(GH) in Skin Fibroblast

0013-7227/98/$03.00/0 Endocrinology Copyright © 1998 by The Endocrine Society

Vol. 139, No. 1 Printed in U.S.A.

Activation of the Signal Transducers and Activators of Transcription Signaling Pathway by Growth Hormone (GH) in Skin Fibroblasts from Normal and GH Binding Protein-Positive Laron Syndrome Children JAMES S. FREETH, CORINNE M. SILVA, ANDREW J. WHATMORE, PETER E. CLAYTON

AND

Endocrine Sciences Research Group (J.S.F., A.J.W., P.E.C.), Department of Medicine, University of Manchester, Stopford Building, Oxford Road, Manchester, M13 9PT, United Kingdom; and Division of Endocrinology (C.M.S.), Department of Internal Medicine, University of Virginia Health Sciences Center, Charlottesville, Virginia 22908 ABSTRACT We have previously described two families (H and M) with GH binding protein-positive Laron Syndrome (LS), proposed to have one or more post GHR signaling defects. In the present study, we have examined whether the signal transducers and activators of transcription (STAT) pathway is activated by GH in skin fibroblast cultures established from these LS children, to determine the level(s) at which GH insensitivity has occurred. On immunoblots, both normal and LS fibroblasts express JAK2 and STATs 1, 3, and 5. GH induced rapid tyrosine phosphorylation of a protein at approximately 93 kDa in normal fibroblasts, and Western blotting with STAT-specific antibodies revealed STAT5 activation (phosphorylation) by GH. To determine further the identity and the DNA binding characteristics of the STAT proteins that were activated by GH, EMSAs were performed using three DNA elements known to bind STAT proteins; m67, the high affinity c-sis-inducible element (SIE), the interferon response element (IRE), and the lactogenic hormone-responsive region (LHRR). GH failed to induce protein binding to the SIE or IRE in normal skin fibroblasts but did induce the

formation of a specific complex with the LHRR. Induction by GH of this LHRR/protein complex, which could be supershifted partially by anti-STAT1 antisera and completely by anti-STAT5 antisera, was transient, maximal between 10 and 30 min and reduced by 60 min. GH also induced distinct LHRR/protein complexes in mouse 3T3F442A fibroblasts and in human IM-9 lymphocytes, but supershift analysis revealed that these complexes contained STAT5 but not STAT1. Whereas no binding to the LHRR was observed in GH-treated H fibroblasts, GH induced binding to this element in M fibroblasts. These results demonstrate that 1) the JAK-STAT pathway is activated by GH in normal fibroblasts and that STATs 1 and 5 have a role in GH-dependent signaling in these cells; 2) GH activation of DNA/STAT binding is cell type- and species-specific; and 3) GH failed to activate the STAT pathway in H fibroblasts but induced STAT signaling in M fibroblasts, indicating that the site of GH resistance in the latter is likely to be located within another GH signaling pathway. These fibroblast cultures therefore provide unique models with which to further our understanding of the mechanisms of human GH signaling. (Endocrinology 139: 20 –28, 1998)

G

H EXERTS its diverse biological effects on tissue growth and metabolism by binding to specific receptors on target tissues (1). The GH receptor (GHR), a member of the cytokine receptor superfamily, is a 130-kDa membrane-spanning glycoprotein that lacks any known intrinsic enzymatic activity (2, 3). The Janus tyrosine kinase JAK2 is now known to be activated by its association with a dimerized GHR following ligand binding (4, 5). Once activated, JAK2 is capable of tyrosine autophosphorylation and inducing the tyrosine phosphorylation of the GHR and a number of intracellular substrates. These include SHC (Src homology containing) proteins (6), leading to the activation of the mitogen activated protein (MAP) kinases (7–9), and insulin receptor substrates (IRS)-1 and -2 (10 –12). In addition, in cultured cell lines and in rat hepatic tissue, several members of the STAT family of transcription factors, STATs 1, 3, and 5, may be recruited by the activated GHR/JAK2 complex and

tyrosine phosphorylated by JAK2 (13–20). These proteins form homo- or hetero-dimers that migrate into the nucleus, bind to DNA, and activate the transcription of specific genes. However, as yet, little is known regarding the intracellular mechanisms of GH signaling in primary human tissues. The rare condition of congenital GH insensitivity (Laron Syndrome, LS) is usually caused by molecular defects within the extracellular domain of the GHR (21). Because the serum GH binding protein (GHBP) shares sequence homology with this portion of the receptor (22), and in humans is probably generated by proteolysis of the GHR (23), LS is often associated with absent or low levels of serum GH binding activity. However, 25% of European LS patients have normal or elevated levels of functional serum GHBP (24). We have previously described four girls with LS from two unrelated families with normal levels of serum GHBP, where a failure to identify GHR gene mutations that account for their GH insensitivity has implicated abnormalities within the intracellular GH signaling pathway (25). We and others have shown that normal skin fibroblasts express GHRs (26) and specifically bind GH (25, 27) and have demonstrated activation of DNA synthesis and IGFBP-3

Received March 17, 1997. Address all correspondence and requests for reprints to: Dr. P. E. Clayton, Endocrine Sciences Research Group, Department of Medicine, University of Manchester, Stopford Building, Oxford Road, Manchester, M13 9PT, United Kingdom. E-mail: [email protected].

