help button home button Endocrine Society Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology, doi:10.1210/en.2006-1612
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
148/5/2309    most recent
Author Manuscript (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
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 Makanji, Y.
Right arrow Articles by Robertson, D. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Makanji, Y.
Right arrow Articles by Robertson, D. M.
Endocrinology Vol. 148, No. 5 2309-2316
Copyright © 2007 by The Endocrine Society

Inhibin A and B in Vitro Bioactivities Are Modified by Their Degree of Glycosylation and Their Affinities to Betaglycan

Yogeshwar Makanji, Craig A. Harrison, Peter G. Stanton, Radha Krishna and David M. Robertson

Prince Henry’s Institute of Medical Research (Y.M., C.A.H., P.G.S., D.M.R.) and Department of Obstetrics and Gynaecology (Y.M.), Monash University, Clayton, Victoria 3168, Australia; and DSL-Beckman Coulter (R.K.), Webster, Texas 77598

Address all correspondence and requests for reprints to: David M. Robertson, Ph.D., Prince Henry’s Institute of Medical Research, PO Box 5152, Clayton, Victoria 3168, Australia. E-mail: david.robertson{at}princehenrys.org.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Inhibin A and B, important regulators of normal function in tissues of the reproductive axis, are glycosylated at either Asn268 or Asn268 and Asn302 in the {alpha}-subunit to produce 31- and 34-kDa isoforms, respectively. In this study, glycosylated isoforms of recombinant human inhibin A and B were purified from conditioned medium using immunoaffinity chromatography and reversed-phase HPLC. The masses of the purified inhibin preparations were determined by several inhibin immunoassays, and their in vitro bioactivities were based on suppression of FSH release by rat pituitary cells in culture. Based on a ratio of in vitro bioactivity to immunoactivity (B:I ratio), the monoglycosylated 31-kDa inhibin A was 5-fold more potent than the diglycosylated 34-kDa inhibin A (B:I ratio, 1.22 ± 0.15 vs. 0.24 ± 0.05; P < 0.001, respectively). The 31-kDa inhibin B was significantly (P < 0.001) more potent (1.75 ± 0.29) than the 34-kDa form (1.08 ± 0.20). Because inhibin biological activity is dependent upon interactions with the coreceptor betaglycan, the effect of inhibin glycosylation on betaglycan binding was assessed. Analogous to the pattern of in vitro bioactivity, 31-kDa inhibin A was 12-fold more active (IC50, 0.68 nM) than the 34-kDa isoform (IC50, 8.2 nM) at displacing [125I]inhibin A from COS7 cells expressing betaglycan. However, the 1.6-fold difference in bioactivity of the inhibin B isoforms was not matched by differences in their affinities for betaglycan. It is concluded that glycosylation of Asn302 of the {alpha}-subunit of inhibin A and B results in a decrease in bioactivity, and the effect on inhibin A, at least, is explained by its reduced affinity to betaglycan.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
THE STRUCTURALLY RELATED proteins of the TGF-ß superfamily, which include inhibins, activins, and bone morphogenetic proteins (BMPs), control diverse cellular processes such as proliferation, apoptosis, homeostasis, differentiation, and endocrine function (1). Inhibin A and inhibin B are typical members of the TGF-ß superfamily in that they possess a cysteine knot scaffold and are secreted as disulfide-linked, dimeric proteins composed of a glycosylated {alpha}-subunit and one of two ß-subunits (ßA or ßB, respectively) (2). Inhibins play important roles in the regulation of fertility based on their dual inhibitory action on the process of folliculogenesis in the ovary (3, 4) and FSH secretion by the pituitary (5, 6). In the female, inhibin B is primarily a product of granulosa cells of the small developing antral follicles (7), whereas inhibin A is produced by the dominant follicle and corpus luteum (8). Inhibin B, produced by Sertoli cells of the testis (9), is the primary inhibin form in the adult human male, although inhibin A has been identified in other species (e.g. rams) (10).

Functionally, inhibins antagonize the actions of TGF-ß ligands that use activin and BMP type II receptors as part of their signaling complex (11). Recent studies have shown that inhibin A effects are dependent upon interactions with betaglycan, a cell-surface proteoglycan that also acts as a TGF-ß2 coreceptor (12). Betaglycan binds inhibin A directly and promotes the formation of a stable high-affinity complex involving activin or BMP type II receptors (11). Sequestration of type II receptors in this way prevents their interactions with signaling ligands such as activins or BMPs. In vitro studies (13) indicate that betaglycan also binds inhibin B and potentiates its association with activin type II receptors. However, based on competition binding studies on mouse Leydig, Sertoli (14), and adrenocortical cell lines (15), it is generally believed that inhibin B is a weak inhibin A agonist.

Mutagenesis studies have identified the inhibin-binding site on betaglycan (amino acids 608–620) (16); however, little is known about the regions of inhibin involved in this interaction. Activin A and B, which are homodimers of inhibin ßA and ßB subunits, respectively, do not bind betaglycan, suggesting that binding is mediated by the {alpha}-subunit of inhibin. The human inhibin {alpha}-subunit is synthesized as a precursor consisting of a 43-amino-acid Pro region, a 171-amino-acid {alpha}N region, and a 134-amino-acid {alpha}C region (17). After association with a ß-subunit precursor and cleavage by proprotein convertases, mature inhibin is formed comprising the {alpha}C region disulfide linked to either a ßA or ßB subunit. In humans, unlike other species, two molecular mass isoforms of mature inhibin A and B (31 and 34 kDa) are observed (18). This molecular weight heterogeneity is due to the presence of two N-linked glycosylation sites at Asn268 and Asn302. Mason et al. (19) showed that Asn268 is always glycosylated (producing 31-kDa inhibin A or B) whereas glycosylation of Asn302 is differentially regulated (producing 34-kDa inhibin A or B). The presence of a unique, regulated glycosylation site in the human {alpha}-subunit suggests that the activity of human inhibin A and B may be under more stringent control than inhibins from other species.

