Insulin-Like Growth Factor (IGF)-I Stimulates Cell Proliferation and Induces IGF Binding Protein (IGFBP)-3 and IGFBP-5 Gene Expression in Cu
http://www.100md.com
内分泌学杂志 2005年第7期
University Children’s Hospital (D.K., S.C., B.T.), 69120 Heidelberg; and Institute of Molecular Animal Breeding and Biotechnology/Gene Center (A.H., E.W.), 81377 Munich, Germany
Address all correspondence and requests for reprints to: Burkhard T?nshoff, M.D., Ph.D., University Children’s Hospital, Im Neuenheimer Feld 153, 69120 Heidelberg, Germany. E-mail: burkhard_toenshoff@med.uni-heidelberg.de.
Abstract
The bioactivity of IGF-I in the cellular microenvironment is modulated by both inhibitory and stimulatory IGF binding proteins (IGFBPs), whose production is partially under control of IGF-I. However, little is known on the IGF-mediated regulation of these IGFBPs in the growth plate. We therefore studied the effect of IGF-I on IGFBP synthesis and the involved intracellular signaling pathways in two cell culture models of rat growth plate chondrocytes. In growth plate chondrocytes in primary culture, incubation with IGF-I increased the concentrations of IGFBP-3 and IGFBP-5 in conditioned cell culture medium in a dose- and time-dependent manner. Coincubation of IGF-I with specific inhibitors of the p42/44 MAPK pathway (PD098059 or U0126) completely abolished the stimulatory effect of IGF-I on IGFBP-3 mRNA expression but did not affect increased IGFBP-5 mRNA levels. In contrast, inhibition of the phosphatidylinositol-3 kinase signaling pathway by LY294002 abrogated both IGF-I-stimulated IGFBP-3 and -5 mRNA expression. Comparable results regarding IGFBP-5 were obtained in the mesenchymal chondrogenic cell line RCJ3.1C5.18, which does not express IGFBP-3. The IGF-I-induced IGFBP-5 gene expression required de novo mRNA transcription and de novo protein synthesis. These data suggest that IGF-I modulates its activity in cultured rat growth plate chondrocytes by the synthesis of both inhibitory (IGFBP-3) and stimulatory (IGFBP-5) binding proteins. The finding that IGF-I uses different and only partially overlapping intracellular signaling pathways for the regulation of two IGFBPs with opposing biological functions might be important for the modulation of IGF bioactivity in the cellular microenvironment.
Introduction
IGF-I IS A POTENT growth factor in the growth plate, exerting its actions by both endocrine and paracrine/autocrine mechanisms. It is known from cell culture (1, 2, 3) and gene knockout experiments (4, 5, 6) that IGF-I stimulates both proliferation and differentiation of growth plate chondrocytes in vitro and in vivo. The action of the IGFs in both the circulation and tissues is tightly regulated by a family of high-affinity IGF binding proteins (IGFBPs). Up to now, six distinct IGFBPs have been identified. Although structurally related, they have individual expression patterns and exert different functions including stimulation or inhibition of IGF bioactivity as well as IGF-independent actions (7, 8, 9).
We previously studied the biological activities of various intact IGFBPs in a growth cartilage model of proliferating growth plate chondrocytes in primary cultures (10, 11). IGFBP-1, -2, -4, and -6 act exclusively as growth inhibitors on IGF-dependent cell proliferation, whereas the biological activity of IGFBP-3 is complex. It has an IGF-independent antiproliferative effect and also inhibits IGF-dependent chondrocyte proliferation under coincubation conditions. Under preincubation conditions, however, IGFBP-3 enhances the IGF-I responsiveness of growth plate chondrocytes by its ability to associate with the cell membrane, in which it presumably facilitates IGF-I receptor binding (11). Intact IGFBP-5, on the other hand, enhances IGF-I-induced chondrocyte proliferation, apparently by its association with the cell membrane in the C-terminal domain, thereby better presenting IGF-I to its receptor (10).
The synthesis of the IGFBPs in various tissues is under partial control of IGF-I. IGF-I exerts its biological effect by binding to the transmembrane type 1 IGF receptor, whose activation leads to the extensive tyrosyl-phosphorylation of insulin receptor substrate-1, which acts as a docking protein for the downstream signal transduction pathways (12, 13). Two canonical pathways, the phosphatidylinositol-3 kinase (PI-3 kinase) and the p42/44 MAPK pathway, have been reported previously to mediate the mitogenic, differentiating and antiapoptotic response to IGF-I (14), but the relative contributions to the diverse cellular actions of IGF-I vary according to the cell type (15). It is therefore necessary to study IGF-I signaling in individual tissues to determine the role of each pathway in that specific cell type.
Little is known on the regulation of IGFBPs by the IGFs in the growth cartilage and the underlying intracellular signaling mechanisms. We therefore examined in the present study the effect of IGF-I on IGFBP synthesis in two different cell culture models of growth plate chondrocytes. Furthermore, we investigated the involved IGF-I-activated signaling pathways by use of specific pharmacological inhibitors. We report here that IGF-I modulates its activity in rat growth plate chondrocytes by stimulating the synthesis of both inhibitory (IGFBP-3) and stimulatory (IGFBP-5) binding proteins. The IGF-I-induced cell proliferation and IGFBP-3 mRNA expression are mediated through both the p42/44 MAPK and PI-3 kinase pathway, whereas the IGF-I-induced IGFBP-5 mRNA expression depends only on a functional PI-3 kinase pathway.
Materials and Methods
Materials
Recombinant human IGF-I was purchased from Bachem (Heidelberg, Germany); [3H]thymidine (25 Ci/mmol) and 32P-labeled uridine 5-triphosphate were obtained from Amersham Pharmacia Biotech (Freiburg, Germany). Radioiodination of IGF-II with 125I (>2000 Ci/mmol of protein) was performed using a standard chloramine T method (10). Briefly, the peptide was dissolved in 0.25 M sodium phosphate, and Na125I (0.2 mCi; Amersham, Braunschweig, Germany) and 10 μl chloramine-T (2.5 mg/ml in phosphate buffer) were added. After incubation, the reaction was stopped by the addition of Na2S2O5. The radiolabeled peptide was purified on Sep-Pak C18 cartridges (Waters-Millipore, Eschborn, Germany). The homogeneity and the specific activity of the radiolabeled peptide were controlled on the HPLC system 322 (Kontron, Neufahrn, Germany).
PBS, HEPES, penicillin-streptomycin, Ham’s F-12, and DMEM were obtained from Seromed Biochrom KG (Berlin, Germany). BSA was purchased from Sigma-Aldrich Chemicals (Deisenhofen, Germany). Clostridium collagenase (EC 3.4.24.3), deoxyribonuclease [DNase I (EC 3.1.21.1)], and trypan blue were from Roche Diagnostics GmbH (Mannheim, Germany). MEM was purchased from cc Pro (Neustadt, Germany); fetal calf serum from Paa Laboratories (Pasching, Austria); and ascorbic acid, ?-glycerophosphate, and dexamethasone from Sigma (Taufkirchen, Germany). Monoclonal antibodies against p44/42 MAPK and phospho-p44/42 MAPK were obtained from Cell Signaling (Frankfurt a. M., Germany). The antibody against hIGFBP-5 was purchased from Santa Cruz Biotechnologies (Heidelberg, Germany); antirat IGFBP-4 and antirat IGFBP-5 antibodies were from GroPep (Adelaide, Australia).
Cell cultures
Epiphyseal chondrocytes from 60- to 80-g Sprague Dawley rats (Charles River, Kieslegg, Germany) were isolated and cultured, as described previously (10, 16). This study was approved by the Institutional Animal Care and Use Committee (35–9185.81/102/98). Pooled growth plates from four to eight animals were digested with clostridial collagenase (0.12% wt/vol) and bacterial DNase (0.02% wt/vol) in F-12 medium. Viability, determined after isolation and at the end of each experiment by the trypan blue exclusion technique, always exceeded 90%. Dissociated cells were counted using a Neubauer chamber (Scheik, Hofheim, Germany).
Cells were cultured in monolayers and in 96-well plates for proliferation assays (Nunc, Wiesbaden, Germany), as described previously (10, 16). The F-12/DMEM (1:1) medium contained a nominal calcium concentration of 1.2 mM, and the medium was supplemented with 10 mM HEPES, 100 μg/ml streptomycin, and 10% fetal calf serum. In previous studies using the same culture system, we demonstrated that the majority of cells after the first passage expressed typical markers for proliferative growth plate chondrocytes (16).
RCJ3.1C5.18 (RCJ) cells (kindly provided by Dr. Anna Spagnoli, Department of Pediatrics, Vanderbilt University Medical Center, Nashville, TN) were grown at 37 C in humidified 5% CO2 atmosphere in MEM (with Earle’s salts) supplemented with 1 mM N-acetyl-L-glutamine, 10 mM HEPES, 100 U/ml penicillin-streptomycin, 2 mM sodium pyruvate, 15% heat-inactivated fetal bovine serum, and 10–7 M dexamethasone and studied within 25 passages (17). Cells were plated at a density of 6 x 104 cells/well in 6-well dishes for RNA extraction and protein isolation. Peptide hormones were dissolved in PBS and added every day unless indicated otherwise. The inhibitor LY294002 (Calbiochem, San Diego, CA) was dissolved in 20% dimethylsulfoxide in ethanol following the instructions of the manufacturer. PD098059 (Calbiochem) and U0126 (Promega, Mannheim, Germany) were dissolved in dimethylsulfoxide following the instructions of the manufacturer. The inhibition of MAPK kinase by PD098059 and U0126 is very selective, but U0126 has a 100-fold higher affinity for MEK than PD098059 (18).
[3H]thymidine incorporation assay
The extent of thymidine incorporation into DNA was determined as uptake of radioactivity in precipitated material, as described previously (10, 16). Briefly, chondrocytes were grown in 96-well plates to subconfluence for 3 d; thereafter, the cultures were changed to serum-free medium. After 12 h, various concentrations of inhibitors with or without IGF-I were added to the medium and the incubation was continued for further 48 h. The rate of chondrocyte proliferation was assessed by incubating the cells with 3 μCi/ml [3H]thymidine for the final 4 h. Subsequently, cells were rinsed twice with PBS. The radiolabeled DNA was precipitated by acetic acid and dissolved in 1 M NaOH. [3H]thymidine incorporation into the acid-extractable pool as a measure of DNA synthesis was determined by scintillation counting.
Real-time RT-PCR.
For analysis of IGFBP-5 mRNA abundance, cells were cultured in differentiating medium until d 7, serum-starved for 12 h and stimulated for an additional 12 h with IGF-I in the presence or absence of inhibitors. RNA was isolated by using the RNeasy minicolumns (QIAGEN, Hilden, Germany), and 1 μg was reverse transcribed using Moloney leukemia virus reverse transcriptase and oligo(dt)/random hexamer primers (10:1) from Applied Biosystems (Darmstadt, Germany). For quantitative analysis, real-time RT-PCR was performed using the Abi 7000 (Applied Biosystems) according to the manufacturer’s protocol. The following set of primers was chosen: hIGFBP-5, forward, 5'-TCAAGATCGAGAGAGACTCCCG-3', reverse, 5'-TTCACTGCTTCAGCCTTCAGC; rIGFBP-5, forward, 5'-AGTCGTGTGGCGTCTACACTGA-3', reverse, 5'-TTTGCTCGCCGTAGCTCTTTT-3'; and rat 18S by the Primer Express program, forward, 5'-AGTTGGTGGAGCGATTTGTC-3', reverse, 5'-GCTGAGCCAGTTCAGTGTAGC-3' (Applied Biosystems). The software provided from the company allowed the quantitative detection of fluorescence by the incorporation of the substance SYBR green into the amplification products. Amplification was performed in the presence of Universal Mastermix (PE Applied Biosystems) with SYBR green to detect PCR products at the end of each amplification step, and results were analyzed as already reported (19).
