| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
School of Biomedical Sciences and Institute for Molecular Bioscience (Y.W., R.B., M.J.W.) and School of Molecular & Microbial Science (Y.Z.Z.), University of Queensland, Brisbane, Queensland 4072, Australia; and School of Life Science (J.M.H.), Queensland University of Technology, Brisbane, 4000 Queensland, Australia
Address all correspondence and requests for reprints to: Professor M. J. Waters, E-mail: m.waters{at}mailbox.uq.edu.au.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
To gain insight into how MAb 263 can activate the GHR, this study has sought to map the antibody binding surface on the ECD of the rabbit GHR, to which MAb 263 binds most effectively. There are currently two standard approaches to epitope mapping. The first involves screening against a random peptide library or against a series of overlapping peptides corresponding to the sequence of the target protein, as used, for example, with the bovine (b) GHR (18) or the EPO receptor (19). The second approach involves swapping homologous segments between species, where one species of the target is immunoreactive and the other is not, followed by alanine scanning within differing segments, as in the case of the interferon 
receptor 
-chain (20). Both methods suffer from the disadvantage that they focus on linear sequences and are of limited utility for determining complex conformational epitopes. We have used a new mapping procedure that is ideally suited to conformational epitopes and that provides a great deal of information without undue effort. The procedure involves random PCR mutagenesis of the entire protein by controlled nucleotide depletion to generate a library, followed by expression screening in yeast with the MAb of interest, in parallel with an anticarboxy-terminal tag antibody to eliminate mutations that prevent expression of the full-length protein.
Because there is no crystal structure for MAb 263, the resulting GHR epitope has been subjected to molecular dynamics docking with an isotypic bivalent MAb for which the crystal structure is known. This has shown that the docked MAb can align the receptor subunits in a manner similar to GH itself.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
. After eliminating clones with residue mutations that may cause breaking or creation of disulfide bonds (C83R, C94S, C108Y, and Y100C) (21) and mutations that would affect stability of the ß sandwich structure of domain 2 [W139 and F225 (22)], 32 sequences with a single amino acid mutation remained, and from these 20 different epitope residues were identified (Table 1). In addition, 46 sequences with double or triple mutations were used to confirm these single mutations. As shown in Fig. 2, almost all of the epitope residues were discontinuously distributed, with only one 3-residue cluster (S88-A89-G90). Eleven of the 20 residues were concentrated between E79 and F96, with the others being scattered throughout the center of the ECD sequence. A sequence comparison of rabbit (rb) GHR ECD with other species (Fig. 3) showed that the MAb 263 epitope residues are 100% conserved for rabbit, pig, human, monkey, panda, bovine, and ovine, and 95% conserved for rat and mouse (Fig. 3).
|
|
|
|
shows the crystal structure of pGH(rbGHR)2 trimeric complex obtained by homology modeling from a human counterpart (3HHR). The complex consists of one molecule of GH and two molecules of GHR ECD, each of which comprise two distinct ß-sandwich modules, domains 1 and 2. The MAb 263 epitope residues are located within domain 1, being particularly concentrated in the D and E strands and the ß-turn loop between them (residues E79, W80, E82, D85, Y86 S88, A89, G90, N92, Y95, and F96, indicated in red). This ß-loop is located on the top and side of domain 1 and is stabilized by a disulfide bond between Cys83 and Cys94 (21). A number of other epitope residues (S102, F123, S124, I128, S40, and S47, indicated in blue) are located adjacent to, but not overlapping, the GH-binding site (23). These residues are located on the loops separated by ß-strands but come into close spatial proximity as a result of folding of the loops, maintained by two disulfide bonds (Cys108–Cys122 and Cys38–Cys48) (21, 23). Based on alanine scanning, we previously reported involvement of residues within the Cys38–Cys 48 disulfide loop in the MAb 263 epitope (24).
|
).
|
). A control guinea pig anti-pGHBP polyclonal antibody did remain immunoreactive, although the reaction intensity was somewhat decreased in the disulfide-cleaved pGHBP relative to untreated pGHBP. Treatment of pGHR ECD with sodium dodecyl sulfate (SDS) alone had little effect on the interaction of the pGHBP with these antibodies.
