| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Unité de Recherches sur l’Endocrinologie du Développement (Institut National de la Santé et de la Recherche Médicale) (C.B., J.-Y.P., N.J., N.D.), 92140 Clamart, France; and Biogen, Inc. (H.V.V., C.E., B.P., A.R.R., R.L.C.), Cambridge, Massachusetts 02142
Address all correspondence and requests for reprints to: Dr. Richard L. Cate, Biogen, Inc., 14 Cambridge Center, Cambridge, Massachusetts 02142. E-mail: Richard.Cate{at}biogenidec.com.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Members of the TGF-ß family are translated as larger precursor proteins that are proteolytically cleaved to generate N-terminal and C-terminal homodimers. Consistent with this paradigm, AMH undergoes proteolytic processing at a monobasic cleavage site to generate a 110-kDa N-terminal dimer and a 25-kDa C-terminal dimer (3). Proteolytic processing is inefficient in vivo but can be driven to completion in vitro using plasmin (3) or a kex2/subtilisin-like endoprotease (4). Processing has been shown to be necessary as mutations that block cleavage at the monobasic site destroy biological activity (4, 5).
The homology of AMH to other members of the TGF-ß family is restricted to the C-terminal fragment, the bioactive domain of the molecule (6, 7). Like other members of this superfamily, the C-terminal domain of AMH presumably interacts with a transmembrane serine/threonine kinase receptor complex, containing type I and type II components. The type II receptor, AMHRII, has been clearly identified (8, 9) and is AMH specific. Although the identity of the type I receptor is uncertain, three type I receptors of the bone morphogenetic protein (BMP) family, activin receptor-like kinase (ALK)-2 (10, 11), ALK-3 (12), and ALK-6 (13) are strong candidates. The BMP-specific cytoplasmic effectors Smads 1, 5, and 8 also mediate the effects of AMH (reviewed in Ref. 14). Thus, mutations within the C-terminal domain of AMH may affect receptor binding and downstream signaling events.
The role of the N-terminal domain is less clear. The N-terminal domain, although not directly involved in function, has been shown to aid the folding, disulfide bond formation, and export of the mature activin and TGF-ß1 homodimers (15). In addition, the N-terminal domain of many TGF-ß family members remains associated with its C-terminal domain in a noncovalent complex. In the case of TGF-ß, the N-terminal domain, also called the latency-associated peptide (LAP), renders the C-terminal fragment latent because of this interaction (16). In contrast, the N-terminal domain of AMH enhances the bioactivity of the C-terminal domain (7). Thus, mutations within the N-terminal domain could affect either biosynthesis or bioactivity of AMH.
Because mutations leading to PMDS occur in various regions of the AMH protein, their study may provide insight into the regions that are important for biosynthesis and biological activity. Here, we have studied the biosynthesis and secretion of wild-type AMH and of seven PMDS proteins, containing mutations in either the N- or C-terminal domain. We find that mutations in both domains can have profound effects on protein stability and folding, and that mutant proteins lacking the C-terminal domain are secreted more rapidly than full-length AMH. We also present a molecular model of the AMH C-terminal domain that provides insights into how mutations within this domain could affect biosynthesis and function.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
). Three of the mutations affect the N-terminal fragment of the AMH molecule: 350–354del interrupts translation after proline residue 113; R194C is a missense mutation; and 2277–2292del causes premature termination close to the end of the N-terminal domain. The other four mutations, V477A, Q496H, H506Q, and C525Y, all missense, are located within the C-terminal domain. All patients were homozygous for the mutation, except one, carrying H506Q on one allele and 350–354del on the other. The seven mutations were reproduced by site-directed mutagenesis of the wild-type AMH cDNA and expressed in COS cells. Two other mutations included in Table 1 were also generated and characterized: R451T, which inactivates the cleavage site and consequently prevents the release of the C-terminal domain (3), and S452stop, which terminates translation at the cleavage site and encodes only the N-terminal domain. Although not found in PMDS patients, the R451T and S452stop mutations were valuable in allowing us to interpret some of the effects seen with the natural mutations, in addition to providing insight into the biosynthesis of wild-type AMH.
|
also shows that AMH could be detected in the serum of two PMDS patients with the V477A and Q496H mutations, but not in patients with the R194C and H506Q mutations. The AMH serum level of the Q496H mutant was similar to that of wild-type AMH, whereas the level of the V477A mutant protein was approximately 5-fold lower. Not all patients could be tested, because AMH is detectable in patient serum only before puberty (17).