Ribonuclease protection assay
The amplification products of IGFBP-3 (292 nt) and IGFBP-5 (296 nt) obtained by conventional PCR were cloned into pBlueScript II KS (Stratagene, Techno-Path, Limerick, Ireland), sequenced (MWG Biotech AG, Ebersberg, Germany), and in vitro transcribed using 32P-labeled uridine 5-triphosphate (2.5 μM) and unlabeled nucleotides (Amersham Pharmacia Biotech). Lyophilized samples of total RNA (10 μg for IGFBPs and 50 ng for 18S rRNA) were hybridized with the radiolabeled antisense riboprobes at 45 C overnight. The free probe RNA was digested followed by proteinase K treatment. After phenolization and ethanol precipitation, hybrids were eluted in RNA loading buffer and separated on a sequencing gel. The ODs of the autoradiographs were analyzed by a commercially available computer program (Bio-Rad Laboratories, Inc., Vilber Lourmat, France). Protected bands were quantified densitometrically and normalized against 18S rRNA (19).
Western ligand blotting and immunoblotting
Cells were maintained in serum-free medium for 12 h and incubated with vehicles, peptides, or inhibitors as indicated. The collected conditioned medium was concentrated by trichloracetic acid precipitation, as described previously (20). Equal aliquots of concentrated conditioned medium were loaded on a 12% gel for nonreducing SDS-PAGE (10% acrylamide bis) and transferred to nitrocellulose membranes. Distinct membranes were blocked with 1% fish gelatin and incubated with [125I]-IGF-II (106 cpm/blot) for ligand blotting. All of the incubation and washing steps were performed at 4 C, as previously described (20). The same membranes were incubated overnight at 4 C with a polyclonal rabbit antibody against IGFBP-4 (GroPep, Adelaide, Australia; dilution 1:1000) for Western blotting, incubated for 1 h with the secondary antibody, and washed extensively over a period of 30 min with Tris-buffered saline and Tween 20 0.05% (TBS-T). For signal detection, autoradiography (X-OMAT AR film, Amersham Pharmacia Biotech) or a chemiluminescent detection system (ECL Western blotting detection reagents, Amersham Pharmacia Biotech) and Hyperfilm ECL film (Kodak, Stuttgart, Germany) were used.
Immunoprecipitation
One hundred microliters of packed A/G and agarose beads (Santa Cruz) were washed three times in PBS and then incubated with 15 μl IGFBP-5 antibody (GroPep; dilution 1:1000) overnight at 4 C. Two milliliters conditioned medium mixed with benzamidin (3 mM), aprotinin (10 μg/ml), leupeptin (10 μg/ml), phenylmethylsulfonyl fluoride (1 mM), and NaV03 (1 mM) were 10 times concentrated by centrifugation in a centricon-10 tube (Millipore/Amicon GmbH, Schwalbach, Germany). Two hundred microliters of concentrated conditioned medium were incubated with 15 μl of packed beads for 6 h at 4 C. The pellet was washed three times with 10 mM Tris-HCl buffer (pH 7.5) to remove IGFBPs nonspecifically precipitated with the antibody. The immunoprecipitated pellet was then resuspended in 20 μl sample sodium dodecyl sulfate (SDS)-loading buffer plus ?-mercaptoethanol [63 mmol Tris-HCl (pH 6.8); 10% glycerol; 2.1% SDS; 0.005% bromphenol blue] and boiled. Samples were subjected to Western immunoblotting by using the antibody against IGFBP-5 (GroPep; dilution 1:1000).
Statistical analysis
Data are given as mean ± SE. All data were examined for normal and non-Gaussian distribution by the Kolmogorov-Smirnov test. For comparison among normally distributed groups, one-way ANOVA followed by pairwise multiple comparison (Student-Newman-Keuls method) was used. For nonnormally distributed data, the nonparametric Kruskal-Wallis test followed by an all pairwise multiple comparison (Dunnett’s method) was used. P < 0.05 was considered statistically significant.
Results
IGF-I-induced chondrocyte proliferation is mediated through both the p42/44 MAPK and PI-3 kinase pathways
Under baseline conditions, inhibition of the p42/44 MAPK pathway by U0126 or PD098059 and inhibition of the PI-3 kinase pathway by LY294002 did not affect chondrocyte proliferation (Table 1). Incubation of chondrocytes with IGF-I (60 ng/ml) for 48 h significantly enhanced cell proliferation. We observed that the IGF-I-stimulated cell proliferation was completely suppressed by coincubation with the PI-3 kinase pathway inhibitor LY294002 and the two p42/44 MAPK pathway inhibitors U0126 or PD098059. Taken together, these data indicate that the IGF-I-stimulated cell proliferation depends on the p42/44 MAPK and PI-3 kinase pathways.
TABLE 1. Effect of the p42/44 MAPK inhibitors PD098059 and U0126 and effect of LY294002, a PI-3 kinase pathway inhibitor, on basal and IGF-I-stimulated DNA synthesis, as assessed by [3H]thymidine incorporation
Characterization of IGFBPs in conditioned cell culture medium
First, we sought to identify which IGFBPs are synthesized under baseline conditions by growth plate chondrocytes in primary culture. Conditioned cell culture medium was collected and subjected to Western ligand blotting, immunoblotting, and immunoprecipitation. By Western ligand blotting using [125I]-IGF-II as a ligand, IGFBPs with a molecular mass of 40–45, 35, 32, and 28 kDa were detected (Fig. 1A). After exposure to exogenous IGF-I (60 ng/ml) for 6–24 h, the intensity of the 40- to 45-kDa band (IGFBP-3) increased maximally approximately 2-fold, the intensity of the 32-kDa band (IGFBP-5) approximately 5-fold, and the intensity of the 28-kDa band (IGFBP-4) approximately 2-fold in a time-dependent manner. By immunoblotting with an antibody directed against human and rat IGFBP-4, the 28-kDa band already detected by [125I]-IGF-II ligand blotting and a 24-kDa band were identified as IGFBP-4 (Fig. 1B). It is known from previous investigations that the 24-kDa band corresponds to the nonglycosylated form of IGFBP-4 and the 28-kDa band to the glycosylated form (21). The 32-kDa band was identified as IGFBP-5 by immunoprecipitation with an antibody directed against human and rat IGFBP-5 (Fig. 1C). The identity of the 40- to 45-kDa band was not further characterized by immunoblotting because it is known that this band identified by IGF ligand blotting in conditioned medium of growth plate chondrocytes represents IGFBP-3 (22). The 35-kDa band, most likely representing IGFBP-2 (23), increased only slightly in response to IGF-I (Fig. 1A).
FIG. 1. Characterization of IGFBP proteins in conditioned cell culture medium of growth plate chondrocytes in primary culture. A, Western ligand blot of IGFBPs in conditioned cell culture medium under baseline conditions and after exposure to IGF-I (60 ng/ml) for the indicated time period, after 12 h of serum starvation. The concentrated samples were subjected to SDS-PAGE on a 12.5% acrylamide gel, and a Western ligand blot was performed by use of [125I]-IGF-II as the radioligand, as described in Materials and Methods. The molecular mass markers are indicated on the ordinate. B, Western immunoblot of IGFBP-4 in conditioned medium of growth plate chondrocytes treated with IGF-I (60 ng/ml) for 6 h after 12 h of serum starvation. The concentrated samples were subjected to SDS-PAGE on a 12.5% acrylamide gel, transferred to filters, and then immunodetected with IGFBP-4 antiserum, as described in Materials and Methods. C, Immunoprecipitation of IGFBP-5. Two hundred microliters of concentrated conditioned cell culture medium from growth plate chondrocytes treated with IGF-I (60 ng/ml) for 6 h after 12 h of serum starvation were incubated with 15 μl packed beads for 6 h at 4 C. The immunoprecipitated pellet was resuspended in 20 μl SDS loading buffer. The supernatants were subjected to Western immunoblotting using a polyclonal IGFBP-5 antibody.
IGF-I enhances IGFBP-3 and -5 mRNA expression and protein concentration in a dose- and time-dependent manner
Under baseline conditions, rat growth plate chondrocytes expressed mRNA species for IGFBP-2 to -6, whereas IGFBP-1 mRNA could not be detected by multiplex RT-PCR (data not shown). In the following series of experiments, we focused on IGFBP-3 and IGFBP-5 because these two IGFBPs have opposing biological functions in growth plate chondrocytes in primary culture and their synthesis increased the most in response to exogenous IGF-I. Incubation of cells with IGF-I led to a dose- and time-dependent increase of IGFBP-3 mRNA expression and protein content in conditioned medium; the extent of stimulation (2- to 4-fold) was comparable on the mRNA and protein level (Tables 2 and 3), suggesting that there was no major posttranscriptional or posttranslational modification of this IGFBP, e.g. by proteolysis. In contrast, we observed a sharp (up to 14-fold) increase of IGFBP-5 mRNA expression in an IGF dose- and time-dependent manner, whereas the increase of IGFBP-5 protein content (2-fold) was only moderate (Tables 2 and 3).
TABLE 2. IGF-I induces the respective mRNA abundance and protein concentration of IGFBP-3 and IGFBP-5 in conditioned cell culture medium in a dose-dependent manner
TABLE 3. IGF-I induces the respective mRNA abundance and protein concentration of IGFBP-3 and IGFBP-5 in conditioned cell culture medium in a time-dependent manner
IGF-I-induced IGFBP-3 mRNA expression is mediated through both the p42/44 MAPK and PI-3 kinase pathways, whereas the IGF-I-induced IGFBP-5 mRNA expression is mediated only through the PI-3 kinase pathway
Next, we sought to determine the intracellular signaling pathways, by which IGF-I differentially regulates IGFBP-3 and IGFBP-5 expression. We used two pharmacological inhibitors of the p42/44 MAPK pathway, U0126 and PD098059. Both inhibitors at the indicated concentrations were capable of suppressing phosphorylation of ERK1/2 but did not alter total ERK1/2 concentration in chondrocyte cell lysates (Fig. 2). Under baseline conditions, inhibition of the p42/44 MAPK pathway by U0126 did not alter IGFBP-3 or IGFBP-5 mRNA expression (Fig. 3). However, coincubation of IGF-I with U0126 completely suppressed IGF-I-induced IGFBP-3 mRNA expression, whereas IGF-I-induced IGFBP-5 mRNA expression was unaffected (Fig. 3). Comparable results were obtained by use of the other p42/44 MAPK pathway inhibitor, PD098059 (Fig. 4). For these experiments, the same concentrations of U0126 and PD098059 that were able to suppress ERK1/2 phosphorylation were used, indicating that the inhibition of the p42/44 MAPK pathway was effective. In contrast, coincubation of IGF-I with the PI-3 kinase inhibitor LY294002 abolished the stimulatory effect of IGF-I on IGFBP-3 and IGFBP-5 mRNA expression (Fig. 5).
FIG. 2. Effects of the specific p42/44 MAPK inhibitors U0126 and PD098059 on the phosphorylation of ERK1/2. Subconfluent chondrocytes in monolayer culture were synchronized in serum-free medium for 12 h. Medium was changed and the cells were incubated with IGF-I (60 ng/ml) with or without the respective inhibitor for 6 h. Unphosphorylated and phosphorylated ERK1/2 were measured by Western immunoblotting of cell extracts, using specific monoclonal antibodies. Similar results were obtained in two different experiments.
FIG. 3. Inhibition of the p42/44 MAPK pathway by U0126 blocks specifically the IGF-I-induced IGFBP-3 mRNA expression but not IGF-I-induced IGFBP-5 mRNA expression. A, Effect of U0126 at the indicated concentrations on IGFBP-3 gene expression in response to IGF-I (60 ng/ml). B, Effect of U0126 at the indicated concentrations on IGFBP-5 mRNA expression in response to IGF-I (60 ng/ml). Subconfluent chondrocytes in monolayer culture were synchronized in serum-free medium for 12 h. Medium was changed and the cells were incubated with IGF-I and the respective inhibitor for 6 h. An RNase protection assay was performed, using 32P-labeled rat IGFBP and 18S rRNA probes as described in Materials and Methods. Two representative autoradiographs of IGFBP-3 and IGFBP-5 are depicted. Autoradiographs from two independent experiments were quantified by computed densitometry. The columns represent the mean.
