Glucocorticoids induce glutamine synthetase expression in human osteoblastic cells: a novel observation in bone

Bone 34 (2004) 320 – 329 www.elsevier.com/locate/bone

Glucocorticoids induce glutamine synthetase expression in human osteoblastic cells: a novel observation in bone A. Olkku, a,* P.V.N. Bodine, b A. Linnala-Kankkunen, c and A. Mahonen a a

b

Department of Medical Biochemistry, University of Kuopio, Kuopio, Finland Women’s Health Research Institute, Wyeth Research, Collegeville, PA 19426, USA c Department of Biochemistry, University of Kuopio, Kuopio, Finland Received 16 May 2003; revised 28 August 2003; accepted 26 September 2003

Abstract Glucocorticoids have marked effects on bone metabolism, and continued exposure of skeletal tissue to excessive amounts of these steroids results in osteoporosis. Therefore, in the present proteomic study, we characterized the potential effects of glucocorticoids on protein expression in human osteoblastic cells. Using two-dimensional gel electrophoresis and mass spectrometry, we identified an increased expression of glutamine synthetase (GS) in dexamethasone (Dex)-treated human MG-63 osteosarcoma cells. GS is an enzyme catalyzing the conversion of glutamate and ammonia to glutamine. Intracellular and extracellular glutamate levels may be important in cell signalling mediated by glutamate transporters and receptors which have recently been found in bone cells. The induction of GS protein by Dex was accompanied by an increase in mRNA level and enzyme activity. Dex induction of GS was also mediated by glucocorticoid receptors (GRs) because it was blocked by the GR antagonist RU-38486. In addition, Dex induction of GS expression was partially blocked by cyclohexamide indicating that it at least partly required new protein synthesis. GS induction by Dex was not associated with apoptosis as determined by Bax/Bcl-2 ratio and DNA staining. In addition to MG-63 cells, Dex induction of GS was also observed in human G-292 osteosarcoma cells as well as conditionally immortalized human preosteoblastic (HOB-03-C5) and mature osteoblastic (HOB-03-CE6) cells. However, in two other human osteosarcoma cell lines, SaOS-2 and U2-OS, GS expression was not affected by Dex. This observation may be explained by the lower levels of GR protein in these cells. In summary, this is the first report of the regulation of GS expression by glucocorticoids in bone cells. The role of GS in bone cell metabolism and glucocorticoid action on the skeleton is not yet known, but as a modulator of intracellular glutamate and glutamine levels, it may have an important role in these processes. D 2003 Elsevier Inc. All rights reserved. Keywords: Osteoblast; Glucocorticoid; Proteomics; Glutamine synthetase; Glutamate

Introduction Glucocorticoids, such as cortisol, are steroid hormones whose cellular actions are mediated through nuclear receptors that act as ligand-inducible transcription factors [35]. Glucocorticoids bind to cytosolic glucocorticoid receptors (GRs) that upon binding are activated by removal of heat-shock proteins (Hsps), like Hsp90, from the DNA-binding region of the receptor [3,16]. The active GR forms a homodimer that is rapidly translocated to the

* Corresponding author. Department of Medical Biochemistry, University of Kuopio, PO Box 1627, FIN-70211 Kuopio, Finland. Fax: +35817-2811510. E-mail address: [email protected] (A. Olkku). 8756-3282/$ - see front matter D 2003 Elsevier Inc. All rights reserved. doi:10.1016/j.bone.2003.09.010

nucleus. Within the nucleus, the GR either induces gene transcription by binding to specific glucocorticoid responsive elements (GREs) in the promoter –enhancer regions of target genes, or reduces gene transcription by transrepression. There is also increasing evidence for nongenomic actions of glucocorticoids that are mediated by membrane receptors or other physicochemical interactions [27,38]. Glucocorticoids have well-documented effects on bone metabolism [15,17,28], although their mechanism of action is still poorly understood. The effects of glucocorticoids on bone cells in different experimental models are complex and depend upon the concentration of the steroid, the presence of serum as a source of interacting hormones, species differences, the timing of hormone addition, and on the maturation stage of the osteoblast. Glucocorticoids are

A. Olkku et al. / Bone 34 (2004) 320–329

known to promote the development of the osteoblastic phenotype in osteoprogenitor cells, and in this way contribute to osteogenic differentiation from mesenchymal stem cells. Several studies have shown that physiological concentrations of glucocorticoids induce the differentiation of human bone marrow osteoprogenitor cells into cells that exhibit a mature osteoblast phenotype [2,11,53]. Continued exposure of the skeletal tissue to these steroids, however, results in decreased bone formation by inhibiting proliferation and promoting apoptosis of osteoblastic cells [31,50,52]. Therefore, the high and frequent doses of glucocorticoids used pharmacologically can cause osteoporosis [9,23,30,43]. The aim of this study was to characterize changes in protein expression profiles during glucocorticoid treatment of human osteoblast-like cells. Total proteins from controland dexamethasone (Dex)-treated human MG-63 osteosarcoma cells were analyzed by two-dimensional gel electrophoresis (2DE). Comparison of these 2D-gels revealed one major change. This Dex-inducible protein was in-gel digested with trypsin, and the masses of the resulting peptides were measured with matrix-assisted laser desorption or ionization-time of flight mass spectrometry (MALDI-TOF MS). The obtained masses corresponded to the tryptic peptides of glutamine synthetase (GS). In further studies, we showed that glucocorticoids, by a direct GRmediated mechanism, induced GS expression in several human osteoblastic cell lines. Glucocorticoid-induced GS expression was the outcome of changes in gene transcription and translation as well as protein stabilization. GS is an enzyme that catalyzes the conversion of glutamate and ammonia to glutamine, and is therefore able to regulate intracellular and extracellular glutamate concentrations as well as intracellular glutamine concentration. This novel observation is intriguing, because glutamate signalling has recently been found to be functional in nonneuronal tissues including bone [12,47,48].

