite total levels of bound [125I]inhibin A being equivalent, indicating that the cell-derived complexes did not reflect tracer artifacts.


Figure 2
View larger version (27K):
[in this window]
[in a new window]

  		FIG. 2. Comparison of inhibin A and TGF-β2 binding proteins in inhibin target cell lines. Autoradiographs were obtained after SDS-PAGE of affinity-labeled protein complexes obtained from TM3, TM4, AC, and LβT2 cells after binding [125I]inhibin A for 4 h at room temperature (A) or [125I]TGF-β2 for 2 h at room temperature (B) and then cross-linking bound ligand using BS3. Before the binding assay, cell monolayers had been incubated in serum-free medium alone. Estimated molecular weights of complexes based on the relative migration of markers are shown on the right of each panel. The pattern obtained with the respective tracer (Tr.) preparation is also shown.

 

Figure 9
View larger version (23K):
[in this window]
[in a new window]

  		FIG. 9. Comparison of the effects of TGF-β through direct competition with those as a pretreatment on [125I]inhibin A affinity labeling of cell surface binding proteins. Semiconfluent monolayers of TM3 (A), TM4 (B), and AC (C) cells were either untreated for 18 h and then incubated for 4 h at room temperature with [125I]inhibin A in the absence (c, control) or presence of TGF-β1 (T1) or TGF-β2 (T2) (each 0.4 nM) as direct competitor for binding sites (Compn.) or were pretreated (Pre-trt.) for 18 h with the indicated TGF-β isoform and then washed and incubated for 4 h at room temperature with [125I]inhibin A in the absence of added TGF-β. Other details are the same as for Fig. 2Go.

 
Statistical analyses
Concentration-response curves were fitted and median inhibitory concentrations (IC50) were determined by nonlinear regression, Scatchard plots were fitted by linear regression (with correction for the presence of lower-affinity binding sites), and statistical analyses were performed using GraphPad Prism (version 3.00 for Windows; GraphPad Software Inc., San Diego, CA). Treatment effects on mRNA levels and radiolabeled ligand binding were replicated in separate experiments (see tables and figure legends for details). Mean data for treatments were compared using one-way ANOVA followed by post hoc Tukey’s multiple comparison test. P < 0.05 was considered indicative of a significant difference.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Expression of TGF-β superfamily signal transduction genes by inhibin-receptive cell types
The TM3, TM4, LβT2, and AC cell lines were all found to express mRNA encoding the type II receptors ActRII, ActRIIB, and BMPRII (Fig. 1AGo). However, only TM3 and TM4 cells expressed significant amounts of the type II receptor for TGF-β, TβRII (Fig. 1AGo). Relative levels of expression of each type II receptor mRNA determined by real-time PCR analysis of pooled samples for each cell type (Table 1Go) indicated that the ActRII mRNA level was similar in all the tested cell types, as was that of the more abundant ActRIIB mRNA, whereas BMPRII mRNA was more abundant in AC cells than the other cell types. TM4 cells expressed more TβRII mRNA than TM3 cells. Betaglycan mRNA was expressed by all four cell lines (Fig. 1BGo), with LβT2 cells expressing the lowest level (Table 1Go).


Figure 1
View larger version (18K):
[in this window]
[in a new window]

  		FIG. 1. Ethidium bromide-stained 1.7% agarose gels showing the presence of betaglycan (BG) and TGF-β superfamily binding and signaling molecules in mouse TM3, TM4, LβT2, and AC cells. In each case, RT-PCR was performed on pooled replicate total RNA samples, each extracted from the respective cell type cultured in the presence of 10% AS in the absence of treatment, and each band qualitatively shows the product from 40 cycles of PCR amplification. For each set of amplifications, a negative control was included where the reverse transcriptase (+RT) was omitted from the reaction (–RT), and no PCR products were obtained, indicating that the products were not amplified from contaminating genomic DNA. Molecular weight markers were run in the first lane of each gel, the size of each amplified product in base pairs is indicated on the right side, and the identities of representative products were confirmed by DNA sequencing. RT-PCR products shown are derived from the mRNA species encoding the following: A, ActRII and -IIB, BMPRII, and TβRII; B, betaglycan; C, ALK 1–7; D, Smad2 and Smad3.

 

View this table:
[in this window]
[in a new window]

  		TABLE 1. Representative levels (femtograms amplicon per microgram total RNA) of mRNA encoding TGF-β superfamily type I and II receptors, Smads 2 and 3, and betaglycan

 
At the qualitative level, TM3, TM4, LβT2, and AC cells expressed mRNA encoding the type I receptors ALK4 and ALK5 (Smad2/3 signaling) and ALK2, ALK3, and ALK6 (Smad1/5/8 signaling), whereas LβT2, AC, and to a minor extent, TM4 cells expressed mRNA encoding ALK7 (Smad2/3 signaling) (Fig. 1CGo). Expression of mRNA for ALK1 (Smad1/5/8 signaling) was evident in TM3, LβT2, and to a minor extent, TM4 cells but was not detected in AC cells (Fig. 1CGo). Quantitative comparisons by real-time PCR (Table 1Go) indicated that TM3 cells expressed much more ALK1 mRNA, and AC cells expressed more ALK2 mRNA but less ALK6 mRNA compared with the other cell lines, whereas the four cell lines expressed similar levels of ALK3 and ALK5 mRNAs and of the less abundant ALK4 mRNA. TM4, AC, and LβT2 cells expressed low levels of activin B/nodal-signaling ALK7 mRNA, which was at or below the detection limit in TM3 cells (Table 1Go). Smad2 and Smad3 were expressed in all four cell types (Fig. 1DGo), with highest levels of Smad2 mRNA in TM4 cells, whereas the less abundant Smad3 mRNA was lowest in AC cells (Table 1Go).

The patterns of expression of these binding and signaling molecules indicate that all four cell lines should bind inhibin and TGF-β and respond to activin (and BMP), but only TM3 and TM4 cells should transduce signals from TGF-β, because AC and LβT2 cells lack expression of the essential type II receptor, TβRII.

Binding proteins for inhibin A and TGF-β1 on inhibin-receptive cells
[125I]Inhibin A variously affinity-labeled protein complexes of more than 200, 148, 145, 142, 118, 110, 105, 95, 74–80, 50, and 40 kDa in TM3, TM4, AC, and LβT2 cells (e.g. Fig. 2AGo; also see Refs. 14 , 16 , and 17). Complexes of more than 200, 105, 95, 74–80, and 50 kDa were common to all four cell lines, whereas AC and LβT2 cells lacked the 148- and 142-kDa complexes, TM3 cells lacked the 145-kDa complex, LβT2 cells lacked the 40-kDa complex, and the 118-kDa complex was only (and inconsistently) seen in AC cell lysates (Fig. 2AGo; also see Ref. 17).

The patterns of affinity labeling with radioiodinated TGF-β isoforms were less complex than those seen with inhibin A in the same cell types. Using a similar strategy, [125I]TGF-β2 intensely affinity-labeled protein complexes of more than 200, 140, 95, and 68–75 kDa and faintly labeled complexes of 100, 85, and 40 kDa in TM3 and TM4 cells (Fig. 2BGo). In contrast, AC cells principally displayed [125I]TGF-β2-labeled complexes of more than 200 kDa (Fig. 2BGo). Labeling with [125I]TGF-β1 gave similar patterns (comparative data not shown; see Fig. 6Go).


