help button home button Endocrine Society JCEM
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Journal of Clinical Endocrinology & Metabolism , doi:10.1210/jc.2005-2558
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a related Letter to the Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
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 Fang, P.
Right arrow Articles by Rosenfeld, R. G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Fang, P.
Right arrow Articles by Rosenfeld, R. G.
Related Collections
Right arrow Neuroendocrinology and Pituitary
Right arrow Metabolism
The Journal of Clinical Endocrinology & Metabolism Vol. 91, No. 4 1526-1534
Copyright © 2006 by The Endocrine Society

A Mutant Signal Transducer and Activator of Transcription 5b, Associated with Growth Hormone Insensitivity and Insulin-Like Growth Factor-I Deficiency, Cannot Function as a Signal Transducer or Transcription Factor

Peng Fang1, Eric M. Kofoed1, Brian M. Little, Xiangdong Wang, Richard J. M. Ross, Stuart J. Frank, Vivian Hwa and Ron G. Rosenfeld

Department of Pediatrics (P.F., E.M.K., B.M.L., V.H., R.G.R.), Oregon Health and Science University, Portland, Oregon 97239-3098; Division of Clinical Sciences (R.J.M.R.), Sheffield University, Sheffield S5 7AU, United Kingdom; Departments of Medicine (X.W., S.J.F), Cell Biology (S.J.F.), and Physiology (S.J.F.), University of Alabama at Birmingham and Birmingham Veterans Affairs Medical Center (S.J.F.), Birmingham, Alabama 35294; Lucile Packard Foundation for Children’s Health (R.G.R.), Palo Alto, California 94304; and Department of Pediatrics (R.G.R.), Stanford University, Stanford, California, 94305-2038

Address all correspondence and requests for reprints to: Dr. Vivian Hwa, Department of Pediatrics, NRC5, Oregon Health and Science University, 3181 Southwest Sam Jackson Park Road, Portland, Oregon 97239-3098. E-mail: hwav{at}ohsu.edu.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Context: A natural missense mutation in the signal transducer and activator of transcription (STAT) 5b gene was recently identified in association with a female patient presenting with severe growth failure and immune dysfunction. The mutation results in an alanine to proline substitution at residue 630 (A630P) in the src-homology-2 domain, a region essential for docking of STATs to phospho-tyrosines on activated receptors, STAT dimerization, and stabilization of phospho-STAT-DNA interactions.

Objective: The purpose of this study was to explore the molecular mechanisms underlying the GH insensitivity and IGF-I deficiency caused by the A630P-mutated STAT5b.

Results: In reconstitution experiments using HEK293 cells, both GH and interferon-{gamma} were unable to activate mutant STAT5b (A630P), as demonstrated by lack of immunodetectable phospho-tyrosyl-STAT5b (A630P) and inability to drive luciferase reporter activity. However, the Src family of nonreceptor kinases [constitutively active v-src and epithelial growth factor-induced c-src] tyrosine-phosphorylated STAT5b(A630P). The v-src-induced phospho-STAT5b(A630P) translocated to the nucleus but, unlike wild-type Stat5b, was unable to bind DNA.

Conclusions: The A630P mutation disrupts the src-homology-2 architecture such that: 1) mutant STAT5b most likely cannot dock to phospho-tyrosines on ligand-activated receptors; and 2) stable interactions with DNA are prevented. Because STAT5b (A630P) is an inefficient signal transducer and transcription factor, the detrimental impact on signaling pathways important for normal growth and immunity explains, in part, the complex clinical phenotype of GH insensitivity and immune dysfunction.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
THE JANUS KINASE (JAK)-signal transducers and activators of transcription (STAT) signaling pathways, activated by multiple cytokine and growth factors, are fundamental to the regulation of genes involved in cell proliferation, immune surveillance, tumor suppression, and other biological activities. Ligand association with cytokine receptors results in JAK-induced tyrosine phosphorylation of the receptors, which act as docking sites for cytosolic proteins including the STATs. The STATs are subsequently phosphorylated on a single tyrosine by the JAKs, dimerize, and translocate to the nucleus, where they bind to DNA and participate in transcriptional regulation of targeted genes.

In mammals there are seven STATs (STAT1, -2, -3, -4, -5a, -5b, and -6). Although structurally similar, the unique properties of each STAT have been supported by gene disruption studies in rodent models (1). Corresponding mutations in humans, however, are rare and have been identified only in the STAT1 gene (2, 3) and, recently, in the STAT5b gene (4, 5). The autosomal recessive STAT5b mutations were identified in two unrelated female patients diagnosed with GH insensitivity syndrome, both of whom also presented with some degree of immune dysfunction. One of the STAT5b mutations was a frame-shift, which resulted in ablation of total STAT5b expression due to early termination of protein synthesis (5), and the other was a missense mutation that altered Ala to Pro at residue 630 (A630P) (4). STAT5b(A630P) was readily detectable in patient cells, albeit less well than wild-type STAT5b (4).

Ala 630 is within the highly conserved src-homology 2 (SH2) domain of STAT5b (Fig. 1AGo). The SH2 domain in STAT proteins has three well-characterized functions (1): 1) it permits docking of the STATs to phospho-tyrosines on activated receptors; 2) it permits phosphorylated STATs to hetero- or homodimerize before translocation to the nucleus; and 3) it stabilizes STAT-DNA interactions. Architecturally similar to other SH2 domains, the core of the STAT SH2 domain consists of an antiparallel ß-sheet flanked by two {alpha} helices (6). The strictly conserved arginine, which recognizes the phosphate group of phosphotyrosines, is in strand-ßB, and residues in the loop connecting strands-ßB and -ßC provide the supporting structure for phosphate binding (6). The proline substitution at residue 630 within strand-ßC is predicted to disrupt the ßC-structure, thereby impairing the phosphate binding and possibly destabilizing STAT5b structure.