|
, even at high concentration, this Fab was unable to produce GHR activation (as indicated by increased cell proliferation), whereas the intact MAb was able to do so. However, the Fab fragment was still able to compete for binding with the GHR (32.0 ± 3% inhibition of specific [125I]bGH binding to rabbit liver membranes compared with 32.6 ± 2.5% for MAb 263 at 12.5 nM, assayed according to Ref. 27). This was verified by showing that the Fab acts as an antagonist for human GH (hGH)-stimulated proliferation in the cells stably expressing rbGHR (Fig. 7B). When the Fab fragments were cross-linked with antimouse Fab in the bioassay, the proliferative response was reconstituted (Fig. 7A).
|
. Molecular dynamic simulation under constraints from the bound IgG resulted in distortion of the rbGHR as shown in Fig. 8B, producing marked torsion of the lower domain of the rbGHR. Additionally, there was noticeable compression of the upper domains of the ECD. However, the receptor subunit alignment in the minimized model is similar to that seen with GH bound in the trimeric complex. This is shown by overlay of these two forms in Fig. 8B.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
-chain, recognized by the neutralizing MAb A6, which comprises 22 residues (20). It is of interest that the agonist MAb 263 epitope corresponds to the weakest epitope (epitope 3) identified in a peptide scan of polyclonal antisera to the bGHR ECD, which could explain the difficulty in obtaining agonist antibodies to the GHR (18). In addition to MAb 263, only two other monoclonal antibodies recognizing different regions of the GHR ECD molecule have been reported to be GH agonists for full-length GHR, despite the dozens of MAbs that have been studied. An initial report by Wang et al. (16) found that of three antirat GHR ECD MAbs, only one, 2C3, was able to promote the growth of hypophysectomized rats, although in vitro proliferative activities were not assessed. In vivo proliferative potency of this antibody [0.5 g weight gain/d with a 160-µg dose (16)] was comparable to that reported for MAb 263 in hypophysectomized rats [1.1 g weight gain/d for a 200-µg dose (15)], which is similar to that achieved with 3 µg hGH/d, a 10-fold lower molar amount (15). 2C3 acted as a competitor for GH binding, leading these authors to propose that a precise targeting to the hormone binding site is needed for agonism (16). However, a subsequent study by the same group with at least eight MAbs to the rat GHR ECD identified others that competed strongly for GH binding but did not possess agonist activity (25). Thus, a precise targeting at the GH-binding sites does not appear to be sufficient for a GH agonist antibody, which is in agreement with the conclusions of our previous in vitro study with a panel of 14 GHR MAbs, several of which acted as competitors for hGH binding without acting as agonists (4). Indeed, based on our modeling, it would be difficult to correctly align the receptor subunits through binding to an epitope within the hormone-binding site, because of the size and inflexibility of the IgG protein. The third agonist MAb to the receptor ECD, an IgM raised against the human GH-binding protein (GHBP) by Mellado et al. (3), was the only one in a series that was found to stimulate the proliferation of human IM-9 lymphocytes expressing endogenous GHRs. The stimulation of proliferation was similar in extent to that found for MAb 263 with Ba/F03 lines expressing full-length GHR (4). These authors found that this MAb (GHR05) did not compete for hGH binding and stated that it was directed to the hinge domain between cytokine homology modules 1 and 2, although no information was provided to support this conclusion. Provocatively, GHR05 was claimed to discriminate between agonist (hGH) and antagonist (G120R hGH) bound forms of the hGH receptor on the cell surface, although it did not show such discrimination with the purified human GHR ECD. With the exception of the finding that targeting the GH binding site is not necessary for agonist activity, it is difficult to draw definite conclusions from these published studies with agonist GHR ECD antibodies.
The finding that the MAb 263 epitope does not occupy the GH-binding sites is in line with the previous observations that GH was unable to abolish binding of MAb 263 to GHR ECD (26) and that MAb 263 was able to mimic GH action either in the absence or in the presence of the hormone (15). However, a close spatial relationship of the antigenic determinant to the GH-binding region exists (there are seven epitope-residues closely adjacent to the GH-binding region), and this presumably accounts for the partial inhibition of GH binding to the GHR by MAb 263 when added before the radiolabeled hormone (26, 27).
