Platelet-derived Growth Factor and Lysophosphatidic Acid Inhibit Growth Hormone Binding and Signaling via a Protein Kinase C-dependent Pathway*

Growth hormone (GH) regulates body growth and metabolism. GH exerts its biological action by stimulating JAK2, a GH receptor (GHR)-associated tyrosine kinase. Activated JAK2 phosphorylates itself and GHR, thus initiating multiple signaling pathways. In this work, we demonstrate that platelet-derived growth factor (PDGF) and lysophosphatidic acid (LPA) down-regulate GH signaling via a protein kinase C (PKC)-dependent pathway. PDGF substantially reduces tyrosyl phosphorylation of JAK2 induced by GH but not interferon-γ or leukemia inhibitory factor. PDGF, but not epidermal growth factor, decreases tyrosyl phosphorylation of GHR (by approximately 90%) and the amount of both total cellular GHR (by approximately 80%) and GH binding (by approximately 70%). The inhibitory effect of PDGF on GH-induced tyrosyl phosphorylation of JAK2 and GHR is abolished by depletion of 4β-phorbol 12-myristate 13-acetate (PMA)-sensitive PKCs with chronic PMA treatment and is severely inhibited by GF109203X, an inhibitor of PKCs. In contrast, extracellular signal-regulated kinases 1 and 2 and phosphatidylinositol 3-kinase appear not to be involved in this inhibitory effect of PDGF. LPA, a known activator of PKC, also inhibits GH-induced tyrosyl phosphorylation of JAK2 and GHR and reduces the number of GHR. We propose that ligands that activate PKC, including PDGF, LPA, and PMA, down-regulate GH signaling by decreasing the number of cell surface GHR through promoting GHR internalization and degradation and/or cleavage of membrane GHR and release of the extracellular domain of GHR.

Growth hormone (GH) 1 is a circulating peptide hormone secreted from the anterior pituitary. It is the primary hormone known to stimulate postnatal longitudinal bone growth and increase body mass (1,2). GH deficiency or dysfunction of the GH receptor (GHR) leads to dwarfism, whereas GH in excess results in gigantism or, in adults, acromegaly. In addition to stimulating body growth, GH regulates a variety of other biological functions, including body metabolism and the immune response (2). Clinically, GH has long been used to treat children with short stature. More recently, it has been used to prevent muscle wasting in AIDS patients (3,4). Clinical studies indicate it greatly increases donor site wound healing in human patients (5).
GH exerts its diverse biological functions via GHR. Upon GH binding, GHR binds and activates JAK2, a cytoplasmic tyrosine kinase (6,7). Activated JAK2 then phosphorylates itself and GHR on multiple tyrosines, generating binding sites for other signaling proteins containing Src homology 2 or phosphotyrosine interacting domains, including signal transducers and activators of transcription (8 -13), Shc (14), and SH2-B (15,16). Recruitment of these signaling molecules to GHR⅐JAK2 complexes and the subsequent tyrosyl phosphorylation of these proteins by JAK2 initiate a variety of signaling pathways, leading to the multiple cellular responses responsible for the diverse actions of GH (7,17).
In contrast to GHR that is a member of the cytokine receptor family, the PDGF receptor is a receptor tyrosine kinase. A variety of signaling pathways, including the MEK/MAP kinase, the phosphatidylinositol 3 (PI-3)-kinase, and the protein kinase C (PKC) pathways (7, 18 -21), has been shown to be activated by both GH and PDGF. Despite some shared signaling pathways, PDGF and GH are thought to regulate different functions even in the same cell type. Whereas GH has been implicated as a differentiation factor for a variety of cells (22)(23)(24), PDGF has been shown to be a competence factor that causes 3T3 cells to progress from a quiescent G 0 state to the G 1 phase of the cell cycle (25,26). In 3T3-F442A cells, GH has been shown to dampen the mitogenic effect of PDGF and insulin (27). Because PDGF and GH can have opposing effects, even in the same cell, and receptors for GH and PDGF are coexpressed in many tissues and cell types, including brain (28 -32), liver (29 -34), bone (35)(36)(37), kidney (29 -31, 38 -40), and muscle (29,30,(41)(42)(43)(44)(45), we hypothesized that PDGF and GH cross-talk and modulate each others' action at the cellular level. Cross-talk among different hormones, cytokines, and growth factors has been shown to be an important mechanism regulating the magnitude and specificity of cellular responses and to be involved in many physiological and/or pathological processes. For instance, tumor necrosis factor-␣ (TNF-␣) causes insulin resistance by inhibiting insulin signaling at the cellular level (46,47).