Figure 1
View larger version (42K):
[in this window]
[in a new window]
  		FIG. 1. Mutant STAT5b(A630P) is not activated by GH in primary dermal fibroblasts. A, Schematic secondary structure of the human STAT5b(A630P) protein. Numbers indicate amino acid residue. P, Proline due to a missense mutation; Y, tyrosine that is normally phosphorylated; ND, N-terminal domain; CCD, coiled-coil domain; DBD, DNA-binding domain; L, linker domain; TAD, transcription activation domain. B, CF (normal) and PF (STAT5b mutation) infected with adenovirus vector (Ad-v) or adenovirus carrying the rabbit GH receptor (Ad-rGHR), were treated with GH (500 ng/ml) for 30 min and cell lysates analyzed by Western immunoblot (WIB) for phosphorylation of STAT5 ({alpha}-pSTAT5). Overexpression of rGHR was confirmed by immunodetection with {alpha}GHR. Note endogenous GHR was not immunodetected due to low concentrations of endogenous GHR in these cells. Total STAT5b and STAT5a were immunodetected as indicated.

 
Significantly, in primary dermal fibroblasts carrying STAT5b(A630P), after stimulation by GH or interferon (IFN)-{gamma}, both of which normally activate the JAK-STAT5b signaling pathways, phospho-STAT5 was not detected, and the STAT5b target gene, IGF-I, clearly was dysregulated (4, 7). The precise explanation for reduced IGF-I gene transcription remained uncertain because it could be the result of aberrant STAT5b(A630P) function(s) or simply reflect the presence of decreased levels of unstable STAT5b(A630P). The current studies provide evidence that, although mutant STAT5b(A630P) can be phosphorylated by non-receptor-associated kinases such as src, the phosphorylated STAT5b(A630P) is still unable to activate STAT5b-dependent target genes. This, together with new data demonstrating the inability of the mutant protein to be phosphorylated by GH and IFN-{gamma}, is consistent with the hypothesis that the A630P mutation most likely prevents docking to activated cytokine receptors and prevents stable protein-DNA complex formation.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Antibodies

Antibodies used were as follows: anti-phospho-tyrosine-STAT5 (Cell Signaling Technology, Beverly, MA); anti-STAT5a (L-20) and anti-STAT5b (G-2) (Santa Cruz Biotechnology, Santa Cruz, CA); goat sera, anti-FLAG M2, and anti-FLAG-M2-agarose gel (Sigma, St. Louis, MO); antimouse IgG and antirabbit IgG (Amersham-Pharmacia Biotech, Uppsala, Sweden); Hoechst 33342, fluorescein-conjugated goat antimouse, or antirabbit IgG (Molecular Probes, Eugene, OR). Anti-GH receptor (GHR) polyclonal antibody (GHRcyt-AL47) was generated as previously described (8).

Cell culture

Human primary dermal fibroblasts, normal (CF) and carrying mutant STAT5b(A630P) (PF), have been described previously (4). COS-7 cells (American Type Culture Collection, Manassas, VA), HEK293 (ATCC), and HEK293 stably transfected with the human GH receptor gene, HEK293(hGHR) (9), were maintained as recommended. Cells were treated with 100 U/ml IFN-{gamma} (Roche, Mannheim, Germany), 100 ng/ml human recombinant epithelial growth factor (EGF) (Sigma, St. Louis, MO), or 500 ng/ml recombinant human GH (generous gift from Genentech, Inc., South San Francisco, CA), as previously described (7).

Immunocytochemical analysis

Poly-D-lysine-coated eight-chamber slides (Becton Dickinson, Bedford, MA) were seeded with 6000 HEK293 or HEK239(hGHR) cells/chamber, and transfected for 24 h, treated with GH (500 ng/ml) for 30 min before processing as previously described (10). Appropriate primary (anti-FLAG or anti-pY-STAT5 at 1:500 dilution) and secondary antibodies were applied, and the nucleus was stained with Hoechst (1:1000 dilution). Immunofluorescence was observed as previously described (7, 10).

Plasmids and adenovirus constructs

N-terminally FLAG-tagged STAT5b (F-STAT5b) and STAT5b(A630P) (F-A630P) constructs have been described previously (7). Constructs carrying v-src or EGF receptor (EGFR) cDNAs were gifts from Drs. Brian Druker and Gayle Clinton, respectively (Oregon Health and Science University). The luciferase reporter construct, 8 x GH response element (GHRE) from the rat Spi2.1 gene in pGL2 (pGHRE-LUC), was a gift from Drs. Joachim Woelfle and Peter Rotwein (Oregon Health and Science University). The generation of adenovirus carrying the rabbit recombinant GH receptor, Ad-rGHR, was similar to that previously reported (11): rabbit GHR cDNA (a gift of W. Wood, Genentech), was N-terminally HA tagged, and resultant N-HA-GHR fragment (XbaI-PmeI) was subcloned into the XbaI and EcoRV sites of the adenoviral shuttle vector, pAdTrack-CMV. Final construction of Ad-rGHR followed the AdEasy system (Q-Biogene, Carlsbad, CA) protocol.

Transfection experiments

HEK293 and HEK293(hGHR) cells (~60% confluent), plated on poly-D-lysine-coated plates or slides (Becton Dickinson, San Jose, CA), were transiently transfected with pcDNA3.1 (1 µg/well), F-STAT5b (1 µg/well), F-A630P (3 µg/well), v-src (0.5 µg/well), or EGFR (0.5 µg/well) as indicated, using TransIT-LT1 (Mirus, Madison, WI). Transfections were performed in duplicates, at least three independent times, unless otherwise indicated. A higher concentration of F-A630P plasmid was necessary to generate final F-A630P protein concentrations immunologically equivalent to that of F-STAT5b protein. After 18–24 h of transfection, cells were washed, serum starved for 6 h, and treated for 30 min as indicated.

Adenovirus infections of primary dermal fibroblasts

CF and PF cells were seeded at 5 x 105 cells per 100-mm plate and grown to approximately 70% confluency. Infections with adenovirus constructs (vector, Ad-V; or Ad-rGHR) were carried out at a multiplicity of infection of 500, in {alpha}MEM supplemented with 5% fetal bovine serum. After 48 h of infection, the cultures were treated for 1 h with 500 ng/ml GH. Cells lysates were analyzed for expression and phosphorylation of recombinant proteins as indicated.