Effects of AMH Mutations on Intracellular Biosynthesis
We first analyzed the AMH mutations by in vitro translation and transient expression in COS cells. All of the in vitro translated proteins were of the expected size (Fig. 1A), with the exception of the C525Y protein, which appeared slightly smaller than expected. Because the in vitro translation products were radiolabeled with L-[35S]methionine, the lower intensity of the two frameshift mutant proteins, 350–354del and 2277–2292del, is probably due to the reduced number of methionine residues in these truncated proteins.
|
). Figure 1C shows the Western blot analysis of the cell lysates, and the results are summarized in Table 2. Only Mab 10.6 recognized all the AMH mutant proteins. The two frameshift mutations, 350–354del and 2277–2292del, produced truncated proteins of the predicted size (Table 1), whereas four of the missense mutations, R451T, V477A, Q496H, and H506Q, produced proteins similar in size to the wild-type AMH protein. The C525Y missense mutation produced a protein smaller than expected, similar in size to the 2277–2292del protein, which was not detected by Mab C, indicating that it might be missing the C-terminal domain. The R194C missense mutation also produced an anomalous protein as characterized by Western blotting, with two bands that were only detected by Mab 10.6. The amount of intracellular AMH detected by Mab 10.6 was similar for most of the mutant proteins, except for mutant proteins 350–354del and R194C, which were expressed at lower levels.
|
, intracellular wild-type AMH was readily detected with the Mab 10.6. A level of staining similar to that of wild-type AMH was observed in cells expressing the R451T, V477A, Q496H, and H506Q mutant proteins. In contrast, the level of staining of the 350–354del, R194C, 2277–2292del, and C525Y mutant proteins was much lower than that of wild-type AMH. The results of the experiments shown in Fig. 1D are summarized in Table 2.
Effects of AMH Mutations on Secretion
To determine whether the mutant proteins were secreted, a Western blot analysis of culture medium from COS cells expressing each of the seven AMH mutants was performed using the N-terminal domain-specific Mab 10.6 (Fig. 2A and Table 2). Bands comparable in intensity to wild-type AMH were observed for 2277–2292del, Q496H, and C525Y. Mutant proteins 350–354del, R194C, and H506Q were completely absent from the culture medium, and the V477A mutant protein was present at a low level. AMH levels in the culture medium were also measured with a commercial ELISA assay using Mabs N2 and C (Table 2). Consistent with the Western blot analysis (Fig. 2A), a normal level of AMH was detected for mutant Q496H, no AMH was detected for mutants 350–354del, R194C, and H506Q, and a lower level was detected for the V477A mutant. There was a good correlation between AMH serum levels of prepubertal patients and the AMH levels in the COS cell culture medium for mutants R194C, V477A, Q496H, and H506Q (Tables 1 and 2).
|
), yet proteins were detectable by Western blotting with Mab 10.6 (Fig. 2A). The 2277–2292del mutant protein is missing the C-terminal domain and cannot be detected by the ELISA that includes a C-terminal domain-specific antibody (Mab C) in addition to an N-terminal domain-specific antibody (Mab N2). Like the intracellular form, the secreted C525Y mutant protein was smaller than expected and similar in size to the 2277–2292del mutant protein (Fig. 2A). An immunoprecipitation experiment using the N-terminal domain-specific Mab 10.6 indicated that the C525Y protein was either identical in size or slightly smaller than the 2277–2292del protein (Fig. 2B). Furthermore, the C525Y mutant protein failed to be immunoprecipitated by Mab C (Fig. 2B) and was not detected by the ELISA (Table 2). The inability of the C-terminal domain-specific Mab to detect the C525Y protein, together with the size of the protein, indicates that it is truncated at the C terminus and places the site of truncation at, or close to, the monobasic cleavage site. In Fig. 2B, the C-specific Mab failed to detect a free C-terminal domain in the culture medium of C525Y-expressing COS cells. This indicates that, if the C-terminal domain is released from the C525Y protein by proteolysis, then it is either not secreted or not in a conformation that the Mab C antibody can recognize.