FIG. 4. Inhibition of the p42/44 MAPK pathway by PD098059 blocks specifically the IGF-I-induced IGFBP-3 mRNA expression but not IGF-I-induced IGFBP-5 mRNA expression. A, Effect of PD098059 at the indicated concentrations on IGFBP-3 mRNA expression in response to IGF-I (60 ng/ml). B, Effect of PD098059 at the indicated concentrations on IGFBP-5 gene expression in response to IGF-I (60 ng/ml). Subconfluent chondrocytes in monolayer culture were synchronized in serum-free medium for 12 h. Medium was changed and the cells were incubated with IGF-I and the respective inhibitor for 6 h. An RNase protection assay was performed, using 32P-labeled rat IGFBP and 18S rRNA probes as described in Materials and Methods. Two representative autoradiographs of IGFBP-3 and IGFBP-5 are depicted. Autoradiographs from two independent experiments were quantified by computed densitometry. The columns represent the mean.
FIG. 5. Inhibition of the PI-3 kinase pathway by LY294002 blocks both the IGF-I-induced IGFBP-3 and IGFBP-5 mRNA expression. A, Effect of LY294002 at the indicated concentrations on IGFBP-3 gene expression in response to IGF-I (60 ng/ml). B, Effect of LY294002 at the indicated concentrations on IGFBP-5 gene expression in response to IGF-I (60 ng/ml). Subconfluent chondrocytes in monolayer culture were synchronized in serum-free medium for 12 h. Medium was changed and the cells were incubated with IGF-I and the respective inhibitor for 6 h. An RNase protection assay was performed, using 32P-labeled rat IGFBP and 18S rRNA probes as described in Materials and Methods. Two representative autoradiographs of IGFBP-3 and IGFBP-5 are depicted. Autoradiographs from three independent experiments were quantified by computed densitometry. The columns represent the mean ± SE. Statistics by ANOVA. *, P < 0.05 vs. control; #, P < 0.05 vs. IGF-I.
To determine, whether the same mechanism for IGF-I-induced IGFBP-5 gene expression is also operative in other cell culture models of the growth plate, similar experiments were performed in the mesenchymal chondrogenic cell line RCJ, which are growth plate chondrocytes derived from fetal rat calvaria (24, 25, 26). An advantage of this cell line is that it does not express IGF-I; therefore, the action of this hormone can be studied without interference from endogenous IGFs (17). RCJ cells also do not express IGFBP-3 (17). In accordance with our observations in growth plate chondrocytes in primary culture, IGF-I induced IGFBP-5 gene expression to a comparable extent (Fig. 6). Coincubation of IGF-I with the PI-3 kinase inhibitor LY294002 abolished the stimulatory effect of IGF-I on IGFBP-5 mRNA expression, whereas IGF-I-induced IGFBP-5 mRNA expression was unaffected by coincubation with the p42/44 MAPK pathway inhibitor U0126 (Fig. 6, A and Fig. B). Comparable results were obtained on the level of protein expression (Fig. 6C).
FIG. 6. IGF-I-induced IGFBP-5 gene expression and protein concentration in conditioned culture medium are mediated through the PI-3 kinase but not through the MAPK/ERK1/2 pathway. Subconfluent RCJ cells in monolayer culture were synchronized in serum-free medium for 12 h and incubated with 100 ng/ml of IGF-I in the presence or absence of specific inhibitors of the different signaling pathways [LY294002 (LY) for the PI-3 kinase pathway and U0126 for the MAPK/ERK1/2 pathway] at the indicated concentrations. Control cells were cultured without IGF-I in the absence and presence of the same concentration of inhibitors. After 12 h, total RNA was extracted and real-time RT-PCR performed, and conditioned medium was collected and Western immunoblot analysis was performed. Membranes were probed with antibodies against IGFBP-5. Representative autoradiographs of a total of three experiments are shown. The columns represent the mean ± SE from three independent experiments. Statistics by ANOVA. *, P < 0.05 vs. control; #, P < 0.05 vs. IGF-I.
IGF-I-induced IGFBP-5 gene expression requires de novo mRNA transcription and de novo protein synthesis
The increased abundance of steady-state IGFBP-5 mRNA in response to IGF-I could be due to increased gene transcription and/or decreased mRNA decay. The first possibility was investigated by use of the transcription blocker actinomycin D. Subconfluent RCJ cells were serum deprived for 12 h. Actinomycin D with and without IGF-I was added to the cell culture medium and after 12 h total mRNA was extracted. Whereas actinomycin D alone did not affect IGFBP-5 mRNA abundance, it completely abolished the 3-fold increase of IGFBP-5 in response to IGF-I, suggesting that IGF-I enhances IGFBP-5 synthesis by stimulation of IGFBP-5 de novo gene transcription (Fig. 7A). Coincubation with cycloheximide, an inhibitor of protein synthesis, also abolished IGF-I-induced IGFBP-5 mRNA abundance (Fig. 7B), suggesting a mechanism also involving de novo protein synthesis. Furthermore, both actinomycin and cycloheximide abrogated the increased IGFBP-5 protein content in conditioned cell culture medium in response to IGF-I (Fig. 7C).
FIG. 7. IGF-I-induced IGFBP-5 mRNA expression requires de novo mRNA transcription and de novo protein synthesis. Subconfluent RCJ cells in monolayer culture were synchronized in serum-free medium for 12 h and stimulated by IGF-I (100 ng/ml) alone or in combination with either actinomycin D (A) or cycloheximide (B). After 12 h, RNA was extracted and IGFBP-5 mRNA expression was assessed by real-time RT-PCR. The columns represent the mean ± SE of three independent experiments. Statistics by ANOVA. *, P < 0.05 vs. control; #, P < 0.05 vs. IGF-I. C, Conditioned culture medium of RCJ cells treated as described above was collected and IGFBP-5 protein concentration was determined by Western immunoblot analysis. The membrane was probed with specific antibodies against IGFBP-5. Representative experiments of a total of two are shown.
Taken together, these data indicate that the IGF-I-stimulated cell proliferation and IGFBP-3 expression is mediated both through the p42/44 MAPK and PI-3 kinase pathway, whereas only the PI-3 kinase pathway is necessary for stimulation of IGFBP-5 mRNA expression by IGF-I.
Discussion
We observed that the effect of IGF-I on cell proliferation in rat growth plate chondrocytes in primary culture is mediated both through the p42/44 MAPK and PI-3 kinase pathway. This is the first report on the intracellular signal transduction pathways, which are used by IGF-I for stimulation of growth plate chondrocyte proliferation. The IGF receptor signaling cascade consists of two main signal transduction pathways, the p42/44 MAPK pathway and the PI-3 kinase pathway (27). Whereas the p42/44 MAPK pathway mainly mediates mitogenic signals, the PI-3 kinase pathway is thought to affect both growth and differentiation (15). The effects of both signaling pathways are cell type specific and general predictions, which signaling pathway is involved in specific responses to IGF-I in a particular cell type cannot be made. For example, whereas in L6A1 myoblasts the mitogenic response to IGF-I is mediated primarily through the p42/44 MAPK pathway (28), an active PI-3 kinase, but not an active p42/44 MAPK pathway, is required to convey the mitogenic message of IGF-I in breast cancer cells (29).
We have shown here that the p42/44 MAPK and PI-3 kinase pathways do not act independently but are in fact interdependent for IGF-I-mediated cell proliferation because the mitogenic effect of IGF-I was abolished completely by inhibition of each individual pathway with the respective pharmacological inhibitor. Consistent with this observation, it was recently reported that IGF-I signals mitogenesis and survival in osteoblastic cells through parallel, partly overlapping intracellular pathways involving PI-3 kinase, MAPK/ERK1/2, and G? subunits (14). These data indicate that IGF-I exerts its mitogenic effect in chondrocytes and osteoblasts by the use of comparable intracellular signaling pathways. In contrast to our finding, Kuemmerle et al. (30) observed in human intestinal smooth muscle cells that both the PI-3 kinase and p42/44 MAPK pathways contribute independently to IGF-I-mediated cell proliferation because inhibition of each individual pathway only partially reduced the response to IGF-I. Hence, the respective contribution of the PI-3 kinase and p42/44 MAPK pathway to IGF-I-mediated cell proliferation appears to be cell type specific.
The biological activity of IGF-I is tightly regulated by six distinct IGFBPs, whose production is cell type and tissue specific and partially under control of IGF-I. We observed by use of RT-PCR that rat growth plate chondrocytes in primary culture synthesize under baseline conditions mRNA species for IGFBP-2 to -6, and we detected by ligand blot analysis of conditioned cell culture medium IGFBP-3, IGFBP-2, glycosylated IGFBP-4, and to a lesser extent IGFBP-5. Exogenous IGF-I enhanced IGFBP-3 and IGFBP-5 on the level of mRNA and protein expression in a dose- and time-dependent manner. There was also a moderate increase of IGFBP-4 and a 35-kDa band, most likely representing IGFBP-2, in response to IGF-I. Regarding IGFBP-5, there was a large discrepancy between the extent of IGFBP-5 mRNA expression in response to IGF-I (up to 13-fold) and the corresponding protein concentration in conditioned cell culture medium (up to 2-fold), suggesting posttranscriptional or posttranslational regulation of IGFBP-5 abundance. A comparable discrepancy between IGF-I-induced IGFBP-5 mRNA abundance and the corresponding protein concentration was also observed in rat articular chondrocytes (31) and rabbit costal chondrocytes (32) and ascribed to increased IGFBP-5 protease activity in the cell culture medium.
Other investigators have previously studied the regulation of IGFBPs by the IGFs in cell culture models of the growth plate. Consistent with our observations, de los Rios and Hill (33) reported that isolated epiphyseal growth plate chondrocytes from the ovine fetus release under basal conditions IGFBP-2, IGFBP-3, IGFBP-4, and IGFBP-5; mRNA species encoding these IGFBPs plus IGFBP-6 were identified by Northern blot analysis. Exposure to IGF-I or IGF-II caused an increase in expression and release of IGFBP-3 and an increase of IGFBP-5 protein, whereas, in contrast to our finding, IGFBP-5 mRNA abundance was not enhanced. The authors speculated that this discrepancy between IGFBP-5 mRNA expression and protein concentration might be due to increased release of IGFBP-5 from the cell membrane in response to the IGFs. Hence, the differences in IGFBP-5 mRNA expression in response to IGF-I in various cell culture models of chondrocytes appear to be dependent on species differences.
We observed that IGF-I enhances both the synthesis of IGFBP-3, which has mainly an inhibitory effect in cell culture models of the growth plate (11, 17), and the synthesis of IGFBP-5, which is a stimulatory IGFBP in this tissue (10). Interestingly, IGF-I uses different signaling pathways to regulate the synthesis of these IGFBPs with contrasting biological functions. The IGF-I-induced IGFBP-3 expression is mediated both through the p42/44 MAPK and PI-3 kinase pathway. Similarly as observed for IGF-I-mediated cell proliferation, the p42/44 MAPK and PI-3 kinase pathways do not act independently for IGF-I-enhanced IGFBP-3 expression but are in fact interdependent because the stimulatory effect of IGF-I on IGFBP-3 synthesis was abolished completely by inhibition of each individual pathway with the respective pharmacological inhibitor. In contrast, the IGF-I-induced IGFBP-5 expression in this cell culture model exclusively depends on the PI-3 kinase pathway. During blockade of the p44/42 MAPK pathway, IGF-I-induced IGFBP-5 synthesis was not affected, but increased IGFBP-5 synthesis in the absence of IGF activity due to a blocked p44/42 MAPK pathway was not capable to functionally counteract the blocked IGF activity on cell proliferation. Further studies are required to link these observations with studies examining the effects of altered IGFBP expression on IGF activity in growth plate chondrocytes. We hypothesize that IGF-I modulates its action by inducing both inhibitory and stimulatory IGFBPs to establish a functional equilibrium.
Consistent with our observations in growth plate chondrocytes in primary culture, exogenous IGF-I also stimulated IGFBP-5 gene expression in RCJ cells. We report here for the first time that the IGF-I-induced IGFBP-5 gene expression in a cell culture model of growth plate chondrocytes requires de novo mRNA transcription and de novo protein synthesis. Also in RCJ cells, IGF-I-induced IGFBP-5 gene expression was mediated through the PI-3 kinase but not through the p42/44 MAPK pathway. These congruent data indicate that our observations regarding the mechanism of IGF-I-induced IGFBP-5 synthesis apply to growth plate chondrocytes in general.