Inhibition of tyrosine phosphatases with sodium orthovanadate

Serum-starved HEK293(hGHR)-transfected cells were pretreated with 1 mM sodium orthovanadate (Na3VO4, Sigma) for 1 h and then treated with fresh serum-free HEK media containing 1 mM Na3VO4 and GH (500 ng/ml) for the times indicated before cell lysis. Mock treatment was included as a control. HEK293 cells cotransfected with v-src and STAT5b (wild-type or mutant) were similarly processed.

Western immunoblot analysis

Cell lysates were solubilized in Triton X-100 lysis buffer as previously described (7). Equal quantities of protein (protein assay; Bio-Rad Laboratories, Hercules, CA), 30 µg, were size fractionated on reducing 7% or 13% SDS-polyacrylamide gels and electroblotted onto nitrocellulose membranes. Western blots were processed with the appropriate primary and secondary antibodies, following manufacturers’ protocols, and visualized by enhanced chemiluminescence (PerkinElmer Life Sciences, Inc., Boston, MA). In some cases, immunoprecipitation (IP) of total cell lysates (~500 µg), using anti-FLAG-M2-agarose beads, preceded Western immunoblot analysis.

Luciferase reporter assays

Luciferase (pGHRE-LUC)-transfected HEK293 or HEK293(hGHR) cells were analyzed for reporter activity using the luciferase assay system (Promega Corp., Madison, WI). Input DNA was a total of 3 µg/well, adjusted accordingly with pcDNA3.1:0.5 µg v-src, 1.0 µg pGHRE-LUC, 2 µg F-A630P, 0.5 µg F-STAT5b, and 0.5 µg EGFR. After treatment with GH, EGF, or IFN{gamma} for 24 h, cells were lysed and luciferase activities (normalized to total protein concentration) measured with a luminometer (Wallac, Inc., Gaithersburg, MD). Each experiment was performed at least two independent times, in triplicate. The results are reported as relative fold induction ± SD relative to untreated conditions.

EMSA

Target DNA for pSTAT5b binding was the rat spi2.1 GHRE: 5'-ACGCTTCTACTAATCCATGTTCTGAGAAATCATCCAGTCTGCCCA-3'. The assay was performed with 1.5 µg (normal EMSA) or 2.5 µg (supershift assays) nuclear extract, 4 pmol nonbiotinylated target DNA, and 20 fmol 5'-biotinylated target DNA following the manufacturer’s instructions (LightShift chemiluminescent EMSA kit; Pierce, Rockford, IL). Nuclear extracts were generated as described previously (12) and stored at –80 C. For supershift assay, excess (2 µg) anti-FLAG M2 monoclonal antibody or mouse IgG (Sigma-Aldrich, St. Louis, MO) was included in the reactions. The DNA-protein complexes were resolved on native 5% polyacrylamide gels and detected as recommended by the manufacturer.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
GH-induced phospho-STAT5 is not detected in primary dermal fibroblasts carrying STAT5b(A630P)

In normal primary dermal fibroblasts (CF cells), both the GH and IFN{gamma} signaling systems preferentially tyrosine phosphorylate STAT5b over STAT5a (4, 7). Phosphorylation of STAT5b by GH, however, is considerably less robust than that induced by IFN{gamma} (7) due, presumably, to low numbers of GHRs. Total GHR concentrations were increased, therefore, by infecting fibroblasts with adenovirus carrying rabbit GHR (Ad-rGHR). Upon GH treatment, phospho-STAT5 (pSTAT5) was consistently immunodetected in Ad-rGHR-infected normal fibroblasts, compared with fibroblasts infected with adenovirus control (Fig. 1BGo). Primary dermal fibroblasts carrying the STAT5b(A630P) mutation (PF cells) were similarly infected with the adenovirus constructs, and, as shown in Fig. 1BGo, pSTAT5 still was not detectable by immunoblotting, despite overexpression of GHR.

FLAG-tagged mutant STAT5b (F-A630P) is not activated by GH in HEK293(hGHR)

The failure to detect pSTAT5 in the PF cells (Fig. 1BGo) suggested that STAT5b(A630P) could not be phosphorylated or that phosphorylated STAT5b(A630P) could not be detected because expression of the mutant protein was low (PF cells) in comparison with wild-type STAT5 (CF cells). To demonstrate definitively whether GH can induce phosphorylation of mutant STAT-5b(A630P), reconstitution experiments were performed in HEK293 cells stably transfected with hGHR. In HEK293(hGHR) cells, N-terminally FLAG-tagged mutant (F-A630P) and wild-type (F-STAT5b) STAT5b were overexpressed to immunologically equivalent amounts (Fig. 2AGo) and response to GH treatment evaluated. Immunoblot analysis of vector (pcDNA3.1)-transfected cell lysates showed that GH treatment induced tyrosine phosphorylation of endogenous STAT5. When F-STAT5b was overexpressed, the band corresponding to pSTAT5 was intensified, suggesting F-STAT5b was also phosphorylated. In contrast, in cell extracts in which F-A630P was overexpressed, the pSTAT5 band was poorly detectable and similar to cells transfected with vector (Fig. 2AGo). IP of F-STAT5b and F-A630P confirmed GH-induced phosphorylation of F-STAT5b but not F-A630P (Fig. 2BGo).


Figure 2
View larger version (33K):
[in this window]
[in a new window]
  		FIG. 2. F-A630P is not activated by GH in reconstitution experiments. HEK293(hGHR) cells were transfected with plasmids pcDNA3.1 (Invitrogen Life Technologies, Carlsbad, CA), F-STAT5b, or F-A630P and treated with GH for 30 min or as indicated. A, Cell lysates were collected and analyzed by Western immunoblot (WIB). Antibodies used are indicated. B, Recombinant FLAG-tagged proteins were IP from 500 µg of total protein before WIB analysis. C, Transfected cells were pretreated with 1 mM Na3VO4 or media for 1 h before treatment with GH (with or without Na3VO4). All 0 time points also reflect no GH treatment. Cell lysates were IP with {alpha}FLAG before WIB analysis. D, GHRE(Spi2.1)-luciferase reporter activity. Cells, cotransfected with pGHRE-LUC, were treated with or without GH for 24 h and cell lysates collected for determination of luciferase activity. Results are reported as relative fold induction, compared with untreated (–GH, given an arbitrary value of 1) ± SD, from four independent experiments, each performed in triplicate.