Figure 6
View larger version (23K):
[in this window]
[in a new window]

  		FIG. 6. Effects of competitors on affinity labeling of TM3 and TM4 cell binding proteins by [125I]TGF-β1 and [125I]TGF-β2. Near-confluent monolayers of cells were incubated for 2 h at room temperature with the indicated tracers in the absence (con) or presence of competitors (each 10 nM) inhibin A (IA), TGF-β1 (Tβ1), or TGF-β2 (Tβ2). Other details are the same as for Fig. 2Go.

 
After affinity labeling cell membrane proteins with [125I]inhibin A, equivalent samples of whole-cell lysates were immunoprecipitated using an antiserum directed against betaglycan. The 142-kDa complexes and a portion of the material of more than 200 kDa, corresponding to betaglycan core protein and glycosylated betaglycan, respectively, were immunoprecipitated (data not shown), as previously described (14, 16, 17). The same antiserum immunoprecipitated complexes of 140 and more than 200 kDa from TM3 and TM4 cells after affinity labeling with either [125I]TGF-β1 or [125I]TGF-β2 (data not shown), confirming the assignment of these species as forms of the inhibin/TGF-β coreceptor, betaglycan. The antiserum immunoprecipitated only [125I]inhibin A- and [125I]TGF-β-affinity-labeled complexes of more than 200 kDa from AC and LβT2 cells (14, 17) (data for [125I]TGF-β not shown), suggesting that betaglycan core protein is poorly represented on the surfaces of these cells. The betaglycan antiserum also immunoprecipitated [125I]inhibin A-labeled complexes of 105, 95, and 74–80 kDa (faint bands) from TM3 cells (see later, Fig. 10Go), consistent in size with coprecipitated type II receptors for the superfamily ligands (105 and 95 kDa) and unknown species (74–80 kDa).


Figure 10
View larger version (23K):
[in this window]
[in a new window]

  		FIG. 10. Comparison of the effects of TGF-β1 (T1) or TGF-β2 (T2) (each 0.4 nM) through direct competition with those as a pretreatment on [125I]inhibin A affinity labeling of betaglycan and associated proteins. Details of the pretreatments, competition for binding, and labeling are as described in Fig. 9Go, except that 40-µl aliquots of whole-cell lysates were subjected to immunoprecipitation using an antiserum against betaglycan. Samples of immunoprecipitates were then separated by 7.5% nonreducing SDS-PAGE and analyzed by autoradiography.

 
Inhibin and TGF-β competition for [125I]inhibin A binding to cells
When applied in direct competition with [125I]inhibin A, unlabeled inhibin A maximally competed for 78–82% of total binding to TM3 (Fig. 3AGo), TM4, AC (Fig. 3BGo), and LβT2 cells in a concentration-dependent manner, with IC50 values of 0.065–0.13 nM (also see Table 2Go and Refs. 14 , 16 , and 17). Competition from inhibin B was not fully characterized in this study because supplies of purified inhibin B were insufficient. Analysis of the limited data according to a single-site binding model showed that low nanomolar concentrations of inhibin B competed for a smaller portion (52–53%) of total [125I]inhibin A binding, with at least 10-fold lower potency (higher IC50) than inhibin A (e.g. Fig. 3Go and Table 2Go). However, the tabulated IC50 values are estimates because a second phase of competition for [125I]inhibin A binding was commonly evident when inhibin B was applied at concentrations above 20 nM (data not shown; detailed in Ref. 17).


Figure 3
View larger version (18K):
[in this window]
[in a new window]

  		FIG. 3. Competition between inhibin and TGF-β isoforms for cell surface binding sites. A–D, Concentration-dependent competition for TM3 (A and C) and AC (B and D) cell binding of [125I]inhibin A (A and B) or [125I]TGF-β1 (C and D) provided by inhibin A ({circ}), inhibin B (bullet), TGF-β1 ({blacktriangleup}), or TGF-β2 ({triangledown}). All cell cultures were maintained overnight in FBS-free medium containing 10% AS before washing, and incubation with radiotracer under the conditions described for Fig. 2Go. Data at each point are single values obtained in one representative experiment and have been normalized to the level of binding in the absence of competitor (equal to 100%). GraphPad Prism software was employed to fit data according to a single-site model in each case. Corresponding aggregate data from replicate experiments are presented in Tables 2Go and 3Go.

 

View this table:
[in this window]
[in a new window]

  		TABLE 2. Inhibition of [125I]inhibin A binding to TM3, TM4, AC, and LβT2 cells provided by inhibin and TGF-β isoforms through direct competition for binding

 
TGF-β1 competed for around 50% of total [125I]inhibin A binding to TM3, TM4, and AC cell lines (e.g. Fig. 3Go, A and B), with average IC50 values around 0.9 nM, whereas it competed for only 42% of total [125I]inhibin A binding to LβT2 cells but with 3-fold greater potency (Table 2Go). TGF-β2 was at least as potent as TGF-β1 in competing for similar proportions of [125I]inhibin A binding sites on TM3, TM4, and AC cells (with IC50 values between 0.43 and 0.66 nM) and in most cases was more potent than inhibin B but was 4- to 10-fold less potent than inhibin A as a competitor (Fig. 3Go and Table 2Go). TGF-β2 competed potently for only a third of total [125I]inhibin A binding sites on LβT2 cells (Table 2Go). As with inhibin B, TGF-β isoforms at concentrations above 20 nM competed for an additional portion of inhibin A binding sites (data not shown), but the quantities of ligand required to fully characterize this phase of binding were considered excessive.

Taking TM3 cells as an example, excess unlabeled inhibin A (10 nM) abolished the [125I]inhibin A affinity labeling of all complexes (data not shown), as previously described (12, 16), whereas TGF-β1 and TGF-β2 as competitors (10 nM) partly blocked the labeling of not only betaglycan complexes of more than 200 and 145 kDa but also all other complexes (Fig. 4Go). As negative controls (12, 16, 17), activin A and activin B had minimal effects (Fig. 4Go).


Figure 4
View larger version (33K):
[in this window]
[in a new window]

  		FIG. 4. Effects of competitors on affinity labeling of TM3 cell binding proteins by [125I]inhibin A. Near-confluent monolayers of TM3 cells were incubated for 4 h at room temperature with [125I]inhibin A in the absence (con) or presence of competitors (each 10 nM) activin A (AA), activin B (AB), TGF-β1 (Tβ1), or TGF-β2 (Tβ2), and then bound [125I]inhibin A was cross-linked to its binding partners using BS3. The autoradiograph was obtained after SDS-PAGE of affinity-labeled protein complexes. Other details are the same as for Fig. 2Go.

 
Scatchard analyses of the inhibin A binding data yielded nonlinear plots that suggested the existence of high-, intermediate-, and low-affinity binding sites on TM3, TM4, AC, and LβT2 cells (e.g. Fig. 5Go, A and B), as previously described (12, 16, 17). This pattern is consistent with the number of different binding protein complexes evident by affinity labeling.


