Gene expression profiling of glucocorticoid ... - Semantic Scholar

175

Gene expression profiling of glucocorticoid-inhibited osteoblasts Nathalie Leclerc1, Cynthia A Luppen1, Vincent V Ho1, Shailender Nagpal1, Joseph G Hacia1, Elisheva Smith2 and Baruch Frenkel1,2 1Department

of Biochemistry and Molecular Biology, Institute for Genetic Medicine, Keck School of Medicine at the University of Southern California, 2250 Alcazar Street, Los Angeles, California 90033, USA

2Department

of Orthopedic Surgery, Institute for Genetic Medicine, Keck School of Medicine at the University of Southern California, 2250 Alcazar Street, Los Angeles, California 90033, USA

(Requests for offprints should be addressed to Baruch Frenkel; Email: [email protected])

Abstract Glucocorticoid (GC) treatment for the management of autoimmune and inflammatory diseases is associated with decreased bone formation and increased risk for fracture. In MC3T3-E1 cell cultures, 0·1 – 1 µM dexamethasone (DEX) arrests development of the osteoblast phenotype when administration commences at a commitment stage around the time of confluency. To gain new insights into GC-induced osteoporosis, we performed microarray-based gene expression analysis of GC-arrested MC3T3-E1 cultures, 2·5 days after the administration of DEX. Of the .12 000 transcripts interrogated, 74 were up-regulated and 17 were down-regulated by at least 2·5-fold (P¯0·05). Some of these genes, such as Mmp13, Serum/GC-regulated kinase and Tieg, have previously been reported as GC-responsive. Others are shown here for the first time to respond to GCs. DEX strongly repressed Krox20/Egr2 at both the mRNA and the protein level. This is especially significant because mice lacking this transcription factor develop osteoporosis. The data also suggest that the bone morphogenetic protein (BMP) pathway, which is involved in regulating bone mass, and other pathways that influence BMP signaling, are abrogated by GCs: (i) DEX increased the mRNA levels of the BMP antagonists Follistatin and Dan; (ii) DEX increased the levels of p21 Rasgap3 and Ptpn16/MKP-1 mRNAs, negative regulators of the MAP kinase pathway; and (iii) DEX decreased Cox mRNA levels. DEX also increased thrombospondin mRNA levels, which negatively regulate bone mass in vivo, as well as the adipocytic marker Fkbp51. These and other observations disclose novel gene targets, whose regulation by GCs in osteoblasts may shed light on and provide new therapeutic approaches to osteoporosis. Journal of Molecular Endocrinology (2004) 33, 175–193

Introduction Glucocorticoids (GCs) are used extensively for the treatment of autoimmune and inflammatory diseases, including arthritis, asthma, multiple sclerosis, systemic lupus erythematosus, inflammatory bowel disease and chronic active hepatitis. In addition, GCs are used in combination with other drugs to reduce inflammation associated with hematologic and other cancers, and to suppress the immune system following organ transplantation. Perhaps the most disturbing side effect of GC treatment is rapid bone loss and increased risk of fracture (Van Staa et al. 2000). Although bone Journal of Molecular Endocrinology (2004) 33, 175–193 0952–5041/04/033–175 © 2004 Society for Endocrinology

resorption has been suggested to contribute to GC-induced osteoporosis (Luckert & Kream 1996, Luckert 1997), it is now clear that the main mechanism underlying long-term GC-induced bone loss is the impairment of osteoblast function and bone formation (Canalis 1996, Manolagas & Weinstein 1999, Rubin & Bilezikian 2002). Regarding their adverse effects on osteoblasts, GCs have been reported to attenuate cell proliferation and differentiation, promote apoptosis, abrogate collagen metabolism, interfere with the autocrine/paracrine action of insulin-like growth factor-I and bone morphogenetic protein-2 (BMP-2), and inhibit the bone-specific transcription Online version via http://www.endocrinology.org