As revealed in the present study, the majority of the MAb 263 epitope residues are discontinuously distributed on ß-turn loop residues 79–96 and the loops between ß-strands of domain 1. These are highly folded structures in domain 1, maintained by three disulfide bonds (21, 23), implying that the MAb 263 recognizes a conformational epitope. This hypothesis was confirmed by immunoblot analysis, which showed that MAb 263 does not react with rbGHBP molecules in which the disulfide bonds are broken, although a polyclonal anti-GHBP antibody did react (Fig. 6
). The view that MAb 263 is a conformation-sensitive antibody is also in line with the previous evidence that MAb263 performs poorly as a probe in immunoblots on which the GHR was reduced (28).
Our mutagenic analysis shows that not only does MAb 263 require disulfide bonds to maintain the conformational epitope, but a number of separate structural elements must be present together for effective binding. Notwithstanding this, because of the likely economic benefit, we have investigated whether immunization of young GH-deficient dwarf rats with the E79–F96 main epitope peptide could result in growth stimulation in the same way that passive immunization of rats with MAb 263 promotes growth (15). We found that immunization with correctly oxidized E79–F96 peptide attached via its amino terminus to either hemocyanin or BSA did not produce additional body weight gain over control immunizations (Wan, Y., and M. J. Waters, unpublished observations). Interestingly, the (low titer) antisera that resulted from these immunizations recognized the disulfide loop peptide, but not the intact receptor, unless this was denatured with 2% SDS. This may reflect the low surface accessibility of the peptide, because most of the epitope residues identified in this study have solvent exposure below 30%. Therefore, while the DE strand epitope loop is accessible to MAb 263, it appears to be difficult for this to react with polyclonal antibodies directed against it. This observation is suggestive of a specific conformational change in the ECD that is induced by MAb 263.
A sequence comparison of rbGHR ECD to those of other species revealed that MAb 263 epitope residues were highly homologous in rabbit, pig, human, monkey, panda, rat, murine, bovine, and ovine (Fig. 2). This explains why MAb 263 is able to react with GHR from multiple species and implies a widely applicable potential for these sequences as immunogens to enhance growth. The poor immunoreactivity of mouse GH receptor for MAb 263 (26) could result from the murine exon 4b insertion of eight residues adjacent to the main epitope and/or the substitution of lysine at residue 99 within the main epitope sequence.
A central issue here is how a knowledge of the epitope for an agonist antibody can assist in understanding the receptor activation process. As previously demonstrated (4), ability of a GHR MAb to dimerize the receptor alone is not sufficient for activation. Indeed, we (Ref. 30 and Brown, R., K. Palethorpe, T. Monks, K. Eidne, and M. J. Waters, unpublished) and others (29) find that at the cell surface, the GH receptor exists as a dimer before GH binding. The issue of activation then becomes: what conformational change is required for receptor activation? The major part of the MAb 263 epitope is located on the side and top of cytokine receptor module 1, a disposition that would enable the antibody to grasp the two receptors in a pincer-like movement on opposite sides. Recent findings have demonstrated that IgG has the conformational flexibility required to coordinate both faces of a single rbGHR molecule (31, 32). Our modeling indicates that this places the receptors in an appropriate alignment for activation of JAK2, i.e. in a position similar to that seen in the hormone-bound complex, involving approximation of the receptor dimerization domains. The lack of activity of the Fab fragment alone suggests that some kind of torsional force may be required to align the two receptors in a signaling-competent manner. This is in agreement with the use of two receptor-binding sites by the GH molecule, and the ability of a cross-linking anti-Fab antibody to elicit receptor activation. However, alternative and complementary mechanisms may be involved: for example, the apparent increase in affinity for MAb 263 after substitution of the key GH-binding residue, W104, could be indicative of a conformational change triggered by steric hindrance between MAb 263 and W104, perhaps inducing a conformational change in the receptor analogous to that induced by GH binding to this important binding residue. A second possibility relates to the increased apparent affinity for MAb 263 seen on removal of the amino-terminal GHR ECD sequence. Recent studies with the ovine GHR chimeras have indicated that the amino-terminal sequence, which is not resolved in the crystal structure, may fold back into the hormone-binding region adjacent to lysine 41 in the first minihelix, and contribute binding residues (33). Such a folding back of the amino terminus is seen with the homologous prolactin receptor (34). In the case of the rabbit GHR studied here, it may be that MAb 263 is able to induce a conformational change in the amino terminus through steric hindrance similar to that induced by the hormone, resulting in receptor activation. Consistent with this, our molecular modeling indicates that binding of an IgG to both subunits of rbGHR imposes spatial restraints (flattening of domain 1) that result in repositioning of the lower cytokine modules. The fact that MAb 263 is a relatively weak agonist compared with hGH may be a result of the lower module alignment not being precisely the same as that observed with GH bound (23).