Pulse chase experiments were performed to compare the secretion rates of the mutant proteins with that of wild-type AMH (Fig. 2C and Table 2). Consistent with the Western blot results of Fig. 2A, the R194C mutant protein was not secreted and the H506Q mutant protein was secreted at a much reduced level (<1% of wild-type AMH). The V477A mutant protein was also secreted at a lower level. For the mutant proteins that were secreted at close to normal levels, two distinct kinetic profiles of secretion could be discerned. The R451T and Q496H mutant proteins were secreted at a relatively slow rate similar to the wild-type protein, with a gradual accumulation of the labeled protein in the medium observed at longer chase times. In contrast, the 2277–2292del and C525Y mutant proteins were secreted at a faster rate, with a maximal level of accumulation of labeled protein observed after a 3-h chase. The similarity in the rate of secretion of the 2277–2292del and C525Y mutant proteins provides further evidence that the C525Y protein is physically similar to the 2277–2292del protein. Because both of these mutant proteins have no C-terminal domain, these data suggest that the longer time needed for secretion of the full-length AMH molecule may be due to the presence of the C-terminal domain.
The Presence of the C-Terminal Domain Affects the Rate of Secretion
To further evaluate the role of the C-terminal domain on the secretion rate, we characterized the secretion of wild-type AMH and the S452stop mutant protein in Chinese hamster ovary (CHO) cells. The S452stop mutant cDNA encodes for the complete N-terminal domain of AMH, but without the 22 out-of-frame amino acids included at the C terminus of the 2277–2292del mutant protein (Table 1). Previous work has indicated that the protein produced by the S452stop mutation is secreted from CHO cells as a dimer (7).
A time course of secretion of wild-type AMH from CHO cells is shown in Fig. 3A. After a 1-h chase, all of the wild-type AMH was found inside the cells. A small amount of AMH was found in the medium after 2 h, and increasing amounts were seen with longer chase times. A significant amount of AMH was retained by the cells even after a 5-h chase. The rate of secretion of the S452stop mutant polypeptide from CHO cells is shown in Fig. 3B. In contrast to the full-length wild-type protein, secretion of the N-terminal domain polypeptide was almost complete by 2 h. A quantification of the radioactive signals in Fig. 3, A and B, clearly shows that the rate of secretion of the N-terminal fragment is faster than wild-type AMH (Fig. 3C). It should also be noted that the N-terminal fragment generated by proteolytic processing was secreted with similar kinetics as the wild-type AMH protein (Fig. 3A; culture medium). In this case, the N-terminal cleavage fragment would still be complexed to the C-terminal domain, unlike the N-terminal fragment encoded by the S452stop mutation.
|
, all AMH retained by the CHO cells after a 20-h chase was sensitive to Endoglycosidase H.
|
). AMH from COS cells was sensitive to Endoglycosidase H and N-glycosidase F in cell lysates, and insensitive to Endoglycosidase H but sensitive to N-glycosidase F in culture medium. To provide further support for the ER being the intracellular location of the wild-type AMH protein, we characterized the intracellular location by confocal microscopy. Figure 4C demonstrates a clear colocalization between AMH and rhodamine 6G, a marker of ER proteins (18). The longer retention time of full-length AMH within the ER, compared with the N-terminal fragment, implies that the wild-type AMH protein folds more slowly than the N-terminal fragment.
Perturbations in Proteolytic Processing and Glycosylation Do Not Affect the Rate of Secretion
The effect of proteolytic processing on the rate of secretion of AMH was investigated using the uncleavable R451T mutant protein. This mutant is secreted as a 140-kDa dimer of two 70-kDa polypeptides that appears indistinguishable from the wild-type 140-kDa dimer, but is resistant to cleavage that is seen at low levels for wild-type AMH and to plasmin-induced cleavage, which converts 100% of the wild-type AMH to the mature form (data not shown). The rate of secretion of the R451T mutant protein from CHO cells was similar to that of wild-type AMH, as indicated by the pulse chase experiment shown in Fig. 5A. No secretion of the mutant polypeptide occurred within 1 h, and 50% was still retained by the cells after a 5-h chase. The data for the R451T mutant protein revealed that the rate of secretion of AMH is not affected by whether or not it can be cleaved.
|
). Although less AMH was produced in the presence of tunicamycin, the same ratio of extracellular to intracellular AMH was observed at 4 h and 10 h, regardless of whether tunicamycin was present or absent. Thus N-linked glycosylation is not required for efficient secretion of AMH.