Figure 5
View larger version (17K):
[in this window]
[in a new window]

  		FIG. 5. Scatchard plots of homologous competition binding data for [125I]inhibin A (A and B) and [125I]TGF-β1 (C and D) binding to cell lines. Binding was performed at 25 C on 0.1–0.4 million cells per well in 48-well cluster dishes in the presence of either 150 pM [125I]inhibin A and graded concentrations of unlabeled inhibin A (0–20 nM) (A and B) or 80 pM [125I]TGF-β1 and graded concentrations of unlabeled TGF-β1 (0–14 nM) (C and D) (see Materials and Methods for additional details). From the plots obtained for [125I]inhibin A binding to TM3 and AC cells (A), three sites were evident, whereas for such binding to TM4 and LβT2 cells (B), two sites were apparent. The following estimated dissociation constants (nM) were deduced as follows: site 1, AC 0.22, TM3 0.16, LβT2 0.068, and TM4 0.055; site 2, AC 1.4, TM3 1.3, LβT2 1.6, and TM4 1.3; site 3, AC 16, TM3 14, LβT2 none, and TM4 none. The matching binding site concentrations (sites per cell) were as follows: site 1, AC 5000, TM3 8000, LβT2 860, and TM4 4,400; site 2, AC 16,000, TM3 22,000, LβT2 1,600, TM4 19,000; site 3, AC 63,000 and TM3 69,000. From the plots obtained for [125I]TGF-β1 binding to TM3 and AC cells (C) and TM4 and LβT2 cells (D), two or three sites were evident, with the following estimated dissociation constants (nM): site 1, AC 0.29, TM3 0.036, TM4 0.086, and LβT2 0.12; site 2, AC 2.2, TM3 1.6, TM4 2.2, and LβT2 2.1; site 3, AC 16, TM3 12, TM4 12, and LβT2 none. The matching binding site concentrations (sites per cell) were as follows: site 1, AC 4000, TM3 5000, TM4 4800, and LβT2 750; site 2, AC 32,000, TM3 25,000, TM4 15,000, and LβT2 8,800; site 3, AC 240,000, TM3 200,000, and TM4 77,000. The data presented in each panel are from single experiments representative of at least two in each case.

 
Inhibin and TGF-β competition for [125I]TGF-β binding to cells
In direct competition with [125I]TGF-β1, unlabeled TGF-β1 competed for a maximum 48–53% of total binding to TM3 (Fig. 3CGo) and TM4 cells, with IC50 values between 0.2 and 0.4 nM (Table 3Go, top half). TGF-β2 competed for similar portions of [125I]TGF-β1 binding to TM3 and TM4 cells but tended to be less potent than TGF-β1 in each case, and the limited data obtained with LβT2 cells followed a similar pattern (Table 3Go). In AC cells, TGF-β isoforms competed for more than 50% of total [125I]TGF-β1 binding, but TGF-β2 was the more potent isoform (Fig. 3DGo and Table 3Go).


View this table:
[in this window]
[in a new window]

  		TABLE 3. Inhibition of [125I]TGF-β1 and [125I]TGF-β2 binding to TM3, TM4, AC, and LβT2 cells provided by inhibin and TGF-β isoforms through direct competition for binding

 
Binding was also investigated using [125I]TGF-β2 as tracer in a limited series of experiments, because TGF-β2 depends on betaglycan for its binding to signaling receptors and so might be more subject to competition from inhibin for shared binding sites. TGF-β1 competed for around 60% of total [125I]TGF-β2 binding to TM3, TM4, and AC cells (Table 3Go, lower half), with lower potency (IC50 values between 1 and 3.6 nM) than was obtained when [125I]TGF-β1 was the tracer (Table 3Go; statistically significant difference in the case of AC cells, P < 0.001). TGF-β2 competed for a similar or greater proportion of total [125I]TGF-β2 binding to these cell types than did TGF-β1, with lower potency than TGF-β1, and TGF-β2 was also a significantly less potent competitor of [125I]TGF-β2 than of [125I]TGF-β1 in TM3 and TM4 cells (Table 3Go; P < 0.01).

In contrast to the substantial competition provided by TGF-β isoforms for [125I]inhibin A binding, 100-fold molar excesses of inhibin A and inhibin B competed for less than 17% of total [125I]TGF-β1 and [125I]TGF-β2 binding to TM3, TM4, AC, and LβT2 cells, which made IC50 determinations unreliable, but estimates of these generally exceeded 5 nM (Table 3Go). TM4 cells provided the exception, yielding inhibin A IC50 values of 0.4 nM or lower, but the extent of this competition was also minor (Table 3Go).

As with inhibin binding, Scatchard analyses of the TGF-β1 binding data for inhibin target cell lines yielded nonlinear plots, revealing complexity in the binding interactions (Fig. 5Go, C and D) and consistent with the multiple binding proteins observed by affinity labeling. Scatchard plots were devolved into three phases, suggesting the presence of high-, intermediate-, and low-affinity binding sites for TGF-β1 on TβRII-expressing TM3 (Fig. 5CGo) and TM4 (Fig. 5DGo) cells. In contrast, TβRII-deficient AC (Fig. 5CGo) and LβT2 cells (Fig. 5DGo) yielded essentially biphasic plots, suggesting the high-affinity binding sites for TGF-β were scarce in each but displayed substantial binding to the intermediate- and, in the case of AC cells, low-affinity sites.

Inhibin A (10 nM) minimally competed for [125I]TGF-β1 or [125I]TGF-β2 affinity labeling of membrane proteins on TM3 and TM4 cells (Fig. 6Go). At the same concentration, TGF-β1 abolished the labeling of all protein species other than those incorporating betaglycan (>200 and 140 kDa complexes), and labeling of those complexes was greatly diminished (Fig. 6Go). Competing unlabeled TGF-β2 (10 nM) was partly effective at blocking the affinity labeling of TM3 and TM4 cell proteins by [125I]TGF-β2 but preferentially competed for labeling of the 68- to 75-kDa complexes when the tracer was [125I]TGF-β1 (Fig. 6Go).

In summary, inhibins at or less than 10 nM poorly blocked specific TGF-β binding to inhibin target cell types, but TGF-β isoforms in the same physiological concentration range competed for up to 62% of specific inhibin A binding to the same cell types by blocking access to all the binding proteins. TGF-β2 was the more potent competitor for inhibin binding and for TGF-β1 binding to AC cells, whereas TGF-β1 was the more potent competitor for TGF-β binding in all other cases.

Effects of inhibin and TGF-β on expression of betaglycan by cell lines
Because TGF-β is known to up- or down-regulate betaglycan expression in nonreproductive tissues (19, 20, 21), the effects of inhibin and TGF-β on the expression of their common coreceptor were examined in the TβRII-expressing gonadal and TβRII-deficient AC and LβT2 cell lines. The levels of betaglycan mRNA were reduced by only between 17 and 18% of control in TM3 and TM4 cells, respectively (P > 0.05) and changed little in AC and LβT2 cells after overnight treatment with inhibin A (1 nM) (Fig. 7Go). In contrast, both TGF-β1 and TGF-β2 significantly decreased the level of betaglycan mRNA (Fig. 7Go) in a concentration-dependent manner (data not shown) by as much as 74% of control in TM3 cells (IC50 values between 10 and 50 pM) and 49% of control in TM4 cells (IC50 values around 130 pM) (differs from 0%, P < 0.01 in each case). However, neither TGF-β isoform significantly affected betaglycan mRNA levels in AC or LβT2 cells (Fig. 7Go). At 200 pM, TGF-β1 decreased the level of betaglycan mRNA by an average 59% of control in TM3 cells (data not shown).