20

GH ACTIVATION OF STAT SIGNALING IN SKIN FIBROBLASTS

gene expression by GH (25, 28, 29). Skin fibroblast cultures established from three of the GHBP-positive LS girls are also capable of normal GH binding but are insensitive to GH at the levels of DNA synthesis and IGFBP-3 gene expression (25). In the present study, we have assessed the ability of GH to activate the STAT signaling pathway in both normal and GHBP-positive LS fibroblasts. Materials and Methods 1) Reagents Recombinant human GH (hGH) was obtained from Pharmacia and Upjohn (Milton Keynes, UK). All oligonucleotides were synthesized by Genosys Biotechnologies Inc. (Pampisford, UK). The polyclonal antiphosphotyrosine antibody and the polyclonal anti-STAT1 antibody (ap91) for use in Western blotting have been described previously (5). The anti-STAT3 monoclonal antibody was purchased from Transduction Laboratories (Lexington, KY; S21320), and the anti-JAK2 polyclonal antibody was purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA) (SC-294). The anti-STAT5B polyclonal antibody used in Western blotting, which was raised against the C-terminus of STAT5B, was produced by Dr. C. M. Silva (unpublished data). The following antibodies were used for supershift analysis: anti-ISGF-3 (STAT91/84) monoclonal antibody was purchased from Transduction Laboratories (G16920) and has been described previously (14). Anti-STAT5B polyclonal antibody (C17), which cross-reacts with human STAT5A and STAT5B, was obtained from Santa Cruz Biotechnology, Inc. (SC-835X). Anti-IGF-I receptor monoclonal antibody was from Oncogene Science (Uniondale, NY), anti-Soy bean polyclonal antibody was from Sigma (S2519), and antiphosphotyrosine antibody was from Chemicon International Ltd. (Harrow, UK). Rabbit IgG was purchased from Vector Laboratories (Peterborough, UK).

2) Cell culture and preparation of cell lysates Fibroblast cultures were established from skin biopsies taken from three healthy children of normal stature (NI, NII, and NIII) and from three children with GHBP-positive Laron Syndrome (HI, MI, and MII), as described previously (25). Cells were maintained in a 37 C, 5% CO2 incubator in Costar 75 cm2 flasks in DMEM (Sigma) supplemented with 10% FBS, 1 mm l-glutamine, 50 IU/ml penicillin, and 50 mg/ml streptomycin. Upon reaching monolayer confluence, cells were split by treatment with 1 g/liter trypsin/0.4 g/liter EDTA solution to give a final concentration of 5 3 104 cells/ml. At 3 days, cells were washed twice with PBS and incubated for a further 24 h in DMEM containing 0.1% BSA before treatment with or without hGH (200 ng/ml, unless stated otherwise) at 37 C for the indicated times. The cells were then rinsed twice in ice-cold PBS to terminate the incubations. For Western blotting, cells were scraped on ice into detergent buffer (10 mm Tris, pH 7.4, 150 mm NaCl, 1 mm EGTA, 1 mm EDTA, 1% Triton X-100, 0.5% NP40) containing protease inhibitors (0.2 mm Na3VO4, 0.2 mm phenylmethylsulfonyl fluoride, 10 mg/ml leupeptin, and 10 mg/ml pepstatin A). Cell lysates were quick frozen in liquid nitrogen and stored at 270 C. Before use, lysates were centrifuged at 10,000 3 g for 30 min at 4 C. For JAK2 immunoprecipitation, an aliquot of the supernatant was incubated with 1:100 of the JAK2 antibody, before the addition of protein A-agarose (Boehringer Mannheim) for a further 1 h at 4 C. The agarose pellets were washed twice in detergent buffer, and proteins that were specifically bound were removed by boiling the pellet in 1 3 Laemmli sample buffer (30). For EMSA, cells were scraped on ice into an equal volume of 150 mm NaCl/10 mm Tris, pH 8.0, and lysis buffer (0.1% NP40, 1 mm EDTA, 1 mm DTT, 20% glycerol) containing protease inhibitors (1 mm Na3VO4, 0.5 mm phenylmethylsulfonyl fluoride, 1 mg/ml leupeptin, 1 mg/ml aprotinin), and incubated for 1 h at 4 C with occasional mixing. Cell lysates were then microfuged (10,000 3 g, 10 min, 4 C), and supernatants were immediately stored at 270 C until use. 3T3-F442A fibroblasts were kindly provided by Dr. N. Anderson (Department of Surgery, University of Manchester, UK), and were grown in DMEM (4.5 mg/ml glucose, GIBCO, BRL, Paisley, UK) supplemented with 2 mm l-glutamine and 10% FBS. Cells were incubated

21

in DMEM/0.1% BSA/0.5% FBS for 24 h before treatment with or without 200 ng/ml hGH for 15 min. Cell lysates were prepared as described above. Human IM-9 cells were purchased from European Collection of Cell Cultures (Salisbury, UK) and were grown in suspension in RPMI 1640 medium, supplemented with 1 mm l-glutamine and 10% FBS. Cells were incubated in RPMI 1640/0.1% BSA for 24 h before treatment with or without hGH (200 ng/ml) for 15 min IM-9 cells were then pelleted by centrifugation at 1000 3 g for 3 min, washed in ice-cold PBS, and lysed as described above. Protein concentration of cell lysates was determined spectrophotometrically using Bio-Rad (Hemel Hempstead, UK) protein assay dye. All experiments were performed on a minimum of two independent lysate preparations from each skin fibroblast line.