Molecular Modeling of the AMH C-Terminal Domain
To further investigate the nature of the four mutations within the C-terminal domain, we generated a three-dimensional model of the C-terminal domain dimer by comparative modeling using the Modeler module (19) of the InsightII software package (Accelrys, Inc., San Diego, CA). The x-ray structures of BMP2 (20, 21) and BMP7 (22, 23) were used as templates in the calculation of the model, together with the multiple sequence alignment shown in Fig. 6A. Figure 6B shows a view of this model with residues V477, Q496, H506, and C525 highlighted. The model reveals the cysteine knot motif that is characteristic of all TGF-ß family members, as well as the characteristic 
-helices and ß-sheets. V477 is located at the interface between the two ß-sheets of an AMH monomer, Q496 is located in the pre-helix loop between residues 490 and 505, and H506 is located at the N-terminal end of the -helix, at the interface between two AMH monomers. C525 forms the interchain disulfide linking the two monomers. The potential effects of the four mutations that alter these amino acids on the structure of the C-terminal domain are discussed below.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Two of the N-terminal domain gene mutations, 350–354del and 2277–2292del, are frameshift mutations and generate truncated proteins that are missing the C-terminal domain, which explains their lack of bioactivity. The shorter 350–354del mutant protein was not secreted, in contrast to the 2277–2292del mutant protein, which encodes almost the entire N-terminal domain (Fig. 2A). Pulse-chase experiments (Fig. 2C) showed that the 2277–2292del mutant protein was secreted at a faster rate than wild-type AMH in COS cells. This was confirmed by comparing the rates of secretion of wild-type AMH and the N-terminal fragment, encoded by the S452stop mutant cDNA, in CHO cells, which clearly demonstrated the faster secretion of the N-terminal fragment (Fig. 3). The relatively slow kinetics of secretion of the full-length AMH molecule resemble that of TGF-ß1 (25) and activin (26). For AMH, we have shown that before secretion, the full-length AMH protein is retained within the ER (Fig. 4), implying the intracellular protein has not yet attained its proper conformation. Studies on TGF-ß and activin have shown that the C-terminal domain cannot dimerize and be secreted when expressed in cells by itself, but can be secreted as a dimer when the N-terminal domain is supplied in trans (15). When combined with the results of Gray and Mason (15), our studies suggest that the folding of the C-terminal domain may be the rate-limiting step in the secretion of AMH.
Constam and Robertson (27) have shown that secretion of BMP4 into the culture medium of transfected COS cells is minimal after 20 h, but can be enhanced by cotransfection with proteolytic enzymes, leading to the suggestion that inefficient processing is responsible for the poor secretion. This does not appear to be the reason for the slow secretion of AMH, although it is true that processing of the AMH precursor is very inefficient, with only 5–10% of the secreted protein being cleaved. In Fig. 3A, the N-terminal cleavage product accumulated in the culture medium with the same kinetics as the full-length protein. If processing permitted faster export, one would expect the cleaved AMH to be released from the cells at a faster rate than the full-length protein. Also, the uncleavable mutant, R451T, was secreted at a rate indistinguishable from that of the wild-type protein (Fig. 5A), providing further evidence that processing does not affect the rate of secretion.
The third N-terminal domain mutation, R194C, failed to be secreted (Fig. 2A) and was present intracellularly at a lower level than wild-type AMH (Fig. 1, C and D). The mutant R194C protein could not be detected with Mab N2 or Mab C (Fig. 1C), indicating an abnormal conformation of the mutant protein. Because the mutation introduces an extra cysteine into the protein, some improper disulfide bonding may have occurred. The reduced intracellular level of the R194C mutant protein compared with wild-type AMH (Fig. 1D) suggests that the improperly folded protein may be unstable. Interestingly, a comparable mutation in the N-terminal domain of TGF-ß1, R218C, leads to premature activation of the latent TGF-ß1 complex and to Camurati-Engelmann disease (28, 29). This mutation does not affect the ability of the protein to fold, but rather the ability of the N-terminal domain to associate with the C-terminal domain and render it latent.