In summary, the present study has determined for the first time the precise epitope residues for MAb 263 and mapped them onto the crystal structure model of the receptor molecule. The data explain several important features of the monoclonal antibody and reveal potential mechanisms underlying its agonist action.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
-factor for extracellular secretion in the yeast host. The rbGHR ECD coding sequence in this plasmid was used as a template for random PCR mutagenesis after denaturation in a buffer containing 1 M NaOH and 1 mM EDTA. The template and the primers (YaN21: 5'-AGC ACA AAT AAC GGG TTA TTG-3' and YcC21: 5'-TAC AGA CGC GTG TAC GCA TGT-3', Sigma Chemical Co., St. Louis, MO) were then subjected to PCR using a protocol in which the concentrations of each of the four dNTPs was individually reduced sequentially from 200 µM to 100 µM, 50 µM, 25 µM, 12.5 µM, 6.25 µM, and finally 3.125 µM, so as to induce mutagenesis by dNTP depletion. For each reaction the PCR products were religated into Yep and cloned in E. coli (XL-Blue I). Ten clones from each concentration of the four dNTPs were then sequenced to determine the optimal nucleotide concentration at which more than 80% single or double mutations were produced.
Creating a rbGHR ECD Mutant Library for MAb 263 Epitope Screening
The selected mutants from reactions with the optimal concentration of each dNTP were transformed into the BJ3505 yeast, and the positive clones were selected on complete supplement mixture minus tryptophan (45 11–012, BIO 101, Inc., Vista, CA) with 2% glucose. More than 5000 colonies were then picked into 96-well microplates to establish a library of rbGHR ECD mutants. The colonies from this library were replica plated onto cellulose acetate filters in 144-mm culture dishes using a multichannel replica plater. After 3 d in culture, the filters with yeast colonies were placed onto a sheet of Hybond-C extra membrane (RPN137E, Amersham, Buckinghamshire, UK) on complete supplement mixture without glucose to allow the protein products secreted by the yeast colonies on the filter to be transferred onto the Hybond membrane. After the membranes were blocked with 6% nonfat milk in PBS with 0.03% Tween 20, they were screened with 2.5 µg/ml of MAb 263 (Agen, Acacia Ridge, Australia), and bound antibody was detected using a horseradish peroxidase-conjugated antimouse antibody together with enhanced chemiluminescence Western blotting detection reagents (RPN 2106; Amersham). The resulting films were scanned into digital images, and the negative dots were determined with image analysis software (Mocha.1.4; Jandel Scientific, San Rafael, CA). In each film the average integrated gray density of three negative control dots was taken as a threshold by which the computer automatically recognized the dots with an integrated gray density higher than the threshold value as "positive." The negative colonies were then confirmed with MAb 263 and selected for C-terminal tag screening.
Screening of MAb 263-Negative Colonies for C-Terminal Tag
To eliminate prematurely terminated or unexpressed mutants from the screen, colonies negative for MAb 263 were screened with a monoclonal antibody to C-terminal tag (8E7/55; Queensland Institute of Medical Research, Queensland, Australia) in the manner described above. Plasmids were then isolated from C tag-positive, MAb 263 negative stock colonies and sequenced by ABI autosequencing (Applied Biosystems Inc., Columbia, MD). The resulting sequences were analyzed using Sequencher 3.1.1 and MacVector 6.5.1 to identify the specific residues responsible for the MAb 263 epitope.
Mapping of MAb 263 Epitope onto The Crystal Structure Model of GHR ECD
MAb 263 epitope residues were mapped onto a homology-modeled crystal structure of the GH(rbGHR)2 trimeric complex. The model was created by homology modeling from the structure of the human counterpart [PDB 3HHR (23)] using the First Approach Mode and then the Optimize Mode of SWISS-MODE (35, 36) after manual alignment with SWISS-pdb Viewer Version 3.7.1.