Figure 7
View larger version (23K):
[in this window]
[in a new window]

  		FIG. 7. Regulation of betaglycan mRNA levels by TGF-β and inhibin isoforms. Semiconfluent monolayers of TM3, TM4, AC, and LβT2 cells were incubated for 24 h with the indicated TGF-β superfamily members (each 1 nM), and then total RNA was extracted, and betaglycan mRNA content was estimated by real-time RT-PCR with correction for GAPDH mRNA content. Data represent the average result ± SEM obtained after normalization to the matching control (equal to 1, indicated by the broken horizontal line) from replicate independent experiments (n = 4–9, except in the case of LβT2, where only the experiments with inhibin A were replicated). a, Significantly different from control level, P < 0.01.

 
TGF-β pretreatments suppress subsequent inhibin binding to TM3 and TM4 cells
We next determined whether the suppression of betaglycan mRNA expression by pretreatment of cells for 18 h with TGF-β has consequences for their binding of inhibin during the following 4 h (after the removal of TGF-β). Pretreatment of the TβRII-deficient AC and LβT2 cells with graded concentrations of TGF-β1 progressively decreased their subsequent binding of [125I]inhibin A by maxima of only 28 and 26%, respectively, of the total binding in untreated cultures (cf. 49 and 41%, respectively, when applied as a direct competitor; compare data in Tables 4Go and 2Go), with between 3- and 4-fold lower apparent IC50 values (e.g. Fig. 8AGo and Table 4Go). This is taken to reflect the extent of inhibition arising from the residual occupation of inhibin binding sites on the cells by TGF-β1 applied during the pretreatment period. For comparison, the competition curve obtained with TGF-β1 in the same binding experiment is shown in Fig. 8Go as open symbols and a dotted line. In contrast, TGF-β1 pretreatment suppressed subsequent [125I]inhibin A binding in the TβRII-expressing TM4 cells by 41% (Fig. 8BGo and Table 4Go), similar to the level of suppression when acting as a direct competitor, but with a 9-fold lower IC50 (0.092 nM; Table 4Go; significantly different from the competition IC50, P < 0.001). Furthermore, TGF-β1 pretreatment maximally suppressed subsequent [125I]inhibin A binding to TM3 cells by significantly more (64%, cf. 50% when applied as a direct competitor; P < 0.001), with 17-fold greater potency (IC50 of 0.054 nM; P < 0.001; Fig. 8CGo and Table 4Go). Qualitatively similar patterns were observed when cells were pretreated with TGF-β2 (data not shown). The difference in TGF-β2 potency between pretreatment (IC50 of 0.14 ± 0.04 nM, Table 4Go) and competition effects (IC50 of 0.66 ± 0.10 nM, Table 2Go) was statistically significant in the case of TM4 cells (P < 0.01), and the difference in the maximal effect was statistically significant in the case of TM3 cells [pretreatment brought about 65 ± 2% suppression of binding (Table 4Go) compared with 46 ± 3% suppression through competition (Table 2Go), P < 0.01]. Another difference was the greater potency of TGF-β1 than TGF-β2 pretreatment at suppressing subsequent [125I]inhibin A binding to TM3 and TM4 cells (Table 4Go; statistically significant in TM3 cells, P < 0.001). This was the reverse of the pattern seen when the TGF-β isoforms were directly competing for binding (see Table 2Go), although the latter differences were not statistically significant (P > 0.05).


View this table:
[in this window]
[in a new window]

  		TABLE 4. Effects of pretreating TM3, TM4, AC, and LβT2 cells for 18 h with inhibin and TGF-β isoforms on their subsequent binding of [125I]inhibin A

 

Figure 8
View larger version (10K):
[in this window]
[in a new window]

  		FIG. 8. Comparison of the direct effects of TGF-β1 as a competitor with its indirect effects as a pretreatment on the binding of [125I]inhibin A. Semiconfluent monolayers of AC (A), TM4 (B), and TM3 (C) cells were either untreated (open symbols, broken lines) and then incubated for 4 h at room temperature with [125I]inhibin A in the absence or presence of graded concentrations of TGF-β1 as direct competitor for binding sites, or treated (closed symbols, solid lines) for 18 h with graded concentrations of TGF-β1 and then washed and incubated for 4 h at room temperature with [125I]inhibin A in the absence of added TGF-β. Other details are the same as for Fig. 3Go. Aggregate data obtained from multiple replicate pretreatment experiments are presented in Table 4Go.

 
Pretreatment of TM3 and TM4 cells for 18 h with inhibin A suppressed subsequent binding of [125I]inhibin A by 61 and 48% of total binding, respectively (Table 4Go). However, the average extent and potency of inhibition (IC50 values of 0.16 and 0.31 nM, respectively; Table 4Go) was in each case less than that observed when inhibin A was directly competing for binding of [125I]inhibin A (see Table 2Go), consistent with residual occupation of binding sites by the pretreatment. Inhibin B also caused substantially less suppression of inhibin A binding when used as a pretreatment than when acting as a direct competitor but did so with lower estimated IC50 values (compare Tables 2Go and 4Go), and a similar pattern was observed in AC cells after treatments with either inhibin A or B (Tables 2Go and 4Go).

Suppression of inhibin A binding reflects decreased interaction with betaglycan
In comparison with control TM3 cell cultures, those pretreated for 18 h with TGF-β1 or TGF-β2 (0.4 nM) showed reductions in the subsequent affinity labeling of all protein complexes by [125I]inhibin A, whereas direct competition for the binding of [125I]inhibin A by the same concentration of each TGF-β isoform was minimal (Fig. 9AGo). A similar pattern was evident in TM4 cells pretreated with 0.4 nM TGF-β1 or TGF-β2 (Fig. 9BGo), although the extent of the differences in affinity labeling was not as great as in the TM3 cells. In contrast, neither pretreatment nor direct competition with TGF-β isoforms (each 0.4 nM) affected [125I]inhibin A affinity labeling of proteins on the surfaces of AC (Fig. 9CGo) or LβT2 cells (data not shown).

Pretreatment of TM3 and TM4 cells with TGF-β isoforms not only decreased the [125I]inhibin A affinity labeling of all complexes in the total cell lysate but also greatly reduced the amounts of betaglycan species (142 and >200 kDa) that could be immunoprecipitated with betaglycan antiserum (Fig. 10Go).

In summary, pretreatment of TM3 and TM4 cells with TGF-β isoforms at or less than 1 nM suppressed subsequent inhibin binding to multiple inhibin binding proteins, and betaglycan species in particular, but AC and LβT2 cells showed little response to pretreatments with either TGF-β isoform in the same concentration range.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
The present studies were undertaken to examine whether interactions occur between inhibin and TGF-β in several inhibin-responsive cell lines chosen to model Leydig, Sertoli, and adrenocortical cells as autocrine/paracrine targets, and pituitary gonadotropes as the classical endocrine target of inhibin action. The results show that TGF-β isoforms unilaterally reduce inhibin A binding to its target cells through two distinct mechanisms, both involving the inhibin/TGF-β coreceptor betaglycan. First, in cells that express TβRII, low concentrations of TGF-β suppress betaglycan expression, with concomitant reductions in the affinity labeling of all the binding proteins, including betaglycan, by radiolabeled inhibin A. Second, in cells that express betaglycan, whether or not they can transduce TGF-β signals via TβRII, TGF-β isoforms compete directly for inhibin A binding. The latter interactions require TGF-β1 concentrations an order of magnitude higher than those that suppress betaglycan expression. TGF-β2 is relatively more potent than TGF-β1 as a competitor and relatively less potent as a suppressor of betaglycan expression, so the concentration differential for the dual effects of TGF-β2 is lower. (The extent to which TGF-β modifies the binding of inhibin B is yet to be explored.) In contrast, inhibins negligibly modify betaglycan expression in the absence of added agonists (activin and BMP) and compete poorly for binding of TGF-β1 or TGF-β2. The results therefore suggest that under conditions in which endogenous betaglycan expression is subject to physiological regulation by TGF-β, the action of inhibin A will be modified in a cell context-dependent manner by TGF-β, but not vice versa.