Materials and methods

321

Osteosarcoma cells were seeded at 12 000 –18 000 cells/ cm2 depending on the experiment and cultured in a medium containing 7% FBS. After overnight incubation, the cells were washed with phosphate buffered saline (PBS) and the medium was changed into a medium containing 2% charcoal-treated FBS to eliminate the effects of endogenous steroid hormones. Steroid hormones, RU38486, and cyclohexamide (Sigma, St. Louis, MO, USA) were added to the cultures in ethanol. Riluzole (Sigma) was diluted into DMSO. Control cultures contained the same amounts of ethanol or DMSO with the final concentrations being 0.1%. Glutamine-free DMEM was used for the glutamine-deprivation experiments. HOB cell experiments were conducted according to Bodine et al. [6,7] at 39jC for 48 h, that is, in conditions where the cells revert to the normal phenotype. Two-dimensional gel electrophoresis Osteoblastic cells were washed with PBS before lysing them in buffer containing 9.8 M urea, 2% CHAPS, 100 mM DTT, and 0.2% (w/v) Bio-Lytes 3/10 (Bio-Rad Laboratories, Hercules, CA, USA). Protein concentration was determined using Bradford assay (Bio-Rad). 2DE was performed essentially according to O’Farrell [36] using Bio-Rad ProteanR IEF Cell and 17 cm immobilized pH gradient (IPG) ReadyStripsk (pH 5 – 8). For the first dimension separation, the protein samples (20 –100 Ag) were introduced to the IPG strips by passive rehydration with rehydration buffer [9.8 M urea, 2% CHAPS, 15 mM DTT, 0.2% (w/v) Bio-Lytes 3/10, trace of bromophenol blue] during overnight incubation. Isoelectric focusing was conducted for a total of about 85 000 Vh. The second dimension separation was a standard sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) with 12% resolving gel overlaid by 4% stacking gel. Before running the second dimension, the IPG strips were equilibrated by soaking in buffer containing 6 M urea, 0.375 M Tris (pH 8.8), 2% SDS, 20% glycerol twice for 10 min; 2% DTT (w/v) was added to the equilibration buffer for the first and 2.5% (w/v) iodoacetamide for the second wash.

Cell culture Silver staining MG-63, G-292, SaOS-2, and U2-OS human osteosarcoma cell lines were obtained from the American Type Culture Collection (ATCC, Rockville, MD, USA). The cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM, Gibco, Grand Island, NY, USA) supplemented with 7% fetal bovine serum (FBS), 2 mM L-glutamine, 100 U/ml penicillin, and 0.1 mg/ml streptomycin at 37jC in a humidified atmosphere of 5% CO2 in air. The conditionally immortalized human preosteoblasts (HOB-03-C5) and mature osteoblasts (HOB-03-CE6) were cultured as previously described [6,7]. These cells express a transformed phenotype when cultivated at 34jC, but revert to a normal phenotype at 39jC.

After electrophoresis, the gels were silver stained with a mass spectrometry-compatible method: the gels were fixed with 40% ethanol/5% acetic acid for overnight, washed with water, sensitized with 0.02% sodium thiosulphate, incubated 30 min with chilled 0.1% silver nitrate, and developed with 0.04% formaldehyde in 2% sodium carbonate; 1% acetic acid was used to stop the reaction. In-gel protein digestion and peptide recovery Protein spots were excised from silver-stained gels and in-gel digested with modified trypsin. A control piece of gel

322

A. Olkku et al. / Bone 34 (2004) 320–329

was cut from a blank region and processed in parallel with the samples. After the gel pieces were excised and shrunk by dehydration in acetonitrile, they were dried in a vacuum centrifuge. The gel pieces were swollen in a digestion buffer containing 25 mM NH4HCO3 and 0.05 Ag/Al trypsin (sequencing grade, Promega, Madison, WI, USA) in an icecold bath. Additional buffer without trypsin was added, if necessary, to keep the gel pieces wet during enzymatic cleavage (37jC, overnight). Peptides were extracted with three changes of 50% acetonitrile/5% TFA and two changes of H2O by shaking at room temperature and then dried to volumes <10 Al. MALDI analysis The recrystallized matrix material (a-cyano-4-hydroxytrans-cinnamic acid, Sigma) was dissolved in 30% acetonitrile/5% TFA to obtain a saturated solution. Tryptic peptides were mixed with matrix (1:1) and 1 Al of this mixture was deposited on a stainless steel sample target. All mass spectra were obtained using positive ion mode on a time-of-flight mass spectrometer, model Biflexk III (Bruker-Franzen Analytik, Bremen, Germany) equipped with multiprobe inlet. The detector signals were transferred to the XACQ program on a SUN workstation (Sun Microsystems Inc., Palo Alto, CA, USA). Saved data were