 
One explanation for the lack of detectable GH-induced phospho-F-A630P (p-F-A630P) is that the A630P mutation rendered the protein more vulnerable to dephosphorylation. To determine whether this was the case, transfected cells were preexposed to Na3VO4, a general tyrosine phosphatase inhibitor, before GH treatment. In the absence of Na3VO4, phosphorylation of wild-type F-STAT5b was significantly reduced 18 h after GH treatment (Fig. 2CGo), but the presence of Na3VO4 prolonged and enhanced detectable GH-induced p-F-STAT5b. Unlike p-F-STAT5b, however, Na3VO4 treatment did not enhance detection of p-F-A630P. Clearly, the lack of detectable p-STAT5b(A630P) was not due to a more rapid dephosphorylation process.

The activity of GH-induced p-F-STAT5b was measured by luciferase reporter assays, using the response element, GHRE, from the rat Spi2.1 gene, which is specific for STAT5b (13, 14). GH induced a modest 1.8-fold increase in luciferase activity in cells transfected with pcDNA3.1, whereas a greater than 20-fold induction in reporter activity was observed in GH-treated cells transfected with F-STAT5b (Fig. 2DGo). As predicted, in cells transfected with F-A630P, GH-induced reporter activity was similar to that of vector control (Fig. 2DGo). Expression of mutant and wild-type STAT5b in these assays was equivalent by immunoblot analysis (data not shown).

F-A630P is not activated by IFN{gamma} in HEK293

We previously demonstrated that IFN{gamma} could not induce phosphorylation of STAT5b(A630P) in primary fibroblasts, resulting in dysregulation of the IGF-I target gene (7). These earlier observations are extended in reconstitution studies using HEK293 cells. IFN{gamma} treatment, like GH treatment, resulted in tyrosine phosphorylation of wild-type F-STAT5b but not of mutant F-A630P (Fig. 3AGo). Furthermore, whereas IFN{gamma}-induced p-F-STAT5b was able to drive luciferase reporter activity, F-A630P could not (Fig. 3BGo).


Figure 3
View larger version (25K):
[in this window]
[in a new window]
  		FIG. 3. F-A630P is not activated by the IFN{gamma} signaling pathway in reconstitution experiments. HEK293 cells were transfected with plasmids pcDNA3.1, F-STAT5b, or F-A630P and treated with IFN{gamma} (100 U/ml) for 30 min. A, Cell lysates were collected and recombinant FLAG-tagged proteins were IP from 500 µg total protein before Western immunoblot (WIB) analysis. B, GHRE(Spi2.1)-luciferase reporter activity. pGHRE-LUC cotransfected cells were treated with or without IFN{gamma} for 24 h and cell lysates collected for determination of luciferase activity. Results are reported as relative fold induction to untreated (–IFN{gamma}, given an arbitrary value of 1) ± SD, from three independent experiments, each performed in triplicate.

 
Phosphorylation of F-A630P by src-mediated systems

Both GH and IFN{gamma} activation of the JAK-STAT5b signaling pathways require initial recruitment of STAT5b, which, via their SH2 domain, dock to phospho-tyrosines (pY) on the receptors. The receptor-associated JAKs subsequently tyrosine phosphorylate the docked STAT5b at Y699. Since STAT5b(A630P) could not be phosphorylated by activated-receptor-JAK2 systems, we asked whether STAT5b(A630) could be phosphorylated through non-receptor-mediated mechanism(s), and, if phosphorylated, could it function as a transcription factor. To evaluate this possibility, a constitutively active nonreceptor tyrosine kinase, v-src, was cotransfected with F-STAT5b or F-A630P in HEK293 cells, and pSTAT5 analyzed. As shown in Fig. 4AGo, the presence of v-src resulted in phosphorylation of not only F-STAT5b but also mutant F-A630P, although p-F-A630P was reproducibly less well immunodetected, compared with that of wild type. Pretreatment of transfected cells with Na3VO4 did not further enhance detectable p-F-A630P (Fig. 4BGo). These results support the hypothesis that Y699 on the mutant protein is available for phosphorylation, although efficiency may be lowered.


Figure 4
View larger version (37K):
[in this window]
[in a new window]
  		FIG. 4. F-STAT5b and F-A630P are both phosphorylated by v-src and the EGF signaling pathway. HEK293 cells were transfected with plasmids pcDNA3.1, F-STAT5b, or F-A630P or cotransfected with v-src or EGFR as indicated. A, Cell lysates were collected, FLAG-tagged protein IP from 500 µg total protein, and analyzed by Western immunoblot (WIB). Antibodies used are indicated. B, Transfected cells were treated with 1 mM Na3VO4 or media for 6 h before cell lysate collection. Cell lysates were IP with {alpha}-FLAG before WIB analysis. C, Cells were cotransfected with EGFR and treated with or without EGF (100 ng/ml) for 30 min before collecting cell lysates. FLAG-tagged proteins were IP from 500 µg total protein and analyzed by WIB.

 
It is possible, however, that even proteins not normally phosphorylated become phosphorylated when v-src is highly overexpressed. We therefore examined the EGF signaling pathway, which, unlike the GH and IFN{gamma} pathways, can induce tyrosine phosphorylation of STAT5b by activating endogenous c-src (15, 16). Coexpression of EGFR and F-STAT5b in HEK293 cells resulted in a low level of detectable p-F-STAT5b, which was significantly enhanced with EGF treatment (Fig. 4CGo). No p-F-A630P was detected when EGFR was coexpressed with F-A630P, but on addition of EGF, p-F-A630P was readily and reproducibly immunodetected, albeit consistently at levels lower than those observed with F-STAT5b (Fig. 4CGo). Identical results were obtained in reconstitution experiments using COS-7 cells, which had high levels of endogenous EGFR (data not shown). Altogether, the results suggest that mutant STAT5b(A630P) can be tyrosine phosphorylated under limited situations.