The four mutations within the C-terminal domain are all missense point mutations. To assist in evaluating the effects of these mutations on the structure of the C-terminal domain, a molecular model was generated using the structures of BMP2 and BMP7 as templates (Fig. 6B). V477 is located at the interface between the two ß-sheets of an AMH monomer, and part of a small hydrophobic packing core with the side chains of L471 and L536. The V477 side chain does not point toward the interaction interface with the type II receptor as found in the TGF-ß3/TGF-ß receptor II complex (30). Although the effect of mutating V477 to alanine is unlikely to affect receptor binding directly, perturbation of the small hydrophobic core could affect the folding or stability of mature AMH, which may explain the low level of the V477A mutant protein detected in COS cell medium and in patient serum (Fig. 2A and Tables 1 and 2). An unresolved issue is whether the V477A protein is biologically inactive or whether the reduced level of this protein causes PMDS in this patient.
Q496 is located in the prehelix loop between residues 490 and 505. This loop in BMP2 makes an important contact with the type I receptor, BMPRI-A (also known as ALK-3) (21, 31). Mutation of Q496 to a histidine may disturb the interaction with the AMH type I receptor directly and damage the function of the AMH protein. Given that Q496 is a solvent-exposed residue in a loop of the AMH model structure, it is unlikely that mutation to histidine would cause any folding or stability problems. This notion agrees with our experimental results, in which the Q496H mutant is expressed and secreted normally from COS cells (Fig. 2A and Table 2) and was detected at normal levels in PMDS patient serum (Table 1).
H506 is located at the N-terminal end of the -helix, which consists of residues 506–518, at the interface between two AMH monomers. It is unlikely to be involved in a direct interaction with the receptor(s). Mutation of H506 to glutamine could disrupt proper association of the dimer or could destabilize the structure at the interface of the dimer, possibly explaining the absence of this mutant in the culture medium and in patient serum (Fig. 2A and Tables 1 and 2). Proteins retained in the ER because they have not achieved a proper conformation may either form undegradable aggregates (32) or undergo retrotranslocation into the cytoplasm, before degradation in the proteasome (33, 34). The level of the H506Q mutant protein inside the cell was similar to that of the wild-type protein, as indicated by Western blot analysis (Fig. 1C) and confocal microscopy (Fig. 1D), suggesting that it may undergo retrotranslocation. This may also be true for the V477A mutation, which has properties similar to the H506Q mutation.
The properties of the C525Y mutant are hard to reconcile. In the molecular model, cysteine 525 forms an interchain disulfide bond linking the two monomers. The C525Y mutation precludes the formation of this important disulfide bond and may destabilize the protein. Although conserved in most members of the TGF-ß family, BMP15 and GDF9 lack this conserved cysteine but still form functional noncovalent dimers (35). For this reason, the loss of a cysteine at this position in AMH might not affect dimerization. Surprisingly, the C525Y mutation generates a protein that is truncated at the C terminus at or near the monobasic cleavage site (Figs. 1C, 2A, and 2B). This could be the result of proteolysis or arrested translation. The C525Y in vitro translation product appears to be smaller than expected (Fig. 1A), but larger than the C525Y protein secreted by COS cells (Fig. 2A), suggesting that the C525Y mutation does not arrest translation near position 451 of the monobasic cleavage site. One possibility is that the mutation arrests translation at position 525, and that this truncated protein subsequently undergoes further proteolysis at or close to the monobasic cleavage site. Alternatively, the C525Y mutation may directly render the protein more susceptible to proteolysis. AMH is unique among TGF-ß family members in that proteolytic cleavage at the monobasic site is inefficient. If the C-terminal domain is released through proteolysis, then the C525Y mutation must also render the C-terminal domain unstable or incapable of assuming a conformation that can be recognized by the Mab C antibody, because efforts to detect it have failed (Fig. 2B).