3) Western blotting Western blotting was performed as described previously (5). Lysates or immunoprecipitates were fractionated through a 7.5% polyacrylamide gel according to the method of Laemmli (30) and electrophoretically transferred to nitrocellulose membranes. Membranes were blocked and then incubated with a polyclonal antiphosphotyrosine antibody or the specific JAK/STAT antibodies (see Reagents). Filters were washed, incubated with donkey antirabbit whole antibody conjugated to horseradish peroxidase (Amersham), and washed again. Antibody binding was detected using an Amersham ECL kit.

4) EMSAs The following oligonucleotides were synthesized: 1) The lactogenic hormone responsive region (LHRR) from the bovine b-casein gene sense: 59-gatcAGATTTCTAGGAATTCAAATC antisense: 59-gatcGATTTGAATTCCTAGAAATCT 2) m67, the high affinity form of the c-sis inducible element (SIE) from the human c-fos gene sense: 59-agctTCATTTCCCGTAAATCCCTA antisense: 59-agctTAGGGATTTACGGGAAATGA 3) The interferon responsive element (IRE) from the human interferon regulatory factor 1 gene sense: 59-gatcCAGCCTGATTTCCCCGAAATGACG antisense: 59-gatcCGTCATTTCGGGGAAATCAGGCTG Sense and antisense oligonucleotides of each DNA element were annealed, end-labeled using polynucleotide kinase (Boehringer Mannheim) and [g-32P]ATP (4500 Ci/mmol, ICN), and purified on a G-25 Quick Spin Column (Boehringer Mannheim, Lewes, UK). EMSA was performed by adding 10 ml (20 mg) protein lysate and 5 ml H2O (or antibody for supershift experiments [1.25 mg], or unlabeled double-stranded oligonucleotide for competition experiments) to 8 ml 4 3 EMSA buffer (10% glycerol, 1% Ficoll, 0.06% NP40, 0.2 mg/ml BSA, 8 mm spermidine, 4 mm EDTA, 2 mm DTT, 350 mm KCl, 16 mm HEPES, pH 7.8) containing 3 mg polydeoxyinosinic-deoxycytidylic acid, and the mixture incubated at room temperature for 15 min 1 ml (1 ng) of 32Plabeled oligonucleotide (;1 3 105 cpm) was added to each reaction to give a final volume of 24 ml, and the mixture incubated at 30 C for 30 min The samples were immediately electrophoresed on a 4.5% polyacrylamide gel containing 2.5% glycerol and 0.25 3 TBE (22.5 mm Tris borate, 0.5 mm EDTA, pH 8.0) at 200 V. Gels were fixed, dried, and autoradiographed for 1 to 8 days.

5) RT-PCR Identification of mRNAs for STAT5A and B (18) was achieved by RT-PCR. Cell pellets were lysed with 0.5 ml RNAzol B (Tel-Test Inc., Friendswood, TX), and total cellular RNA was isolated according to manufacturer’s instructions. Preparation of cDNA from total RNA was achieved using an RT system (Promega Corporation, Madison, WI). Human STAT5A and B were amplified from cDNA by 35 cycles of PCR using a Techne Unit Progene thermocycler (Techne Ltd., Cambridge, UK) (60 sec denaturation at 95 C, 45 sec annealing at 58 C, and 60 sec elongation at 72 C). 1) Human STAT5A-specific primers sense: 59-CCTTCTTGTTGCGCTTTAGTG-39 antisense: 59-TCGAGTACATGGTCAGGGTTC-39

22


Endo • 1998 Vol 139 • No 1

2) Human STAT5B specific primers sense: 59-GTGAGGCGCTCAACATGAAAT-39 antisense: 59-AAGCTGAAGATGGAGAGGTCG-39 PCR products were electrophoresed on a 1.5% agarose gel containing ethidium bromide and visualized by UV light.