Src-induced p-F-A630P can translocate to the nucleus but cannot bind DNA

To determine whether p-F-A630P was capable of translocating to the nucleus, immunocytochemical and cellular fractionation studies were performed on HEK293 cells cotransfected with v-src and vector, wild-type, or mutant STAT5b. Immunofluorescent staining of cells indicated that pSTAT5 was not detectable when v-src was cotransfected with vector only (Fig. 5AGo, top panels). In the absence of v-src, pSTAT5 was also not immunodetected, although FLAG-tagged wild-type and mutant STAT5b were observed in the cytoplasm of cells overexpressing the proteins (data not shown). When v-src was cotransfected with F-STAT5b, pSTAT5 was readily detected and appeared to be nuclear (Fig. 5AGo, middle panels). Surprisingly, similar observations were made in cells cotransfected with v-src and mutant F-A630P (Fig. 5AGo, bottom panels).


Figure 5
View larger version (29K):
[in this window]
[in a new window]
  		FIG. 5. v-src-induced phospho-F-A630P can translocate to the nucleus. Cells were transfected with pcDNA3.1, F-STAT5b, or F-A630P and processed for immunocytochemical analysis as described in Materials and Methods or for Western immunoblot (WIB) analysis. A, Immunofluorescence of HEK293 cells cotransfected with v-src. Primary antibodies used are indicated at the top of the panels. B, Immunofluorescence of transfected HEK293(hGHR) cells, treated with or without GH (500 ng/ml) for 30 min. C, WIB of nuclear and cytoplasmic fractionations of cell extracts of HEK293 cotransfected with v-src (see Materials and Methods). FLAG-tagged proteins were IP from approximately 700 µg total protein before WIB analysis as indicated. NE, Nuclear extract; CE, cytoplasmic extract.

 
The nuclear localization of v-src-activated p-F-A630P was not due simply to overexpresssion of F-A630P because F-A630P was not detected in the nucleus of GH-treated HEK293(hGHR) cells overexpressing F-A630P but was readily detected in the cytoplasm of the cells (Fig. 5BGo). Under the same conditions, wild-type F-STAT5b was immunodetected in the nucleus only after GH treatment (Fig. 5BGo).

The results from immunocytochemical studies were confirmed by cellular fractionation studies. Immunoblot analysis of F-STAT-5b or F-A630P, IP from fractionated HEK293 cells cotransfected with v-src, indicated both p-F-STAT-5b and p-F-A630P were present in the nuclear fractions, although proportionally less p-F-A630P was detected. We concluded from these experiments that v-src-induced p-F-A630P can translocate to the nucleus.

We next evaluated whether v-src-mediated p-F-A630P could bind DNA. Analysis by EMSA demonstrated that F-STAT5b gel shifted GHRE in the presence of v-src (Fig. 6AGo, lane 12). The same gel-shifted band was detected when F-STAT5b was activated by GH (Fig. 6AGo, lane 3) and absent in unstimulated F-STAT5b (Fig. 6AGo, lanes 1, 2, 10, and 11). Specificity of binding was indicated by loss of the gel-shifted band in competition experiments with unlabeled GHRE oligonucleotides (Fig. 6AGo, lanes 4 and 13). Furthermore, supershifting of the respective band was achieved with anti-Flag antibody (Fig. 6BGo, lanes 2, 4) but not with excess mouse IgG (Fig. 6BGo, lane 5). In stark contrast, v-src-activated F-A630P did not gel shift the GHRE (Fig. 6AGo), suggesting that, despite the unexpected ability to translocate to the nucleus, p-F-A630P was unable to stably bind DNA and therefore unlikely to be able to act as a transcription factor.


Figure 6
View larger version (47K):
[in this window]
[in a new window]
  		FIG. 6. v-src-phosphorylated F-STAT5b, but not F-A630P, binds to GHRE and drives GHRE(Spi2.1)-luciferase reporter activity. HEK293 and HEK293(hGHR) cells were transfected and processed, and EMSAs or luciferase assays were performed. A, Cells were treated with GH or cotransfected with v-src, as indicated. Plasmid constructs transfected (F-STAT5b or F-A630P) and nonlabeled GHRE (competitor) are indicated. Arrow indicates the gel-shifted band. B, Specificity of F-STAT5b-GHRE complex (lower arrow) was determined by supershifting (upper arrow) with anti-FLAG antibody or mouse IgG (2 µg/reaction). C, EGFR cotransfected cells (as described in Fig. 4Go) were treated with or without EGF for 24 h and cell lysates collected for determination of luciferase activity. Results are reported as relative fold induction to untreated (–EGF, given an arbitrary value of 1) ± SD from two independent experiments, each performed in triplicate. D, Luciferase activity was determined from v-src-cotransfected cells (see Fig. 4Go). Results (fold induction ± SD) are relative to pcDNA3.1 basal activity (set arbitrarily as 1), performed three independent times, in triplicate.

 
Src-induced p-F-A630P cannot drive GHRE(Spi2.1) luciferase-reporter activity

The inability of p-F-A630P to act as a transcription factor was confirmed in vitro using luciferase reporter assays (Fig. 6Go, C and D). In vector-transfected HEK293 cells, only a modest 1.5-fold increase in luciferase activity was observed on EGF treatment or with v-src. In the presence of F-STAT5b, EGF or v-src treatment resulted in a 7-fold increase in luciferase activity relative to untreated conditions. In contrast, neither EGF treatment nor v-src could induce F-A630P to drive luciferase activity above that of background (Fig. 6Go, and D).


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
The identification of a natural missense mutation in STAT5b associated with severe growth failure and immune dysfunction (4) pointed to the biological importance of STAT5b but, simultaneously, raised fundamental questions concerning the impact(s) this unique mutation had on STAT5b function(s). In the STAT5b–/– mouse model, targeted disruption of the STAT5b gene ablated expression of STAT5b, and the effect of absence of STAT5b was loss of the sexual dimorphic body growth rates characteristic of normal mice, with STAT5b–/– male mice reduced to the size of female mice and displaying marked changes in liver gene expression, whereas the size of female mice remained unaltered (17). Aberrancy in immune function was subsequently noted (18, 19, 20, 21, 22), although the most severe immunological changes were those observed in mice lacking both STAT5a and STAT5b (23, 24). Clearly the human phenotype is somewhat different from that reported in the rodent knockout model because the human female patient was both severely growth retarded and immunologically impaired. Whether this reflects true interspecies differences of STAT5 function, or is the consequence of the specific mutation involved, remains to be resolved, although it is of note that the second case of STAT5b mutation reported, also in a female, had a similar growth and immune phenotype (5).