The results confirm that the inhibin-binding TM3, TM4, AC, and LβT2 cell lines express the inhibin/TGF-β coreceptor betaglycan (mRNA and protein) (12, 14, 16, 17), as do their normal counterparts (Leydig, Sertoli, adrenocortical, and gonadotrope cells, respectively) (29, 30, 31, 32). These cell lines were also shown to express the inhibin/activin/BMP binding and signaling proteins ActRII, ActRIIB, and BMPRII, most or all of the ALKs, and Smads 2 and 3, in some cases confirming the findings of others (14, 16, 33, 34). It is notable that AC and LβT2 cells strongly express ALK5 in the absence of TβRII but in the presence of BMPRII, providing the basis for responsiveness to GDF-9, but not TGF-β (35). TM3 cells also express the alternate, Smad1/5/8-signaling TGF-β receptor ALK1, for which no function in Leydig cells has yet been identified.

Betaglycan is ideally placed to mediate interaction between the inhibin and TGF-β families of hormones. The first point at which interaction occurs, competition for binding at the cell surface, has been examined by others, mostly through the constitutive expression of transfected betaglycan in the absence or presence of ActRII, ActRIIB, or BMPRII (9, 10). TGF-β competes with inhibin for binding to the more membrane-proximal of two binding sites for TGF-β in the extracellular domain of betaglycan (9, 16, 18). In the absence of type II receptors, the relative affinities of TGF-β isoforms and inhibin A when competing for [125I]inhibin A binding to betaglycan decrease in the rank order TGF-β2 is greater than inhibin A is greater than TGF-β1 (18). Individual amino acid residues of betaglycan contribute differentially to the binding of each ligand because specific mutated forms of betaglycan selectively bind only TGF-β or inhibin (36). The presence of type I and II receptor proteins alongside betaglycan complicates the ligand binding kinetics. For instance, in the presence of TGF-β receptors, TGF-β occupation of both binding sites on betaglycan promotes the formation of higher-order complexes whereby one betaglycan molecule couples with four type I and four type II receptors, with differences in affinities for superficially similar interactions (37). On the other hand, inhibin affinity for betaglycan is increased 3-fold by the presence of ActRII/IIB but is decreased 1.7-fold by the presence of BMPRII (9, 36). Inhibin A binding to cells that constitutively overexpress both betaglycan and ActRII (9) and to ovine pituitary cells in primary culture (13) was originally reported to resist competition from TGF-β. The present studies employed cell lines that endogenously express the relevant binding proteins at regulated levels, in different combinations and ratios (Table 1Go), to characterize and compare the competition between inhibin and TGF-β for [125I]inhibin A binding. The results show a uniform pattern of competition in TM3, TM4, and AC cells whereby TGF-β isoforms compete for more than half of the specific binding sites for inhibin A, with potencies decreasing in the rank order inhibin A is more than TGF-β2 is more than or equal to TGF-β1, whereas TGF-β2 is the most potent competitor in LβT2 cells. Nanomolar concentrations of TGF-β are required for substantial competition to occur. Nevertheless, such competition has functional consequences. We previously found that the actions of inhibin A on LβT2 gonadotropes to block activin-responsive promoter-driven reporters (ovine FSH-β subunit and GnRH receptor) are relieved by nanomolar concentrations of TGF-β, at least partly through blockade of inhibin access to betaglycan (14). The present results suggest that this form of antagonism of inhibin by TGF-β could occur more generally in various cell types that are targets for local inhibin actions.

Inhibin A and B competition for binding of [125I]TGF-β1 and -β2 to TM3, TM4, AC, and LβT2 cell lines was minor. This agrees with earlier findings for other cell types that similarly express betaglycan and provides little evidence for the involvement of more promiscuous binding proteins in the observed competition (e.g. a distinct 70- to 74-kDa protein species on GH3 rat pituitary somatotropes that competitively binds inhibin, activin, and TGF-β) (38).

A critical finding of the present work is that suppression of betaglycan expression by TGF-β isoforms potently inhibits inhibin interaction with some of its target cells. The intact signaling pathway for TGF-β is exquisitely sensitive to activation by the ligands (TGF-β1 EC50 values between 5 and 35 pM) (39). The TGF-β1 pretreatment IC50 values for suppressing inhibin A binding to TM3 and TM4 cells (54 and 92 pM, respectively) agree with the Kd values estimated for high-affinity TGF-β1 binding sites on TM3 and TM4 cells (36 and 86 pM, respectively) and are consistent with actions mediated by this pathway. Previous studies of the interaction between inhibin and TGF-β primarily employed expression vectors that constitutively overexpress inhibin binding proteins, most notably betaglycan, making it impossible to detect interactions that reflect the physiological regulation of betaglycan via its native promoter. However, earlier studies had revealed that TGF-β regulates the expression of critical components in its signaling pathway, including betaglycan (19, 20, 21). Consistent with those findings, the cloned human and rat betaglycan gene promoters include consensus sequences for Smad3 and Smad4 binding (40, 41), and the inhibitory function of one of these has been confirmed through coupling to a reporter (20). TGF-β down-regulates betaglycan expression in several cell types (19, 20, 42) but up-regulates expression in others (21). The present study extends the range of cell types in which TGF-β isoforms suppress endogenous betaglycan expression to include Leydig-like TM3 and Sertoli-like TM4 cells but not AC and LβT2 cells, reflecting their respective levels of endogenous TβRII expression.

Pretreatment of TM3 cells with TGF-β suppresses betaglycan mRNA levels and subsequent inhibin A binding to similar extents, and the same is true for TM4 cells. However, residual occupation of receptors and binding proteins by slowly dissociating bound ligand potentially complicates the interpretation of [125I]inhibin binding data involving pretreatment of cells with TGF-β superfamily members. Attempts to strip residual TGF-β (and inhibin) from their binding proteins using dilute acid (43) severely limited the subsequent binding of [125I]inhibin A (Farnworth, P., unpublished observations), so this procedure was abandoned. Nevertheless, the results obtained after treatment of TM3, TM4, and AC cells with inhibin and of AC cells with TGF-β, treatments that have minor or no effects on betaglycan expression in these cells, demonstrate that the net residual effects of bound ligand are generally minor when compared with those observed in TβRII-expressing cells. A second issue is that pretreatment of cells with TGF-β can result in disappearance of betaglycan, TβRII, and ALK5 from the cell surface through rapid internalization of occupied receptor complexes (44). Ligand-promoted betaglycan internalization was not examined in the present study but may have contributed to the major reductions in inhibin binding to membrane proteins, and affinity labeling of betaglycan, obtained after TGF-β pretreatment of TM3 and TM4 cells.