In this study, we demonstrate that PDGF, but not epidermal growth factor (EGF), dramatically inhibits GH-stimulated ty-rosyl phosphorylation of JAK2 and GHR and rapidly reduces the number of both total and cell surface GHR. Depletion or inhibition of PMA-sensitive PKCs blocks the action of PDGF. In contrast, the MEK/MAP kinase and the PI 3-kinase pathways appear not to be involved in this action of PDGF. LPA, a G protein-coupled receptor that, like PDGF, activates PKC (20,21,48), similarly inhibits GH-stimulated tyrosyl phosphorylation of JAK2 and GHR and reduces the number of GHR by a PKC-dependent pathway. We conclude that PDGF and LPA down-regulate GH signaling via a PKC-dependent pathway by at least in part reducing the level of GHR. It is likely that other ligands that activate PMA-sensitive PKCs down-regulate GH signaling in a similar fashion.

EXPERIMENTAL PROCEDURES
Materials-Recombinant hGH was a gift of Lilly. hGH was iodinated by the Reproductive Sciences Training Grant Core Facility at the University of Michigan Medical School to a specific activity of ϳ2,000 Ci/nmol. Recombinant murine EGF was from Collaborative Biomedical Products. Recombinant human PDGF-BB was from Intergen. Recombinant murine LIF was from R & D Systems. Recombinant murine IFN-␥, LPA, and PMA were from Sigma. Protein A-agarose was from Repligen. Aprotinin, leupeptin, and Triton X-100 were purchased from Roche Molecular Biochemicals. Enhanced chemiluminescence (ECL) detection system was from Amersham Pharmacia Biotech. Wortmannin and bisindolylmaleimide I (GF109203X) were from Calbiochem. Anti-JAK2 antiserum (␣JAK2) was raised in rabbits against a synthetic peptide corresponding to amino acids 758 -776 (49) and was used at a dilution of 1:500 for immunoprecipitation and 1:15,000 for immunoblotting. Polyclonal anti-GHBP (␣GHBP) recognizing the extracellular domain of GHR (50) was a gift of Dr. W. R. Baumbach (American Cyanamid, Princeton, NJ) and was used at a dilution of 1:250 for immunoprecipitation and 1:5,000 for immunoblotting. Polyclonal antiphosphorylated (Ser-473), active Akt (␣pAkt) was from New England Biolabs Inc. and was used at a dilution of 1:1,000 for immunoblotting. Polyclonal anti-phosphorylated (Thr-183 and Tyr-185), active MAP kinase (␣active MAP kinase) was from Promega and was used at a dilution of 1:5,000 for immunoblotting. Monoclonal anti-phosphotyrosine antibody 4G10 (␣PY) was purchased from Upstate Biotechnology Inc. and was used at a dilution of 1:7,500 for immunoblotting. The stock of 3T3-F442A cells was provided by H. Green (Harvard University, Cambridge, MA).
Immunoprecipitation and Immunoblotting-3T3-F442A fibroblasts were grown on 100-mm tissue culture dishes in Dulbecco's modified Eagle's medium supplemented with 8% calf serum, 1 mM L-glutamine, 100 units/ml penicillin, 100 g/ml streptomycin, and 0.25 g/ml amphotericin B. Cells were deprived of serum overnight in the same growth medium except that 1% bovine serum albumin (BSA) was substituted for the calf serum. The deprived cells were treated for various times with the indicated drugs and/or ligands at 37°C and then rinsed three times with 10 mM sodium phosphate, pH 7.4, 150 mM NaCl, 1 mM Na 3 VO 4 . The cells were solubilized in lysis buffer (50 mM Tris, pH 7.5, 0.1% Triton X-100, 150 mM NaCl, 2 mM EGTA, 1 mM Na 3 VO 4 , 1 mM phenylmethylsulfonyl fluoride, 10 g/ml aprotinin, 10 g/ml leupeptin), and centrifuged at 14,000 ϫ g for 10 min at 4°C. Proteins in the supernatant were quantified using BCA TM Protein Assay Reagent (Pierce). The supernatant was incubated with the indicated antibody on ice for 2 h. The immune complexes were collected on protein A-agarose (50 l) during 1 h incubation at 4°C. The beads were washed 3 times with washing buffer (50 mM Tris, pH 7.5, 0.1% Triton X-100, 150 mM NaCl, 2 mM EGTA) and boiled for 5 min in a mixture (80:20) of lysis buffer and SDS-polyacrylamide gel electrophoresis (PAGE) sample buffer (250 mM Tris-HCl, pH 6.8, 10% SDS, 10% ␤-mercaptoethanol, 40% glycerol, 0.01% bromphenol blue). The eluted proteins were separated by SDS-PAGE (5-12% gradient gel) followed by immunoblotting with the indicated antibody using the ECL detection system.