In the current study, the two glycosylated isoforms of recombinant human inhibin A and inhibin B (designated as 31- and 34-kDa for convenience) were highly purified and characterized in terms of their immunological and biological activities and their binding to betaglycan. The results show that the diglycosylated 34-kDa forms of inhibin A and B were significantly less potent than the corresponding 31-kDa isoforms.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Purification procedure
Conditioned media from CHO cells producing recombinant human inhibin A and B were obtained from DSL-Beckman Coulter (Webster, TX). Initial purifications were achieved by affinity chromatography with mouse antihuman inhibin {alpha}-subunit (R1) antibody immobilized to a NHS (N-hydroxysuccinimide)-activated Sepharose Fast Flow matrix. Inhibin A was then fractionated by reversed-phase (RP)-HPLC (Vydac C-18 preparative column) using a 60-min 0–100% acetonitrile gradient in 0.1% trifluoroacetic acid. The inhibin A-containing fractions and the immunopurified inhibin B fractions were sent to Melbourne for further purification. The 31- to 34-kDa inhibin A and B were then purified by RP-HPLC (Brownlee Aquapore RP-300, 4.6 x 100 mm) using a linear 20–75% acetonitrile gradient in 0.2% heptafluorobutyric acid over 120 min at 1 ml/min. The purity of these preparations was assessed by silver stain on one-dimensional (1D) SDS-PAGE, Western blots using monoclonal antibodies to {alpha} and either ßA or ßB subunits, absence of activin A by activin A ELISA, and Western blots for activin A and B.

Preparations
The First International Standard for Inhibin A [international reference preparation (IRP), human recombinant, 91/624, 30–34 kDa] (20) was obtained from the National Institute of Biological Standards and Control (Potters Bar, UK) and expressed in terms of its nominal vial content (5 µg). This preparation was used as reference preparation for both inhibin A and B in the in vitro bioassay, {alpha}C subunit immunofluorometric assay (IFMA), inhibin RIA, and inhibin A ELISA. The inhibin preparations for immunoassay and in vitro bioassay were diluted in 10 mM phosphate buffer (pH 7.4) containing 154 mM NaCl and 0.1% BSA, aliquoted, and stored at –80 C. Throughout this study, unless otherwise stated, the immunoactivities of inhibin A and B are determined by inhibin A and B ELISAs, respectively.

Analyses
In vitro bioassay using rat pituitary cells in culture.
The method of Scott et al. (21) was employed, except FSH release rather than FSH content was used as assay endpoint. Inhibin A (91/624) was used as reference preparation in all assays of inhibin A and B. Rat FSH was determined by a specific rat FSH IFMA as previously described (22) employing reagents provided by A. Grootenhuis and J. Verhagen of N.V. Organon (The Netherlands). The sensitivity of the assay was 12.5 pg/well (inhibin A IRP). The between-assay variation based on the repeated measurement of a purified inhibin preparation was 18.7% (n = 8). The mean index of precision ({lambda}) (23) was –0.078.

Immunoassays
Inhibin A ELISA.
The inhibin A ELISA (DSL, Webster, TX) was used as described (24) employing kit reagents provided by Oxford Bio-Innovation Ltd. (Upper Heyford, UK). The ELISA used the ßA subunit antibody (E4) as capture antibody and {alpha}-subunit antibody (R1) as label. Inhibin A (91/624) was used as reference preparation. The sensitivity of the assay was 2 pg/ml. The between-assay variation was 13.1% (n = 4). The mean index of precision ({lambda}) was –0.048.

Inhibin B ELISA.
The inhibin B ELISA (DSL) was used as described (24) employing kit reagents provided by Oxford Bio-Innovation. The ELISA used the ßB subunit antibody (C5) as capture antibody and {alpha}-subunit antibody (R1) as label. A commercial recombinant human inhibin B preparation (R&D Systems Inc., Minneapolis, MN) was used as reference preparation. This preparation was calibrated in terms of its {alpha}C subunit content against the inhibin A IRP in the {alpha}C subunit IFMA. The potency of the R&D inhibin B preparation using this calibration was 49% of the stated manufacturer’s value. The basis for this difference is unclear. The between-assay variation was 13.8% (n = 5), and the sensitivity was 8 pg/ml. The mean index of precision ({lambda}) was –0.027.

Inhibin {alpha}-subunit immunofluorometric assay.
This IFMA measures the {alpha}C subunit content of inhibin A and B preparations (24), using PO no. 23 as the capture antibody and the R1 antibody as label and with the inhibin A IRP as reference preparation. These antibodies were directed to epitopes within the {alpha}C region of the inhibin {alpha}-subunit. The inhibin A IRP and inhibin A and B preparations were reduced and alkylated as previously described (25) and assayed in an {alpha}C inhibin IFMA (24). The sensitivity was 13 pg/ml. The mean index of precision ({lambda}) was –0.016.

Inhibin RIA (Monash RIA).
As an alternative immunoassay procedure, inhibin A and B were measured by RIA (26) using iodinated recombinant human 31- plus 34-kDa inhibin A as tracer, inhibin IRP (91/624) as reference preparations, and rabbit antiserum (no. 1989) raised against bovine inhibin A as antibody. Sensitivity was 750 pg/ml. The mean index of precision ({lambda}) was –0.035.

Activin A ELISA.
The activin A ELISA (DSL) was used employing kit reagents provided by Oxford Bio-Innovation.