processed with the XMASS 5.0 program (Bruker Daltonics, Bremen, Germany). Calibration for the peptides was performed using angiotensin (1046.5 Da) as an internal standard and the resulting peptide masses were analyzed with PeptIdent (ExPASy Molecular Biology Server, Swiss Institute of Bioinformatics, Switzerland). The Mascot peptide mass fingerprint program (Matrix Science Ltd, UK) was used to calculate the significance for the match. Immunoblotting Cells for total protein extraction were lysed in TXM buffer [10 mM Tris –HCl (pH 7.4), 1 mM MgCl2, 20 AM ZnCl2, 0.02% NaN3, 0.1% Triton X-100] and sonicated on ice for 2  20 s. The lysate was centrifuged at 13 200 rpm for 30 min 4jC, and the protein concentration of the clear supernatant was determined using Bradford assay (Bio-Rad). Protein extracts (10 – 50 Ag) were separated on SDS-PAGE using 12% acrylamide gels. Proteins were electrotransferred to polyvinylidene difluoride (PVDF) membranes (Boehringer Mannheim Gmbl, Germany). Immunodetection was performed using rabbit polyclonal antibodies to human GS, GR, Bax, and Bcl-2 (Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA). The secondary anti-rabbit antibody was alkaline phosphatase conjugated, and the protein patterns were detected with nitroblue tetrazolium and 5-bromo-4-

Fig. 1. (A) A two-dimensional, silver-stained gel of protein extracts from human MG-63 cells. For the first dimension, 40 Ag of total protein was separated on an IPG strip (pH 5 – 8) followed by a standard SDS-PAGE for the second dimension. The inset shows the region of interest that is displayed in panel B. (B) Differentially expressed protein pattern after 24-h treatment with 100 nM of Dex. In further studies, the protein identified by the arrow was determined to be glutamine synthetase (GS) by MALDI-TOF MS and PeptIdent peptide mass analyzing program (MW 42 kDa, pI 6.4).

A. Olkku et al. / Bone 34 (2004) 320–329 Table 1 Identification of the Dex-inducible protein spot Mass submitted/matched

Peptide sequence, modifications 335

GYFEDR340 HQYHIR286 358 TCLLNETGDEPFQYK372 341 RPSANCDPFSVTEALIR357 300 LTGFHETSNINDFSAGVANR319 281

1 CAMa (C) 1 CAMa (C)

786.26/786.34 853.38/853.44 1815.17/1814.83 1933.40/1932.96 2150.52/2150.03

Protein sequence of glutamine synthetase as retrieved from Swiss-Prot (accession no. P15104). Sequence coverage 17.2%. a CAM = carboxamidomethylation.

chloro-3-indolyl phosphate. Each experiment was repeated four to six times with similar results. Statistical analysis was performed by the Student’s t test with the Microsoft Excel XP program. Nuclear fragmentation assay with Hoechst staining Morphological assessment of apoptotic cells was performed using Hoechst 33258 (Sigma) staining. Cells grown on glass coverslips were washed in PBS and fixed with 4% paraformaldehyde in PBS for 10 min. Cells

323

were permeabilized in PBS containing 0.1% Triton X-100 for 10 min, and nuclei were stained with Hoechst dye (10 Ag/ml, 10 min) and observed with fluorescence microscopy (Nikon FXA Microphot, Nikon Corp., Tokyo, Japan). RT-PCR Total cellular RNA was isolated by the guanidinium thiocyanate method of Chomczynski and Sacchi [14]. Total RNA (1 Ag) was reverse transcribed using MuLVreverse transcriptase (MBI Fermentas Inc., Amherst, NY, USA) and oligo16 (dT) primer. First, RNA was denatured at 65jC for 5 min followed by cDNA-synthesis for 1 h in 37jC. The reaction was stopped at 70jC for 8 min. The synthetized cDNA was used as a template for a PCR reaction with Cy5-labelled primer pairs for both GS [51] and glyceraldehyde phosphate dehydrogenase (GAPDH) [44] in the same reaction tube. GAPDH was used as an internal standard to normalize the mRNA expression. The PCR products were quantified by ALFexpress DNA Analysing System (Amersham Biosciences, Buckinghamshire, UK). Each RNA isolation, cDNA synthesis, as well as PCR reactions were repeated independently two to three times with similar results. Statistical analysis were

Fig. 2. (A) Induction of GS protein expression (42 kDa) by glucocorticoids in MG-63 cells. The cells were treated with vehicle, 100 nM Dex, 100 nM hydrocortisone (HC), and 100 nM prednisolone (Pred) for 24 h. (B) Time-dependent expression of GS protein after treatment of MG-63 cells with 100 nM of Dex. (C) Dose-dependent expression of GS protein after 24-h treatment with Dex in MG-63 cells. (D) GS protein expression after treatment of various human osteoblastic cells with 100 nM of Dex for 24 h. The HOB cells were cultured for 48 h to revert them to the normal phenotype. After treatments, the cells were processed for Western blot analysis, and each blot is represented also as relative densitometric units with means F SEM (n = 6). In each experiment, P values reached the most significant level (P < 0.001).

324

A. Olkku et al. / Bone 34 (2004) 320–329

Table 2 GS enzyme activity. Amol/min/g (mean F SEM) Cell line

control

MG-63a G-292a SaOs-2a U2-Osa

17.2 F 11.0 F 12.6 F 5.0 F

HOB-03-C5b HOB-03-CE6b

9.7 F 0.8 10.5 F 1.2

1.9 0.8 1.7 0.8

100 nM Dex

P

242.7 60.2 15.1 4.6

<0.001 <0.001 n.s. n.s.