In this report, we confirm and extend our previous observations regarding the inability of ligand-dependent signaling pathways to activate STAT5b(A630P) (4, 7) and provide new data demonstrating that, via ligand-independent mechanisms, STAT5b(A630P) can be phosphorylated and translocate to the nucleus but remains functionally impaired as a transcription factor. Activation of STAT5b by type I and type II cytokine receptors, represented by the GH and IFN{gamma} receptors, respectively, requires the recruitment of STAT5b to phospho-tyrosines on ligand-activated receptors before phosphorylation of STAT5b by JAK2. The lack of detectable phospho-tyrosyl-STAT5b(A630P) in both reconstitution experiments and native state suggests that the A630P mutation affects these initial steps in signal transduction. Evidence in support of this hypothesis comes from two observations: 1) STAT5b(A630P) can be tyrosine phosphorylated by non-receptor-associated constitutively active v-src; and 2) Na3VO4 treatment does not enhance detection of GH-induced phospho-STAT5b(A630P), indicating that an increased vulnerability to dephosphorylation is not the explanation for lack of detectable phopho-STAT5b(A630P). Attempts to determine whether mutant STAT5b(A630P) could dock to GH-activated GHR (using reciprocal co-IP techniques) were inconclusive because the wild-type STAT5b did not consistently co-IP with GHR (data not shown). Nevertheless, the combination of a disrupted SH2 domain and phosphorylation by JAK2-independent, nonreceptor kinase mechanisms would support the concept that STAT5b(A630P) is unlikely to dock normally to ligand-activated receptors.

The tyrosine phosphorylation of STAT5b(A630P) by src-mediated systems is consistent with cumulative reports demonstrating activation of STAT5a/b via JAK2-independent mechanisms involving nonreceptor kinases (15, 25, 26, 27, 28). The p-STAT5b(A630P) detected reflected phosphorylation at Y699 because the commercially available anti-pSTAT-5 antibody we used detects STAT5 only when phosphorylated at Y694 (STAT-5a) or the equivalent Y699 (STAT-5b) (29). Indeed, all published reports indicate Y699 as the main site phosphorylated by the src family of kinases (16, 26, 27). It is probable that tyrosines other than Y699 were also phosphorylated in STAT5b(A630P), as has been demonstrated for rat STAT5b, in which v-src additionally phosphorylated Y724 and Y679 (26) and, for human STAT5b, EGF-induced phosphorylation of Y725, Y740, and Y743, in addition to Y699 (16, 29). The biological consequences of phosphorylated tyrosines other than Y699, however, are unclear, although alterations in the pattern of nuclear localization and gene expression have been suggested (26).

One consequence of src-mediated phosphorylation of mutant STAT5b(A630P) was its translocation to the nucleus, although whether this occurs in vivo remains to be established. Our results indicate that the nuclear translocation signal, proposed to be either in the DNA binding domain (30) or the N-terminal region (31), was unaffected by the A630P mutation. Furthermore, the results implied that, according to accepted paradigm, homodimers of p-STAT5b(A630P) formed before translocation. It is unclear, at present, whether STAT5b(A630P) could form dimers through the accepted mechanism involving reciprocal binding of pY via the SH2 domain. Interestingly, SH2-independent dimerization mechanisms were suggested in a recent study demonstrating that nonphosphorylated STAT4 formed dimers through its N domain (32).

Nuclear p-F-STAT5b(A630P) could not drive gene expression because, unlike the phosphorylated wild-type, the mutant STAT5b appeared to be unable to bind DNA. In gel-shift assays, only activated F-STAT5b bound to the GHRE, which could be further supershifted with anti-FLAG antibody. The inability of p-F-STAT5b(A630P) to bind DNA in steady-state in vitro assays is consistent with the proposed role of the SH2 domain in stabilizing STAT-DNA interactions (6).

The A630P mutation rendered STAT5b(A630P) poorly detected in PF cells due, most probably, to the considerably shortened half-life of STAT5b(A630P) (3 h), compared with wild-type STAT5b (>24 h) (33). We have now demonstrated that, even if STAT5b(A630P) is expressed more abundantly, the mutant protein remains functionally impaired. Interestingly, overexpressing wild-type F-STAT5b in PF cells can re-regenerate the GH signaling pathway (data not shown), suggesting that STAT5b(A630P) does not appear to exert an obvious negative effect on the function(s) of wild-type STAT5b. This would be consistent with the heterozygous state of the parents of the patient, who are of normal stature and with no apparent immune complications (4).

Finally, the importance of STAT5b for human statural growth has recently been verified by the identification of a second case of mutant STAT5b associated with severe growth retardation and GH insensitivity/IGF deficiency (5). The identified frameshift mutation ablated detectable STAT5b as a consequence of early termination of protein synthesis. The phenotype of immune dysfunction in both cases indicated that STAT5b is also critical for normal immunity. Altogether, STAT5b(A630P) appears to be functionally equivalent to an absence of the STAT5b protein.

In conclusion, the A630P mutation disrupted the SH2 architecture such that the mutant STAT5b: 1) most likely cannot dock normally to phospho-tyrosines on ligand-activated receptors, thus preventing phosphorylation by these systems; and 2) cannot form stable STAT5b-DNA interactions, even if phosphorylated and nuclear localized. Because STAT5b(A630P) has proven to be an inefficient signal transducer and transcription factor, the detrimental impacts on signaling pathways important for normal growth and immunity explain, in part, the complex clinical phenotype of GH insensitivity/IGF deficiency and immune dysfunction.


   Footnotes
 
This work was supported by National Institutes of Health Grants CA58110 (to R.G.R.) and DK46395 (to S.J.F.), and a grant from Pharmacia, Inc. (to R.G.R.). P.F. is supported by a postdoctoral traineeship award from the Prostate Cancer Research Program, Department of Defense.

P.F., E.M.F., B.M.L., X.W., R.J.M.R., S.J.F., V.H., and R.G.R. have nothing to declare.