Recombinant GHBP
Recombinant porcine GHBP was expressed in E. coli after mutating codon 21 (GGG to GGT), codon 25 (ACA to ACC), codon 28 (GTC to GTG), codon 29 (CTT to CTG), codon 30 (GTC to GTG), codon 31 (AGA to CGT), codon 38 (AGA to CGT), and codon 174 (AGA to CGT) to facilitate bacterial expression. The modified coding sequence for residues 1–238 was ligated into pET20b(+) vector and expressed in E. coli strain BL21(DE3). These cells were incubated in 500 ml of Terrific Broth medium by shaking at 200 rpm at 37 C in 2-liter flasks to an A600 of 0.8, after which isopropyl-ß-D-thiogalactopyranoside (1.0 mM) was added. The cells were incubated for an additional 3 h and harvested, after which phenylmethylsulfonylfluoride was added to 1 mM, and the paste was stored frozen at -70 C for less than 4 wk. In a modification of the method of Sakal et al. (37), the inclusion body pellet obtained from 2 liters of bacterial culture by French press lysis was solubilized in 400 ml of 4.5 M urea buffered with 10 mM Tris base (pH 10.4), L-cysteine was added to 0.1 mM, and the clear solution was stirred at 4 C for 48 h. This solution was then dialyzed for 48 h against five changes of 10 mM Tris-HCl (pH 8.1), at 4 C. The solution was subsequently loaded at 120 ml/h onto a Q-Sepharose column (1.6 x 20 cm), preequilibrated with 10 mM Tris-HCl (pH 8.1) at 4 C. Elution was carried out with a discontinuous NaCl gradient (0–0.4 M) in the same buffer. Fractions eluting between 0.05 and 0.25 M NaCl were analyzed by 12.5% SDS-PAGE in the presence or absence of reducing agents, and fractions judged to be greater than 95% purity were pooled.
Immunoreactivity of GHR ECD After Reduction and Alkylation of Disulfide Bonds
To determine the effect of breaking the three disulfide bonds in domain 1 on MAb 263 immunoreactivity, 20 µl recombinant GHBP (20 µg/ml in PBS) were boiled for 5 min either alone or with 2% SDS, with or without 100 mM dithiothreitol (DTT). The resulting samples were then placed onto Hybond-C extra membrane. After immersion in 75 mM iodoacetamide for 30 min, the membrane was washed with PBS, and then blotted with MAb 263 (2.5 µg/ml) or a guinea pig anti-pGHBP polyclonal antibody (1:600).
Purification of MAb 263 and Bioassay
Ba/F03 proliferation assay was carried out as previously described (4) using protein G-purified MAb 263 that had been bound to a 1 x 10 cm protein G Sepharose column in 20 mM sodium phosphate, pH 7.0, and after extensive washing with this buffer, eluted with 0.1 M glycine (pH 2.8) into 0.1 M sodium phosphate, pH 8.2. Purified MAb 263 was dialyzed extensively against PBS before use, as were other antibodies. The antimouse Fab (catalog no. M 6898) was obtained from Sigma Chemical Co. and was used at a final concentration of 3 µM.
Preparation of Fab Fragment
Digestion of Protein G-purified MAb 263 (10 mg/ml) with papain-agarose (catalog no. P4406, Sigma Chemical Co.) was performed in 0.1 M Na-phosphate buffer, pH 7.0, containing 10 mM cysteine and 2 mM EDTA at an enzyme-substrate ratio of 1 U to 2 mg for 3 h at 37 C. The digested solution was then loaded at 0.4 ml/min onto the protein G Sepharose column, preequilibrated with sodium phosphate buffer, pH 7.0, and eluted with the same buffer at 4 C to separate the Fab fragment from bound Fc and undigested antibody. The eluted fractions were analyzed by 12.5% SDS-PAGE in the presence or absence of reducing agents, and fractions containing the 55-kDa Fab fragment were concentrated by ultrafiltration. The resulting solution was then purified on a Superdex 200 fast protein liquid chromatography column to remove the high molecular weight material, and the peak eluting around 50 kDa was pooled and again concentrated by ultrafiltration. SDS-PAGE analysis of this material showed two bands corresponding to heavy- and light-chain fragments at 27 kDa and 28 kDa. In nonreducing SDS gels, a major band was observed at 55 kDa, with a minor band at 28 kDa.