F F F F

19.2 4.4 0.7 0.6

19.5 F 0.9 36.4 F 2.3

detected at this pI range with silver staining (Fig. 1B). This Dex-inducible protein spot was excised from six duplicate gels and digested in-gel with modified trypsin. The exact masses of the resulting peptides were determined by MALDI-TOF mass spectrometry and compared to the theoretical peptide maps (Table 1). Only the masses of peptides with good isotopic distribution were used. No missed cleavage sites by trypsin were included, and only

<0.001 <0.001

Assay measures GS catalyzed g-glutamyltransferase activity. n.s.: not significant. a Cells treated with vehicle or 100 nM Dex for 24 h (n = 6). b Cells treated with vehicle or 100 nM Dex for 48 h (n = 4).

performed by the Student’s t test with the Microsoft Excel XP program. GS activity assay GS activity of control and Dex-treated cells was assayed according to Santoro et al. [46] with slight modifications. Due to the more rigid cell wall of these cells, the samples were sonicated on ice for 2  20 s to completely lyse the cells. The protein concentrations were measured by Bradford assay (Bio-Rad). The experiments were repeated independently two to three times with similar results. Statistical analysis was performed by the Student’s t test with the Microsoft Excel XP program.

Results Characterization of protein pattern changes in dexamethasone-treated MG-63 cells In the present study, we were interested in the potential effects of Dex, a synthetic glucocorticoid, on protein expression patterns in human osteoblastic cells. Two-dimensional gels of control and Dex-treated MG-63 cells were compared to characterize differentially expressed proteins at a pI range from 5 to 8 (Fig. 1A). One major change was Table 3 Expression of GS mRNA in MG-63, SaOS-2, HOB-03-C5, and HOB-03CE6 cells Densitometric units (mean F SEM) Cell line

control

100 nM Dex

P

MG-63a SaOs-2a

0.15 F 0.03 0.20 F 0.02

0.78 F 0.07 0.21 F 0.05

<0.001 n.s.

HOB-03-C5b HOB-03-CE6b

0.23 F 0.02 0.28 F 0.03

0.48 F 0.03 0.80 F 0.05

<0.001 <0.001

n.s.: not significant. a Cells treated with vehicle or 100 nM Dex for 3 h (n = 6). b Cells treated with vehicle or 100 nM Dex for 48 h (n = 8).

Fig. 3. (A) Effect of the glucocorticoid receptor antagonist, RU-38486, on the Dex induction of GS protein in MG-63 cells. Control and 100 nM Dex samples were treated for 24 h. For samples containing the antagonist, the cells were treated first for 1 h with 10 AM RU-38486 and either collected or 100 nM Dex was added to the culture medium for 23-h. The cells were then processed for Western blot analysis. (B) Expression of glucocorticoid receptor in osteoblastic cells. The cells were cultured under experimental conditions for 48 h before being processed for Western blot analysis. (C) Effect of a protein synthesis inhibitor on Dex-induced GS protein expression in MG-63 cells. The cells were treated with vehicle, 10 AM cyclohexamide (Chx), 100 nM Dex, or both for 24 h and then processed for Western blot analysis, and each blot is represented also as relative densitometric units with means F SEM (n = 4). In the Chx experiment, P values reached the most significant level (P < 0.001).

A. Olkku et al. / Bone 34 (2004) 320–329

one match was received from the analysis. According to the Mascot search engine, the identification was statistically significant ( P < 0.05, window of error 250 ppm). The expected pI and MW of the identified protein corresponded with its position in the 2D gels. The matching protein, glutamine synthetase (GS), and the Dex induction of it were confirmed by Western blotting using anti-GS antibody, which recognizes the 42-kDa subunit of the enzyme. Glucocorticoids selectively increase both protein and mRNA levels of GS in human osteoblastic cells GS induction was observed by the natural glucocorticoid, cortisol, and by two synthetic compounds, dexamethasone

325

and prednisolone (100 nM) (Fig. 2A). Dex was used in the following studies. With 100 nM of Dex, the maximum response in GS protein level was reached within 12 h, and the increase was estimated to be about 12-fold compared to the control level in MG-63 cells (Fig. 2B). The elevated expression level was maintained with a slight decrease at least for 72 h. The GS protein induction was also dose dependent: after a 24-h treatment, GS protein level increased with 10 nM Dex and was maximal with 100 nM (Fig. 2C). GS protein expression was also determined using other human osteoblastic cell lines. In G-292 cells and in conditionally immortalized human osteoblast cells lines representing preosteoblastic cells (HOB-03-C5) and mature osteo-

Fig. 4. (A) Expression of apoptotic marker proteins Bax and Bcl-2 in MG-63 cells. The cells were treated with vehicle, 100 nM Dex, and 50 AM riluzole for 72 h and then processed for Western blot analysis. (B) Riluzole-induced morphological changes in MG-63 cells (72 h) observed with fluorescence microscopy. Hoechst staining revealed nuclear fragmentation and chromatin condensation in riluzole-treated cultures, which were absent in control and 100 nM Dex-treated MG-63 cells. Scale bar = 50 Am. (C) Effect of riluzole-inhibited glutamate release on GS protein expression in MG-63 cells. The cells were treated with vehicle, 100 nM Dex, 50 AM riluzole, or both for 24 h and then processed for Western blot analysis. Each blot is represented also as relative densitometric units with means F SEM (n = 4). In each experiment, P values reached the most significant level (P < 0.001).