Other investigators have studied the intracellular signaling pathways that are involved in the IGF-I-mediated expression of IGFBP-3 and IGFBP-5. Consistent with our observation, Koedam et al. (32) reported that the IGF-induced IGFBP-5 expression in rabbit costal chondrocytes requires the activation of PI-3 kinase; the p42/44 MAPK pathway and the pathways involved in the IGF-I-mediated IGFBP-3 expression were not investigated in this study. In agreement with our observation, the stimulation of IGFBP-5 mRNA expression by IGF-I is mediated exclusively via the PI-3 kinase pathway in vascular smooth muscle cells (34) and primary Schwann cells (35). Contrasting observations were made in rat intestinal smooth muscle cells, in which the IGF-I-mediated IGFBP-5 induction was mediated exclusively via the p42/44 MAPK pathway (36). Regarding the regulation of IGFBP-3 expression by IGF-I, only two previous studies with different results have been published. In mammary epithelial cells, the IGF-induced IGFBP-3 expression is mainly regulated via the PI-3 kinase pathway and only partly via the p42/44 MAPK pathway (37). In human intestinal smooth muscle cells, the IGF-I-enhanced IGFBP-3, -4, and -5 mRNA expression is mediated both via the p42/44 MAPK and PI-3 kinase pathway (38). Hence, the intracellular signaling pathways involved in the IGF-I-mediated gene expression of IGFBP-3 and IGFBP-5 are cell type specific.
In conclusion, our data demonstrate that IGF-I modulates its activity in cultured rat growth plate chondrocytes by the synthesis of both inhibitory (IGFBP-3) and stimulatory (IGFBP-5) binding proteins. IGF-I stimulates cell proliferation and IGFBP-3 mRNA expression through both the p42/44 MAPK and PI-3 kinase pathways, whereas IGF-I-induced IGFBP-5 expression depends only on the PI-3 kinase pathway. The finding that IGF-I uses different and only partially overlapping intracellular signaling pathways for the regulation of two IGFBPs with opposing biological functions might be important for the regulation of IGF bioactivity in the cellular microenvironment.
Acknowledgments
We thank Dr. Anna Spagnoli (Vanderbilt University Medical Center, Nashville, TN) for generously providing the RCJ3.1C5.18 cell line. Furthermore, we thank Ralph Witzgall for technical support with the RNase protection assay and Ludger St?ndker for providing radiolabeled IGF-II.
References
Ohlsson C, Nilsson A, Isaksson OG, Lindahl A 1992 Effect of growth hormone and insulin-like growth factor-I on DNA synthesis and matrix production in rat epiphyseal chondrocytes in monolayer culture. J Endocrinol 133:291–300
Olney RC, Wang J, Sylvester JE, and Mougey EB 2004 Growth factor regulation of human growth plate chondrocyte proliferation in vitro. Biochem Biophys Res Commun 317:1171–1182
Reinecke M, Schmid AC, Heyberger-Meyer B, Hunziker, EB, Zapf J 2000 Effect of growth hormone and insulin-like growth factor (IGF-I) on the expression of IGF-I messenger ribonucleic acid and peptide in rat tibial growth plate and articular chondrocytes in vivo. Endocrinology 141:2847–2853
Yakar S, Rosen CJ, Beamer WG, Ackert-Bicknell CL, Wu Y, Liu JL, Ooi GT, Setser J, Frystyk J, Boisclair YR, LeRoith D 2002 Circulating levels of IGF-1 directly regulate bone growth and density. J Clin Invest 110:771–781
Wang J, Zhou J, Bondy CA 1999 IGF1 promotes longitudinal bone growth by insulin-like actions augmenting chondrocyte hypertrophy. FASEB J 13:1985–1990
Wang J, Zhou J, Cheng CM, Kopchick JJ, Bondy CA 2004 Evidence supporting dual, IGF-I-independent and IGF-I-dependent, roles for GH in promoting longitudinal bone growth. J Endocrinol 180:247–255
Baxter RC 2000 Insulin-like growth factor (IGF)-binding proteins: interactions with IGFs and intrinsic bioactivities. Am J Physiol Endocrinol Metab 278:E967–E976
Jones JI, Clemmons DR 1995 Insulin-like growth factors and their binding proteins: biological actions. Endocr Rev 16:3–34
Schneider MR, Wolf E, Hoeflich A, Lahm H 2003 IGF-binding protein-5: flexible player in the IGF system and effector on its own. J Endocrinol 172:423–440
Kiepe D, Andress DL, Mohan S, Standker L, Ulinski T, Himmele R, Mehls O, T?nshoff B 2001 Intact IGF-binding protein-4 and -5 and their respective fragments isolated from chronic renal failure serum differentially modulate IGF-I actions in cultured growth plate chondrocytes. J Am Soc Nephrol 12:2400–2410
Kiepe D, Ulinski T, Powell DR, Durham SK, Mehls O, T?nshoff B 2002 Differential effects of insulin-like growth factor binding proteins-1, -2, -3, and -6 on cultured growth plate chondrocytes. Kidney Int 62:1591–1600
Parrizas M, Saltiel AR, LeRoith D 1997 Insulin-like growth factor 1 inhibits apoptosis using the phosphatidylinositol 3'-kinase and mitogen-activated protein kinase pathways. J Biol Chem 272:154–161
Tsakiridis T, Tsiani E, Lekas P, Bergman A, Cherepanov V, Whiteside C, Downey GP 2001 Insulin, insulin-like growth factor-I and platelet-derived growth factor activate extracellular signal-regulated kinase by distinct pathways in muscle cells. Biochem Biophys Res Commun 288:205–211
Grey A, Chen Q, Xu X, Callon K, Cornish J 2003 Parallel phosphatidylinositol-3 kinase and p42/44 mitogen-activated protein kinase signaling pathways subserve the mitogenic and antiapoptotic actions of insulin-like growth factor I in osteoblastic cells. Endocrinology 144:4886–4893
Petley T, Graff K, Jiang W, Yang H, Florini J 1998 Variation among cell types in the signaling pathways by which IGF-I stimulates specific cellular responses. Horm Metab Res 31:70–76
Klaus G, Merke J, Eing H, Hugel U, Milde P, Reichel H, Ritz E, Mehls O 1991 1,25(OH)2D3 receptor regulation and 1,25(OH)2D3 effects in primary cultures of growth cartilage cells of the rat. Calcif Tissue Int 49:340–348
Spagnoli A, Hwa V, Horton WA, Lunstrum GP, Roberts Jr CT, Chiarelli F, Torello M, Rosenfeld RG 2001 Antiproliferative effects of insulin-like growth factor-binding protein-3 in mesenchymal chondrogenic cell line RCJ3.1C5.18. relationship to differentiation stage. J Biol Chem 276:5533–5540
Favata MF, Horiuchi KY, Manos EJ, Daulerio AJ, Stradley DA, Feeser WS, Van Dyk DE, Pitts WJ, Earl RA, Hobbs F, Copeland RA, Magolda RL, Scherle PA, Trzaskos JM 1998 Identification of a novel inhibitor of mitogen-activated protein kinase kinase. J Biol Chem 273:18623–18632
Homme M, Schmitt CP, Himmele R, Hoffmann GF, Mehls O, Schaefer F 2003 Vitamin D and dexamethasone inversely regulate parathyroid hormone-induced regulator of G protein signaling-2 expression in osteoblast-like cells. Endocrinology 144:2496–2504
H?flich A, Lahm H, Blum W, Kolb H, Wolf E 1998 Insulin-like growth factor-binding protein-2 inhibits proliferation of human embryonic kidney fibroblasts and of IGF-responsive colon carcinoma cell lines. FEBS Lett 434:329–334
Matsumoto T, Tsurumoto T, Goldring MB, Shindo H 2000 Differential effects of IGF-binding proteins, IGFBP-3 and IGFBP-5, on IGF-I action and binding to cell membranes of immortalized human chondrocytes. J Endocrinol 166:29–37
Olney RC, Smith RL, Kee Y, Wilson DM 1993 Production and hormonal regulation of insulin-like growth factor binding proteins in bovine chondrocytes. Endocrinology 133:563–570
Borromeo V, Bramani S, Holder AT, Carter C, Secchi C, Beattie J 1996 Growth hormone stimulates the secretion of insulin-like growth factor binding protein-2 (IGFBP-2) by monolayer cultures of sheep costal growth plate chondrocytes. Mol Cell Biochem 162:145–151
Grigoriadis AE, Heersche JN, Aubin JE 1996 Analysis of chondroprogenitor frequency and cartilage differentiation in novel family of clonal chondrogenic rat cell lines. Differentiation 60:299–307
Lunstrum GP, Keene DR, Weksler NB, Cho Y-J, Cornwall M, Horton WA 1999 Chondrocyte differentiation in a rat mesenchymal cell line. J Histochem Cytochem 47:1–6
McDougall S, Fu YH, Lowe GN, Williams A, Polendo R, Benya PD, Iida-Klein A, Fang MA, Hahn TJ 1996 Surface adhesion-mediated regulation of chondrocyte-specific gene expression in the nontransformed RCJ3.1C5.18 rat chondrocyte cell line. J Bone Miner Res 11:1130–1138
LeRoith D 2000 Insulin-like growth factor I receptor signaling—overlapping or redundant pathways? Endocrinology 141:1287–1288[Free Full Text]
Coolican SA, Samuel DS, Ewton DZ, McWade FJ, Florini JR 1997 The mitogenic and myogenic actions of insulin-like growth factors utilize distinct signaling pathways. J Biol Chem 272:6653–6662
Dufourny B, Alblas J, van Teeffelen HA, van Schaik FM, van der Burg B, Steenbergh PH, Sussenbach JS 1997 Mitogenic signaling of insulin-like growth factor I in MCF-7 human breast cancer cells requires phosphatidylinositol 3-kinase and is independent of mitogen-activated protein kinase. J Biol Chem 272:31163–31171
Kuemmerle JF, Bushman TL 1998 IGF-I stimulates intestinal muscle cell growth by activating distinct PI 3-kinase and MAP kinase pathways. Am J Physiol 275:G151–G158
Matsumoto T, Gargosky SE, Oh Y, Rosenfeld RG 1996 Transcriptional and post-translational regulation of insulin-like growth factor-binding protein-5 in rat articular chondrocytes. J Endocrinol 148:355–369
Koedam JA, Hoogerbrugge CM, Van Buul-Offers SC 2000 Differential regulation of IGF-binding proteins in rabbit costal chondrocytes by IGF-I and dexamethasone. J Endocrinol 165:557–567
De Los Rios P, Hill DJ 2000 Expression and release of insulin-like growth factor binding proteins in isolated epiphyseal growth plate chondrocytes from the ovine fetus. J Cell Physiol 183:172–181
Duan C, Liimatta MB, Bottum OL 1999 Insulin-like growth factor (IGF)-I regulates IGF-binding protein-5 gene expression through the phosphatidylinositol 3-kinase, protein kinase B/Akt, and p70 S6 kinase signaling pathway. J Biol Chem 274:37147–37153
Cheng HL, Shy M, Feldman EL 1999 Regulation of insulin-like growth factor-binding protein-5 expression during Schwann cell differentiation. Endocrinology 140:4478–4485
Xin X, Hou YT, Li L, Schmiedlin-Ren P, Christman GM, Cheng HL, Bitar KN, Zimmermann EM 2004 IGF-I increases IGFBP-5 and collagen 1(I) mRNAs by the MAPK pathway in rat intestinal smooth muscle cells. Am J Physiol Gastrointest Liver Physiol 286:G777–G7783
Sivaprasad U, Fleming J, Verma PS, Hogan KA, Desury G, Cohick WS 2004 Stimulation of insulin-like growth factor (IGF) binding protein-3 synthesis by IGF-I and transforming growth factor- is mediated by both phosphatidylinositol-3 kinase and mitogen-activated protein kinase pathways in mammary epithelial cells. Endocrinology 145:4213–4221
Kuemmerle JF 2000 Endogenous IGF-I regulates IGF binding protein production in human intestinal smooth muscle cells. Am J Physiol Gastrointest Liver Physiol 278:G710–G717(Daniela Kiepe1, Sonia Cia)
Address all correspondence and requests for reprints to: Burkhard T?nshoff, M.D., Ph.D., University Children’s Hospital, Im Neuenheimer Feld 153, 69120 Heidelberg, Germany. E-mail: burkhard_toenshoff@med.uni-heidelberg.de.