To assess the total amount of GHR, 3T3-F442A cells were solubilized in the above lysis buffer supplemented with 1% SDS and boiled for 5 min. The concentration of proteins in the cell lysates was determined using BCA TM Protein Assay Reagent. Proteins (50 g) in the lysates were boiled for 5 min in a mixture (80:20) of lysates and SDS-PAGE sample buffer, separated by SDS-PAGE, and subjected to immunoblotting using ␣GHBP.

125
I-hGH Binding Assay-Confluent 3T3-F442A fibroblasts in 6-well plates were deprived of serum overnight and stimulated with the indicated ligands for 45 min. The cells were washed with Krebs-Ringer phosphate (KRP) buffer (128 mM NaCl, 6.7 mM KCl, 1 mM CaCl 2 , 2.6 mM MgSO 4 , and 10 mM Na 2 HPO 4 , pH 7.4) containing 1% BSA and then incubated in KRP, 1% BSA containing 125 I-hGH (80,000 cpm/well) overnight at 4°C. The cells were then washed with cold KRP and solubilized in 1 ml of 1 M NaOH for counting radiation. The concentration of protein in the cell lysates was determined using BCA Protein Assay Reagent. 125 I-hGH binding activity was normalized to the protein level and expressed as mean Ϯ S.E.
For quantification, immunoblots were scanned using an Agfa ArcusII scanner and Fotolook SA software (Mortsel, Belgium). The resulting image was analyzed using multi-analyst image analysis software from Bio-Rad.

RESULTS
PDGF Rapidly Inhibits GH-induced Tyrosyl Phosphorylation of JAK2 and GHR and Reduces the Amount of GHR-To examine whether PDGF cross-talks with GH, 3T3-F442A cells, which express endogenous receptors for PDGF (51,52) and GH (53), were deprived of serum overnight, preincubated with 25 ng/ml PDGF for 20 min, and stimulated with 50 ng/ml GH for an additional 10 min. JAK2 was immunoprecipitated with ␣JAK2 and immunoblotted with ␣PY. The level of tyrosyl phosphorylation of JAK2 was used to assess the extent of activation of JAK2 by GH (6,16). PDGF substantially inhibits GH-induced tyrosyl phosphorylation of JAK2 ( Because GHR is a physiological substrate of JAK2, we tested whether PDGF alters tyrosyl phosphorylation of GHR in response to GH. 3T3-F442A cells were deprived of serum overnight and pretreated with PDGF for 20 min prior to stimulation with 50 ng/ml GH for 10 min. GHR was immunoprecipitated with ␣GHBP that recognizes the extracellular domain of GHR and immunoblotted with ␣PY or ␣GHBP. PDGF dramatically inhibits GHinduced tyrosyl phosphorylation of GHR ( We believe the mobility shift is caused by phosphorylation of GHR, presumably on tyrosines (6,54,55).
PDGF-induced inhibition of GH signaling is very rapid. Pretreatment with PDGF for 2 min results in significant inhibition of GH-dependent tyrosyl phosphorylation of JAK2 ( Fig. 2A,  lane 3) and GHR (Fig. 2B, lane 3). The inhibition of tyrosyl phosphorylation of GHR and JAK2 reaches a maximal level within 15 min pretreatment with PDGF (Fig. 2, A and B). This rapid onset makes it unlikely that PDGF-induced inhibition of GH signaling involves new gene expression or synthesis of new proteins. PDGF-induced inhibition of GH-dependent tyrosyl phosphorylation of JAK2 (Fig. 2C) and GHR (Fig. 2D) is dependent on the dose of PDGF. It is detectable at 1 ng/ml and reaches a maximum with 25 ng/ml PDGF (Fig. 2, C and D).