326

A. Olkku et al. / Bone 34 (2004) 320–329

blasts (HOB-03-CE6), the GS protein was elevated by Dex. The induction in these cells was weaker when compared to MG-63 cells, being about 6-fold (Fig. 2D). These immortal cell lines were cultured for a longer period (48 h) due to the retrieval of the normal phenotype during this time. However, in two other human osteosarcoma cell lines, SaOS-2 and U2OS, the GS protein expression was not affected during the 72h Dex treatment. The basal expression level of GS was very low and approximately the same in all these cell lines. Also, the enzyme activity of GS was increased accordingly in MG63, G-292, and HOB cells (Table 2). The proliferation of MG63 cells was decreased by Dex, but the growth of SaOS-2 and U2-OS cells did not respond to the Dex treatment (data not shown). Furthermore, with glutamine deprivation (16 h), the basal level of GS in MG-63 cells was increased by about 3fold, but there was no significant change in Dex-induced GS expression level with or without glutamine (data not shown). Observed with RT-PCR, the GS mRNA level was upregulated in MG-63 cells by about 5-fold after 100 nM Dex administration for 3 h. GS mRNA levels remained constant in SaOS-2 cells with the same treatment. HOB cells, reverted to the normal phenotype at 39jC, also responded at the transcriptional level to Dex with a 2- to 3-fold increase in GS mRNA (Table 3). Also here the HOB cells were cultured for 48 h. Osteoblastic cells have variable levels of glucocorticoid receptors which mediate Dex induction of GS RU-38486, a selective glucocorticoid receptor (GR) antagonist [20], completely blocked the Dex-induced increase in GS protein expression in MG-63 cells (Fig. 3A). However, treatment with the antagonist by itself had no effect on GS expression. Further, the various osteoblastic cells had different basal levels of GR as determined by Western blot analysis with an anti-GR antibody (Fig. 3B). The highest expression of GR was seen in MG-63, G-292, and HOB cells, which is consistent with the level of GS inducibility by glucocorticoids observed in these cells. Cyclohexamide inhibits the dexamethasone-mediated induction of GS To determine whether the Dex-induced GS expression needed new protein synthesis, the cultured MG-63 cells were treated with 10 AM cyclohexamide and 100 nM Dex for 24 h (Fig. 3C). Accordingly, cyclohexamide treatment resulted in partial, approximately 70% inhibition of GS induction by Dex. The same effect was also seen earlier at the 8-h time point (data not shown). Glucocorticoid-induced GS expression is not directly linked to apoptosis Riluzole, an inhibitor of glutamate release, has been shown to induce apoptosis in osteoblastic cells, including

MG-63 cells [21]. In our experiments, we wanted to see whether the use of riluzole had any effect on the GS expression in osteoblastic cells as it increases the level of the substrate for this enzyme. Riluzole (50 AM, 72 h) decreased the expression of anti-apoptotic protein Bcl-2 thus increasing the Bax/Bcl-2 ratio, which is an indicator of apoptotic status (Fig. 4A). In addition, Hoechst staining revealed nuclear fragmentation and chromatin condensation (Fig. 4B) at the same time point. These apoptotic changes were not observed in control or 100-nM Dex-treated MG-63 cells under the same experimental conditions. Under the apoptotic conditions, GS expression was not induced in MG-63 cells by riluzole. GS expression was only elevated when the cells were treated with Dex in the absence or presence of riluzole (Fig. 4C).

Discussion Glucocorticoids have marked effects on bone metabolism, and continued exposure of skeletal tissue to excessive amounts of these steroids results in osteoporosis. However, the actions of these steroids on gene expression in skeletal cells are complex, and they seem to be dependent on the stage of osteoblast growth and differentiation. Therefore, in the present proteomic study, glucocorticoid-induced changes in protein expression pattern were characterized in human osteoblastic cells. Two-dimensional gels of total proteins (pI 5 – 8) isolated from human MG-63 osteoblastic cells were compared to characterize Dex-induced changes. The 2DEgels were highly reproducible allowing reliable comparison of protein expression patterns between separate runs. Using the described methods, one major change in the protein expression pattern observed in MG-63 osteoblastic cells was identified as glutamine synthetase. GS, an enzyme catalyzing conversion of glutamate to glutamine, has a central role in amino acid metabolism and pH regulation in mammals, and its expression is regulated both posttranscriptionally and posttranslationally [19,24,26]. GS expression is reported to be cell-type specific and induced by glucocorticoids in different cell types [1,22,33,45]. The specific glucocorticoid responsive elements (GRE) have been characterized in chicken [54] and rat GS genes [10], but not in human genes. This glucocorticoid response of GS is, however, a novel observation in human bone cells. In the present study, the synthetic glucocorticoid, Dex, time and dose dependently increased transcription and translation of GS in human osteoblastic MG-63 cells. The induction was also observed with two other glucocorticoids, cortisol and prednisolone. Dex induction of GS was also seen in human G-292 osteosarcoma cells and in two conditionally immortalized human osteoblastic cell lines, the preosteoblastic HOB-03-C5 cells and the mature osteoblastic HOB-03-CE6, which represent normal cell phenotype. Dex-induced change in GS protein expression was also reflected by increased activity of the enzyme in all these