The significance of our observations is that suppression of betaglycan expression by TGF-β in responsive cell types could allow low concentrations of TGF-β to rescue these cells from the antagonism of other TGF-β superfamily ligand actions by inhibin. Attempts to demonstrate this consequence using activin-responsive promoter-luciferase reporter constructs have not been successful, perhaps reflecting the influences of other inhibitory pathways (e.g. induction of inhibitory Smads 6 and 7) that are almost certainly activated during pretreatment of TM3 and TM4 cells with TGF-β in our static bioassay. Whether a transient phase of TGF-β action can result in the blockade of inhibin action and facilitation of activin action in a more dynamic bioassay, and for how long, remains to be determined. However, inhibin A is capable of sequestering not only activin type II receptors, ActRII/IIB, but also BMP type II receptor, BMPRII (9, 10). This theoretically should allow inhibin to block the actions of many ligands other than activin (2). Therefore, an alternative consequence for the blockade of inhibin binding through TGF-β suppression of betaglycan is the facilitation of Smad1/5/8-mediated BMP/GDF action rather than Smad2/3-mediated activin/nodal action.

The fact that TGF-β isoforms inhibit the affinity labeling of betaglycan by [125I]inhibin A was expected. Less predictably, TGF-β also disrupts the affinity labeling of other naturally occurring binding proteins by [125I]inhibin A. Although these other binding proteins have not been identified, most are not (co-)immunoprecipitated by betaglycan antiserum, so presumably are neither related to betaglycan nor closely associated with it in the cell membrane. The pattern suggests a breakdown of the cooperativity that betaglycan confers on several ligand binding processes. When the level of betaglycan expression is increased in cells, total binding of TGF-β to its type I and II receptors substantially increases, binding of inhibin to ActRII/ActRIIB is enhanced, and interaction of inhibin with BMPRII becomes possible (7, 9, 10). TGF-β, whether through suppression of betaglycan expression or direct occupation of the common binding site on betaglycan, decreases the number of betaglycan molecules available to inhibin, which would undo or limit the cooperative processes, thereby disrupting the interaction of [125I]inhibin A with its multiple binding partners in concert.

Our results with the LβT2 and AC cell lines suggest that normal pituitary gonadotropes and fetal adrenocortical cells might lack the TβRII-mediated mechanism for potent TGF-β suppression of inhibin binding in vivo. The finding of minimal TβRII mRNA expression by LβT2 gonadotropes confirms previous findings (24, 45), and a human adrenocortical carcinoma cell line (SW-13), like the AC cell line, lacks TβRII (46). However, TβRII expression by normal rat pituitary gonadotropes has been found by some (45), but not others (25), and FSH-secreting pituitary tumors mostly express TβRII (47, 48). In addition, although TβRII expression in the normal adrenal cortex has not been reported, the NCI-H295 adrenocortical cell line expresses TβRII mRNA (49), and normal adrenocortical cells respond to TGF-β stimulation in vitro (50, 51). With respect to TβRII expression, TM3 and TM4 cells mimic not only their normal counterparts but also Sertoli cell lines and testicular tumors (31, 32, 52, 53). Therefore, lack of the TβRII-mediated mechanism for suppression of inhibin binding is probably uncommon. Nevertheless, neoplastic transformation of many cell types is accompanied by a loss of TβRII or betaglycan expression (42, 54). It remains to be determined whether altered responsiveness to inhibin contributes to the phenotype in such circumstances.

In summary, our results indicate that inhibin effects on its endocrine and paracrine/autocrine target cells under physiological conditions are subject to multiple levels of regulation, including transcriptional control of betaglycan expression, competitive ligand binding of this coreceptor, and functionality of specific signaling pathways. These different regulatory processes are independent but form a complex interacting system of control, because different subsets of them are operative in different cell types. For example, in cells that lack TGF-β signaling (such as adrenocortical carcinoma cells and gonadotropes, due to the absence of TβRII), inhibin actions are affected only by concentrations of TGF-β that compete for betaglycan binding. In contrast, in cells with a functional TGF-β signaling pathway (such as Leydig and Sertoli cell types), low levels of TGF-β reduce betaglycan expression, thereby decreasing the level of receptor occupancy by inhibin, which leads to decreased inhibin interactions with the activin/BMP/GDF receptor systems. In stark contrast, inhibin poorly competes for TGF-β binding and barely influences betaglycan expression in the studied cell lines. From these results, we conclude that the net action of inhibin under physiological conditions will vary with cell context, because it is determined by the prevailing local levels of TGF-β through dual mechanisms of antagonism involving regulation of betaglycan availability in TGF-β-responsive cell types and competition for binding to betaglycan on the cell surface.


   Acknowledgments
 
We thank Assoc. Prof. David Robertson and colleagues for providing purified inhibin A and B; Prof. Ilpo Huhtaniemi (Imperial College, London, United Kingdom), Dr. Nafis Rahman (University of Turku, Turku, Finland), Dr. Pam Mellon (University of California, San Diego, CA), and Dr. Kate Loveland (Monash Institute of Medical Research, Clayton, Victoria, Australia) for kindly providing the cell lines; and Dr. Ann Drummond for contributions to the receptor RT-PCR studies.


   Footnotes
 
This study was funded by Program Grants (Reg Key Nos. 983212 and 241000) and a Fellowship for J.K.F. (Reg Key No. 198705) from the National Health and Medical Research Council of Australia.

Disclosure Statement: The authors have no potential conflicts of interest to declare.

First Published Online July 26, 2007

Abbreviations: ActRII, Activin receptor type II; ALK, activin receptor-like kinase; AS, artificial serum; BMP, bone morphogenetic protein; BMPRII, BMP receptor type II; FBS, fetal bovine serum; GDF, growth/differentiation factor; TβRII, TGF-β type II receptor.

Received February 2, 2007.