The salient features of the seven PMDS mutants are summarized in Table 2. Three of the mutants (highlighted in yellow), 350–354del, 2277–2292del, and C525Y, are missing all or part of the C-terminal domain and are therefore incapable of interacting with the AMH receptor(s). Three of the mutants (highlighted in blue), R194C, V477A, and H506Q, have potential problems with folding. Whereas R194C is not secreted, H506Q and V477A are secreted at a low level. Finally, mutant Q496H (highlighted in red) appears to be completely normal with regard to biosynthesis and secretion, but may not be able to interact appropriately with the type I receptor. Our study is the first one addressing the functional defects of AMH mutant proteins. Because natural mutations of the other members of the TGF-ß superfamily are extremely rare, this work may provide new insight into other members of the TGF-ß superfamily.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
DNA Transfection
COS-7 cells were cultured at 37 C under 5% CO2 in DMEM with Glutamax-I, 4.5 g/liter D-glucose, and 4 mg/liter pyridoxine-HCl (Life Technologies, Gaithersburg, MD) and containing 10% fetal bovine serum (Life Technologies), 100 U/ml penicillin, and 100 µg/ml streptomycin (Life Technologies). Twenty-four hours before transfection, COS-7 cells were seeded in six-well plates at 3 x 105 or in 100-mm plates at 106 cells. Transient transfection was performed with the diethylaminoethyl-dextran/chloroquine method as previously described (8).
The CHO cell line L2–58, which produces wild-type human AMH, has already been described (7). CHO cell lines L7–118 and L9–16C, which stably express the S452stop and the R451T mutant AMH proteins, respectively, were generated in a similar way. Briefly, CHO cells lacking the dihydrofolate reductase gene (38) were propagated on 100-mm dishes in (+)-MEM (Life Technologies) containing 10% fetal bovine serum. After 24 h, the cells (
50% confluent) were transfected with 10 µg linearized plasmid DNA /dish by the calcium phosphate procedure (39) with minor modifications. Forty-eight hours after transfection, cells were transferred to (-)-MEM (Life Technologies) lacking ribonucleosides and deoxyribonucleosides and containing 10% dialyzed fetal bovine serum. After 2 wk, CHO cell clones expressing AMH mutant proteins were identified with an ELISA, which employs an N-terminal domain-specific antibody, Mab 10.6, and a polyclonal antibody raised against the full-length AMH protein (40). All three CHO cell lines expressed in the range of 0.1–0.5 mg protein/liter of culture medium per d.
In Vitro Translation
The mutant AMH cDNAs in pCDM8 (20 ng/µl) were incubated for 90 min in a transcription-translation-coupled rabbit reticulocyte lysate system (Promega Corp.) in the presence of T7 RNA polymerase and 20 µCi L-[35S]methionine (1000 Ci/mmol; Amersham Pharmacia Biotech, Piscataway, NJ), following the manufacturer’s protocol. Reaction mixtures (5 µl) were resolved by SDS-PAGE on 4–20% gradient gels (Bio-Rad Laboratories, Hercules, CA) and analyzed by autoradiography.
Immunoprecipitation
COS cells were transfected with wild-type and mutant AMH cDNAs and incubated for various times in the appropriate medium. The cells were washed and lysed as previously described (41). AMH was recovered from cell lysates or culture medium by immunoprecipitation using anti-AMH Mab 10.6 or Mab C (5µg/ml), followed by adsorption to protein A-sepharose beads (Amersham Pharmacia Biotech) for 90 min at 4 C with agitation. The immunoprecipitates were washed five times with cold PBS, treated with electrophoresis sample buffer, and analyzed by Western blotting or fluorography, as indicated.
Western Blot Analysis
The protein concentration of cell lysates and culture medium was determined using the BCA assay (Pierce Chemical Co., Rockford, IL). Immunoprecipitates (8 µg total protein) of cell lysate, or culture medium were loaded on 7.5% or 4–20% SDS-PAGE (Bio-Rad Laboratories) under reducing conditions [3% (vol/vol) 2ß-mercaptoethanol] and transferred onto a Protran BA85 nitrocellulose membrane (Schleicher & Schuell, Keene, NH) as previously described (41). AMH was detected using anti-AMH Mabs at 1 µg/ml and horseradish peroxidase-labeled antimouse IgG at 1:5000 (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA). Bands were visualized with the enhanced chemiluminescence detection reagent (Amersham Pharmacia Biotech).