First Published Online February 7, 2006

1 P.F. and E.M.K. contributed equally to this work. Back

Abbreviations: A630P, Alanine to proline substitution at residue 630; CF, primary dermal fibroblasts, normal; EGF, epithelial growth factor; EGFR, EGF receptor; GHR, GH receptor; GHRE, GH response element; hGHR, human GHR; IFN, interferon; IP, immunoprecipitation or immunoprecipitated; JAK, Janus kinase; PF, mutant STAT5b(A630P); p-F-A630P, phospho-F-A630P; pSTAT5, phospho-STAT5; SH2, src-homology 2; STAT, signal transducers and activators of transcription.

Received November 28, 2005.

Accepted February 1, 2006.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Levy DE, Darnell Jr JE 2002 STATs: transcriptional control and biological impact. Nat Rev Mol Cell Biol 3:651–662[CrossRef][Medline]
  2. Dupuis S, Dargemont C, Fieschi C, Thomassin N, Rosenzweig S, Harris J, Holland SM, Schreiber RD, Casanova J-L 2001 Impairment of mycobacteril but not viral immunity by a germline human STAT1 mutation. Science 293:300–303[Abstract/Free Full Text]
  3. Dupuis S, Jouanguy E, Al-Hajjar S, Fieschi C, Al-Mohsen IZ, Al-Jumaah S, Yang K, Chapgier A, Eidenschenk C, Eid P, Ghonaium AA, Tufenkeji H, Frayha H, Al-Rayes H, Schreiber RD, Gresser I, Casanova J-L 2003 Impaired response to interferon-{alpha}/ß and lethal viral disease in human STAT1 deficiency. Nat Genet 33:388–391[CrossRef][Medline]
  4. Kofoed EM, Hwa V, Little B, Woods KA, Buckway CK, Tsubaki J, Pratt KL, Bezrodnik L, Jasper H, Tepper A, Heinrich J, Rosenfeld RG 2003 Growth-hormone insensitivity (GHI) associated with a STAT-5b mutation. N Engl J Med 349:1139–1147[Free Full Text]
  5. Hwa V, Little B, Adiyaman P, Kofoed EM, Pratt KL, Ocal G, Berberoglu M, Rosenfeld RG 2005 Severe growth hormone insensitivity resulting from total absence of signal transducer and activator of transcription 5b. J Clin Endocrinol Metab 90:4260–4266[Abstract/Free Full Text]
  6. Chen X, Vinkemeier U, Zhao Y, Jerzalmi D, James E, Darnell J, Kuriyan J 1998 Crystal structure of a tyrosine phosphorylated STAT-1 dimer bound to DNA. Cell 93:827–839[CrossRef][Medline]
  7. Hwa V, Little B, Kofoed EM, Rosenfeld RG 2004 Transcriptional regulation of insulin-like growth factor-I (IGF-I) by interferon-{gamma} (IFN-{gamma}) requires Stat-5b. J Biol Chem 279:2728–2736[Abstract/Free Full Text]
  8. Zhang Y, Guan R, Jiang J, Kopshick JJ, Black RA, Baumann G, Frank SJ 2001 Growth hormone (GH)-induced dimerization inhibits phorbol ester-stimulated GH receptor proteolysis. J Biol Chem 276:24565–24573[Abstract/Free Full Text]
  9. Maamra M, Finidori J, Von Laue S, Simon S, Justice S, Webster J, Dower S, Ross R 1999 Studies with a growth hormone antagonist and dual-fluorescent confocal microscopy demonstrate that the full-length human growth hormone receptor, but not the truncated isoform, is very rapidly internalized independent of JAK2-Stat5 signaling. J Biol Chem 274:14791–14798[Abstract/Free Full Text]
  10. Wilson EM, Oh Y, Hwa V, Rosenfeld RG 2001 Interaction of IGF-binding protein-related protein 1 with a novel protein, neuroendocrine differentiation factor, results in neuroendocrine differentiation of prostate cancer cells. J Clin Endocrinol Metab 86:4504–4511[Abstract/Free Full Text]
  11. Cowan JW, Wang X, Guan R, He K, Jiang J, Baumann G, Black RA, Wolfe MS, Frank SJ 2005 Growth hormone receptor is a target for presenilin-dependent {gamma}-secretase cleavage. J Biol Chem 280:19331–19342[Abstract/Free Full Text]
  12. Dignam JD, Martin PL, Shastry BS, Roeder RG 1983 Eukaryotic gene transcription with purified components. Methods Enzymol 101:582–598[Medline]
  13. Bergad PL, Shih HM, Towle HC, Schwarzenberg SJ, Berry SA 1995 Growth hormone induction of hepatic serine protease inhibitor 2.1 transcription is mediated by a Stat5-related factor binding synergistically to two {gamma}-activated sites. J Biol Chem 270:24903–24910[Abstract/Free Full Text]
  14. Thomas MJ, Gronowski AM, Berry SA, Bergad PL, Rotwein P 1995 Growth hormone rapidly activates rat serine protease inhibitor 2.1 gene transcription and induces a DNA-binding activity distinct from those of Stat1, -3, and -4. Mol Cell Biol 15:12–18[Abstract]
  15. Olayioye MA, Beuvink I, Horsch K, Daly JM, Hynes NE 1999 ErbB receptor-induced activation of Stat transcription factors is mediated by src tyrosine kinases. J Biol Chem 274:17209–17218[Abstract/Free Full Text]
  16. Kloth MT, Laughlin KK, Biscardi JS, Boerner JL, Parsons SJ, Silva CM 2003 STAT5b, a mediator of synergism between c-Src and the epidermal growth factor receptor. J Biol Chem 278:1671–1679[Abstract/Free Full Text]
  17. Udy GB, Towers RP, Snell RG, Wilkins RJ, Park SH, Ram PA, Waxman DJ, Davey HW 1997 Requirement of STAT5b for sexual dimorphism of body growth rates and liver gene expression. Proc Natl Acad Sci USA 94:7239–7244[Abstract/Free Full Text]
  18. Teglund S, McKay C, Schuetz E, van Deursen JM, Stravopodis D, Wang D, Brown M, Bodner S, Grosveld G, Ihle JN 1998 Stat5a and Stat5b proteins have essential and nonessential, or redundant, roles in cytokine responses. Cell 93:841–850[CrossRef][Medline]
  19. Imada K, Bloom ET, Nakajima H, Horvath-Arcidiacono JA, Udy GB, Davey HW, Leonard WJ 1998 Stat5b is essential for natural killer cell-mediated proliferation and cytolytic activity. J Exp Med 188:2067–2074[Abstract/Free Full Text]
  20. Moriggl R, Sexl V, Piekorz R, Topham D, Ihle JN 1999 Stat5 activation is uniquely associated with cytokine signaling in peripheral T cells. Immunity 11:225–230[CrossRef][Medline]
  21. Moriggl R, Topham DJ, Teglund S, Sexl V, McKay C, Wang D, Hoffmeyer A, van Deursen J, Sangster MY, Bunting KD, Grosveld GC, Ihle JN 1999 Stat5 is required for IL-2-induced cell cycle progression of peripheral T cells. Immunity 10:249–259[CrossRef][Medline]
  22. Kagami S, Nakajima H, Kumano K, Suzuki K, Suto A, Imada K, Davey HW, Saito Y, Takatsu K, Leonard WJ, Iwamoto I 2000 Both Stat5a and Stat5b are required for antigen-induced eosinophil and T-cell recruitment into the tissue. Blood 95:1370–1377[Abstract/Free Full Text]
  23. Shelburne CP, McCoy ME, Piekorz R, Sexi V, Roh K-H, Jacobs-Helber SM, Gillespie SR, Bailey DP, Mimonsef P, Mann MN, Kashyap M, Wright HV, Chong HJ, Bouton LA, Barnstein B, Ramirez CD, Bunting KD, Sawyer S, Lantz CS, Ryan JJ 2003 Stat5 expression is critical for mast cell development and survival. Blood 102:1290–1297[Abstract/Free Full Text]
  24. Snow JW, Abraham N, Ma MC, Herndier BG, Pastuszak AW, Goldsmith MA 2003 Loss of tolerance and autoimmunity affecting multiple organs in STAT5A/5B-deficient mice. J Immunol 171:5042–5050[Abstract/Free Full Text]
  25. Kazansky AV, Kabotyanski EB, Wyszomierski SL, Manicini MA, Rosen JM 1999 Differential effects of prolactin and src/abl kinases on the nuclear translocation of STAT5b and STAT5a. J Biol Chem 274:22484–22492[Abstract/Free Full Text]
  26. Kabotyanski EB, Rosen JM 2003 Signal transduction pathways regulated by prolactin and src result in different conformations of activated Stat5b. J Biol Chem 278:17218–17227[Abstract/Free Full Text]
  27. Klejman A, Schreiner SJ, Nieborowska-Skorska M, Slupianek A, Wilson M, Smithgall TE, Skorski T 2002 The Src family kinase Hck couples BCR/ABL to STAT5 activation in myeloid leukemia cells. EMBO J 21:5766–5774[CrossRef][Medline]
  28. Guren TK, Odegard J, Abrahamsen H, Thoresen GH, Susa M, Andersson Y, Ostby E, Christoffersen T 2003 EGF receptor-mediated, c-Src-dependent, activation of Stat5b is downregulated in mitogenically responsive hepatocytes. J Cell Physiol 196:113–123[CrossRef][Medline]
  29. Kloth MT, Catling AD, Silva CM 2002 Novel activation of STAT5b in response to epidermal growth factor. J Biol Chem 277:8693–8701[Abstract/Free Full Text]
  30. Herrington J, Rui L, Luo G, Yu-Lee L-Y, Carter-Su C 1999 A functional DNA binding domain is required for growth hormone-induced nuclear accumulation of Stat5b. J Biol Chem 274:5138–5145[Abstract/Free Full Text]
  31. Zeng R, Aoki Y, Yoshida A, Arai K-I, Watanabe S 2002 Stat5b shuttles between cytoplasm and nucleus in a cytokine-dependent and -independent manner. J Immunol 168:4567–4575[Abstract/Free Full Text]
  32. Ota N, Brett TJ, Murphy TL, Fremont DH, Murphy KM 2004 N-domain-dependent nonphosphorylated STAT4 dimers required for cytokine-driven activation. Nat Immunol 5:208–215[CrossRef][Medline]
  33. Chia DJ, Subbian E, Buck TM, Hwa V, Rosenfeld RG, Skach WR, Shinde U, Rotwein P2006 Aberrant folding of a mutant STAT5b causes growth hormone insensitivity and proteasomal dysfunction. J Biol Chem 281:6552–6558