Abstract
The bioactivity of IGF-I in the cellular microenvironment is modulated by both inhibitory and stimulatory IGF binding proteins (IGFBPs), whose production is partially under control of IGF-I. However, little is known on the IGF-mediated regulation of these IGFBPs in the growth plate. We therefore studied the effect of IGF-I on IGFBP synthesis and the involved intracellular signaling pathways in two cell culture models of rat growth plate chondrocytes. In growth plate chondrocytes in primary culture, incubation with IGF-I increased the concentrations of IGFBP-3 and IGFBP-5 in conditioned cell culture medium in a dose- and time-dependent manner. Coincubation of IGF-I with specific inhibitors of the p42/44 MAPK pathway (PD098059 or U0126) completely abolished the stimulatory effect of IGF-I on IGFBP-3 mRNA expression but did not affect increased IGFBP-5 mRNA levels. In contrast, inhibition of the phosphatidylinositol-3 kinase signaling pathway by LY294002 abrogated both IGF-I-stimulated IGFBP-3 and -5 mRNA expression. Comparable results regarding IGFBP-5 were obtained in the mesenchymal chondrogenic cell line RCJ3.1C5.18, which does not express IGFBP-3. The IGF-I-induced IGFBP-5 gene expression required de novo mRNA transcription and de novo protein synthesis. These data suggest that IGF-I modulates its activity in cultured rat growth plate chondrocytes by the synthesis of both inhibitory (IGFBP-3) and stimulatory (IGFBP-5) binding proteins. The finding that IGF-I uses different and only partially overlapping intracellular signaling pathways for the regulation of two IGFBPs with opposing biological functions might be important for the modulation of IGF bioactivity in the cellular microenvironment.
Introduction
IGF-I IS A POTENT growth factor in the growth plate, exerting its actions by both endocrine and paracrine/autocrine mechanisms. It is known from cell culture (1, 2, 3) and gene knockout experiments (4, 5, 6) that IGF-I stimulates both proliferation and differentiation of growth plate chondrocytes in vitro and in vivo. The action of the IGFs in both the circulation and tissues is tightly regulated by a family of high-affinity IGF binding proteins (IGFBPs). Up to now, six distinct IGFBPs have been identified. Although structurally related, they have individual expression patterns and exert different functions including stimulation or inhibition of IGF bioactivity as well as IGF-independent actions (7, 8, 9).
We previously studied the biological activities of various intact IGFBPs in a growth cartilage model of proliferating growth plate chondrocytes in primary cultures (10, 11). IGFBP-1, -2, -4, and -6 act exclusively as growth inhibitors on IGF-dependent cell proliferation, whereas the biological activity of IGFBP-3 is complex. It has an IGF-independent antiproliferative effect and also inhibits IGF-dependent chondrocyte proliferation under coincubation conditions. Under preincubation conditions, however, IGFBP-3 enhances the IGF-I responsiveness of growth plate chondrocytes by its ability to associate with the cell membrane, in which it presumably facilitates IGF-I receptor binding (11). Intact IGFBP-5, on the other hand, enhances IGF-I-induced chondrocyte proliferation, apparently by its association with the cell membrane in the C-terminal domain, thereby better presenting IGF-I to its receptor (10).
The synthesis of the IGFBPs in various tissues is under partial control of IGF-I. IGF-I exerts its biological effect by binding to the transmembrane type 1 IGF receptor, whose activation leads to the extensive tyrosyl-phosphorylation of insulin receptor substrate-1, which acts as a docking protein for the downstream signal transduction pathways (12, 13). Two canonical pathways, the phosphatidylinositol-3 kinase (PI-3 kinase) and the p42/44 MAPK pathway, have been reported previously to mediate the mitogenic, differentiating and antiapoptotic response to IGF-I (14), but the relative contributions to the diverse cellular actions of IGF-I vary according to the cell type (15). It is therefore necessary to study IGF-I signaling in individual tissues to determine the role of each pathway in that specific cell type.
Little is known on the regulation of IGFBPs by the IGFs in the growth cartilage and the underlying intracellular signaling mechanisms. We therefore examined in the present study the effect of IGF-I on IGFBP synthesis in two different cell culture models of growth plate chondrocytes. Furthermore, we investigated the involved IGF-I-activated signaling pathways by use of specific pharmacological inhibitors. We report here that IGF-I modulates its activity in rat growth plate chondrocytes by stimulating the synthesis of both inhibitory (IGFBP-3) and stimulatory (IGFBP-5) binding proteins. The IGF-I-induced cell proliferation and IGFBP-3 mRNA expression are mediated through both the p42/44 MAPK and PI-3 kinase pathway, whereas the IGF-I-induced IGFBP-5 mRNA expression depends only on a functional PI-3 kinase pathway.
Materials and Methods
Materials
Recombinant human IGF-I was purchased from Bachem (Heidelberg, Germany); [3H]thymidine (25 Ci/mmol) and 32P-labeled uridine 5-triphosphate were obtained from Amersham Pharmacia Biotech (Freiburg, Germany). Radioiodination of IGF-II with 125I (>2000 Ci/mmol of protein) was performed using a standard chloramine T method (10). Briefly, the peptide was dissolved in 0.25 M sodium phosphate, and Na125I (0.2 mCi; Amersham, Braunschweig, Germany) and 10 μl chloramine-T (2.5 mg/ml in phosphate buffer) were added. After incubation, the reaction was stopped by the addition of Na2S2O5. The radiolabeled peptide was purified on Sep-Pak C18 cartridges (Waters-Millipore, Eschborn, Germany). The homogeneity and the specific activity of the radiolabeled peptide were controlled on the HPLC system 322 (Kontron, Neufahrn, Germany).
PBS, HEPES, penicillin-streptomycin, Ham’s F-12, and DMEM were obtained from Seromed Biochrom KG (Berlin, Germany). BSA was purchased from Sigma-Aldrich Chemicals (Deisenhofen, Germany). Clostridium collagenase (EC 3.4.24.3), deoxyribonuclease [DNase I (EC 3.1.21.1)], and trypan blue were from Roche Diagnostics GmbH (Mannheim, Germany). MEM was purchased from cc Pro (Neustadt, Germany); fetal calf serum from Paa Laboratories (Pasching, Austria); and ascorbic acid, ?-glycerophosphate, and dexamethasone from Sigma (Taufkirchen, Germany). Monoclonal antibodies against p44/42 MAPK and phospho-p44/42 MAPK were obtained from Cell Signaling (Frankfurt a. M., Germany). The antibody against hIGFBP-5 was purchased from Santa Cruz Biotechnologies (Heidelberg, Germany); antirat IGFBP-4 and antirat IGFBP-5 antibodies were from GroPep (Adelaide, Australia).
Cell cultures
Epiphyseal chondrocytes from 60- to 80-g Sprague Dawley rats (Charles River, Kieslegg, Germany) were isolated and cultured, as described previously (10, 16). This study was approved by the Institutional Animal Care and Use Committee (35–9185.81/102/98). Pooled growth plates from four to eight animals were digested with clostridial collagenase (0.12% wt/vol) and bacterial DNase (0.02% wt/vol) in F-12 medium. Viability, determined after isolation and at the end of each experiment by the trypan blue exclusion technique, always exceeded 90%. Dissociated cells were counted using a Neubauer chamber (Scheik, Hofheim, Germany).
Cells were cultured in monolayers and in 96-well plates for proliferation assays (Nunc, Wiesbaden, Germany), as described previously (10, 16). The F-12/DMEM (1:1) medium contained a nominal calcium concentration of 1.2 mM, and the medium was supplemented with 10 mM HEPES, 100 μg/ml streptomycin, and 10% fetal calf serum. In previous studies using the same culture system, we demonstrated that the majority of cells after the first passage expressed typical markers for proliferative growth plate chondrocytes (16).
RCJ3.1C5.18 (RCJ) cells (kindly provided by Dr. Anna Spagnoli, Department of Pediatrics, Vanderbilt University Medical Center, Nashville, TN) were grown at 37 C in humidified 5% CO2 atmosphere in MEM (with Earle’s salts) supplemented with 1 mM N-acetyl-L-glutamine, 10 mM HEPES, 100 U/ml penicillin-streptomycin, 2 mM sodium pyruvate, 15% heat-inactivated fetal bovine serum, and 10–7 M dexamethasone and studied within 25 passages (17). Cells were plated at a density of 6 x 104 cells/well in 6-well dishes for RNA extraction and protein isolation. Peptide hormones were dissolved in PBS and added every day unless indicated otherwise. The inhibitor LY294002 (Calbiochem, San Diego, CA) was dissolved in 20% dimethylsulfoxide in ethanol following the instructions of the manufacturer. PD098059 (Calbiochem) and U0126 (Promega, Mannheim, Germany) were dissolved in dimethylsulfoxide following the instructions of the manufacturer. The inhibition of MAPK kinase by PD098059 and U0126 is very selective, but U0126 has a 100-fold higher affinity for MEK than PD098059 (18).
[3H]thymidine incorporation assay
The extent of thymidine incorporation into DNA was determined as uptake of radioactivity in precipitated material, as described previously (10, 16). Briefly, chondrocytes were grown in 96-well plates to subconfluence for 3 d; thereafter, the cultures were changed to serum-free medium. After 12 h, various concentrations of inhibitors with or without IGF-I were added to the medium and the incubation was continued for further 48 h. The rate of chondrocyte proliferation was assessed by incubating the cells with 3 μCi/ml [3H]thymidine for the final 4 h. Subsequently, cells were rinsed twice with PBS. The radiolabeled DNA was precipitated by acetic acid and dissolved in 1 M NaOH. [3H]thymidine incorporation into the acid-extractable pool as a measure of DNA synthesis was determined by scintillation counting.
Real-time RT-PCR.
For analysis of IGFBP-5 mRNA abundance, cells were cultured in differentiating medium until d 7, serum-starved for 12 h and stimulated for an additional 12 h with IGF-I in the presence or absence of inhibitors. RNA was isolated by using the RNeasy minicolumns (QIAGEN, Hilden, Germany), and 1 μg was reverse transcribed using Moloney leukemia virus reverse transcriptase and oligo(dt)/random hexamer primers (10:1) from Applied Biosystems (Darmstadt, Germany). For quantitative analysis, real-time RT-PCR was performed using the Abi 7000 (Applied Biosystems) according to the manufacturer’s protocol. The following set of primers was chosen: hIGFBP-5, forward, 5'-TCAAGATCGAGAGAGACTCCCG-3', reverse, 5'-TTCACTGCTTCAGCCTTCAGC; rIGFBP-5, forward, 5'-AGTCGTGTGGCGTCTACACTGA-3', reverse, 5'-TTTGCTCGCCGTAGCTCTTTT-3'; and rat 18S by the Primer Express program, forward, 5'-AGTTGGTGGAGCGATTTGTC-3', reverse, 5'-GCTGAGCCAGTTCAGTGTAGC-3' (Applied Biosystems). The software provided from the company allowed the quantitative detection of fluorescence by the incorporation of the substance SYBR green into the amplification products. Amplification was performed in the presence of Universal Mastermix (PE Applied Biosystems) with SYBR green to detect PCR products at the end of each amplification step, and results were analyzed as already reported (19).