Results

The effect of GH on JAK-STAT signaling has been examined in fibroblasts from normal children and three children with GHBP-positive LS (HI, MI, and MII), and compared with that found in mouse 3T3-F442A fibroblasts and human IM-9 lymphocytes. By Western blotting of total cell lysates with specific antibodies (STATs 1, 3, and 5), and by JAK2 immunoprecipitation followed by Western blotting with antiserum against JAK2, we have confirmed that JAK2 and STATs 1, 3, and 5 were present in normal and LS fibroblasts and IM-9 cells (Fig. 1A). cDNAs generated from normal, HI, MI, and MII fibroblasts and IM-9 cells were amplified with primers specific for hSTAT5A and hSTAT5B. Single PCR products of 400 bp and 865 bp, respectively, were observed in all cell types (Fig. 1B), confirming the presence of both isoforms. Western blotting using an antiphosphotyrosine antibody was performed on lysates prepared from control or GHtreated normal fibroblasts and IM-9 cells. GH treatment of normal fibroblasts induced the tyrosine phosphorylation of a protein band of similar molecular mass to the 93-kDa phosphorylated protein (STAT5) observed in GH-treated IM-9 cells (18) (Fig. 2A). Tyrosine phosphorylation of this protein, characteristic of the STATs, was maximal after 15 min of GH treatment and still present after 30 min. Direct immunoblotting of lysates using a STAT5B antibody revealed a GHinduced partial band shift to a higher mol mass in IM-9 cells (18) and in normal fibroblasts, representing a phosphorylated form of STAT5B (Fig. 2B). No GH-induced phosphorylation shift in either STAT1 or STAT3 was detected in either cell type (Fig. 2B). To establish further the identity, and to investigate the DNA binding characteristics, of the STAT protein(s) activated by GH in normal fibroblasts, EMSAs were performed by incubating 32P-labeled oligonucleotides (SIE, IRE, and LHRR) with total lysates prepared from untreated cells and from cells treated with GH for 10, 30, and 60 min (Fig. 3A). Whereas lysates from GH-treated 3T3-F442A cells formed two distinct complexes with the SIE, consistent with the studies of Meyer et al. (13), no GH-induced binding to this element was observed in normal skin fibroblasts (n 5 5

FIG. 1 A, Expression of JAK2 and STATs 1, 3, and 5 by normal and Laron (LS) fibroblasts, compared with IM-9 lymphocytes. Total cell lysates prepared from untreated IM-9 cells, normal (N) fibroblasts, and LS (MII) fibroblasts were analyzed by JAK2 immunoprecipitation followed by JAK2 Western blotting, or by direct Western blotting with antisera to STATs 1, 3 and 5. B, PCR amplification from cDNA using primers specific to sequences within hSTAT5A and hSTAT5B in IM-9 cells (lane 2), normal fibroblasts (NI, lane 3; NII, lane 4) and Laron fibroblasts (HI, lane 5; MI, lane 6; MII, lane 7). The arrow indicates STAT5A transcripts (400 bp), and the arrowhead indicates STAT5B transcripts (865 bp). Lane 1 contained no cDNA. The positions of the primers relative to the human STAT5A and B proteins are also shown.


23

FIG. 2. A, GH-induced tyrosine phosphorylation of one or more 93-kDa proteins in IM-9 cells (large arrow) and in normal fibroblasts (small arrow). Total cell lysates prepared from IM-9 cells treated without (2) or with 200 ng/ml GH for 15 min (1), and from normal (NIII) fibroblasts treated without (2) or with 200 ng/ml GH for 5, 15, or 30 min, were analyzed by Western blotting using an antiphosphotyrosine antibody. The 97-kDa molecular mass marker is indicated. B, Anti-STAT Western blotting of total cell lysates from untreated (2) and GH-treated (200 ng/ml, 10 min) (1) normal fibroblasts (N), compared with IM-9 cells. No bands were observed when the membrane was blotted with antipreimmune STAT5B (top panel). A GH-induced STAT5B partial band shift to a higher molecular weight (representing phosphorylated STAT5; STAT5P) was observed in both normal fibroblasts and IM-9 cells. In contrast, no GH-induced STAT1 or STAT3 band shift (i.e. phosphorylation) was observed in either cell type when the membrane was blotted with antiSTAT1 or anti-STAT3 antisera.

experiments). Furthermore, whereas GH treatment of 3T3F442A cells resulted in the formation of a single complex with the IRE, GH failed to induce binding to this element in normal fibroblasts (n 5 2 experiments). However, GH did induce the formation of a distinct complex with the LHRR in these cells. This complex, which was of similar mobility to that formed in GH-treated 3T3-F442A cells (20), was induced

within 10 min of GH treatment and reduced within 60 min (n 5 3 experiments). When normal fibroblasts were treated with a range of GH doses (2–20,000 ng/ml) for 15 min, optimal DNA-protein binding was induced at 2000 ng/ml GH (Fig. 3B). The formation of this complex was inhibited by an excess of unlabeled LHRR in skin and 3T3-F442A fibroblasts and in IM-9 lymphocytes, confirming the specificity of

24


Endo • 1998 Vol 139 • No 1

FIG. 3. A, EMSAs of GH-activated binding to three GAS-like elements; m67 SIE, IRE and LHRR. 3T3-F442A cells were treated without or with 200 ng/ml GH for 15 min at 37 C, and normal fibroblasts were treated without or with 200 ng/ml GH for 10, 30, and 60 min at 37 C, as indicated. Total cell lysates were prepared, incubated with 32 P-labeled SIE, IRE or LHRR, and DNA-protein complexes analyzed on 4.5% polyacrylamide gels. The arrows indicate the GH-induced DNA-protein complexes. B, EMSA of LHRR binding to total cell lysates from normal fibroblasts treated without or with 2, 20, 200, 2000, or 20,000 ng/ml GH for 15 min. The GH-induced LHRR-protein complex is indicated by the arrow.