PDGF also reduces the level of GHR in a time-and dose-dependent manner. The reduction is substantial at 2 min, and maximal within 15 min stimulation with PDGF (Fig. 2B, lower  panel). The reduction of GHR is detected at a concentration as low as 1 ng/ml PDGF (Fig. 2D, lane 3, lower panel). 2 Thus, the PDGF-induced inhibition of GH-stimulated tyrosyl phosphorylation of JAK2 (Fig. 2, A and C, upper panel) and GHR (Fig. 2, B and D, upper panel) correlates in magnitude, time, and dose with the PDGF-induced reduction of GHR (Fig. 2, B and D,  lower panel). This correlation suggests that PDGF inhibits GH signaling by reducing the amount of GHR present in cells.
PDGF Decreases the Number of GHR on the Cell Surface Available to Bind to GH-Because PDGF does not interfere with the activation of JAK2 by IFN-␥ (Fig. 1C) or LIF (Fig. 1D), and PDGF-induced inhibition of tyrosyl phosphorylation of JAK2 and GHR correlates with the reduction in the level of cellular GHR (Fig. 2), we reasoned that PDGF does not directly inhibit JAK2 but rather reduces the number of GHR on the cell surface available to bind GH. To examine whether PDGF decreases the number of GHR on the cell surface available to bind GH, 3T3-F442A cells were deprived of serum overnight, treated for 45 min with 25 ng/ml PDGF, 125 ng/ml EGF, or 200 nM PMA, and subjected to a GH binding assay at 4°C as described under "Experimental Procedures." PMA has been shown to reduce GH binding in 3T3-F442A cells (57) and in IM9 cells (58) and was used as a positive control. PDGF reduced GH binding by approximately 70% (Fig. 3). PMA decreased GH binding to a slightly greater extent (by approximately 75%). In contrast, EGF did not affect GH binding.
The PDGF-induced Reduction of GHR Requires neither the ERK Cascade nor PI 3-Kinase-To verify the reduction of total cellular GHR by PDGF described in Figs. 1 and 2, 3T3-F442A cells were deprived of serum overnight and treated with 25 ng/ml PDGF or 125 ng/ml EGF for 40 min. Cells were lysed in lysis buffer containing 1% SDS and boiled for 5 min. No residual cell pellet was observed following this treatment. The whole cell lysates were subjected to SDS-PAGE. Proteins in the cell lysates were immunoblotted with ␣GHBP (Fig. 4A, lanes 2-4).
To verify the migration of GHR, ␣GHBP immunoprecipitate was also analyzed on the same SDS-PAGE gel (Fig. 4A, lane 1). PDGF reduced the amount of total cellular GHR by approximately 75% (Fig. 4A, lanes 3 versus 2). In contrast, EGF does not affect the level of GHR (Fig. 4A, lane 4).
PDGF activates multiple signaling molecules and pathways, including the MEK/ERK cascade, the PI 3-kinase/Akt pathway, and the PLC␥/PKC pathway. To determine which molecules and pathways are involved in PDGF-induced inhibition of GH signaling, 3T3-F442A cells were treated with 25 ng/ml PDGF or 125 ng/ml EGF for 40 min, and proteins in cell lysates were immunoblotted with ␣PY (Fig. 4B, top panel) antibody to the activated form of mitogen-activated protein kinase which recognizes the activated, dually phosphorylated form of ERKs 1 and 2 (Fig. 4B, middle panel) or antibody to the activated form of Akt phosphorylated on Ser-473 which is the site phosphorylated by PI 3-kinase (Fig. 4B, bottom panel). PDGF stimulates 2 Proteins in the tight band migrating with apparent M r ϳ90,000 in Fig. 2, B and D, lower panel, are believed not to be functional GHR, because they are not tyrosyl-phosphorylated in response to GH (Fig. 2,  B and D, upper panel). tyrosyl phosphorylation of two proteins migrating with apparent molecular weights of approximately 175,000 and 145,000 (Fig. 4B, lane 2, top panel), sizes appropriate for the PDGF receptor and PLC␥, respectively. In contrast, EGF stimulates tyrosyl phosphorylation of a protein migrating with apparent molecular weight 160,000, a size appropriate for the EGF receptor (Fig. 4B, lane 3, top panel). PDGF activates both ERKs 1 and 2 (Fig. 4B, lane 2, middle panel) and Akt (Fig. 4B, lane 2,  bottom panel). EGF activates ERKs 1 and 2 to an extent similar to that observed with PDGF (Fig. 4B, lane 3, middle panel) but does not activate Akt (Fig. 4B, lane 3, bottom panel). Because PDGF but not EGF reduces the amount of GHR and inhibits GH-induced tyrosyl phosphorylation of JAK2 and GHR, it seems unlikely that ERKs 1 and 2 are involved in the inhibition of GH signaling by PDGF.