Accepted for publication July 19, 2007.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Welt C, Sidis Y, Keutmann H, Schneyer A 2002 Activins, inhibins, and follistatins: from endocrinology to signaling. A paradigm for the new millennium. Exp Biol Med 227:724–752[Abstract/Free Full Text]
  2. Mazerbourg S, Sangkuhl K, Luo CW, Sudo S, Klein C, Hsueh AJ 2005 Identification of receptors and signaling pathways for orphan bone morphogenetic protein/growth differentiation factor ligands based on genomic analyses. J Biol Chem 280:32122–32132[Abstract/Free Full Text]
  3. Herpin A, Lelong C, Favrel P 2004 Transforming growth factor-β-related proteins: an ancestral and widespread superfamily of cytokines in metazoans. Dev Comp Immunol 28:461–485[CrossRef][Medline]
  4. Chang H, Brown CW, Matzuk MM 2002 Genetic analysis of the mammalian transforming growth factor-β superfamily. Endocr Rev 23:787–823[Abstract/Free Full Text]
  5. Hsueh AJ, Dahl KD, Vaughan J, Tucker E, Rivier J, Bardin CW, Vale W 1987 Heterodimers and homodimers of inhibin subunits have different paracrine action in the modulation of luteinizing hormone-stimulated androgen biosynthesis. Proc Natl Acad Sci USA 84:5082–5086[Abstract/Free Full Text]
  6. Farnworth PG, Stanton PG, Wang Y, Escalona R, Findlay JK, Ooi GT 2006 Inhibins differentially antagonize activin and bone morphogenetic protein action in a mouse adrenocortical cell line. Endocrinology 147:3462–3471[Abstract/Free Full Text]
  7. Lopez-Casillas F, Wrana JL, Massague J 1993 Betaglycan presents ligand to the TGF β signaling receptor. Cell 73:1435–1444[CrossRef][Medline]
  8. del Re E, Babitt JL, Pirani A, Schneyer AL, Lin HY 2004 In the absence of type III receptor, the transforming growth factor (TGF)-β type II-B receptor requires the type I receptor to bind TGF-β2. J Biol Chem 279:22765–22772[Abstract/Free Full Text]
  9. Lewis KA, Gray PC, Blount AL, MacConell LA, Wiater E, Bilezikjian LM, Vale W 2000 Betaglycan binds inhibin and can mediate functional antagonism of activin signalling. Nature 404:411–414[CrossRef][Medline]
  10. Wiater E, Vale W 2003 Inhibin is an antagonist of bone morphogenetic protein signaling. J Biol Chem 278:7934–7941[Abstract/Free Full Text]
  11. del Re E, Sidis Y, Fabrizio DA, Lin HY, Schneyer A 2004 Reconstitution and analysis of soluble inhibin and activin receptor complexes in a cell-free system. J Biol Chem 279:53126–53135[Abstract/Free Full Text]
  12. Farnworth PG, Harrison CA, Leembruggen P, Chan KL, Stanton PG, Ooi GT, Rahman NA, Huhtaniemi IT, Findlay JK, Robertson DM 2001 Inhibin binding sites and proteins in pituitary, gonadal, adrenal and bone cells. Mol Cell Endocrinol 180:63–71[CrossRef][Medline]
  13. Hertan R, Farnworth PG, Fitzsimmons KL, Robertson DM 1999 Identification of high affinity binding sites for inhibin on ovine pituitary cells in culture. Endocrinology 140:6–12[Abstract/Free Full Text]
  14. Ethier JF, Farnworth PG, Findlay JK, Ooi GT 2002 Transforming growth factor-β modulates inhibin A bioactivity in the LβT2 gonadotrope cell line by competing for binding to betaglycan. Mol Endocrinol 16:2754–2763[Abstract/Free Full Text]
  15. Farnworth PG, Robertson DM, de Kretser DM, Burger HG 1988 Effects of 31 kilodalton bovine inhibin on follicle-stimulating hormone and luteinizing hormone in rat pituitary cells in vitro: actions under basal conditions. Endocrinology 122:207–213[Abstract/Free Full Text]
  16. Harrison CA, Farnworth PG, Chan KL, Stanton PG, Ooi GT, Findlay JK, Robertson DM 2001 Identification of specific inhibin A-binding proteins on mouse Leydig (TM3) and Sertoli (TM4) cell lines. Endocrinology 142:1393–1402[Abstract/Free Full Text]
  17. Farnworth PG, Wang Y, Leembruggen P, Ooi GT, Harrison C, Robertson DM, Findlay JK 2006 Rodent adrenocortical cells display high affinity binding sites and proteins for inhibin A, and express components required for autocrine signalling by activins and bone morphogenetic proteins. J Endocrinol 188:451–465[Abstract/Free Full Text]
  18. Esparza-Lopez J, Montiel JL, Vilchis-Landeros MM, Okadome T, Miyazono K, Lopez-Casillas F 2001 Ligand binding and functional properties of betaglycan, a co-receptor of the transforming growth factor-β superfamily. Specialized binding regions for transforming growth factor-β and inhibin A. J Biol Chem 276:14588–14596[Abstract/Free Full Text]
  19. Norgaard P, Spang-Thomsen M, Poulsen HS 1996 Expression and autoregulation of transforming growth factor β receptor mRNA in small-cell lung cancer cell lines. Br J Cancer 73:1037–1043[Medline]
  20. Lopez-Casillas F, Riquelme C, Perez-Kato Y, Ponce-Castaneda MV, Osses N, Esparza-Lopez J, Gonzalez-Nunez G, Cabello-Verrugio C, Mendoza V, Troncoso V, Brandan E 2003 Betaglycan expression is transcriptionally up-regulated during skeletal muscle differentiation. Cloning of murine betaglycan gene promoter and its modulation by MyoD, retinoic acid, and transforming growth factor-β. J Biol Chem 278:382–390[Abstract/Free Full Text]
  21. Worapamorn W, Haase HR, Li H, Bartold PM 2001 Growth factors and cytokines modulate gene expression of cell-surface proteoglycans in human periodontal ligament cells. J Cell Physiol 186:448–456[CrossRef][Medline]
  22. Rahman NA, Kiiveri S, Siltanen S, Levallet J, Kero J, Lensu T, Wilson DB, Heikinheimo MT, Huhtaniemi IT 2001 Adrenocortical tumorigenesis in transgenic mice: the role of luteinizing hormone receptor and transcription factors GATA-4 and GATA-6. Reprod Biol 1:5–9[Medline]
  23. Bailey JS, Rave-Harel N, McGillivray SM, Coss D, Mellon PL 2004 Activin regulation of the follicle-stimulating hormone β-subunit gene involves Smads and the TALE homeodomain proteins Pbx1 and Prep1. Mol Endocrinol 18:1158–1170[Abstract/Free Full Text]
  24. Makanji Y, Harrison CA, Stanton PG, Krishna R, Robertson DM 2007 Inhibin A and B in vitro bioactivities are modified by their degree of glycosylation and their affinities to betaglycan. Endocrinology 148:2309–2316[Abstract/Free Full Text]
  25. Suszko MI, Woodruff TK 2006 Cell-specificity of transforming growth factor-β response is dictated by receptor bioavailability. J Mol Endocrinol 36:591–600[Abstract/Free Full Text]
  26. Drummond AE, Dyson M, Thean E, Groome NP, Robertson DM, Findlay JK 2000 Temporal and hormonal regulation of inhibin protein and subunit mRNA expression by post-natal and immature rat ovaries. J Endocrinol 166:339–354[Abstract]
  27. de Winter JP, ten Dijke P, de Vries CJ, van Achterberg TA, Sugino H, de Waele P, Huylebroeck D, Verschueren K, van den Eijnden-van Raaij AJ 1996 Follistatins neutralize activin bioactivity by inhibition of activin binding to its type II receptors. Mol Cell Endocrinol 116:105–114[CrossRef][Medline]
  28. Buzzard JJ, Farnworth PG, De Kretser DM, O’Connor AE, Wreford NG, Morrison JR 2003 Proliferative phase Sertoli cells display a developmentally regulated response to activin in vitro. Endocrinology 144:474–483[Abstract/Free Full Text]
  29. Vanttinen T, Kuulasmaa T, Liu J, Voutilainen R 2002 Expression of activin/inhibin receptor and binding protein genes and regulation of activin/inhibin peptide secretion in human adrenocortical cells. J Clin Endocrinol Metab 87:4257–4263[Abstract/Free Full Text]