Ribonuclease protection assay
The amplification products of IGFBP-3 (292 nt) and IGFBP-5 (296 nt) obtained by conventional PCR were cloned into pBlueScript II KS (Stratagene, Techno-Path, Limerick, Ireland), sequenced (MWG Biotech AG, Ebersberg, Germany), and in vitro transcribed using 32P-labeled uridine 5-triphosphate (2.5 μM) and unlabeled nucleotides (Amersham Pharmacia Biotech). Lyophilized samples of total RNA (10 μg for IGFBPs and 50 ng for 18S rRNA) were hybridized with the radiolabeled antisense riboprobes at 45 C overnight. The free probe RNA was digested followed by proteinase K treatment. After phenolization and ethanol precipitation, hybrids were eluted in RNA loading buffer and separated on a sequencing gel. The ODs of the autoradiographs were analyzed by a commercially available computer program (Bio-Rad Laboratories, Inc., Vilber Lourmat, France). Protected bands were quantified densitometrically and normalized against 18S rRNA (19).
Western ligand blotting and immunoblotting
Cells were maintained in serum-free medium for 12 h and incubated with vehicles, peptides, or inhibitors as indicated. The collected conditioned medium was concentrated by trichloracetic acid precipitation, as described previously (20). Equal aliquots of concentrated conditioned medium were loaded on a 12% gel for nonreducing SDS-PAGE (10% acrylamide bis) and transferred to nitrocellulose membranes. Distinct membranes were blocked with 1% fish gelatin and incubated with [125I]-IGF-II (106 cpm/blot) for ligand blotting. All of the incubation and washing steps were performed at 4 C, as previously described (20). The same membranes were incubated overnight at 4 C with a polyclonal rabbit antibody against IGFBP-4 (GroPep, Adelaide, Australia; dilution 1:1000) for Western blotting, incubated for 1 h with the secondary antibody, and washed extensively over a period of 30 min with Tris-buffered saline and Tween 20 0.05% (TBS-T). For signal detection, autoradiography (X-OMAT AR film, Amersham Pharmacia Biotech) or a chemiluminescent detection system (ECL Western blotting detection reagents, Amersham Pharmacia Biotech) and Hyperfilm ECL film (Kodak, Stuttgart, Germany) were used.
Immunoprecipitation
One hundred microliters of packed A/G and agarose beads (Santa Cruz) were washed three times in PBS and then incubated with 15 μl IGFBP-5 antibody (GroPep; dilution 1:1000) overnight at 4 C. Two milliliters conditioned medium mixed with benzamidin (3 mM), aprotinin (10 μg/ml), leupeptin (10 μg/ml), phenylmethylsulfonyl fluoride (1 mM), and NaV03 (1 mM) were 10 times concentrated by centrifugation in a centricon-10 tube (Millipore/Amicon GmbH, Schwalbach, Germany). Two hundred microliters of concentrated conditioned medium were incubated with 15 μl of packed beads for 6 h at 4 C. The pellet was washed three times with 10 mM Tris-HCl buffer (pH 7.5) to remove IGFBPs nonspecifically precipitated with the antibody. The immunoprecipitated pellet was then resuspended in 20 μl sample sodium dodecyl sulfate (SDS)-loading buffer plus ?-mercaptoethanol [63 mmol Tris-HCl (pH 6.8); 10% glycerol; 2.1% SDS; 0.005% bromphenol blue] and boiled. Samples were subjected to Western immunoblotting by using the antibody against IGFBP-5 (GroPep; dilution 1:1000).
Statistical analysis
Data are given as mean ± SE. All data were examined for normal and non-Gaussian distribution by the Kolmogorov-Smirnov test. For comparison among normally distributed groups, one-way ANOVA followed by pairwise multiple comparison (Student-Newman-Keuls method) was used. For nonnormally distributed data, the nonparametric Kruskal-Wallis test followed by an all pairwise multiple comparison (Dunnett’s method) was used. P < 0.05 was considered statistically significant.
Results
IGF-I-induced chondrocyte proliferation is mediated through both the p42/44 MAPK and PI-3 kinase pathways
Under baseline conditions, inhibition of the p42/44 MAPK pathway by U0126 or PD098059 and inhibition of the PI-3 kinase pathway by LY294002 did not affect chondrocyte proliferation (Table 1). Incubation of chondrocytes with IGF-I (60 ng/ml) for 48 h significantly enhanced cell proliferation. We observed that the IGF-I-stimulated cell proliferation was completely suppressed by coincubation with the PI-3 kinase pathway inhibitor LY294002 and the two p42/44 MAPK pathway inhibitors U0126 or PD098059. Taken together, these data indicate that the IGF-I-stimulated cell proliferation depends on the p42/44 MAPK and PI-3 kinase pathways.
TABLE 1. Effect of the p42/44 MAPK inhibitors PD098059 and U0126 and effect of LY294002, a PI-3 kinase pathway inhibitor, on basal and IGF-I-stimulated DNA synthesis, as assessed by [3H]thymidine incorporation
Characterization of IGFBPs in conditioned cell culture medium
First, we sought to identify which IGFBPs are synthesized under baseline conditions by growth plate chondrocytes in primary culture. Conditioned cell culture medium was collected and subjected to Western ligand blotting, immunoblotting, and immunoprecipitation. By Western ligand blotting using [125I]-IGF-II as a ligand, IGFBPs with a molecular mass of 40–45, 35, 32, and 28 kDa were detected (Fig. 1A). After exposure to exogenous IGF-I (60 ng/ml) for 6–24 h, the intensity of the 40- to 45-kDa band (IGFBP-3) increased maximally approximately 2-fold, the intensity of the 32-kDa band (IGFBP-5) approximately 5-fold, and the intensity of the 28-kDa band (IGFBP-4) approximately 2-fold in a time-dependent manner. By immunoblotting with an antibody directed against human and rat IGFBP-4, the 28-kDa band already detected by [125I]-IGF-II ligand blotting and a 24-kDa band were identified as IGFBP-4 (Fig. 1B). It is known from previous investigations that the 24-kDa band corresponds to the nonglycosylated form of IGFBP-4 and the 28-kDa band to the glycosylated form (21). The 32-kDa band was identified as IGFBP-5 by immunoprecipitation with an antibody directed against human and rat IGFBP-5 (Fig. 1C). The identity of the 40- to 45-kDa band was not further characterized by immunoblotting because it is known that this band identified by IGF ligand blotting in conditioned medium of growth plate chondrocytes represents IGFBP-3 (22). The 35-kDa band, most likely representing IGFBP-2 (23), increased only slightly in response to IGF-I (Fig. 1A).
FIG. 1. Characterization of IGFBP proteins in conditioned cell culture medium of growth plate chondrocytes in primary culture. A, Western ligand blot of IGFBPs in conditioned cell culture medium under baseline conditions and after exposure to IGF-I (60 ng/ml) for the indicated time period, after 12 h of serum starvation. The concentrated samples were subjected to SDS-PAGE on a 12.5% acrylamide gel, and a Western ligand blot was performed by use of [125I]-IGF-II as the radioligand, as described in Materials and Methods. The molecular mass markers are indicated on the ordinate. B, Western immunoblot of IGFBP-4 in conditioned medium of growth plate chondrocytes treated with IGF-I (60 ng/ml) for 6 h after 12 h of serum starvation. The concentrated samples were subjected to SDS-PAGE on a 12.5% acrylamide gel, transferred to filters, and then immunodetected with IGFBP-4 antiserum, as described in Materials and Methods. C, Immunoprecipitation of IGFBP-5. Two hundred microliters of concentrated conditioned cell culture medium from growth plate chondrocytes treated with IGF-I (60 ng/ml) for 6 h after 12 h of serum starvation were incubated with 15 μl packed beads for 6 h at 4 C. The immunoprecipitated pellet was resuspended in 20 μl SDS loading buffer. The supernatants were subjected to Western immunoblotting using a polyclonal IGFBP-5 antibody.
IGF-I enhances IGFBP-3 and -5 mRNA expression and protein concentration in a dose- and time-dependent manner
Under baseline conditions, rat growth plate chondrocytes expressed mRNA species for IGFBP-2 to -6, whereas IGFBP-1 mRNA could not be detected by multiplex RT-PCR (data not shown). In the following series of experiments, we focused on IGFBP-3 and IGFBP-5 because these two IGFBPs have opposing biological functions in growth plate chondrocytes in primary culture and their synthesis increased the most in response to exogenous IGF-I. Incubation of cells with IGF-I led to a dose- and time-dependent increase of IGFBP-3 mRNA expression and protein content in conditioned medium; the extent of stimulation (2- to 4-fold) was comparable on the mRNA and protein level (Tables 2 and 3), suggesting that there was no major posttranscriptional or posttranslational modification of this IGFBP, e.g. by proteolysis. In contrast, we observed a sharp (up to 14-fold) increase of IGFBP-5 mRNA expression in an IGF dose- and time-dependent manner, whereas the increase of IGFBP-5 protein content (2-fold) was only moderate (Tables 2 and 3).
TABLE 2. IGF-I induces the respective mRNA abundance and protein concentration of IGFBP-3 and IGFBP-5 in conditioned cell culture medium in a dose-dependent manner
TABLE 3. IGF-I induces the respective mRNA abundance and protein concentration of IGFBP-3 and IGFBP-5 in conditioned cell culture medium in a time-dependent manner
IGF-I-induced IGFBP-3 mRNA expression is mediated through both the p42/44 MAPK and PI-3 kinase pathways, whereas the IGF-I-induced IGFBP-5 mRNA expression is mediated only through the PI-3 kinase pathway
Next, we sought to determine the intracellular signaling pathways, by which IGF-I differentially regulates IGFBP-3 and IGFBP-5 expression. We used two pharmacological inhibitors of the p42/44 MAPK pathway, U0126 and PD098059. Both inhibitors at the indicated concentrations were capable of suppressing phosphorylation of ERK1/2 but did not alter total ERK1/2 concentration in chondrocyte cell lysates (Fig. 2). Under baseline conditions, inhibition of the p42/44 MAPK pathway by U0126 did not alter IGFBP-3 or IGFBP-5 mRNA expression (Fig. 3). However, coincubation of IGF-I with U0126 completely suppressed IGF-I-induced IGFBP-3 mRNA expression, whereas IGF-I-induced IGFBP-5 mRNA expression was unaffected (Fig. 3). Comparable results were obtained by use of the other p42/44 MAPK pathway inhibitor, PD098059 (Fig. 4). For these experiments, the same concentrations of U0126 and PD098059 that were able to suppress ERK1/2 phosphorylation were used, indicating that the inhibition of the p42/44 MAPK pathway was effective. In contrast, coincubation of IGF-I with the PI-3 kinase inhibitor LY294002 abolished the stimulatory effect of IGF-I on IGFBP-3 and IGFBP-5 mRNA expression (Fig. 5).
FIG. 2. Effects of the specific p42/44 MAPK inhibitors U0126 and PD098059 on the phosphorylation of ERK1/2. Subconfluent chondrocytes in monolayer culture were synchronized in serum-free medium for 12 h. Medium was changed and the cells were incubated with IGF-I (60 ng/ml) with or without the respective inhibitor for 6 h. Unphosphorylated and phosphorylated ERK1/2 were measured by Western immunoblotting of cell extracts, using specific monoclonal antibodies. Similar results were obtained in two different experiments.
FIG. 3. Inhibition of the p42/44 MAPK pathway by U0126 blocks specifically the IGF-I-induced IGFBP-3 mRNA expression but not IGF-I-induced IGFBP-5 mRNA expression. A, Effect of U0126 at the indicated concentrations on IGFBP-3 gene expression in response to IGF-I (60 ng/ml). B, Effect of U0126 at the indicated concentrations on IGFBP-5 mRNA expression in response to IGF-I (60 ng/ml). Subconfluent chondrocytes in monolayer culture were synchronized in serum-free medium for 12 h. Medium was changed and the cells were incubated with IGF-I and the respective inhibitor for 6 h. An RNase protection assay was performed, using 32P-labeled rat IGFBP and 18S rRNA probes as described in Materials and Methods. Two representative autoradiographs of IGFBP-3 and IGFBP-5 are depicted. Autoradiographs from two independent experiments were quantified by computed densitometry. The columns represent the mean.