Because PDGF appears to be a more potent activator of Akt than EGF (Fig. 4B, bottom panel), a downstream effector of PI 3-kinase (59), we tested whether PI 3-kinase plays a role in the inhibition of GH signaling by PDGF. 3T3-F442A cells were pretreated for 20 min with 200 nM wortmannin, a potent inhibitor of PI 3-kinase, incubated with 25 ng/ml PDGF for 20 min, and then stimulated for 10 min with 50 ng/ml GH. Pro-teins in cell lysates were immunoprecipitated with ␣JAK2 or ␣GHBP and immunoblotted with ␣PY, ␣JAK2, or ␣GHBP. Wortmannin is unable to inhibit the effect of PDGF on GHinduced tyrosyl phosphorylation of JAK2 (Fig. 4C, upper panel) or GHR (Fig. 4D, upper panel). To verify the inhibition of PI 3-kinase by wortmannin, proteins in cell lysates were immunoblotted with antibody recognizing specifically the phosphorylated, active form of Akt. GH stimulates phosphorylation of Akt (Fig. 4E, lane 2). PDGF and GH together stimulate phosphorylation of Akt to a much greater extent (Fig. 4E, lane 3). Wortmannin completely blocks phosphorylation of Akt by GH and PDGF (Fig. 4E, lane 4). These data indicate that PI 3-kinase and signaling molecules downstream of PI 3-kinase are unlikely to be involved in the inhibition of GH signaling by PDGF.
PKCs Are Required for Down-regulation of GH Signaling by PDGF-In 3T3-F442A cells, PDGF stimulates a more robust and sustained tyrosyl phosphorylation of PLC␥ compared with EGF (Fig. 4B, top panel, and data not shown). Because PLC␥ is an upstream activator of PKC, and activation of the PKC pathway by PMA has been shown to decrease GH binding (57,58,60), we hypothesized that PKC-initiated signaling events play a role in the inhibition of GH signaling by PDGF. To test this hypothesis, PMA-sensitive isoforms of PKC were depleted from 3T3-F442A cells by incubating cells with PMA (200 nM) for 25 h. The treated cells were then incubated with 25 ng/ml PDGF for 20 min, followed by 50 ng/ml GH for 10 min. JAK2 was immunoprecipitated with ␣JAK2 and immunoblotted with ␣PY (Fig. 5A, upper panel) or ␣JAK2 (Fig. 5A, lower panel). PDGF and PMA dramatically inhibit the tyrosyl phosphorylation of JAK2 induced by GH (Fig. 5A, lanes 3 and 4), consistent with the data in Figs. 1 and 2. Depletion of PMA-sensitive PKCs abolishes the ability of either PDGF or PMA to inhibit GH-stimulated tyrosyl phosphorylation of JAK2 (Fig. 5A, lanes  7 and 8, upper panel). Treatment with PMA for 25 h does not increase basal tyrosyl phosphorylation of JAK2 (Fig. 5A, lane 5, upper panel) but slightly increases the amount of JAK2 (Fig.  5A, lower panel). Similarly, when PMA-sensitive PKCs are depleted, GH-stimulated tyrosyl phosphorylation of GHR is not affected by either PDGF or PMA (Fig. 5B, lanes 7 and 8).