Molecular Modeling
The structure of an IgG specific for HIV gp120 (Pdb id 1 hzh) was engineered in silico to preferentially bind to the acidic epitope of rabbit GHR as follows: complimentarity determining regions were scanned for residues forming theoretical contacts with gp120 as described (31, 32). Any acidic residues in these regions were substituted with lysine to force interaction with the acidic epitope mapped to residues 79–100 of the rbGHR. An Fab fragment was then separated from the rest of the molecule for use in docking studies. rbGHR as modeled above was prepared for docking with this modified fragment as follows: bound hormone was removed, side chain valences were satisfied by addition of hydrogen atoms, and incomplete loop regions were rebuilt from loop libraries. This refined structure was then used as a probe in a soft docking simulation conducted within the program BiGGER [Biomolecular complex Generation with Global Evaluation and Ranking (38)]. Soft docking simulates protein flexibility by relaxing the van der Waals energy term used to rank docked solutions. Results from BiGGER were clustered and further analyzed in Chemera, a specialized graphics program integrated with BiGGER. Solutions were scored according to geometric fit and electrostatic complementarity, and the top 50 solutions were clustered to produce a theoretical docked single Fab/rbGHR complex. The single Fab fragment was transformed using the build crystallographic symmetry facility of SPDBV3.5 to produce a mirror image Fab to coordinate with the second rbGHR subunit. The remaining IgG molecule was then rejoined with the Fab fragments using long false bonds and subjected to global minimization within the program ChemSite Pro. Energy minimization was conducted at constant temperature (27 C) for a total of 10 psec with time steps of 1 fsec; solvent was not included.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Abbreviations: b, Bovine; dNTP, Deoxynucleotide triphosphate; DTT, dithiothreitol; ECD, extracellular domain; G-CSF, granulocyte-colony stimulating factor; GHBP, GH-binding protein; GHR, GH receptor; h, human; MAb, monoclonal antibody; p, porcine; rb, rabbit; SDS, sodium dodecyl sulfate.
Received for publication April 30, 2003. Accepted for publication July 25, 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
receptor -chain by alanine scanning mutagenesis and random mutagenesis with phage display. Biochemistry 39:15674–15685[CrossRef][Medline]
This article has been cited by other articles:
 |
|
 |
|
  J. Frystyk, C. M. Andreasen, and S. Fisker Determination of Free Growth Hormone J. Clin. Endocrinol. Metab., August 1, 2008; 93(8): 3008 - 3014. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. L. Conway-Campbell, J. W. Wooh, A. J. Brooks, D. Gordon, R. J. Brown, A. M. Lichanska, H. S. Chin, C. L. Barton, G. M. Boyle, P. G. Parsons, et al. Nuclear targeting of the growth hormone receptor results in dysregulation of cell proliferation and tumorigenesis PNAS, August 14, 2007; 104(33): 13331 - 13336. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. E. Bowles, I. Wilkinson, R. A. G. Smith, A. J. G. Moir, H. Montgomery, and R. J. M. Ross Membrane Reinsertion of a Myristoyl-Peptidyl Anchored Extracellular Domain Growth Hormone Receptor Endocrinology, February 1, 2007; 148(2): 824 - 830. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M J Waters, H N Hoang, D P Fairlie, R A Pelekanos, and R J Brown New insights into growth hormone action J. Mol. Endocrinol., February 1, 2006; 36(1): 1 - 7. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Landsman and D. J. Waxman Role of the Cytokine-induced SH2 Domain-containing Protein CIS in Growth Hormone Receptor Internalization J. Biol. Chem., November 11, 2005; 280(45): 37471 - 37480. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Jiang, X. Wang, K. He, X. Li, C. Chen, P. P. Sayeski, M. J. Waters, and S. J. Frank A Conformationally Sensitive GHR [Growth Hormone (GH) Receptor] Antibody: Impact on GH Signaling and GHR Proteolysis Mol. Endocrinol., December 1, 2004; 18(12): 2981 - 2996. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Wan, A. McDevitt, B. Shen, M. L. Smythe, and M. J. Waters Increased Site 1 Affinity Improves Biopotency of Porcine Growth Hormone: EVIDENCE AGAINST DIFFUSION DEPENDENT RECEPTOR DIMERIZATION J. Biol. Chem., October 22, 2004; 279(43): 44775 - 44784. [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 |