FIG. 4. Inhibition of the p42/44 MAPK pathway by PD098059 blocks specifically the IGF-I-induced IGFBP-3 mRNA expression but not IGF-I-induced IGFBP-5 mRNA expression. A, Effect of PD098059 at the indicated concentrations on IGFBP-3 mRNA expression in response to IGF-I (60 ng/ml). B, Effect of PD098059 at the indicated concentrations on IGFBP-5 gene expression in response to IGF-I (60 ng/ml). Subconfluent chondrocytes in monolayer culture were synchronized in serum-free medium for 12 h. Medium was changed and the cells were incubated with IGF-I and the respective inhibitor for 6 h. An RNase protection assay was performed, using 32P-labeled rat IGFBP and 18S rRNA probes as described in Materials and Methods. Two representative autoradiographs of IGFBP-3 and IGFBP-5 are depicted. Autoradiographs from two independent experiments were quantified by computed densitometry. The columns represent the mean.
FIG. 5. Inhibition of the PI-3 kinase pathway by LY294002 blocks both the IGF-I-induced IGFBP-3 and IGFBP-5 mRNA expression. A, Effect of LY294002 at the indicated concentrations on IGFBP-3 gene expression in response to IGF-I (60 ng/ml). B, Effect of LY294002 at the indicated concentrations on IGFBP-5 gene expression in response to IGF-I (60 ng/ml). Subconfluent chondrocytes in monolayer culture were synchronized in serum-free medium for 12 h. Medium was changed and the cells were incubated with IGF-I and the respective inhibitor for 6 h. An RNase protection assay was performed, using 32P-labeled rat IGFBP and 18S rRNA probes as described in Materials and Methods. Two representative autoradiographs of IGFBP-3 and IGFBP-5 are depicted. Autoradiographs from three independent experiments were quantified by computed densitometry. The columns represent the mean ± SE. Statistics by ANOVA. *, P < 0.05 vs. control; #, P < 0.05 vs. IGF-I.
To determine, whether the same mechanism for IGF-I-induced IGFBP-5 gene expression is also operative in other cell culture models of the growth plate, similar experiments were performed in the mesenchymal chondrogenic cell line RCJ, which are growth plate chondrocytes derived from fetal rat calvaria (24, 25, 26). An advantage of this cell line is that it does not express IGF-I; therefore, the action of this hormone can be studied without interference from endogenous IGFs (17). RCJ cells also do not express IGFBP-3 (17). In accordance with our observations in growth plate chondrocytes in primary culture, IGF-I induced IGFBP-5 gene expression to a comparable extent (Fig. 6). Coincubation of IGF-I with the PI-3 kinase inhibitor LY294002 abolished the stimulatory effect of IGF-I on IGFBP-5 mRNA expression, whereas IGF-I-induced IGFBP-5 mRNA expression was unaffected by coincubation with the p42/44 MAPK pathway inhibitor U0126 (Fig. 6, A and Fig. B). Comparable results were obtained on the level of protein expression (Fig. 6C).
FIG. 6. IGF-I-induced IGFBP-5 gene expression and protein concentration in conditioned culture medium are mediated through the PI-3 kinase but not through the MAPK/ERK1/2 pathway. Subconfluent RCJ cells in monolayer culture were synchronized in serum-free medium for 12 h and incubated with 100 ng/ml of IGF-I in the presence or absence of specific inhibitors of the different signaling pathways [LY294002 (LY) for the PI-3 kinase pathway and U0126 for the MAPK/ERK1/2 pathway] at the indicated concentrations. Control cells were cultured without IGF-I in the absence and presence of the same concentration of inhibitors. After 12 h, total RNA was extracted and real-time RT-PCR performed, and conditioned medium was collected and Western immunoblot analysis was performed. Membranes were probed with antibodies against IGFBP-5. Representative autoradiographs of a total of three experiments are shown. The columns represent the mean ± SE from three independent experiments. Statistics by ANOVA. *, P < 0.05 vs. control; #, P < 0.05 vs. IGF-I.
IGF-I-induced IGFBP-5 gene expression requires de novo mRNA transcription and de novo protein synthesis
The increased abundance of steady-state IGFBP-5 mRNA in response to IGF-I could be due to increased gene transcription and/or decreased mRNA decay. The first possibility was investigated by use of the transcription blocker actinomycin D. Subconfluent RCJ cells were serum deprived for 12 h. Actinomycin D with and without IGF-I was added to the cell culture medium and after 12 h total mRNA was extracted. Whereas actinomycin D alone did not affect IGFBP-5 mRNA abundance, it completely abolished the 3-fold increase of IGFBP-5 in response to IGF-I, suggesting that IGF-I enhances IGFBP-5 synthesis by stimulation of IGFBP-5 de novo gene transcription (Fig. 7A). Coincubation with cycloheximide, an inhibitor of protein synthesis, also abolished IGF-I-induced IGFBP-5 mRNA abundance (Fig. 7B), suggesting a mechanism also involving de novo protein synthesis. Furthermore, both actinomycin and cycloheximide abrogated the increased IGFBP-5 protein content in conditioned cell culture medium in response to IGF-I (Fig. 7C).
FIG. 7. IGF-I-induced IGFBP-5 mRNA expression requires de novo mRNA transcription and de novo protein synthesis. Subconfluent RCJ cells in monolayer culture were synchronized in serum-free medium for 12 h and stimulated by IGF-I (100 ng/ml) alone or in combination with either actinomycin D (A) or cycloheximide (B). After 12 h, RNA was extracted and IGFBP-5 mRNA expression was assessed by real-time RT-PCR. The columns represent the mean ± SE of three independent experiments. Statistics by ANOVA. *, P < 0.05 vs. control; #, P < 0.05 vs. IGF-I. C, Conditioned culture medium of RCJ cells treated as described above was collected and IGFBP-5 protein concentration was determined by Western immunoblot analysis. The membrane was probed with specific antibodies against IGFBP-5. Representative experiments of a total of two are shown.
Taken together, these data indicate that the IGF-I-stimulated cell proliferation and IGFBP-3 expression is mediated both through the p42/44 MAPK and PI-3 kinase pathway, whereas only the PI-3 kinase pathway is necessary for stimulation of IGFBP-5 mRNA expression by IGF-I.
Discussion
We observed that the effect of IGF-I on cell proliferation in rat growth plate chondrocytes in primary culture is mediated both through the p42/44 MAPK and PI-3 kinase pathway. This is the first report on the intracellular signal transduction pathways, which are used by IGF-I for stimulation of growth plate chondrocyte proliferation. The IGF receptor signaling cascade consists of two main signal transduction pathways, the p42/44 MAPK pathway and the PI-3 kinase pathway (27). Whereas the p42/44 MAPK pathway mainly mediates mitogenic signals, the PI-3 kinase pathway is thought to affect both growth and differentiation (15). The effects of both signaling pathways are cell type specific and general predictions, which signaling pathway is involved in specific responses to IGF-I in a particular cell type cannot be made. For example, whereas in L6A1 myoblasts the mitogenic response to IGF-I is mediated primarily through the p42/44 MAPK pathway (28), an active PI-3 kinase, but not an active p42/44 MAPK pathway, is required to convey the mitogenic message of IGF-I in breast cancer cells (29).
We have shown here that the p42/44 MAPK and PI-3 kinase pathways do not act independently but are in fact interdependent for IGF-I-mediated cell proliferation because the mitogenic effect of IGF-I was abolished completely by inhibition of each individual pathway with the respective pharmacological inhibitor. Consistent with this observation, it was recently reported that IGF-I signals mitogenesis and survival in osteoblastic cells through parallel, partly overlapping intracellular pathways involving PI-3 kinase, MAPK/ERK1/2, and G? subunits (14). These data indicate that IGF-I exerts its mitogenic effect in chondrocytes and osteoblasts by the use of comparable intracellular signaling pathways. In contrast to our finding, Kuemmerle et al. (30) observed in human intestinal smooth muscle cells that both the PI-3 kinase and p42/44 MAPK pathways contribute independently to IGF-I-mediated cell proliferation because inhibition of each individual pathway only partially reduced the response to IGF-I. Hence, the respective contribution of the PI-3 kinase and p42/44 MAPK pathway to IGF-I-mediated cell proliferation appears to be cell type specific.
The biological activity of IGF-I is tightly regulated by six distinct IGFBPs, whose production is cell type and tissue specific and partially under control of IGF-I. We observed by use of RT-PCR that rat growth plate chondrocytes in primary culture synthesize under baseline conditions mRNA species for IGFBP-2 to -6, and we detected by ligand blot analysis of conditioned cell culture medium IGFBP-3, IGFBP-2, glycosylated IGFBP-4, and to a lesser extent IGFBP-5. Exogenous IGF-I enhanced IGFBP-3 and IGFBP-5 on the level of mRNA and protein expression in a dose- and time-dependent manner. There was also a moderate increase of IGFBP-4 and a 35-kDa band, most likely representing IGFBP-2, in response to IGF-I. Regarding IGFBP-5, there was a large discrepancy between the extent of IGFBP-5 mRNA expression in response to IGF-I (up to 13-fold) and the corresponding protein concentration in conditioned cell culture medium (up to 2-fold), suggesting posttranscriptional or posttranslational regulation of IGFBP-5 abundance. A comparable discrepancy between IGF-I-induced IGFBP-5 mRNA abundance and the corresponding protein concentration was also observed in rat articular chondrocytes (31) and rabbit costal chondrocytes (32) and ascribed to increased IGFBP-5 protease activity in the cell culture medium.
Other investigators have previously studied the regulation of IGFBPs by the IGFs in cell culture models of the growth plate. Consistent with our observations, de los Rios and Hill (33) reported that isolated epiphyseal growth plate chondrocytes from the ovine fetus release under basal conditions IGFBP-2, IGFBP-3, IGFBP-4, and IGFBP-5; mRNA species encoding these IGFBPs plus IGFBP-6 were identified by Northern blot analysis. Exposure to IGF-I or IGF-II caused an increase in expression and release of IGFBP-3 and an increase of IGFBP-5 protein, whereas, in contrast to our finding, IGFBP-5 mRNA abundance was not enhanced. The authors speculated that this discrepancy between IGFBP-5 mRNA expression and protein concentration might be due to increased release of IGFBP-5 from the cell membrane in response to the IGFs. Hence, the differences in IGFBP-5 mRNA expression in response to IGF-I in various cell culture models of chondrocytes appear to be dependent on species differences.
We observed that IGF-I enhances both the synthesis of IGFBP-3, which has mainly an inhibitory effect in cell culture models of the growth plate (11, 17), and the synthesis of IGFBP-5, which is a stimulatory IGFBP in this tissue (10). Interestingly, IGF-I uses different signaling pathways to regulate the synthesis of these IGFBPs with contrasting biological functions. The IGF-I-induced IGFBP-3 expression is mediated both through the p42/44 MAPK and PI-3 kinase pathway. Similarly as observed for IGF-I-mediated cell proliferation, the p42/44 MAPK and PI-3 kinase pathways do not act independently for IGF-I-enhanced IGFBP-3 expression but are in fact interdependent because the stimulatory effect of IGF-I on IGFBP-3 synthesis was abolished completely by inhibition of each individual pathway with the respective pharmacological inhibitor. In contrast, the IGF-I-induced IGFBP-5 expression in this cell culture model exclusively depends on the PI-3 kinase pathway. During blockade of the p44/42 MAPK pathway, IGF-I-induced IGFBP-5 synthesis was not affected, but increased IGFBP-5 synthesis in the absence of IGF activity due to a blocked p44/42 MAPK pathway was not capable to functionally counteract the blocked IGF activity on cell proliferation. Further studies are required to link these observations with studies examining the effects of altered IGFBP expression on IGF activity in growth plate chondrocytes. We hypothesize that IGF-I modulates its action by inducing both inhibitory and stimulatory IGFBPs to establish a functional equilibrium.
Consistent with our observations in growth plate chondrocytes in primary culture, exogenous IGF-I also stimulated IGFBP-5 gene expression in RCJ cells. We report here for the first time that the IGF-I-induced IGFBP-5 gene expression in a cell culture model of growth plate chondrocytes requires de novo mRNA transcription and de novo protein synthesis. Also in RCJ cells, IGF-I-induced IGFBP-5 gene expression was mediated through the PI-3 kinase but not through the p42/44 MAPK pathway. These congruent data indicate that our observations regarding the mechanism of IGF-I-induced IGFBP-5 synthesis apply to growth plate chondrocytes in general.