Epitope Mapping of Mabs against Recombinant Human AMH
Mouse anti-AMH monoclonal antibody Mab 10.6 has been previously described (7). Mabs N2 and C were obtained by immunizing female BALB/c mice with human recombinant AMH. Antibody specificity was determined by probing Western blots containing the AMH N- and C-terminal fragments (200 ng/lane), prepared as described (7). Western blotting was performed as described above.
Metabolic Labeling
COS cells transfected with wild-type or mutant AMH cDNAs were plated in 60-mm dishes 1 d before labeling. After reaching 70% confluency, the cells were washed three times with Hank’s balanced salt solution (Life Technologies). COS cells were incubated with methionine and cysteine-free DMEM (ICN Biomedicals, Costa Mesa, CA) containing 100 µCi/ml L-[35S]methionine and cysteine (1000 Ci/mmol; ICN Biomedicals) for 45 min at 37 C. After labeling, the plates were cultured for different periods of time in the same medium supplemented with unlabeled methionine and cysteine. The supernatants were collected and the cell monolayers were washed twice with ice-cold PBS and lysed with 1 ml lysis buffer (20 mM Tris, pH 7.2; 50 mM NaCl; 0.5% Nonidet P-40; and 0.5% sodium deoxycholate)/dish. AMH was recovered from the supernatants and cell lysates by immunoprecipitation using Mab 10.6 (5 µg/ml), followed by adsorption to protein A-sepharose beads (Amersham Pharmacia Biotech) for 90 min at 4 C. Immunoprecipitates were washed five times with cold PBS. Labeled proteins were eluted from the beads by boiling in Laemmli buffer in nonreducing and/or reducing conditions [3% (vol/vol) 2ß-mercaptoethanol], subjected to SDS-PAGE on 7.5% or 10% gels, and visualized by autoradiography.
Labeling of CHO cells was performed in a similar way except for the following modifications. L-[35S]cysteine at 500 µCi/ml (1000 Ci/mmol; PerkinElmer, Norwalk, CT) was used to label the cells for 15 min. AMH was recovered from the supernatants and cell lysates using Mab 10.6 and antimouse IgG agarose (Sigma-Aldrich Corp., St. Louis, MO), eluted from the beads at 65 C, and treated as above.
Endoglycosidase H, N-Glycosidase F, and Tunicamycin Treatments
In selected studies, CHO cells were pretreated with 10 µg/ml tunicamycin (Sigma-Aldrich Corp.), 1 h before labeling with L-[35S]cysteine. Metabolic labeling was performed as described above except that tunicamycin at 10 µg/ml was included during the pulse and the subsequent chase. In other selected studies, the immunoprecipitates of metabolically labeled intracellular proteins were eluted from protein A-sepharose or antimouse IgG agarose by heating samples at 65 C in 0.1 M Tris (pH 7.5), 0.5% sodium dodecyl sulfate. The protein samples were treated with Endoglycosidase H or N-glycosidase F (Roche Diagnostics, Indianapolis, IN) in the buffer conditions recommended by the manufacturer. The digestion products were analyzed by SDS-PAGE under reducing conditions as described above. Control samples were treated exactly like the test samples but without the addition of the glycosidase.
AMH ELISA
AMH levels in the serum of PMDS patients or in culture medium of COS cells were measured using a commercial ELISA kit (Immunotech-Coulter-Beckman, Marseilles, France). Plates were coated with Mab C (22A2), which recognizes the C-terminal domain of AMH, and probed with biotinylated Mab N2 (11F8), which recognizes the N-terminal domain (42).
Confocal Microscopy Analysis
Transfected COS cells were seeded in four-dish permanox Labtec chambers at 0.5 x 105 cells. After 2 d, cells were washed, fixed, and permeabilized by incubation with methanol/acetone (vol/vol) for 2 min at room temperature. After the cells were blocked with PBS containing 10% goat serum (Life Technologies) at room temperature for 1 h at 4 C, they were incubated overnight with Mab 10.6 in PBS containing 3% BSA, and then washed with PBS three times and incubated with a fluorescein isothiocyanate (FITC)-labeled antimouse IgG (Jackson ImmunoResearch Laboratories) for 2 h at 4 C. After two washes, cells were successively labeled with 0.5 µM rhodamine 6G (Molecular Probes, Eugene, OR) for 30 min at 4 C and with 2 µg/ml Hoechst (Sigma-Aldrich) for 10 min at room temperature.