this complex (Fig. 4A). STAT-specific antibodies were used to determine the composition of the GH-induced LHRRprotein complex. Preincubation of GH-treated lysates with anti-ISGF3 (anti-STAT1) antiserum supershifted the majority of this complex in normal skin fibroblasts, but not in IM-9 (18) or 3T3-F442A cells (Fig. 4, A and B). The antibody to STAT5,

however, supershifted the GH-induced complex in all three cell types (Fig. 4, A and B). A complex that migrated faster through the gel, was competed by excess unlabeled probe, but was not GH responsive, was seen in all cell types (Fig. 4A, arrowheads). The GH-induced LHRR/STAT1/5 complex in normal fibroblasts was not supershifted by IgG or by two


FIG. 4. A, Competition and antibody supershift analysis of GH-induced protein binding to the LHRR. Binding of total cell lysates from untreated and GH-treated (200 ng/ml, 15 min) 3T3F442A (lanes 1–7), IM-9 (lanes 8 –14) and normal (NII) fibroblast cells (lanes 15–21) to the LHRR was analyzed by EMSA. GH-treated lysates were preincubated with 10 3 (lanes 3, 10, 17), 25 3 (lanes 4, 11, 18) or 50 3 (lanes 5, 12, 19) unlabeled LHRR, or with antisera to STAT1 (lanes 6, 13, 20) or STAT5 (lanes 7, 14, 21). The arrows indicate the GHinduced LHRR-protein complexes, and the arrowheads indicate LHRR-protein complexes that were competed by excess unlabeled probe but were not GH responsive. B, Supershift analysis of the GH-induced LHRR-protein complex in normal fibroblasts. Normal fibroblasts (NI) were untreated (lane 1) or treated with 200 ng/ml GH for 15 min (lanes 2– 8), and total cell lysates preincubated with antibodies to STAT1 (lane 3), STAT5 (lane 4), rabbit IgG (lane 5), IGF-I receptor (lane 6), soy bean (lane 7), and phosphotyrosine (lane 8).

25

26


unrelated control antibodies but was inhibited by an antiphosphotyrosine antibody (Fig. 4B). The formation of the LHRR-STAT complex in response to GH was assessed in the LS fibroblasts. In MI and MII cells (Fig. 5A), binding to the LHRR was induced within 10 min of GH treatment and was reduced within 60 min in two independent experiments. The complex was supershifted with both anti-STAT1 and anti-STAT5 antibodies (Fig. 5B). However, no GH-induced binding to the LHRR was observed in HI fibroblasts (Fig. 5A) in four independent experiments. Discussion

GH has been shown to activate the JAK-STAT pathway and c-fos gene transcription in a number of animal and human cell lines, and in rat hepatic tissue in vivo, providing a direct link between ligand binding and activation of gene

FIG. 5. A, EMSA showing GH-induced binding to the LHRR in MI and MII Laron fibroblasts but not in HI Laron fibroblasts. Total cell lysates from HI, MI, and MII fibroblasts treated without or with 200 ng/ml GH for 10, 30, and 60 min were incubated with 32P-labeled LHRR and analyzed on a 4.5% polyacrylamide gel. Lysates from untreated and GH-treated normal (NIII) fibroblasts were used as positive controls. The arrow indicates the GH-induced protein-LHRR complex. B, Supershift analysis of the GH-induced LHRR-protein complex (arrow) in MI Laron fibroblasts. MI cells were untreated (lane 1) or treated with 200 ng/ml GH for 15 min (lanes 2– 4) and lysates preincubated with anti-STAT1 (lane 3) or anti-STAT5 (lane 4) antibodies. The anti-STAT1 supershifted complex is shown by the arrowhead, whereas the broad band (as indicated by the bracket) was induced by anti-STAT5.

Endo • 1998 Vol 139 • No 1

expression. Although GH is capable of inducing low level tyrosine phosphorylation of JAK1 and JAK3 in certain cell lines (19, 31), JAK2 is the primary Janus family member that is activated by GH in all cell systems studied so far (4, 5, 19, 20, 32). In contrast, the involvement of the STAT transcription factors 1, 3, and 5 in GH-dependent signaling appears to be cell type- and/or species-specific. Whereas STAT5, but not STAT1 or STAT3, is tyrosine phosphorylated by GH in the human IM-9 lymphocyte cell line (5, 18), all three STAT proteins are tyrosine phosphorylated by GH in the 3T3F442A mouse fibroblast cell line (15, 18, 20). To further our understanding of the mechanisms of intracellular GH signaling in primary human tissue, the first aim of this study was to characterize the effect of GH on JAK-STAT signaling in skin fibroblast cultures established from children of normal stature. We have demonstrated that the components necessary for