A. Olkku et al. / Bone 34 (2004) 320–329

cell lines. Even though glucocorticoids increased both the transcription and translation of GS, there was only a 5-fold increase in the GS mRNA level and a 12-fold increase in GS protein level after steroid treatment of MG-63 cells. Furthermore, Dex-induced GS protein expression required new protein synthesis, estimated to be about 70% of the induction. In addition, other mechanisms like protein stabilization may explain the remaining 30% of the induction. The actions of glucocorticoids on human bone cells are mediated via specific receptors [25]. We showed using mifepristone (RU-38486), a potent and selective glucocorticoid receptor antagonist [20], that the effect of Dex on GS protein expression requires interaction of the glucocorticoid with its intracellular receptors. A similar finding has also been observed in rat muscle cells by Max et al. [33]. Interestingly, in the present study, GS gene induction by Dex was not observed in two other human osteosarcoma cell lines that were examined, SaOS-2 and U2-OS. The observation that GR levels in these two cell lines are low could, at least partly, explains the lack of GS induction by Dex in these cells [5]. However, because the GR levels in the Dexresponsive MG-63, G-292, and HOB cell lines are not quantitatively in concordance with the GS induction, there must also be other mechanisms regulating the expression of this gene. In addition, the proliferation of either SaOS-2 or U2-OS was not affected by 100 nM Dex treatment during our 96-h treatment, while in MG-63 cells, Dex is known to inhibit cell division [41]. Another possible explanation for this cell-type specific expression is that the cell lines represent different maturational stages with varying responses to steroid hormones. Supporting this hypothesis, the magnitude of hormonal response observed for GS expression in chicken retinal cultures increased markedly during retinal development [37,49]. Recently, it became clear that glutamate signalling is also functional in nonneural tissues and occurs like bone [12,47,48]. However, the importance of glutamate and its metabolism in bone cells is currently unknown. Human osteoblasts, including MG-63 cells, as well as osteoclasts express glutamate receptors that are similar to those expressed by glutamatergic synapses in the central nervous system [13,39]. There are two types of these receptors: ionotropic, such as NMDA and AMPA, as well as metabotropic receptors. In osteoblasts, NMDA receptors appeared to be involved in the process of bone formation, because blockage of these receptors inhibited expression of markers of bone and bone nodule formation in vitro [18]. In addition, osteoblasts express a glutamate transporter, GLAST (EAAT1), that is responsible for cellular glutamate uptake [32]. This transporter molecule is present only in osteoblasts and osteocytes, but not in osteoclasts. As well as GS, Rauen and Wiessner [42] demonstrated that protein expression of GLAST is inducible by glucocorticoids in cultured rat retinal Mu¨ller glial cells. Osteoblasts also possess machinery for regulated, vesicle-mediated glutamate exocytosis, and glutamate release by osteoblasts is found to be dependent

327

upon the differentiation state of the cells [4]. In this concept, Dex-induced GS expression and activity might be able to interfere with the balance between the intracellular glutamate concentration and glutamate release. Genever et al. [21] showed that by blocking glutamate release with riluzole in osteoblastic cells, including MG-63 cells, intracellular glutamate accumulated, leading to inhibition of differentiation and induction of apoptosis. Bcl-2 and Bax proteins are inhibitors and accelerators of apoptosis, respectively [8]. Programmed cell death is also characterized by cytomorphological changes and fragmentation of the cell nucleus. In our experiments, riluzole-induced apoptosis in MG-63 cells was confirmed by increased Bax/Bcl2 ratio and DNA staining. However, Dex did not induce these apoptotic features under the same experimental conditions. Nakashima et al. [34] also did not observe apoptotic events with Dex treatment of MG-63 cells, and in primary rat osteoblast cultures, cortisol treatment inhibited apoptosis [40]. On the other hand, glucocorticoids have also been shown to induce apoptosis in mouse and rat osteoblastic cells in vivo [29,52]. Although riluzole induced apoptosis and accumulation of glutamate, it did not stimulate GS expression, and simultaneous administration of both Dex and riluzole increased GS levels as well. These results indicate that the elevated level of GS expression requires Dex, and that this expression is not directly linked to glucocorticoid-induced apoptosis. In conclusion, for the first time, 2D-PAGE-based proteome analysis of glucocorticoid-treated human osteoblastic cells has demonstrated significant induction of GS expression. Further analysis revealed that this direct GR-mediated induction of GS expression was the outcome of changes in both gene transcriptional and translational events as well as protein stabilization, and was not connected to apoptosis in osteoblastic cells. These results imply that GS induction might interfere with the balance between the intracellular glutamate concentration and glutamate release in osteoblasts, and therefore, may affect glutamate signalling in the bone environment. Further studies are, however, needed to find out the possible role of this enzyme in glucocorticoid-induced outcomes in bone and its possible connection to glucocorticoid-induced osteoporosis.

Acknowledgments We thank Merja Ra¨sa¨nen, Eija Korhonen, Saara Kinnunen, and Heli Pallonen for their excellent technical assistance.

References [1] Abcouwer SF, Lukaszewicz GC, Souba WW. Glucocorticoids regulate glutamine synthetase expression in lung epithelial cells. Am J Physiol 1996;270:L141 – 51.