Coverslips were mounted in Vectashield (Vector Laboratories, Inc., Burlingame, CA), and cells were examined with a Zeiss LSM-510 confocal scanning laser microscope (Jena, Germany) equipped with a 25-mW argon laser and a 1-mW helium-neon laser, using a plan apochromat x36 objective (oil immersion). Green fluorescence was observed with a 505- to 550-nm band-pass emission filter under 488-nm laser illumination, and red fluorescence was observed with a 560 long-pass emission filter under 543-nm laser illumination. Hoechst staining was detected using the HBO mercury lamp (50 W) and a set filter 01 (excitation BP 365/12, beamsplitter FT 395, emission LP 397).
Slices were collected every 0.4 µm along the z-axis. Projections of three medium images for FITC and rhodamine 6G were done for each sample. Simultaneous images corresponding to rhodamine 6G and Hoechst were obtained using the multitracking function of the microscope.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
This work was supported in part by a grant from the Association pour la Recherche sur le Cancer (no. 4253 to N.D.).
Abbreviations: ALK, Activin receptor-like kinase; AMH, anti-Müllerian hormone; BMP, bone morphogenetic protein; CHO, Chinese hamster ovary; ER, endoplasmic reticulum; FITC, fluorescein isothiocyanate; Mab, monoclonal antibody; PMDS, persistent Müllerian duct syndrome.
Received for publication September 15, 2003. Accepted for publication November 25, 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
-1 antitrypsin deficiency alleles and severe cystic fibrosis lung disease. Thorax 53:1022–1024This article has been cited by other articles:
 |
|
 |
|
  K. L. Walton, Y. Makanji, M. C. Wilce, K. L. Chan, D. M. Robertson, and C. A. Harrison A Common Biosynthetic Pathway Governs the Dimerization and Secretion of Inhibin and Related Transforming Growth Factor {beta} (TGF{beta}) Ligands J. Biol. Chem., April 3, 2009; 284(14): 9311 - 9320. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Wu, S. Wan, S. Pujar, M. E. Haskins, D. H. Schlafer, M. M. Lee, and V. N. Meyers-Wallen A Single Base Pair Mutation Encoding a Premature Stop Codon in the MIS Type II Receptor Is Responsible for Canine Persistent Mullerian Duct Syndrome J Androl, January 1, 2009; 30(1): 46 - 56. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Catteau-Jonard, S. P. Jamin, A. Leclerc, J. Gonzales, D. Dewailly, and N. di Clemente Anti-Mullerian Hormone, Its Receptor, FSH Receptor, and Androgen Receptor Genes Are Overexpressed by Granulosa Cells from Stimulated Follicles in Women with Polycystic Ovary Syndrome J. Clin. Endocrinol. Metab., November 1, 2008; 93(11): 4456 - 4461. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. E. Kevenaar, J. S. E. Laven, S. L. Fong, A. G. Uitterlinden, F. H. de Jong, A. P. N. Themmen, and J. A. Visser A Functional Anti-Mullerian Hormone Gene Polymorphism Is Associated with Follicle Number and Androgen Levels in Polycystic Ovary Syndrome Patients J. Clin. Endocrinol. Metab., April 1, 2008; 93(4): 1310 - 1316. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
U. Goglia, C. Toncini, M. Giusti, C. Gastaldi, C. Ambruosi, S. Sola, J.-L. Ravetti, G. Carmignani, F. Minuto, and D. Ferone A Unique Association of Clinical "Persistent Mullerian Duct Syndrome" and Syringoid Carcinoma of the Perineal-Scrotal Skin: A Consequence of Urologic Surgery? J Androl, January 1, 2008; 29(1): 15 - 19. [Full Text] [PDF]
|
|
|
|
 |
|
 N. Josso, C. Belville, N. di Clemente, and J.-Y. Picard AMH and AMH receptor defects in persistent Mullerian duct syndrome Hum. Reprod. Update, July 1, 2005; 11(4): 351 - 356. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. X. Liao, R. K. Moore, and S. Shimasaki Functional and Molecular Characterization of Naturally Occurring Mutations in the Oocyte-secreted Factors Bone Morphogenetic Protein-15 and Growth and Differentiation Factor-9 J. Biol. Chem., April 23, 2004; 279(17): 17391 - 17396. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