Deglycosylation of inhibin A and inhibin B.
The 31-kDa inhibin A or B (40 µg) was lyophilized and reconstituted in 50 mM sodium acetate buffer (pH 6.5) containing 10 mM EDTA and BSA (10 µg) in a total volume of 200 µl. The sample was incubated with 50 U N-glycosidase F (Roche Diagnostics, Mannheim, Germany) at 37 C for 12 h. At the end of the incubation, the deglycosylated sample was fractionated by RP-HPLC using a gradient of acetonitrile in heptafluorobutyric acid as described above. Evidence of deglycosylation was assessed by changes in molecular mass on Western blots using the R1 antibody to detect the {alpha}-subunit.

Western transfer.
Samples were fractionated by SDS-PAGE on 4–10% Tris-Tricine gels. The proteins were electrotransferred onto nitrocellulose membranes and the membranes blocked overnight at 4 C in 5% skim milk. The membranes were then probed with the following mouse monoclonal antibodies: R1 ({alpha}-subunit), E4 (ßA subunit), and C5 (ßB subunit, in the presence of 6% H2O2). The membranes were then probed (1 h at room temperature in the dark) with a sheep antimouse serum conjugated to the fluorescent dye IR800 (Rockland Immunochemicals, Gilbertsville, PA) and then washed. The membranes were scanned on an infrared imaging system (Odyssey IR imaging scanner; Li-Cor Biosciences, Lincoln, NE).

2D PAGE.
Purified inhibin A and B samples were lyophilized and resuspended in 40 mM Tris/HCl (pH 9.5), 7 M urea, 2 M thiourea, and 1% wt/vol C7 (Sigma Chemical Co., St Louis, MO). Thiol groups were reduced and alkylated as described (27). Proteins were precipitated in 10 vol acetone and resuspended in 7 M urea, 2 M thiourea, 5 mM dithiothreitol, 1% wt/vol C7, 1% Destreak. Samples were passively rehydrated into 11-cm, pH 3–10 IPG (immobilized pH gradient) strips (Proteome Systems, Sydney, Australia) and focused overnight at 10,000 V until reaching 100,000 V-h in an IsoelectrIQ focusing apparatus (Proteome Systems). Second-dimension electrophoresis was performed in 8–16% Tris-glycine gradient gels using precast GelChips (Proteome Systems). Polyacrylamide gels were stained overnight with Sypro Ruby (Invitrogen, Carlsbad, CA) and imaged on a Fuji FLA 5100 fluoroimager.

Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF-MS)
In-gel digestion and extraction of peptide.
Protein spots of interest were excised from 2D PAGE gels, destained, and dehydrated. Porcine trypsin (Promega, Madison, WI) and the gel plugs were incubated at 37 C overnight, desalted, and concentrated using ZipTips (Millipore Corp., Bedford, MA). Adsorbed peptides were eluted and applied to 384-a (384 spots) target plate in the presence of {alpha}-cyano-4-hydroxycinnamic acid in acetonitrile/trifluoroacetic acid.

Protein identification by peptide mass fingerprinting.
Peptide mass fingerprints of tryptic peptides were collected by MALDI-TOF-MS using a Voyager 4700 mass spectrometer (Applied Biosystems, Foster City, CA) in positive ion reflectron mode. Protein identifications were assigned against human sequences entered in the Uniprot or NCBI databases using the MASCOT (http://www.matrixscience.com) or Profound (http://prowl.rockefeller.edu/profound_bin/WebProFound.exe) search engines, respectively. Searching was constrained only by the species. Carbamidomethylation of cysteines and oxidation of methionine residues were taken into account, and a mass tolerance of 0.1 Da was allowed. Miscleavage sites were considered only after an initial search with no missed cleavages. The following criteria were used to evaluate the search: MOWSE (Molecular Weight Search) score, number and intensity of peptides matched, coverage of the candidate protein sequence, and position on the 2D gel.

In several instances, significant matches could not be obtained. In this case, the predicted peptides obtained by trypsin digestion of inhibin {alpha}C subunit (Uniprot entry P05111) were evaluated using Peptide Mass (http://ca.expasy.org/tools/peptide-mass.html), and the collected spectra were manually searched for the presence of the predicted peptides against a spectrum obtained from another area of the original gel.

Transfection of betaglycan in COS7 cells.
COS7 cells were grown in 5% CO2 in a 37 C humidified incubator in complete DMEM containing 10% BSA. Transfection of COS7 cells with truncated betaglycan (BG 576–853) (16) was performed using the lipofectamine 2000 reagent (Invitrogen). Briefly, COS7 cells were plated at 1 x 105 cells per well in 24-well plates. After overnight recovery, cells were transfected with a total of 100 ng DNA per well (10 ng BG 576–853 plus 90 ng pcDNA3.1) according to the manufacturer’s instructions. Plates were incubated at 37 C in 5% CO2 to allow cells to recover and express protein for 48 h before assay.

Inhibin A and inhibin B binding to betaglycan 576–853.
Inhibin A and inhibin B were iodinated by the lactoperoxidase method and purified by gel permeation chromatography (1.5 x 100 cm) in 0.1 M HCl (28). Inhibin A and inhibin B binding was measured in intact cells; 48 h after transfection, cells were washed with binding buffer (DMEM/0.1% BSA) and then incubated in binding buffer with 4 x 104 cpm/well of either 31-kDa [125I]inhibin A or 31-kDa [125I]inhibin B and increasing concentrations of unlabeled inhibin A and inhibin B isoforms. Plates were incubated for 3 h with gentle rocking, and then wells were washed with cold PBS to remove free tracer. Cells were solubilized in 250 µl 1% Triton X-100, and [125I]inhibin A/B bound in each well was determined using a {gamma}-counter. Binding data were analyzed and graphed using the GraphPad Prism program (San Diego, CA).