the activation of JAK-STAT signaling by GH (i.e. JAK2 and STATs 1, 3, and 5) are expressed by normal fibroblasts. Furthermore, by RT-PCR, we have shown that these cells (and the LS fibroblasts) express mRNAs for both STAT5A and STAT5B. When lysates were immunoprecipitated with antiJAK2 or anti-STAT antibodies, and the immune complexes analyzed by Western blotting using an antiphosphotyrosine antibody (or vice versa), no GH-induced JAK2 or STAT tyrosine phosphorylation could be detected in the normal fibroblasts, in contrast to the 3T3-F442A and IM-9 cell lines (data not shown). We were also unable to detect activation of JAK2 in response to GH using a specific JAK2 kinase assay (data not shown). It is therefore likely that the coimmunoprecipitation technique and the kinase assay were not sensitive enough to detect the comparatively small GH responses in the skin fibroblasts. We have, however, shown that GH induced the rapid tyrosine phosphorylation of a protein at approximately 93 kDa in normal fibroblasts that comigrated with the 93-kDa STAT5 band observed in GHtreated IM-9 cells (18). In addition, by direct Western blotting with an anti-STAT5B antibody, we have demonstrated that STAT5 is activated (i.e. phosphorylated) by GH in normal skin fibroblasts. This coincided with the formation of a specific GH-induced protein complex with the LHRR. Because this complex was supershifted completely with STAT5 antisera but only partially with STAT1 antisera, it is possible that STATs 1 and 5 bind to the LHRR as STAT1/5 heterodimers and STAT5/5 homodimers. The inability of the anti-STAT1 antibody to supershift the GH-induced LHRRSTAT5-containing complex in IM-9 and 3T3-F442A cells confirmed that this antibody did not cross-react with STAT5 in normal fibroblasts. Because STAT1, as well as STAT5, appears to play a role in GH-dependent signaling in normal human skin fibroblasts, we cannot explain why GH-induced STAT1 phosphorylation was not detected by immunoblotting (Fig. 2B). The disappearance of the GH-induced LHRR-STAT complex with an antiphosphotyrosine antibody in normal fibroblasts confirms that the bound STAT proteins were indeed tyrosine phosphorylated in response to GH. Similar results have been demonstrated previously in other cell systems (33, 34). Whereas 200 ng/ml GH was sufficient to induce the LHRR-STAT complex in normal fibroblasts, maximal induction of this complex occurred in response to 2000 ng/ml GH. However, at 20,000 ng/ml, induction of the LHRR/STAT complex was reduced, consistent with the observation that high concentrations of GH inhibit GHR dimerization (35, 36). Although GH induced specific binding to the high-affinity SIE (m67) (13) and the IRE in mouse 3T3-F442A fibroblasts, GH failed to induce binding to these elements in normal fibroblasts, indicating species-specific activation of STAT/ DNA binding by GH. The lack of GH-induced protein binding to the SIE in normal fibroblasts is consistent with the failure of GH to activate c-fos transcription in these cells (our unpublished data). The finding that GH induced the formation of a specific protein complex with the LHRR in IM-9 cells, which could be supershifted by STAT5 antisera but not by STAT1 antisera, has been described previously (18). In the same study, Silva et al. demonstrated GH-induced tyrosine phosphorylation of

27

STAT5 in mouse 3T3-F442A cells (18). In accordance with the studies of Han et al. (20), we show that GH induced a specific protein complex with the LHRR in 3T3-F442A cells. Although STAT1 is also known to be tyrosine phosphorylated in response to GH in these cells (13), this complex could only be supershifted with the STAT5 antibody, and not with the STAT1 antibody. These studies in normal human skin fibroblasts, human IM-9 lymphocytes, and mouse 3T3-F442A fibroblasts therefore confirm cell type- and species-specific differences in GH activation of STAT/DNA binding. We have previously described four girls from two families with GHBP-positive LS. The two affected girls from the first family (HI and HII) are heterozygous for a D152H point mutation in exon 6 of the GHR gene, as are the unaffected father and a brother (25). No GHR gene abnormalities were identified in the two affected girls (MI and MII) from the second family. We have therefore hypothesized that their GH insensitivity is due to a defect(s) within the intracellular GH signaling pathway. In addition, we have previously shown that, in contrast to normal fibroblasts, skin fibroblasts established from HI, MI, and MII are insensitive to GH at the levels of DNA synthesis and IGFBP-3 gene expression (25). The second aim of this study was to examine the effect of GH on STAT signaling in these LS fibroblast cultures in an attempt to identify the site(s) of GH resistance in these children. Whereas HI fibroblasts express full-length forms of JAK2, STATs 1, 3, and 5 (data not shown), GH failed to induce STAT binding to the LHRR in these cells. In a parallel study, we have shown that, whereas GH induces the phosphorylation and activation of MAP kinase (MAPK) in normal fibroblasts, GH-dependent activation of MAPK in HI fibroblasts is reduced (37). Recent studies indicate that two signaling pathways to MAPK can be activated by GH in 3T3-F442A fibroblasts, one involving a JAK23 SHC3 Grb23 Sos3ras3 raf3 MEK cascade (9), and the other by the intermediate activation of phosphoinositide 3-OH kinase by a mechanism involving JAK2 and IRS-1 (38). Our findings therefore would suggest that the site of GH resistance in HI lies at the level of JAK2 or between the GHR and JAK2. Further studies are necessary to determine whether the abnormality is intrinsic to JAK2 itself, or resides within another, as yet, unidentified protein involved in the GHR/JAK2 complex. Skin fibroblasts derived from MI and MII expressed the full-length forms of JAK2, STATs 1, 3, and 5. In contrast to HI fibroblasts, GH did induce STAT1 and STAT5 binding to the LHRR in these cells; MI with a time course identical to that seen in normal fibroblasts. In contrast, GH-induced binding and down-regulation occurred more rapidly in MII cells than in MI and normal fibroblasts. As it is likley that the siblings MI and MII have the same signaling defect, we feel that this difference in kinetics is not relevant. Thus, although GH failed to activate DNA synthesis and IGFBP-3 gene expression in MI and MII fibroblasts (25), GH was capable of inducing activation of STAT signaling in these cells. Further studies will determine whether this pathway is functional at the transcriptional level. In the same cells, GH dependent activation of MAPK was significantly reduced compared with normal fibroblasts (37). These data therefore suggest that the intracellular GH signaling defect in MI and MII may lie distal to JAK2 in an ancillary pathway that modulates