Printed in Great Britain

176

N LECLERC

and others ·

Microarray analysis of DEX-treated osteoblasts

factor Runx2 (Chen et al. 1983, Delany et al. 1995a, Linkhart et al. 1996, Lian et al. 1997, Rogatsky et al. 1997, Chang et al. 1998, Weinstein et al. 1998, Luppen et al. 2003b). While some of these effects, such as the induction of apoptosis, were demonstrated in vivo, most of the studies were performed in vitro, alas under conditions that do not fully recapitulate the inhibitory action of GCs on the ultimate osteoblast phenotype – deposition of mineralized extracellular matrix. In fact, in many in vitro studies, GCs surprisingly stimulated osteoblast proliferation and function (Tenenbaum & Heersche 1985, Bellows et al. 1987, Pockwinse et al. 1995). Inhibition of mineralized extracellular matrix formation by GCs was recapitulated in vitro using the MC3T3-E1 cell line and occurred only when GC treatment commenced prior to the seventh day of culture (Lian et al. 1997). In our quest for molecular mechanisms underlying GC-induced osteoporosis, we continued to employ the MC3T3-E1 culture model and identified a narrow developmental window during which cells are susceptible to the adverse effect of GCs. This window encompasses a commitment stage, occurring around the time of confluency, during which the cells undergo a uniquely regulated post-confluent cell cycle. Unlike cell cycle progression prior to confluency, the differentiationrelated cell cycle is abrogated by GCs, concomitant with the inhibition of mineralization. At the commitment stage, GCs activate the enzyme glycogen synthase kinase 3
, a key negative regulator of the Wnt and other signaling pathways (Smith et al. 2002). In order to gain further insights into molecular events associated with the adverse effects of GCs, we employed oligonucleotide microarrays to compare gene expression profiles between MC3T3-E1 cultures that have just undergone commitment and those arrested by DEX. Here we discuss the potential significance of the effects of DEX on subsets of these genes which encode transcription factors, secreted proteins, receptors, signaling molecules, and metabolic enzymes.

Materials and methods Cell culture

MC3T3-E1 cells were plated in 100-mm dishes (300 000 cells/plate) for all experiments with the Journal of Molecular Endocrinology (2004) 33, 175–193

exception of those shown in Fig. 1, where cells were plated in 12-well plates (25 000 cells/well). Cultures were maintained in -MEM (Invitrogen Corp., Carlsbad, CA, USA) containing 10% FBS (Invitrogen Corp.) and 1·5% penicillinstreptomycin (Invitrogen Corp.). To support differentiation, 50 µg/ml ascorbic acid (Sigma) and 10 mM
-glycerophosphate (Sigma) were added to confluent cultures and the medium was then changed every other day. Calcium deposition

Calcium deposition was demonstrated by alizarin red staining or quantitated using the Sigma Diagnostics Kit 587 as previously described (Luppen et al. 2003a,b). Apoptosis

Apoptosis was evaluated using either acridine orange staining as previously described (Mandavilli & Rao 1996) or the Apopercentage dye (Biocolor, http://www.bicolor.co.uk). RNA preparation and cRNA array hybridization

Total RNA was isolated from control and DEX-treated cells using Trizol reagent (Invitrogen Corp.) according to the manufacturer’s instructions, and further cleaned by two successive phenol/chloroform extractions, followed by selective binding to a silica-gel-based membrane using RNeasy spin columns (Qiagen). Double-stranded cDNA was synthesized from 10 µg total RNA using an oligo-dT primer with a 5 -T7 RNA polymerase promoter sequence and the SuperScript II cDNA synthesis kit (LifeTechnologies, Gaithersburg, MD, USA). The cDNA was phenol/chloroform extracted and recovered by ethanol precipitation. Three micrograms double-stranded cDNA were used for the synthesis of biotin-labeled cRNA (Enzo Eukaryotic Gene Expression Labeling Kit; Affymetrix Inc., Sant Clara, CA, USA), which was hybridized to Affymetrix murine MG-U74Av2 gene chips (Affymetrix Inc., Santa Clara, CA, USA). Unbound cRNA was washed, and hybridized cRNA was labeled with a phycoerythrinstreptavidin conjugate. Fluorescent hybridization signals were detected using a Hewlett Packard GeneArray Scanner. www.endocrinology.org