Analysis.
A parallel line bioassay design (23) was used to calculate the biological and immunological activities of the inhibin preparations. These analyses are based on the comparison of dose-response curves of reference and test preparations and include an assessment of linearity, parallelism, precision, and relative activities. With this design, each sample was tested at a minimum of three doses in triplicate from two to six separate assays. Wherever possible, aliquots from the same batch of standard and sample were tested in the respective assays. Inhibin A IRP was used as reference preparation in the inhibin A ELISA and as a common reference preparation for inhibin A and B in the in vitro bioassay, inhibin RIA, and inhibin {alpha}C subunit ELISA. Inhibin B (R&D Systems) was used as the reference preparation in the inhibin B ELISA; its immunoactivity was defined based on its {alpha}C subunit content as determined by inhibin {alpha}C subunit ELISA, using the inhibin A IRP as reference preparation.

Statistical analysis of the differences in B:I ratios of inhibin A and B isoforms were determined using one-way repeated-measures ANOVA. The differences in the mean values among the treatment groups with P < 0.05 were considered significant. This analysis was performed using the statistical software SigmaStat for Windows, version 2.0 (SPSS, San Rafael, CA). All results are presented as mean ± SD.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Purification and characterization of recombinant human inhibin A and inhibin B
Recombinant human inhibin A and B were purified by a combination of immunoaffinity chromatography and RP-HPLC procedures. Immunoreactive fractions of inhibin A and B were analyzed by silver stain and Western blot using antibodies to the {alpha} and either ßA or ßB subunits. The monoglycosylated (31-kDa) and diglycosylated (34-kDa) isoforms of inhibin A (Fig. 1AGo) and inhibin B (Fig. 1BGo) were separated from activin and higher molecular weight species using this procedure. For the sake of clarity, the mono- and diglycosylated forms of inhibin A and B are defined throughout as 31- and 34-kDa, respectively.


Figure 1
View larger version (39K):
[in this window]
[in a new window]

  		FIG. 1. Profile of inhibin A and B immunoactivity after fractionation on RP-HPLC. A, Inhibin A was detected by inhibin A ELISA and by Western analysis using antisera to the {alpha} (R1) and ßA (E4) subunits. Fractions 13–15 and 16–17 were pooled to produce the 31- plus 34-kDa and 31-kDa inhibin A pools used in subsequent analyses. Sufficient yields of 34-kDa inhibin A were obtained by refractionation of the 31- plus 34-kDa inhibin A fractions through the RP-HPLC procedure. B, Immunoactive profiles of 31- and 34-kDa inhibin B were determined by inhibin B ELISA and by Western analysis using antisera to the {alpha} (R1) and ßB (C5) subunits. Fractions 18, 19–20, and 21–25 were pooled to produce the 34-, 34/31-, and 31-kDa inhibin B pools used in subsequent analyses.

 
Subsequently, 31- plus 34-kDa inhibin A and B were treated with N-glycosidase F to generate deglycosylated (28-kDa) forms (Fig. 2Go, A and E). Fractions containing 28-, 31-, 31- plus 34-, or 34-kDa inhibin A (Fig. 2AGo) and inhibin B (Fig. 2EGo) were pooled for further analysis. Under reducing conditions, a series of pI forms of both {alpha}- and ß-subunits of inhibin A and B were identified (Table 1Go). Mass spectrometric analysis of selected spots (Fig. 2Go, C and H) indicated that they represented either the {alpha}-, ßA, or ßB subunits, respectively. The modifications of the various pI isoforms were not identified; however, the pattern of spots was comparable between all inhibin A (Fig. 2Go, B–D) and inhibin B (Fig. 2Go, F–H) isoforms, respectively.


Figure 2
View larger version (36K):
[in this window]
[in a new window]

  		FIG. 2. The 1D and 2D PAGE profiles of 31- and 34-kDa and deglycosylated (28-kDa) inhibin A and B. A and E, 1D PAGE profiles of inhibin A and B preparations under nonreducing conditions, respectively; B–D and F–H, 2D PAGE profiles of 34- and 31-kDa and deglycosylated isoforms of inhibin A and B under reducing conditions, respectively. *, Spots that were identified by mass spectrometry.

 

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

  		TABLE 1. Molecular mass and pI ranges of dimeric inhibin A and B and their subunits

 
In vitro biological activity of inhibin A and inhibin B
In response to endogenous activin, rat pituitary cells in culture synthesize and secrete FSH (29, 30). Previous studies have shown that inhibin A and B can suppress FSH release from these cells in a dose-dependent manner (31). This rat pituitary cell culture system was used to assess the biological activity of inhibin A and B glycosylated isoforms. The mass of the inhibin A and B pools was measured by inhibin A and B ELISAs, using inhibin A IRP and inhibin B (R&D) as standards. In the in vitro bioassay (Fig. 3Go) and ELISAs (data not shown) of 31- and 34-kDa inhibin A and B preparations exhibited parallel dose-response lines compared with the standards.


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

  		FIG. 3. Dose-response curves of FSH suppression by 31-, 34-, and deglycosylated 28-kDa inhibin A (A) and B (B) in the inhibin in vitro bioassay using primary rat pituitary cells in culture. Inhibin A IRP was used as reference preparation. The plots are mean values of two representative experiments.