  30. MacConell LA, Leal AM, Vale WW 2002 The distribution of betaglycan protein and mRNA in rat brain, pituitary, and gonads: implications for a role for betaglycan in inhibin-mediated reproductive functions. Endocrinology 143:1066–1075[Abstract/Free Full Text]
  31. Goddard I, Bouras M, Keramidas M, Hendrick JC, Feige JJ, Benahmed M 2000 Transforming growth factor-β receptor types I and II in cultured porcine Leydig cells: expression and hormonal regulation. Endocrinology 141:2068–2074[Abstract/Free Full Text]
  32. Konrad L, Luers GH, Volck-Badouin E, Keilani MM, Laible L, Aumuller G, Hofmann R 2006 Analysis of the mRNA expression of the TGF-β family in testicular cells and localization of the splice variant TGF-β2B in testis. Mol Reprod Dev 73:1211–1220[CrossRef][Medline]
  33. Ying C, Zhang Z, Ying SY 1995 Expression and localization of activin βA-subunit and activin receptors in TM3, a mouse Leydig cell line. Endocr Res 21:815–824[Medline]
  34. Suszko MI, Lo DJ, Suh H, Camper SA, Woodruff TK 2003 Regulation of the rat follicle-stimulating hormone β-subunit promoter by activin. Mol Endocrinol 17:318–332[Abstract/Free Full Text]
  35. Mazerbourg S, Klein C, Roh J, Kaivo-Oja N, Mottershead DG, Korchynskyi O, Ritvos O, Hsueh AJ 2004 Growth differentiation factor-9 signaling is mediated by the type I receptor, activin receptor-like kinase 5. Mol Endocrinol 18:653–665[Abstract/Free Full Text]
  36. Wiater E, Harrison CA, Lewis KA, Gray PC, Vale WW 2006 Identification of distinct inhibin and transforming growth factor β-binding sites on betaglycan: functional separation of betaglycan co-receptor actions. J Biol Chem 281:17011–17022[Abstract/Free Full Text]
  37. De Crescenzo G, Grothe S, Zwaagstra J, Tsang M, O’Connor-McCourt MD 2001 Real-time monitoring of the interactions of transforming growth factor-β (TGF-β) isoforms with latency-associated protein and the ectodomains of the TGF-β type II and III receptors reveals different kinetic models and stoichiometries of binding. J Biol Chem 276:29632–29643[Abstract/Free Full Text]
  38. Cheifetz S, Ling N, Guillemin R, Massague J 1988 A surface component on GH3 pituitary cells that recognizes transforming growth factor-β, activin, and inhibin. J Biol Chem 263:17225–17228[Abstract/Free Full Text]
  39. Cheifetz S, Bassols A, Stanley K, Ohta M, Greenberger J, Massague J 1988 Heterodimeric transforming growth factor β. Biological properties and interaction with three types of cell surface receptors. J Biol Chem 263:10783–10789[Abstract/Free Full Text]
  40. Wickert L, Abiaka M, Bolkenius U, Gressner AM 2004 Corticosteroids stimulate selectively transforming growth factor (TGF)-β receptor type III expression in transdifferentiating hepatic stellate cells. J Hepatol 40:69–76[Medline]
  41. Ji C, Chen Y, McCarthy TL, Centrella M 1999 Cloning the promoter for transforming growth factor-β type III receptor. Basal and conditional expression in fetal rat osteoblasts. J Biol Chem 274:30487–30494[Abstract/Free Full Text]
  42. Dong M, How T, Kirkbride KC, Gordon KJ, Lee JD, Hempel N, Kelly P, Moeller BJ, Marks JR, Blobe GC 2007 The type III TGF-β receptor suppresses breast cancer progression. J Clin Invest 117:206–217[CrossRef][Medline]
  43. Massague J, Kelly B 1986 Internalization of transforming growth factor-β and its receptor in BALB/c 3T3 fibroblasts. J Cell Physiol 128:216–222[CrossRef][Medline]
  44. Zwaagstra JC, Kassam Z, O’Connor-McCourt MD 1999 Down-regulation of transforming growth factor-β receptors: cooperativity between the types I, II, and III receptors and modulation at the cell surface. Exp Cell Res 252:352–362[CrossRef][Medline]
  45. Gore AJ, Philips DP, Miller WL, Bernard DJ 2005 Differential regulation of follicle stimulating hormone by activin A and TGFβ1 in murine gonadotropes. Reprod Biol Endocrinol 3:73[CrossRef][Medline]
  46. Yamamoto N, Imai J, Watanabe M, Hiroi N, Sugano S, Yoshino G 2006 Restoration of transforming growth factor-β type II receptor reduces tumorigenicity in the human adrenocortical carcinoma SW-13 cell line. Horm Metab Res 38:159–166[CrossRef][Medline]
  47. Fujiwara K, Ikeda H, Yoshimoto T 1995 Immunohistochemical demonstration of TGF-β-receptor type II in human pituitary adenomas. Acta Histochem 97:445–454[Medline]
  48. Alexander JM, Bikkal HA, Zervas NT, Laws Jr ER, Klibanski A 1996 Tumor-specific expression and alternate splicing of messenger ribonucleic acid encoding activin/transforming growth factor-β receptors in human pituitary adenomas. J Clin Endocrinol Metab 81:783–790[Abstract]
  49. Zatelli MC, Rossi R, del Senno L, degli Uberti EC 1998 Role of transforming growth factor β1 (TGF β1) in mediating androgen-induced growth inhibition in human adrenal cortex in vitro. Steroids 63:243–245[CrossRef][Medline]
  50. Stankovic AK, Parker Jr CR 1995 Receptor binding of transforming growth factor-β by human fetal adrenal cells. Mol Cell Endocrinol 109:159–165[CrossRef][Medline]
  51. Mesiano S, Jaffe RB 1997 Role of growth factors in the developmental regulation of the human fetal adrenal cortex. Steroids 62:62–72[CrossRef][Medline]
  52. Olaso R, Pairault C, Habert R 1998 Expression of type I and II receptors for transforming growth factor β in the adult rat testis. Histochem Cell Biol 110:613–618[CrossRef][Medline]
  53. Cardillo MR, Petrangeli E, Ravenna L, Salvatori L, Di Silverio F 1998 Transforming growth factor-β expression in human testicular neoplasms. Anal Quant Cytol Histol 20:461–469[Medline]
  54. Markowitz S, Wang J, Myeroff L, Parsons R, Sun L, Lutterbaugh J, Fan RS, Zborowska E, Kinzler KW, Vogelstein B, Brattain MG, Willson JKV 1995 Inactivation of the type II TGF-β receptor in colon cancer cells with microsatellite instability. Science 268:1336–1338[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 J EndocrinolHome page
Y. Wang, P. K Nicholls, P. G Stanton, C. A Harrison, M. Sarraj, R. B Gilchrist, J. K Findlay, and P. G Farnworth
Extra-ovarian expression and activity of growth differentiation factor 9
J. Endocrinol., September 1, 2009; 202(3): 419 - 430.
[Abstract] [Full Text] [PDF]

Home page
 Mol. Endocrinol.Home page
E. Wiater, K. A. Lewis, C. Donaldson, J. Vaughan, L. Bilezikjian, and W. Vale
Endogenous Betaglycan Is Essential for High-Potency Inhibin Antagonism in Gonadotropes
Mol. Endocrinol., July 1, 2009; 23(7): 1033 - 1042.
[Abstract] [Full Text] [PDF]

Home page
 Mol. Endocrinol.Home page
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]

Home page
 ReproductionHome page
G. A Tarulli, S. J Meachem, S. Schlatt, and P. G Stanton
Regulation of testicular tight junctions by gonadotrophins in the adult Djungarian hamster in vivo
Reproduction, June 1, 2008; 135(6): 867 - 877.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Farnworth, P. G.
Right arrow Articles by Findlay, J. K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Farnworth, P. G.
Right arrow Articles by Findlay, J. K.

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