28


MAPK activity. Our results indicating that in HI the site of GH resistance lies at the level of JAK2 (central to the initiation of GH signal transduction), whereas in MI and MII it appears to be located along another limb of the GH signaling pathways (affecting MAPK but not STAT activation), could be consistent with the severity of the phenotypes of these girls: HI has a classical LS phenotype (Ht. SDS: 26.8), whereas MI and MII have atypical LS phenotypes, and are less severely growth retarded (Ht. SDS: 24 and 23.4, respectively) (25). In summary, we have demonstrated that GH rapidly and transiently activates STAT signaling in normal skin fibroblasts, where STATs 1 and 5 appear to bind to DNA. We have shown that GH activates STAT-DNA binding in skin fibroblasts from two sisters with GHBP-positive LS, indicating a molecular defect elsewhere within the GH signaling network. In contrast, GH failed to activate the STAT pathway in fibroblasts derived from another GHBP-positive LS girl. The GH insensitivity syndrome with normal GHBP is therefore associated with diverse abnormalities within the GH intracellular signaling pathways. References 1. Isaksson OG, Eden S, Jansson JO 1985 Mode of action of pituitary growth hormone on target cells. Annu Rev Physiol 47:483– 499 2. Cosman D, Lyman SD, Idzerda RL, Beckman MP, Park LS, Goodwin RG, March CJ 1990 A new cytokine receptor superfamily. Trends Biochem Sci 15:265–270 3. Bazan JF 1990 Structural design and molecular evolution of a cytokine receptor superfamily. Proc Natl Acad Sci USA 87:6934 – 6938 4. Argetsinger LS, Campbell GS, Yang X, Witthuhn BA, Silvennoinen O, Ihle JN, Carter-Su C 1993 Identification of JAK2 as a growth hormone receptorassociated tyrosine kinase. Cell 74:237–244 5. Silva CM, Lu H, Weber MJ, Thorner MO 1994 Differential tyrosine phosphorylation of JAK1, JAK2, and STAT1 by growth hormone and interferon-g in IM-9 cells. J Biol Chem 269:27532–27539 6. VanderKuur J, Allevato G, Billestrup N, Norstedt G, Carter-Su C 1995 Growth hormone-promoted tyrosyl phosphorylation of SHC proteins and SHC association with Grb2. J Biol Chem 270:7587–7593 7. Winston LA, Bertics PJ 1992 Growth hormone stimulates the tyrosine phosphorylation of 42- and 45-kDa ERK-related proteins. J Biol Chem 267:4747– 4751 8. Campbell GS, Pang L, Miyasaka T, Saltiel AR, Carter-Su C 1992 Stimulation by growth hormone of MAP kinase activity in 3T3–F442A fibroblasts. J Biol Chem 267:6074 – 6080 9. Carter-Su C, Smit L, VanderKuur J, Argetsinger L, Campbell G GH signal transduction. Program of the 10th International Congress of Endocrinology, San Francisco, CA, 1996, S19 –1 (Abstract) 10. Argetsinger LS, Hsu GW, Myers MG Jr, Billestrup N, White MF, Carter-Su C 1995 Growth hormone, interferon-g and leukemia inhibitory factor promoted tyrosyl phosphorylation of insulin receptor substrate-1. J Biol Chem 270:14685–14692 11. Argetsinger LS, Norstedt G, Billestrup N, White MF, Carter-Su C 1996 Growth hormone, interferon-gamma, and leukemia inhibitory factor utilize insulin receptor substrate-2 in intracellular signalling. J Biol Chem 271:29415–29421 12. Carter-Su C, Schwartz J, Smit LS 1996 Molecular mechanisms of growth hormone action. Annu Rev Physiol 58:187–207 13. Meyer DJ, Campbell GS, Cochran BH, Argetsinger LS, Larner AC, Finbloom DS, Carter-Su C, Schwartz J 1994 Growth hormone induces a DNA binding factor related to the interferon-stimulated 91-kDa transcription factor. J Biol Chem 269:4701– 4704 14. Gronowski AM, Rotwein P 1994 Rapid changes in nuclear protein tyrosine phosphorylation after growth hormone treatment in vivo. J Biol Chem 269:7874 –7878 15. Campbell GS, Meyer DJ, Raz R, Levy DE, Schwartz J, Carter-Su C 1995 Activation of acute phase response factor (APRF)/Stat3 transcription factor by growth hormone. J Biol Chem 270:3974 –3979