 
The in vitro bioactivity of the inhibin A and B preparations were then related to their ELISA immunoactivity (as a measure of protein mass) to generate in vitro B:I ratios. Using this approach, a 5- and 1.6-fold decrease in B:I ratio was observed between 31-kDa (1.22 ± 0.16) and 34-kDa (0.24 ± 0.05) inhibin A and 31-kDa (1.75 ± 0.29) and 34-kDa (1.08 ± 0.20) inhibin B, respectively (Tables 2Go and 3Go and Figs. 3Go and 4Go, A and B). To establish whether these differences in B:I ratios were attributable to the immunoassay method employed, immunoactivities were also determined by an established inhibin RIA and by an inhibin {alpha}C subunit IFMA. Using these assays, changes in B:I ratios between 31- and 34-kDa inhibin A and B were similar to those observed with the ELISAs, although the difference between isoforms was less marked with the RIA (Tables 2Go and 3Go). Nonetheless, significant differences in B:I ratios between 34- and 31-kDa inhibin A and B were observed irrespective of the immunoassay method.


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

  		TABLE 2. Inhibin A isoforms: in vitro biological activity (expressed as ratio of in vitro B:I activity) and betaglycan binding

 

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

  		TABLE 3. Inhibin B isoforms: in vitro biological activity (expressed as ratio of in vitro B:I activity) and betaglycan binding

 

Figure 4
View larger version (13K):
[in this window]
[in a new window]

  		FIG. 4. In vitro bioactivity of inhibin A and B isoforms. The bioactivity of 28-, 31-, 34/31-, and 34-kDa inhibin A (A) and B (B) is expressed as the ratio of in vitro bioactivity to immunoactivity as determined in specific inhibin A and B ELISAs. The dashed line represents the B:I ratio of the inhibin A IRP standard. Different letters indicate significance of differences between groups (mean ± SD; P < 0.05; n = 3–8).

 
Deglycosylation of inhibin A and inhibin B
To further investigate the effects of glycosylation on the in vitro bioactivity and immunoactivity of inhibin A and B, 31-kDa isoforms were deglycosylated and purified from non-deglycosylated forms by RP-HPLC (Fig. 2Go, A and E). The molecular mass of deglycosylated inhibin A and B were determined by 2D PAGE to be approximately 28 kDa (Table 1Go). The B:I ratio of 28-kDa inhibin A (2.53 ± 0.28) using the inhibin A ELISA increased approximately 2- and 10-fold compared with non-deglycosylated 31-kDa (1.19 ± 0.15) and 34-kDa (0.24 ± 0.05) forms, respectively (Table 2Go and Fig. 4AGo). This increase was also observed with {alpha}C subunit IFMA, although it was not evident by RIA. In contrast, deglycosylation of inhibin B resulted in a decrease in biological activity compared with the 31-kDa isoform irrespective of the immunoassay method (Table 3Go and Fig. 4BGo).

Differential binding of inhibin A and inhibin B isoforms to betaglycan
Inhibins mediate their action by binding to the coreceptor betaglycan (12). To determine the effect of inhibin glycosylation on betaglycan binding, competition binding assays were performed in which the inhibin A and B isoforms were assessed for their ability to displace [125I]inhibin A or [125I]inhibin B from COS7 cells transfected with betaglycan (576–853). Previous studies have shown that the betaglycan (576–853) construct encompasses the inhibin-binding site (16). Nontransfected COS7 cells, which do not bind inhibin A or B, were used as controls in the binding assays. The 31-kDa inhibin A displaced [125I]inhibin A from betaglycan with an IC50 of 0.68 ± 0.21 nM (Table 2Go and Fig. 5AGo), which is in good agreement with the high-affinity binding (KD, ~0.6 nM) previously observed between inhibin A and betaglycan (12). In contrast, 34-kDa inhibin A was a weak competitor of [125I]inhibin A binding to betaglycan with an IC50 of 8.2 ± 1.0 nM. Although loss of the carbohydrate moiety from Asn302 led to a dramatic increase in inhibin A affinity for betaglycan (cf. 31- and 34-kDa inhibin A in Fig. 5AGo), further deglycosylation (28-kDa inhibin A, Fig. 5AGo) resulted in only a minor increase in the ability of inhibin A to compete with [125I]inhibin A (IC50 of 0.59 ± 0.21 nM) (Table 2Go).


Figure 5
View larger version (16K):
[in this window]
[in a new window]

  		FIG. 5. Binding of inhibin A and inhibin B isoforms to betaglycan. COS7 cells were transfected with betaglycan (576–853) and 48 h later were subjected to competition binding as described under Materials and Methods. A, Dose-response curves of 31-kDa (bullet), 34-kDa ({square}), or 28-kDa ({circ}) inhibin A for the binding of [125I]inhibin A to the transfected cells. B, Dose-response curves of 31-kDa (bullet), 34-kDa ({square}), or 28-kDa ({circ}) inhibin B for the binding of [125I]inhibin B to the transfected cells. The values given are the mean ± SD; n = 3 from one experiment.

 
Although glycosylation of inhibin A is negatively correlated with binding to betaglycan, the same effect was less apparent with the inhibin B isoforms. The 31- and 34-kDa inhibin B displaced [125I]inhibin B from betaglycan similarly (IC50, 39.8 ± 13.0 pM and 36.5 ± 15.8 pM, respectively). However, 28-kDa inhibin B displayed a 3-fold higher affinity (IC50, 12.2 ± 1.1 pM) for betaglycan than the glycosylated isoforms (Table 3Go and Fig. 5BGo).


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
The reproductive hormones inhibin A and B are differentially glycosylated to produce 31- and 34-kDa isoforms. To understand the contribution of glycosylation to inhibin biological activity, the 31- and 34-kDa forms of inhibin A and B were isolated. This study shows that there is a marked difference in the in vitro biological activity of 31- and 34-kDa inhibin A and B. The 34-kDa inhibin A form, which is attributed to diglycosylation, is only 10–20% as active as the 31-kDa (monoglycosylated) form, whereas the 34-kDa inhibin B form is 60% as active as the 31-kDa form.