LPA Down-regulates GH Signaling-Many hormones and growth factors that activate G protein-coupled receptors activate the PKC pathway. We hypothesized that these ligands would down-regulate GH signaling in a fashion similar to PDGF. We tested whether LPA, whose receptor is widely expressed in tissues and cell lines and activates PKCs via G q (61), alters GH-induced tyrosyl phosphorylation of JAK2 and GHR. 3T3-F442A cells were preincubated for 20 min with the indicated concentrations of LPA prior to 50 ng/ml GH stimulation for an additional 10 min. Proteins in cell lysates were immunoprecipitated with ␣JAK2 (Fig. 7A) and ␣GHBP (Fig. 7, B and C) and immunoblotted with ␣PY (Fig. 7, A-C, upper panel), ␣JAK2 (Fig. 7A, lower panel), and ␣GHBP (Fig. 7C, lower  panel). LPA substantially inhibits GH-stimulated tyrosyl phosphorylation of JAK2 (Fig. 7A) and GHR (Fig. 7, B and C, upper   panel). LPA also significantly reduces the amount of total cellular GHR (Fig. 7C, lower panel) and GH binding (data not shown). Depletion of PMA-sensitive PKCs blocks the inhibitory effect of LPA on GH-stimulated tyrosyl phosphorylation of JAK2 and GHR (data not shown). DISCUSSION We report in this work that PDGF and LPA are potent inhibitors of GH signaling. Pretreating cells with PDGF or LPA dramatically inhibits GH-dependent tyrosyl phosphorylation of the tyrosine kinase JAK2, a key enzyme in GH signaling. Because tyrosyl phosphorylation of JAK2 correlates with its activation (6), PDGF and LPA most likely strongly inhibit GH-dependent activation of JAK2. Consistent with this idea, GH-induced tyrosyl phosphorylation of GHR, a physiological substrate of JAK2, is also severely inhibited by PDGF and LPA. Because activation of JAK2 and tyrosyl phosphorylation of JAK2 and GHR are early obligatory steps in GH signaling (7,62), it is likely that most, if not all, downstream signaling events are inhibited by PDGF and LPA.
Another important finding of this work is that PDGF and LPA decrease both GH binding and total cellular GHR. PDGF substantially reduces both GH binding and total cellular GHR. The decrease in GH binding and number of GHR roughly correlates with the inhibition of GH-induced tyrosyl phosphorylation of JAK2 and GHR, suggesting that the reduction of GHR is the primary cause of down-regulation of GH signaling by PDGF and LPA. In agreement with this idea, PDGF does not inhibit the activation of JAK2 by LIF or IFN-␥, suggesting that PDGF and LPA do not directly inhibit JAK2.
When PMA-sensitive isoforms of PKC are depleted by preincubating cells for 25 h with PMA, neither PDGF nor LPA inhibits GH-induced tyrosyl phosphorylation of JAK2 and GHR, indicating that PMA-sensitive PKCs are required for PDGF-and LPA-induced inhibition of GH action. In support of this hypothesis, GF109203X, a potent inhibitor of PKCs, substantially blocks the inhibition of GH-induced tyrosyl phosphorylation of JAK2 and GHR and the reduction in GHR by PDGF. In agreement with an essential role of PKC in down-regulation of GH signaling by PDGF and LPA, activation of PKC by PMA is sufficient to inhibit GH-induced tyrosyl phosphorylation of JAK2 and GHR and reduce the number of GHR. Interestingly, depletion of PMA-sensitive PKCs blocks the ability of both PDGF and PMA to decrease GH signaling, whereas GF109203X, which is capable of almost completely blocking the ability of PMA to inhibit GH signaling, only partially blocks the inhibitory effect of PDGF on GH signaling. One explanation for this apparent discrepancy is that PDGF stimulates a subset of PKCs that are less sensitive to GF109203X. Different PKCs are known to have different sensitivities to GF109203X (63,64).
It is unlikely that the MEK/ERK cascade and the PI 3-kinase pathway contribute to the down-regulation of GH signaling by PDGF and LPA. EGF, which activates ERKs 1 and 2 to a similar extent as PDGF, is unable to either inhibit GH-induced tyrosyl phosphorylation of JAK2 and GHR or reduce the number of GHR. In addition, when ERKs 1 and 2 are inhibited by PD98059, a potent inhibitor of MEK, or partially inhibited by 1 M wortmannin, the inhibitory effect of PDGF on GH signaling is not affected (data not shown). Similarly, inhibition of PI 3-kinase activity with wortmannin completely blocks GH-and PDGF-induced phosphorylation and activation of Akt but does not affect PDGF-induced inhibition of GH signaling.