Endo • 1998 Vol 139 • No 1

16. Gronowski AM, Zhong Z, Wen Z, Thomas MJ, Darnell JE, Rotwein P 1995 In vivo growth hormone treatment stimulates the tyrosine phosphorylation and activation of Stat3. Mol Endocrinol 9:171–177 17. Wood TJJ, Sliva D, Lobie PE, Pircher TJ, Gouilleux F, Wakao H, Gustafsson J-A, Groner B, Norstedt G, Haldose´n L-A 1995 Mediation of growth hormonedependent transcriptional activation by mammary gland factor/Stat 5. J Biol Chem 270:9448 –9453 18. Silva CM, Lu H, Day RN 1996 Characterization and cloning of STAT5 from IM-9 cells and its activation by growth hormone. Mol Endocrinol 10:508 –518 19. Smit LS, Meyer DJ, Billestrup N, Norstedt G, Schwartz J, Carter-Su C 1996 The role of the growth hormone (GH) receptor and JAK1 and JAK2 kinases in the activation of Stats 1, 3, and 5 by GH. Mol Endocrinol 10:519 –533 20. Han Y, Leaman DW, Watling D, Rogers NC, Groner B, Kerr IM, Wood WI, Stark GR 1996 Participation of JAK and STAT proteins in growth hormoneinduced signalling. J Biol Chem 271:5947–5952 21. Rosenfeld RG, Rosenbloom AL, Guevara-Aguirre J 1994 Growth hormone (GH) insensitivity due to primary GH receptor deficiency. Endocr Rev 15:369 –390 22. Leung DW, Spencer SA, Cachianes G, Hammonds GR, Collins C, Henzel WJ, Barnard R, Waters MJ, Wood WI 1987 Growth hormone receptor and serum binding protein: purification, cloning and expression. Nature 330:537–543 23. Sotiropoulos A, Goujon L, Simonin G, Kelly PA, Postel-Vinay MC, Finidori J 1993 Evidence for generation of the growth hormone-binding protein through proteolysis of the growth hormone membrane receptor. Endocrinology 132:1863–1865 24. Savage MO, Blum WF, Ranke MB, Postel-Vinay MC, Cotterill AM, Hall K, Chatelain PG, Preece MA, Rosenfeld RG 1993 Clinical features and endocrine status in patients with growth hormone insensitivity (Laron Syndrome). J Clin Endocrinol Metab 77:1465–1471 25. Freeth JS, Ayling RM, Whatmore AJ, Towner P, Price DA, Norman MR, Clayton PE 1997 Human skin fibroblasts as a model of growth hormone action in growth hormone receptor-positive Laron’s Syndrome. Endocrinology 138:55– 61 26. Oakes SR, Haynes KM, Waters MJ, Herrington AC, Werther GA 1992 Demonstration and localization of growth hormone receptor in human skin and skin fibroblasts. J Clin Endocrinol Metab 75:1368 –1373 27. Murphy J, Vrhovsek E, Lazarus L 1983 Identification and characterization of specific growth hormone receptors in cultured human fibroblasts. J Clin Endocrinol Metab 57:1117–1124 28. Cook JJ, Haynes KM, Werther GA 1988 Mitogenic effects of growth hormone in cultured human fibroblasts. J Clin Invest 81:206 –212 29. Schmid C, Schla¨pfer I, Waldvogel M, Peter M, Zapf PJ, Froesch ER 1993 Production of IGFBP-3 by connective tissue and endothelial cells. Growth Regul 3:65– 67 30. Laemmli UK 1970 Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680 – 685 31. Johnston JA, Kawamura M, Kirken RA, Chen Y-Q, Blake TB, Shibuya K, Ortaldo JR, McVicar DW, O’Shea JJ 1994 Phosphorylation and activation of the JAK3 Janus kinase in response to interleukin-2. Nature 370:151–153 32. Eriksson H, Ridderstråle M, Tornqvist H 1995 Tyrosine phosphorylation of the growth hormone (GH) receptor and Janus tyrosine kinase-2 is involved in the insulin-like actions of GH in primary rat adipocytes. Endocrinology 136:5093–5101 33. 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 34. Sotiropoulos A, Moutoussamy S, Binart N, Kelly PA, Finidori J 1995 The membrane proximal region of the cytoplasmic domain of the growth hormone receptor is involved in the activation of Stat3. FEBS Lett 369:169 –172 35. Fuh G, Cunningham BC, Fukunaga R, Nagata S, Goeddel DV, Wells JA 1992 Rational design of potent antagonists to the human growth hormone receptor. Science 256:1677–1680 36. Ilondo MM, Damholt AB, Cunningham BA, Wells JA, De Meyts P, Shymko RM 1994 Receptor dimerization determines the effects of growth hormone in primary adipocytes and cultured IM-9 lymphocytes. Endocrinology 134:1–7 37. Whatmore AJ, Silva CM, Freeth JS, Clayton PE Growth hormone (GH) signal transduction in cultured fibroblasts from normal, and GH binding protein (GHBP)-positive Laron children. Program of the 10th International Congress of Endocrinology, San Francisco, CA, 1996, OR7– 8 (Abstract) 38. Kilgour E, Gout I, Anderson NG 1996 Requirement for phosphoinositide 3-OH kinase in growth hormone signalling to the mitogen-activated protein kinase and p70s6k pathways. Biochem J 315:517–522