The apparent masses of the inhibin A and B isoforms were initially determined by specific ELISAs that use a common antibody (R1) directed against the {alpha}-subunit and separate ß-subunit-specific antibodies. To ensure the determined masses were not due to differential cross-reactivity of the 31- and 34-kDa inhibin forms in the ELISA, additional immunoassays (inhibin RIA and inhibin {alpha}C subunit IFMA) were used. The antibody used in the RIA (26, 32) is primarily directed to a conformational epitope on the {alpha}-subunit common to both inhibin A and B. The inhibin {alpha}C subunit IFMA detects the presence of the inhibin {alpha}C region of the {alpha}-subunit and thus is applicable to both inhibin A and B on the basis that their {alpha}-subunits are identical (26). The inhibins and inhibin A IRP used as standard were reduced and alkylated before assay. The inhibin RIA and inhibin {alpha}C subunit IFMA confirmed the differences in biological activity (B:I ratios) between the 31- and 34-kDa inhibin forms observed with the ELISAs.

To further investigate the importance of glycosylation on inhibin bioactivity, inhibin A and B preparations were deglycosylated, purified by RP-HPLC, and tested in the respective immunoassays and in vitro bioassay. The B:I ratio of the deglycosylated (28-kDa) inhibin A increased when using either the ELISA or IFMA; however no effects of deglycosylation were seen when using the RIA. Overall, these studies suggest that the decrease in bioactivity between 31- and 34-kDa inhibin A is due to the inhibitory effects of the second glycosylation site in the in vitro bioassay. However, the increase in bioactivity in inhibin A after deglycosylation is probably related more to a decrease in response of the deglycosylated inhibins in the ELISA/IFMA resulting in lower levels of apparent immunoactivity, because no effect was observed with the RIA. Interestingly, deglycosylation of inhibin B resulted in an approximately 40% decrease in bioactivity (B:I ratio) compared with the 31-kDa form regardless of the immunoassay used.

Inhibin A antagonism of activin and related ligands is mediated by the coreceptor betaglycan (11, 12). In combination with betaglycan, inhibin A forms a stable high-affinity complex with activin type II receptors, rendering them unavailable for association with their signaling ligands (12). Chapman et al. (13) have shown that betaglycan also binds inhibin B, although the affinity for activin type II receptors was not significantly enhanced. Because betaglycan mediates inhibin action, the effect of glycosylation on inhibin binding to betaglycan was assessed. The 34-kDa inhibin A had approximately 19-fold lower affinity for betaglycan than the 31-kDa form, suggesting that glycosylation of Asn302 interferes with betaglycan binding. These results provide a mechanism for the difference in in vitro biological activity of 31- and 34-kDa inhibin A. Deglycosylation to 28-kDa inhibin A did not enhance affinity for betaglycan relative to the 31-kDa form. Together these results suggest that betaglycan binds to the {alpha}-subunit of inhibin A and that the region surrounding Asn302 is the likely binding site. In contrast to inhibin A, differences in the bioactivities of the glycosylated forms of inhibin B are not associated with changes in their affinity for betaglycan. Relative to 31-kDa inhibin B, the 34-kDa and 28-kDa (deglycosylated) isoforms are less potent and yet display higher affinities (1.2- and 3.6-fold, respectively) for their coreceptor. These results suggest that inhibin B biological activity may be dependent upon interactions with binding proteins other than betaglycan.

The 2D PAGE analysis of the inhibin A and B isoforms confirmed that the size heterogeneity between the 31- and 34-kDa inhibin forms was attributable to glycosylation of the {alpha}-subunit. In addition, it was evident the inhibin {alpha}-subunit comprised three to five pI forms and the ßA and ßB subunits displayed two four pI forms. Mass spectrometry analysis confirmed that these pI forms were subunits of either inhibin A or B. This charge heterogeneity suggests that the inhibin {alpha}- and ß-subunits may have additional and as yet uncharacterized posttranslational modifications. However, because the distribution of the pI forms was identical between 34- and 31-kDa inhibin A or B, these modifications are unlikely to contribute to the observed differences in biological activity.

Differential glycosylation of proteins is not a common mechanism to regulate bioactivity. However, studies have shown that factors that regulate terminal sialylation and/or sulfation of the oligosaccharide attachments of the pituitary gonadotropins (LH and FSH) also regulate some functional properties of the gonadotropin molecule, such as metabolic clearance and in vivo biopotency (33, 34). It is intriguing that the activity of both inhibin and the protein it regulates, FSH, is determined by the type and extent of oligosaccharide modification. For FSH, differential glycosylation regulates not only the quantity but also the quality of the gonadotropin signal delivered to the gonads in a given physiological or pathological condition (33). It is possible that differential glycosylation similarly determines the quality of the inhibin A and B signal produced by the gonads. Given that loss of {alpha}-subunit expression leads to the development of gonadal tumors (35) and mutations in the {alpha}-subunit have been associated with premature ovarian failure (36), it is suggested that any decreases in the bioactivity of inhibin would have marked effects on reproductive tissues.

A number of studies have shown that 31- and 34-kDa forms of inhibin A and B are found in human follicular fluid (25), pregnancy serum (37), and serum from women undergoing gonadotropin treatment (25, 26). In one study (38), bioactive inhibin A was isolated to high purity from human follicular fluid and shown to consist of two molecular mass forms (29 and 34 kDa) similar to that reported here; however, no further details were reported. However, it is unclear at this stage whether these glycosylated forms have a physiological role because neither changes in their plasma distribution nor their in vivo biological activities have been established, which are areas for future study.

In conclusion, the 31- and 34-kDa forms of inhibin A and B have been isolated to high purity and shown to exhibit differences in in vitro biological activity, which in part can be attributed to their level of glycosylation and the affinity of these inhibin forms to bind to betaglycan, a key accessory binding protein involved in inhibin’s action. These findings are likely to have relevance in the understanding of inhibin in human physiology.