Although it has been reported previously that PMA can regulate levels of GH binding (57,58,60), this is the first report of physiological ligands that down-regulate GH binding and GHR signaling by a PKC-dependent mechanism. Two models have been proposed to explain the inhibition of GH signaling by PMA-sensitive PKCs (57,58,60). In the first model (57,58), activation of PKCs by PMA leads to a redistribution of GHR within the cell, resulting in a reduction of cell surface GHR and an increase in cytoplasmic GHR. Our data do not support the second half of this model (i.e. cytoplasmic GHR are increased in number). However, we cannot exclude the possibility that PMA activation of PKCs increases the rate of internalization of GHR which are then rapidly degraded, resulting in a decrease in the overall number of GHR.
In the second model (60), activation of PKCs by PMA leads to the activation of a protease that cleaves cell surface GHR. This results in the formation of GHR lacking its extracellular domain and release of GH-binding protein (GHBP, the extracellular domain of GHR) from the cell (shedding). Recent evidence suggests that the PKC-regulated protease may be tumor necrosis factor (TNF)-␣-converting enzyme (TACE) or a TACE-like metalloprotease (65). Consistent with a TACE-like metaolloprotease cleaving GHR and producing GHBP, PMA activation of TACE family members has been shown to lead to ectodomain cleavage of the receptors for TNF-␣ and transforming growth factor-␣, the adhesion protein L-selectin, and ␤-amyloid protein precursor (66,67). Also consistent with this second model is the finding that the PMA-induced decrease in GHR binding is dependent upon a region of GHR composed of the extracellular and transmembrane domains (57). The best evidence for the production of GHBP as a by-product of GHR cleavage comes from studies with GHR from humans and rabbits (60, 68 -70), species in which proteolysis is thought to be the major source of GHBP (71). However, a preliminary report indicates that formation of GHBP as a result of proteolytic cleavage of GHR also occurs with GHR from mice (65), a species for which GHBP is also synthesized from the gene for GHR as an alternative splice product (71). If proteolysis accounts for at least some of the PKC-induced decrease in GH binding, then it seems likely that PDGF, LPA, or any other ligand that activates PMA-sensitive PKCs will stimulate the production of GHBP in cells that express GHR. Up to 50% of human serum GH is believed to bind to GHBP (72). The interaction of GH with GHBP is proposed to increase the stability of GH. GHBP has also been reported to have inhibitory (73) as well as potentiating roles in GH signaling (74 -76). Thus, in addition to regulating the ability of an individual cell to respond to GH, ligands such as PDGF and LPA could also increase the local production of GHBP and thereby modulate GH signaling in neighboring cells. In addition, because PMA is known to activate members of the TACE family of proteases, it seems likely that PDGF and LPA might stimulate the cleavage of the receptors for TNF-␣ and transforming growth factor-␣, the adhesion protein L-selectin, and/or the ␤-amyloid protein precursor.
The finding that PDGF and LPA inhibit GH signaling via a PKC-dependent pathway may have important therapeutic implications. Receptors for PDGF and LPA are members of the receptor tyrosine kinase family and the G protein-coupled receptor family, respectively. Many ligands, including numerous chemical compounds and a variety of hormones, cytokines, and growth factors, are able to activate PKCs. It seems likely that any ligand that activates PMA-sensitive PKCs will also downregulate GH action. Receptors for a variety of hormones, growth factors, and cytokines that activate PKC, including LPA and PDGF, are expressed in GH target tissues and cells. It is likely that these ligands constantly modulate GH action in vivo by regulating the abundance of GHR on the plasma membrane through a PKC-dependent pathway. Therefore, it is extremely important during GH replacement therapy to be cognizant of the fact that agents that stimulate PKCs will render a given dose of GH much less effective. Similarly, it is important to be cognizant that continuous treatment with agents that stimulate PKCs may inhibit growth.
In summary, we have shown that PDGF and LPA inhibit GH-induced tyrosyl phosphorylation of JAK2 and GHR. PDGF and LPA appear to down-regulate GH signaling by decreasing the number of GHR. PMA-sensitive PKCs, but not ERKs 1 and 2 nor PI 3-kinase, are required for these actions of PDGF and LPA. We propose that any ligand that actives PMA-sensitive isoforms of PKC inhibits GH signaling in a similar fashion.