328

A. Olkku et al. / Bone 34 (2004) 320–329

[2] Beresford JN, Joyner CJ, Devlin C, Triffitt JT. The effects of dexamethasone and 1,25-dihydroxyvitamin D3 on osteogenic differentiation of human bone marrow stromal cells in vitro. Arch Oral Biol 1994;39:941 – 7. [3] Bertorelli G, Bocchino V, Olivieri D. Heat shock protein interactions with the glucocorticoid receptor. Pulm Pharmacol Ther 1998; 11:7 – 12. [4] Bhangu PS, Genever PG, Spencer GJ, Grewal TS, Skerry TM. Evidence for targeted vesicular glutamate exocytosis in osteoblasts. Bone 2001;29:16 – 23. [5] Bland R, Worker CA, Noble BS, Eyre LJ, Bujalska IJ, Sheppard MC, et al. Characterization of 11h-hydroxysteroid dehydrogenase activity and corticosteroid receptor expression in human osteosarcoma cell lines. J Endocrinol 1999;161:455 – 64. [6] Bodine PVN, Komm BS. Evidence that conditionally immortalized human osteoblasts express an osteocalcin receptor. Bone 1999;25: 535 – 43. [7] Bodine PVN, Green J, Harris HA, Bhat RA, Stein GS, Lian JB, et al. Functional properties of a conditionally phenotypic, estrogen-responsive, human osteoblast cell line. J Cell Biochem 1997;65:368 – 87. [8] Brady HJM, Gil-Go´mez G. Bax. The pro-apoptotic Bcl-2 family member, Bax. Int J Biochem Cell Biol 1998;30:647 – 50. [9] Canalis E. Mechanisms of glucocorticoid action in bone: implications to glucocorticoid-induced osteoporosis. J Clin Endocrinol Metab 1996; 81:3441 – 7. [10] Chandrasekhar S, Souba WW, Abcouwer SF. Identification of glucocorticoid-responsive elements that control transcription of rat glutamine synthetase. Am J Physiol 1999;276:L319 – 31. [11] Cheng SL, Yang JW, Rifas L, Zhang SF, Avioli LV. Differentiation of human bone marrow osteogenic stromal cells in vitro: induction of the osteoblast phenotype by dexamethasone. Endocrinology 1994;134: 277 – 86. [12] Chenu C. Glutamatergic innervation in bone. Microsc Res Tech 2002; 58:70 – 6. [13] Chenu C, Serre CM, Raynal C, Burt-Pichat B, Delmas PD. Glutamate receptors are expressed by bone cells and are involved in bone resorption. Bone 1998;22:295 – 9. [14] Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 1987;162:156 – 9. [15] Cooper MS, Hewison M, Stewart PM. Glucocorticoid activity, inactivity and the osteoblast. J Endocrinol 1999;163:159 – 64. [16] Defranco DB. Role of molecular chaperones in subnuclear trafficking of glucocorticoid receptors. Kidney Int 2000;57:1241 – 9. [17] Delany AM, Dong Y, Canalis E. Mechanisms of glucocorticoid action in bone cells. J Cell Biochem 1994;56:295 – 302. [18] Dobson KR, Skerry TM. The NMDA-type glutamate receptor antagonist MK801 regulates differentiation of rat bone marrow osteoprogenitors and influences adipogenesis. J Bone Miner Res [abstract] 2000; 15:SA211. [19] Eisenberg D, Gill HS, Pfluegl GMU, Rotstein SH. Structure – function relationships of glutamine synthetases. Biochim Biophys Acta 2000; 1477:122 – 45. [20] Gagne D, Pons M, Philibert D. RU 38486: a potent antiglucocorticoid in vitro and in vivo. J Steroid Biochem 1985;23:247 – 51. [21] Genever PG, Skerry TM. Regulation of Spontaneous Glutamate Release Activity in Osteoblastic Cells and its Role in Differentiation and Survival: Evidence for Intrinsic Glutamatergic Signalling in Bone. FASEB J. 2001;15:1586 – 8. [22] Harmon JM, Thompson EB. Glutamine synthetase induction by glucocorticoids in the glucocorticoid-sensitive human leukemic cell line CEM-C7. J Cell Physiol 1982;110:155 – 60. [23] Ishida Y, Heersche JNM. Glucocorticoid-induced osteoporosis: both in vivo and in vitro concentrations of glucocorticoids higher than physiological levels attenuate osteoblast differentiation. J Bone Miner Res 1998;13:1822 – 6. [24] Labow BI, Souba WW, Abcouver SF. Mechanisms governing the

[25]

[26]

[27] [28]

[29]

[30] [31]

[32]

[33]

[34]

[35] [36] [37]

[38] [39]

[40]

[41]

[42]

[43] [44]

[45]

[46]