   Acknowledgments
 
We acknowledge the excellent technical assistance of Enid Pruysers, Sara Goodman, and Karen Chan and Dr. Andrew Stephens for the 2D-PAGE and mass spectrometry analyses. We also acknowledge with thanks Dr. Arijan Grootenhuis and Jos Verhagen of N.V. Organon (The Netherlands) for providing reagents for the rat FSH IFMA.


   Footnotes
 
This work was supported by National Health and Medical Research Council of Australia Program Grant 241000, Project Grant CH 388920, and Research Fellowship DMR 169201.

Disclosure Statement: Y.M., C.A.H., P.G.S., and R.K. have nothing to declare. D.M.R. is an inventor on patents AU85/00119 and AU86/00097.

First Published Online February 1, 2007

Abbreviations: B:I, Bioactivity to immunoactivity; BMP, bone morphogenetic protein; IFMA, immunofluorometric assay; IRP, international reference preparation; 1D, one-dimensional; MALDI-TOF-MS, matrix-assisted liquid desorption/ionization time-of-flight mass spectrometry; RP, reversed-phase.

Received December 1, 2006.

Accepted for publication January 19, 2007.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Shi Y, Massague J 2003 Mechanisms of TGF-ß signaling from cell membrane to the nucleus. Cell 113:685–700[CrossRef][Medline]
  2. Mason AJ, Hayflick JS, Ling N, Esch F, Ueno N, Ying SY, Guillemin R, Niall H, Seeburg PH 1985 Complementary DNA sequences of ovarian follicular fluid inhibin show precursor structure and homology with transforming growth factor-ß. Nature 318:659–663[CrossRef][Medline]
  3. Findlay JK 1993 An update on the roles of inhibin, activin, and follistatin as local regulators of folliculogenesis. Biol Reprod 48:15–23[Abstract]
  4. Magoffin DA, Jakimiuk AJ 1997 Inhibin A, inhibin B and activin A in the follicular fluid of regularly cycling women. Hum Reprod 12:1714–1719[Abstract/Free Full Text]
  5. Robertson DM, Giacometti MS, de Kretser DM 1986 The effects of inhibin purified from bovine follicular fluid in several in vitro pituitary cell culture systems. Mol Cell Endocrinol 46:29–36[CrossRef][Medline]
  6. Woodruff TK, Krummen LA, Lyon RJ, Stocks DL, Mather JP 1993 Recombinant human inhibin A and recombinant human activin A regulate pituitary and ovarian function in the adult female rat. Endocrinology 132:2332–2341[Abstract]
  7. Welt CK, Schneyer AL 2001 Differential regulation of inhibin B and inhibin a by follicle-stimulating hormone and local growth factors in human granulosa cells from small antral follicles. J Clin Endocrinol Metab 86:330–336[Abstract/Free Full Text]
  8. Welt CK, Adams JM, Sluss PM, Hall JE 1999 Inhibin A and inhibin B responses to gonadotropin withdrawal depends on stage of follicle development. J Clin Endocrinol Metab 84:2163–2169[Abstract/Free Full Text]
  9. Marchetti C, Hamdane M, Mitchell V, Mayo K, Devisme L, Rigot JM, Beauvillain JC, Hermand E, Defossez A 2003 Immunolocalization of inhibin and activin {alpha} and ßB subunits and expression of corresponding messenger RNAs in the human adult testis. Biol Reprod 68:230–235[Abstract/Free Full Text]
  10. McNeilly AS, Souza CJ, Baird DT, Swanston IA, McVerry J, Crawford J, Cranfield M, Lincoln GA 2002 Production of inhibin A not B in rams: changes in plasma inhibin A during testis growth, and expression of inhibin/activin subunit mRNA and protein in adult testis. Reproduction 123:827–835[Abstract]
  11. Wiater E, Vale W 2003 Inhibin is an antagonist of bone morphogenetic protein signaling. J Biol Chem 278:7934–7941[Abstract/Free Full Text]
  12. 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]
  13. Chapman SC, Bernard DJ, Jelen J, Woodruff TK 2002 Properties of inhibin binding to betaglycan, InhBP/p120 and the activin type II receptors. Mol Cell Endocrinol 196:79–93[CrossRef][Medline]
  14. 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]
  15. 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]
  16. Wiater E, Harrison CA, Lewis KA, Gray PC, Vale WW 2006 Identification of distinct inhibin and transforming growth factor ß-binding sites on betaglycan: functional separation of betaglycan co-receptor actions. J Biol Chem 281:17011–17022[Abstract/Free Full Text]
  17. Mayo KE, Cerelli GM, Spiess J, Rivier J, Rosenfeld MG, Evans RM, Vale W 1986 Inhibin A-subunit cDNAs from porcine ovary and human placenta. Proc Natl Acad Sci USA 83:5849–5853[Abstract/Free Full Text]
  18. Tierney ML, Goss NH, Tomkins SM, Kerr DB, Pitt DE, Forage RG, Robertson DM, Hearn MT, de Kretser DM 1990 Physicochemical and biological characterization of recombinant human inhibin A. Endocrinology 126:3268–3270[Abstract]
  19. Mason AJ, Farnworth PG, Sullivan J 1996 Characterization and determination of the biological activities of noncleavable high molecular weight forms of inhibin A and activin A. Mol Endocrinol 10:1055–1065[Abstract]
  20. Rose MP, Gaines Das RE 1996 International collaborative study by in vitro bioassays and immunoassays of the First International Standard for Inhibin, Human Recombinant. Biologicals 24:1–18[CrossRef][Medline]
  21. Scott RS, Burger HG, Quigg H 1980 A simple and rapid in vitro bioassay for inhibin. Endocrinology 107:1536–1542[Abstract]
  22. van Casteren JI, Schoonen WG, Kloosterboer HJ 2000 Development of time-resolved immunofluorometric assays for rat follicle-stimulating hormone and luteinizing hormone and application on sera of cycling rats. Biol Reprod 62:886–894[Abstract/Free Full Text]
  23. Finney DJ 1971 Statistical method in biological assay. 2nd ed. London: Charles Griffin
  24. Robertson DM, Stephenson T, Pruysers E, McCloud P, Tsigos A, Groome N, Mamers P, Burger HG 2002 Characterization of inhibin forms and their measurement by an inhibin {alpha}-subunit ELISA in serum from postmenopausal women with ovarian cancer. J Clin Endocrinol Metab 87:816–824[Abstract/Free Full Text]
  25. Robertson DM, Cahir N, Findlay JK, Burger HG, Groome N 1997 The biological and immunological characterization of inhibin A and B forms in human follicular fluid and plasma. J Clin Endocrinol Metab 82:889–896[Abstract/Free Full Text]
  26. Robertson D, Burger HG, Sullivan J, Cahir N, Groome N, Poncelet E, Franchimont P, Woodruff T, Mather JP 1996 Biological and immunological characterization of inhibin forms in human plasma. J Clin Endocrinol Metab 81:669–676[Abstract]
  27. Grinyer J, McKay M, Nevalainen H, Herbert BR 2004 Fungal proteomics: initial mapping of biological control strain Trichoderma harzianum. Curr Genet 45:163–169[CrossRef][Medline]
  28. 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]
  29. Ling N, Ying SY, Ueno N, Shimasaki S, Esch F, Hotta M, Guillemin R 1986 Pituitary FSH is released by a heterodimer of the ß-subunits from the two forms of inhibin. Nature 321:779–782[CrossRef][Medline]
  30. Mason AJ, Berkemeier LM, Schmelzer CH, Schwall RH 1989 Activin B: precursor sequences, genomic structure and in vitro activities. Mol Endocrinol 3:1352–1358[Abstract]
  31. Ying SY, Czvik J, Becker A, Ling N, Ueno N, Guillemin R 1987 Secretion of follicle-stimulating hormone and production of inhibin are reciprocally related. Proc Natl Acad Sci USA 84:4631–4635[Abstract/Free Full Text]
  32. McLachlan RI, Robertson DM, Burger HG, de Kretser DM 1986 The radioimmunoassay of bovine and human follicular fluid and serum inhibin. Mol Cell Endocrinol 46:175–185[CrossRef][Medline]
  33. Ulloa F, Real FX 2001 Differential distribution of sialic acid in {alpha}2,3 and {alpha}2,6 linkages in the apical membrane of cultured epithelial cells and tissues. J Histochem Cytochem 49:501–510[Abstract/Free Full Text]
  34. Stanton PG, Burgon PG, Hearn MT, Robertson DM 1996 Structural and functional characterisation of hFSH and hLH isoforms. Mol Cell Endocrinol 125:133–141[CrossRef][Medline]
  35. Matzuk MM, Finegold MJ, Su JG, Hsueh AJ, Bradley A 1992 {alpha}-Inhibin is a tumour-suppressor gene with gonadal specificity in mice. Nature 360:313–319[CrossRef][Medline]
  36. Shelling AN, Burton KA, Chand AL, van Ee CC, France JT, Farquhar CM, Milsom SR, Love DR, Gersak K, Aittomaki K, Winship IM 2000 Inhibin: a candidate gene for premature ovarian failure. Hum Reprod 15:2644–2649[Abstract/Free Full Text]
  37. Thirunavukarasu P, Stephenson T, Forray J, Stanton PG, Groome N, Wallace E, Robertson DM 2001 Changes in molecular weight forms of inhibin A and pro-{alpha}C in maternal serum during human pregnancy. J Clin Endocrinol Metab 86:5794–5804[Abstract/Free Full Text]
  38. Hasegawa Y, Miyamoto K, Sugino H, Takio K, Inoue M, Ibuki Y 1994 Progress with human and rat inhibin characterization. In: Frontiers in endocrinology: inhibin and inhibin-related proteins. Rome: Ares-Serono Symposia; 5–24