Other investigators have studied the intracellular signaling pathways that are involved in the IGF-I-mediated expression of IGFBP-3 and IGFBP-5. Consistent with our observation, Koedam et al. (32) reported that the IGF-induced IGFBP-5 expression in rabbit costal chondrocytes requires the activation of PI-3 kinase; the p42/44 MAPK pathway and the pathways involved in the IGF-I-mediated IGFBP-3 expression were not investigated in this study. In agreement with our observation, the stimulation of IGFBP-5 mRNA expression by IGF-I is mediated exclusively via the PI-3 kinase pathway in vascular smooth muscle cells (34) and primary Schwann cells (35). Contrasting observations were made in rat intestinal smooth muscle cells, in which the IGF-I-mediated IGFBP-5 induction was mediated exclusively via the p42/44 MAPK pathway (36). Regarding the regulation of IGFBP-3 expression by IGF-I, only two previous studies with different results have been published. In mammary epithelial cells, the IGF-induced IGFBP-3 expression is mainly regulated via the PI-3 kinase pathway and only partly via the p42/44 MAPK pathway (37). In human intestinal smooth muscle cells, the IGF-I-enhanced IGFBP-3, -4, and -5 mRNA expression is mediated both via the p42/44 MAPK and PI-3 kinase pathway (38). Hence, the intracellular signaling pathways involved in the IGF-I-mediated gene expression of IGFBP-3 and IGFBP-5 are cell type specific.
In conclusion, our data demonstrate that IGF-I modulates its activity in cultured rat growth plate chondrocytes by the synthesis of both inhibitory (IGFBP-3) and stimulatory (IGFBP-5) binding proteins. IGF-I stimulates cell proliferation and IGFBP-3 mRNA expression through both the p42/44 MAPK and PI-3 kinase pathways, whereas IGF-I-induced IGFBP-5 expression depends only on the PI-3 kinase pathway. The finding that IGF-I uses different and only partially overlapping intracellular signaling pathways for the regulation of two IGFBPs with opposing biological functions might be important for the regulation of IGF bioactivity in the cellular microenvironment.
Acknowledgments
We thank Dr. Anna Spagnoli (Vanderbilt University Medical Center, Nashville, TN) for generously providing the RCJ3.1C5.18 cell line. Furthermore, we thank Ralph Witzgall for technical support with the RNase protection assay and Ludger St?ndker for providing radiolabeled IGF-II.
References
Ohlsson C, Nilsson A, Isaksson OG, Lindahl A 1992 Effect of growth hormone and insulin-like growth factor-I on DNA synthesis and matrix production in rat epiphyseal chondrocytes in monolayer culture. J Endocrinol 133:291–300
Olney RC, Wang J, Sylvester JE, and Mougey EB 2004 Growth factor regulation of human growth plate chondrocyte proliferation in vitro. Biochem Biophys Res Commun 317:1171–1182
Reinecke M, Schmid AC, Heyberger-Meyer B, Hunziker, EB, Zapf J 2000 Effect of growth hormone and insulin-like growth factor (IGF-I) on the expression of IGF-I messenger ribonucleic acid and peptide in rat tibial growth plate and articular chondrocytes in vivo. Endocrinology 141:2847–2853
Yakar S, Rosen CJ, Beamer WG, Ackert-Bicknell CL, Wu Y, Liu JL, Ooi GT, Setser J, Frystyk J, Boisclair YR, LeRoith D 2002 Circulating levels of IGF-1 directly regulate bone growth and density. J Clin Invest 110:771–781
Wang J, Zhou J, Bondy CA 1999 IGF1 promotes longitudinal bone growth by insulin-like actions augmenting chondrocyte hypertrophy. FASEB J 13:1985–1990
Wang J, Zhou J, Cheng CM, Kopchick JJ, Bondy CA 2004 Evidence supporting dual, IGF-I-independent and IGF-I-dependent, roles for GH in promoting longitudinal bone growth. J Endocrinol 180:247–255
Baxter RC 2000 Insulin-like growth factor (IGF)-binding proteins: interactions with IGFs and intrinsic bioactivities. Am J Physiol Endocrinol Metab 278:E967–E976
Jones JI, Clemmons DR 1995 Insulin-like growth factors and their binding proteins: biological actions. Endocr Rev 16:3–34
Schneider MR, Wolf E, Hoeflich A, Lahm H 2003 IGF-binding protein-5: flexible player in the IGF system and effector on its own. J Endocrinol 172:423–440
Kiepe D, Andress DL, Mohan S, Standker L, Ulinski T, Himmele R, Mehls O, T?nshoff B 2001 Intact IGF-binding protein-4 and -5 and their respective fragments isolated from chronic renal failure serum differentially modulate IGF-I actions in cultured growth plate chondrocytes. J Am Soc Nephrol 12:2400–2410
Kiepe D, Ulinski T, Powell DR, Durham SK, Mehls O, T?nshoff B 2002 Differential effects of insulin-like growth factor binding proteins-1, -2, -3, and -6 on cultured growth plate chondrocytes. Kidney Int 62:1591–1600
Parrizas M, Saltiel AR, LeRoith D 1997 Insulin-like growth factor 1 inhibits apoptosis using the phosphatidylinositol 3'-kinase and mitogen-activated protein kinase pathways. J Biol Chem 272:154–161
Tsakiridis T, Tsiani E, Lekas P, Bergman A, Cherepanov V, Whiteside C, Downey GP 2001 Insulin, insulin-like growth factor-I and platelet-derived growth factor activate extracellular signal-regulated kinase by distinct pathways in muscle cells. Biochem Biophys Res Commun 288:205–211
Grey A, Chen Q, Xu X, Callon K, Cornish J 2003 Parallel phosphatidylinositol-3 kinase and p42/44 mitogen-activated protein kinase signaling pathways subserve the mitogenic and antiapoptotic actions of insulin-like growth factor I in osteoblastic cells. Endocrinology 144:4886–4893
Petley T, Graff K, Jiang W, Yang H, Florini J 1998 Variation among cell types in the signaling pathways by which IGF-I stimulates specific cellular responses. Horm Metab Res 31:70–76
Klaus G, Merke J, Eing H, Hugel U, Milde P, Reichel H, Ritz E, Mehls O 1991 1,25(OH)2D3 receptor regulation and 1,25(OH)2D3 effects in primary cultures of growth cartilage cells of the rat. Calcif Tissue Int 49:340–348
Spagnoli A, Hwa V, Horton WA, Lunstrum GP, Roberts Jr CT, Chiarelli F, Torello M, Rosenfeld RG 2001 Antiproliferative effects of insulin-like growth factor-binding protein-3 in mesenchymal chondrogenic cell line RCJ3.1C5.18. relationship to differentiation stage. J Biol Chem 276:5533–5540
Favata MF, Horiuchi KY, Manos EJ, Daulerio AJ, Stradley DA, Feeser WS, Van Dyk DE, Pitts WJ, Earl RA, Hobbs F, Copeland RA, Magolda RL, Scherle PA, Trzaskos JM 1998 Identification of a novel inhibitor of mitogen-activated protein kinase kinase. J Biol Chem 273:18623–18632
Homme M, Schmitt CP, Himmele R, Hoffmann GF, Mehls O, Schaefer F 2003 Vitamin D and dexamethasone inversely regulate parathyroid hormone-induced regulator of G protein signaling-2 expression in osteoblast-like cells. Endocrinology 144:2496–2504
H?flich A, Lahm H, Blum W, Kolb H, Wolf E 1998 Insulin-like growth factor-binding protein-2 inhibits proliferation of human embryonic kidney fibroblasts and of IGF-responsive colon carcinoma cell lines. FEBS Lett 434:329–334
Matsumoto T, Tsurumoto T, Goldring MB, Shindo H 2000 Differential effects of IGF-binding proteins, IGFBP-3 and IGFBP-5, on IGF-I action and binding to cell membranes of immortalized human chondrocytes. J Endocrinol 166:29–37
Olney RC, Smith RL, Kee Y, Wilson DM 1993 Production and hormonal regulation of insulin-like growth factor binding proteins in bovine chondrocytes. Endocrinology 133:563–570
Borromeo V, Bramani S, Holder AT, Carter C, Secchi C, Beattie J 1996 Growth hormone stimulates the secretion of insulin-like growth factor binding protein-2 (IGFBP-2) by monolayer cultures of sheep costal growth plate chondrocytes. Mol Cell Biochem 162:145–151
Grigoriadis AE, Heersche JN, Aubin JE 1996 Analysis of chondroprogenitor frequency and cartilage differentiation in novel family of clonal chondrogenic rat cell lines. Differentiation 60:299–307
Lunstrum GP, Keene DR, Weksler NB, Cho Y-J, Cornwall M, Horton WA 1999 Chondrocyte differentiation in a rat mesenchymal cell line. J Histochem Cytochem 47:1–6
McDougall S, Fu YH, Lowe GN, Williams A, Polendo R, Benya PD, Iida-Klein A, Fang MA, Hahn TJ 1996 Surface adhesion-mediated regulation of chondrocyte-specific gene expression in the nontransformed RCJ3.1C5.18 rat chondrocyte cell line. J Bone Miner Res 11:1130–1138
LeRoith D 2000 Insulin-like growth factor I receptor signaling—overlapping or redundant pathways? Endocrinology 141:1287–1288[Free Full Text]
Coolican SA, Samuel DS, Ewton DZ, McWade FJ, Florini JR 1997 The mitogenic and myogenic actions of insulin-like growth factors utilize distinct signaling pathways. J Biol Chem 272:6653–6662
Dufourny B, Alblas J, van Teeffelen HA, van Schaik FM, van der Burg B, Steenbergh PH, Sussenbach JS 1997 Mitogenic signaling of insulin-like growth factor I in MCF-7 human breast cancer cells requires phosphatidylinositol 3-kinase and is independent of mitogen-activated protein kinase. J Biol Chem 272:31163–31171
Kuemmerle JF, Bushman TL 1998 IGF-I stimulates intestinal muscle cell growth by activating distinct PI 3-kinase and MAP kinase pathways. Am J Physiol 275:G151–G158
Matsumoto T, Gargosky SE, Oh Y, Rosenfeld RG 1996 Transcriptional and post-translational regulation of insulin-like growth factor-binding protein-5 in rat articular chondrocytes. J Endocrinol 148:355–369
Koedam JA, Hoogerbrugge CM, Van Buul-Offers SC 2000 Differential regulation of IGF-binding proteins in rabbit costal chondrocytes by IGF-I and dexamethasone. J Endocrinol 165:557–567
De Los Rios P, Hill DJ 2000 Expression and release of insulin-like growth factor binding proteins in isolated epiphyseal growth plate chondrocytes from the ovine fetus. J Cell Physiol 183:172–181
Duan C, Liimatta MB, Bottum OL 1999 Insulin-like growth factor (IGF)-I regulates IGF-binding protein-5 gene expression through the phosphatidylinositol 3-kinase, protein kinase B/Akt, and p70 S6 kinase signaling pathway. J Biol Chem 274:37147–37153
Cheng HL, Shy M, Feldman EL 1999 Regulation of insulin-like growth factor-binding protein-5 expression during Schwann cell differentiation. Endocrinology 140:4478–4485
Xin X, Hou YT, Li L, Schmiedlin-Ren P, Christman GM, Cheng HL, Bitar KN, Zimmermann EM 2004 IGF-I increases IGFBP-5 and collagen 1(I) mRNAs by the MAPK pathway in rat intestinal smooth muscle cells. Am J Physiol Gastrointest Liver Physiol 286:G777–G7783
Sivaprasad U, Fleming J, Verma PS, Hogan KA, Desury G, Cohick WS 2004 Stimulation of insulin-like growth factor (IGF) binding protein-3 synthesis by IGF-I and transforming growth factor- is mediated by both phosphatidylinositol-3 kinase and mitogen-activated protein kinase pathways in mammary epithelial cells. Endocrinology 145:4213–4221
Kuemmerle JF 2000 Endogenous IGF-I regulates IGF binding protein production in human intestinal smooth muscle cells. Am J Physiol Gastrointest Liver Physiol 278:G710–G717(Daniela Kiepe1, Sonia Cia)