expression of the enzymes of glutamine metabolism—Glutaminase and glutamine synthetase. J Nutr 2001;131:2467S – 74S. Liesegang P, Romalo G, Sudmann M, Wolf L, Schweikert HU. Human osteoblast-like cells contain specific, saturable, high-affinity glucocorticoid, androgen, estrogen, and 1-alpha, 25-dihydroxycholecalciferol receptors. J Androl 1994;15:194 – 9. Lie-Venema H, Hakvoort TBM, van Hemert FJ, Moorman AFM, Lamers WH. Regulation of the spatiotemporal pattern of expression of the glutamine synthetase gene. Prog Nucleic Acid Res Mol Biol 1998;61:243 – 308. Lipworth BJ. Therapeutic implications of non-genomic glucocorticoid activity. Lancet 2000;356:87 – 9. Lukert BP, Kream BE. Clinical and basic aspects of glucocorticoid action in bone. In: Bilezikian JP, Raisz LG, Rodan G, editors. Principles of Bone Biology. San Diego, CA: Academic Press; 1996. p. 533 – 48. Lynch MP, Capparelli C, Stein JL, Stein GS, Lian JB. Apoptosis during bone-like tissue development in vitro. J Cell Biochem 1998; 68:31 – 49. Manelli F, Giustina A. Glucocorticoid-induced osteoporosis. TEM 2000;11:79 – 85. Manolagas SC, Weinstein RS. New developments in the pathogenesis and treatment of steroid-induced osteoporosis. J Bone Miner Res 1999; 14:1061 – 6. Mason DJ, Suva LJ, Genever PG, Patton AJ, Steuckle S, Hillam RA, et al. Mechanically regulated expression of a neural glutamate transporter in bone: a role for excitatory amino acids as osteotropic agents? Bone 1997;20:199 – 205. Max SR, Thomas JW, Banner C, Vitkovic L, Konagaya M, Konagaya Y. Glucocorticoid receptor-mediated induction of glutamine synthetase in skeletal muscle cells in vitro. Endocrinology 1987;120: 1179 – 83. Nakashima T, Sasaki H, Tsuboi M, Kawakami A, Fujiyama K, Kiriyama T, et al. Inhibitory effect of glucocorticoid of osteoblast apoptosis induced by activated peripheral blood mononuclear cells. Endocrinology 1998;139:2032 – 40. Norman AW, Litwack G. Adrenal corticoids. Hormones. Second ed. San Diego, CA: Academic Press; 1997. p. 281 – 318. O’Farrell PH. High resolution two-dimensional electrophoresis of proteins. J Biol Chem 1975;250:4007 – 21. Patejunas G, Young AP. Developmentally regulated primary glucocorticoid hormone induction of chick retinal glutamine synthetase mRNA. J Cell Biochem 1987;35:205 – 16. Patschan D, Loddenkemper K, Buttgereit F. Molecular mechanisms of glucocorticoid-induced osteoporosis. Bone 2001;29:498 – 505. Patton AJ, Genever PG, Birch MA, Suva LJ, Skerry TM. Expression of an N-methyl-D-aspartate-type receptor by human and rat osteoblasts and osteoclasts suggests a novel glutamate signaling pathway in bone. Bone 1998;22:645 – 9. Pereira RMR, Delany AM, Canalis E. Cortisol inhibits the differentiation and apoptosis of osteoblasts in culture. Bone 2001;28: 484 – 90. Pirskanen A, Mahonen A, Ma¨enpa¨a¨ PH. Modulation of 1,25(OH)2D3induced osteocalcin synthesis in human osteosarcoma cells by other steroidal hormones. Mol Cell Endocrinol 1991;76:149 – 59. Rauen T, Wiessner M. Fine tuning of glutamate uptake and degradation in glial cells: common transcriptional regulation of GLAST1 and GS. Neurochem Int 2000;37:179 – 89. Reid IR. Glucocorticoid-induced osteoporosis. Baillieres Best Pract Res Clin Endocrinol Metab 2000;14:279 – 98. Ryha¨nen S, Pirskanen A, Ja¨a¨skela¨inen T, Ma¨enpa¨a¨ PH. State of methylation of the human osteocalcin gene in bone-derived and other types of cells. J Cell Biochem 1997;66:404 – 12. Saini K, Thomas P, Bhandari B. Hormonal regulation of stability of glutamine synthetase mRNA in cultured 3T3-L1 adipocytes. Biochem J 1990;267:241 – 4. Santoro JC, Harris G, Sitlani A. Colorimetric detection of glutamine

A. Olkku et al. / Bone 34 (2004) 320–329

[47] [48] [49]

[50]

[51]

synthetase-catalyzed transferase activity in glucocorticoid-treated skeletal muscle cells. Anal Biochem 2001;289:18 – 25. Skerry TM, Genever PG. Glutamate signalling in non-neuronal tissues. Trends Pharmacol Sci 2001;22:174 – 81. Skerry TM, Taylor AF. Glutamate signalling in bone. Curr Pharm Des 2001;7:737 – 50. Vardimon L, Fox LE, Moscona AA. Developmental regulation of glutamine synthetase and carbonic anhydrase II in neural retina. Proc Natl Acad Sci U S A 1986;83:9060 – 4. Walsh S, Jordan GR, Jefferis C, Stewart K, Beresford JN. High concentrations of dexamethasone suppress the proliferation but not the differentiation or further maturation of human osteoblast precursors in vitro: relevance to glucocorticoid-induced osteoporosis. Rheumatology 2001;40:74 – 83. Wang Y, Kudoh J, Kubota R, Asakawa S, Minoshima S, Shimizu N.

329

Chromosomal mapping of a family of human glutamine synthetase genes: functional gene (GLUL) on 1q25, pseudogene (GLULP) on 9p13, and three related genes (GLULL1, GLULL2, GLULL3) on 5q33, 11p15, and 11q24. Genomics 1996;37:195 – 9. [52] Weinstein RS, Jilka RL, Parfitt AM, Manolagas SC. Inhibition of osteoblastogenesis and promotion of apoptosis of osteoblasts and osteocytes by glucocorticoids. Potential mechanisms of their deleterious effects on bone. J Clin Invest 1998;102:274 – 82. [53] Wong MM, Rao LG, Ly H, Hamilton L, Tong J, Sturtridge W, et al. Long-term effects of physiologic concentrations of dexamethasone on human bone-derived cells. J Bone Miner Res 1990;5:803 – 13. [54] Zhang HY, Young AP. A single upstream glucocorticoid response element juxtaposed to an AP1/ATF/CRE-like site renders the chicken glutamine synthetase gene hormonally inducible in transfected retina. J Biol Chem 1991;266:24332 – 8.