This article has been cited by other articles:


Home page
 J. Biol. Chem.Home page
Y. Makanji, K. L. Walton, M. C. Wilce, K. L. Chan, D. M. Robertson, and C. A. Harrison
Suppression of Inhibin A Biological Activity by Alterations in the Binding Site for Betaglycan
J. Biol. Chem., June 13, 2008; 283(24): 16743 - 16751.
[Abstract] [Full Text] [PDF]

Home page
 EndocrinologyHome page
P. G. Farnworth, Y. Wang, R. Escalona, P. Leembruggen, G. T. Ooi, and J. K. Findlay
Transforming Growth Factor-{beta} Blocks Inhibin Binding to Different Target Cell Types in a Context-Dependent Manner through Dual Mechanisms Involving Betaglycan
Endocrinology, November 1, 2007; 148(11): 5355 - 5368.
[Abstract] [Full Text] [PDF]

Home page
 Mol. Endocrinol.Home page
M. Antenos, M. Stemler, I. Boime, and T. K. Woodruff
N-Linked Oligosaccharides Direct the Differential Assembly and Secretion of Inhibin {alpha}- and {beta}A-Subunit Dimers
Mol. Endocrinol., July 1, 2007; 21(7): 1670 - 1684.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
148/5/2309    most recent
Author Manuscript (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
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 Makanji, Y.
Right arrow Articles by Robertson, D. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Makanji, Y.
Right arrow Articles by Robertson, D. M.

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