This article has been cited by other articles:


Home page
 J. Clin. Endocrinol. Metab.Home page
L. Moller, L. Dalman, H. Norrelund, N. Billestrup, J. Frystyk, N. Moller, and J. O. L. Jorgensen
Impact of Fasting on Growth Hormone Signaling and Action in Muscle and Fat
J. Clin. Endocrinol. Metab., March 1, 2009; 94(3): 965 - 972.
[Abstract] [Full Text] [PDF]

Home page
 GutHome page
X Han, X Ren, I Jurickova, K Groschwitz, B A Pasternak, H Xu, T A Wilson, S P Hogan, and L A Denson
Regulation of intestinal barrier function by signal transducer and activator of transcription 5b
Gut, January 1, 2009; 58(1): 49 - 58.
[Abstract] [Full Text] [PDF]

Home page
 J. Clin. Endocrinol. Metab.Home page
P. Fang, S. Riedl, S. Amselem, K. L. Pratt, B. M. Little, G. Haeusler, V. Hwa, H. Frisch, and R. G. Rosenfeld
Primary Growth Hormone (GH) Insensitivity and Insulin-Like Growth Factor Deficiency Caused by Novel Compound Heterozygous Mutations of the GH Receptor Gene: Genetic and Functional Studies of Simple and Compound Heterozygous States
J. Clin. Endocrinol. Metab., June 1, 2007; 92(6): 2223 - 2231.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a related Letter to the Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
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 Fang, P.
Right arrow Articles by Rosenfeld, R. G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Fang, P.
Right arrow Articles by Rosenfeld, R. G.
Related Collections
Right arrow Neuroendocrinology and Pituitary
Right arrow Metabolism

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