Microarray analysis of DEX-treated osteoblasts ·

Microarray data analysis

Raw gene expression scores were generated for every gene in every sample using GENECHIP Software v5·0 (Affymetrix Inc.) with the default setups. To normalize the relative hybridization signal for each transcript, multiplicative scaling factors were calculated based on the median intensity of the 60th to 95th percentile of gene expression scores. All scores were set to a minimum value of 100 to minimize noise associated with rare transcripts (Karaman et al. 2003). The averages of normalized data from three control and three DEX-treated samples were calculated, and only differences of at least 2·5-fold with P^0·05 (equal variance two-tail t-test) are reported herein. Reverse transcription and polymerase chain reaction (RT-PCR)

The double-stranded cDNA from above was subjected to gene-specific PCR amplification to confirm expression results for selected genes. The cDNA was PCR-amplified (MJ Research Inc., Waltham, MA, USA) in a total volume of 25 µl containing 5 pmoles of each primer, 1 µCi [-32P]dCTP (Perkin Elmer Life Sciences Inc., Boston, MA, USA), 0·25 nmoles unlabeled dCTP, 1·25 nmoles each of dATP, dTTP, and dGTP, and 0·5 units Taq polymerase (Invitrogen Corp.). For each of the analyzed genes, preliminary experiments were performed with increasing amounts of input cDNA to achieve close-to-linear conditions of signal/input ratio, and equal cDNA input for all samples was based on amplification of ribosomal protein L10A (rpL10A) cDNA, which itself varied by less than 25% between samples. The PCR primers are listed in Table 1. The amplified PCR products were resolved by electrophoresis in a 5% polyacrylamide gel. Following drying, the gels were exposed to a storage phosphor screen (Molecular Dynamics, Sunnyvale, CA, USA) and the signal associated with the products was measured using Storm 840 Phosphor Analyzer and Image Quant Software (Molecular Dynamics). Western blot analysis

Cells were washed and collected in phosphatebuffered saline, centrifuged, and pellets were resuspended in 1·5 packed cell volume of lysis buffer (100 mM Hepes pH 7·5, 500 mM KCl, www.endocrinology.org

N LECLERC

and others 177

5 mM MgCl2, 0·5 mM EDTA, 28% glycerol) containing protease and phosphatase inhibitors (5 mM NaF, 0·1 mM Na3VO4, 5 µg/ml aprotinin, 5 µg/ml leupeptin, 1 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride, 20 µM MG132). The cells were further subjected to successive passes through 18·5-, 20·5-, and 23-gauge needles, followed by centrifugation at 16 000 r.p.m. for 30 min at 4 C to remove cell debris. The protein concentration in the cell extract was determined using the Micro BCA Protein Assay (Pierce, Rockford, IL, USA) and 30 µg were subjected to electrophoresis in a 10% polyacrylamide denaturating gel. The proteins were then transferred onto a nitrocellulose membrane and Krox20 was detected using specific antibodies (1:150 dilution) from Covance (Richmond, CA, USA) and the enhanced chemoluminescence system (Amersham).

Results Dexamethasone specifically inhibits mineralization when administration commences at a commitment stage associated with confluency

We utilized the MC3T3-E1 cell culture model, in which GCs severely inhibit development of the osteoblast phenotype (Smith et al. 2000). As shown by biochemical determination of calcium deposition, DEX inhibited mineralization in a manner strongly dependent on both the concentration and the timing of DEX administration (Fig. 1A). First, consistent with previous reports (reviewed by Ishida & Heersche 1998), DEX was inhibitory only at the pharmacological concentrations of 100 and 1000 nM, whereas 10 nM was stimulatory. Secondly, DEX at 100 and 1000 nM had a maximal inhibitory effect when administered commencing on day 3, as cells became confluent. Consistent with our previous work (Smith et al. 2000), little or no inhibition was observed when DEX was administered after the cells had committed to mineralization, around day 4. Most remarkably, exposure of the cultures to 100 or 1000 nM DEX commencing at early stages – prior to confluency – rendered them resistant to the later inhibitory potential of the drug at the same doses (Fig. 1A). Moreover, the 100 nM dose, which was inhibitory when treatment commenced on days 2–4, was stimulatory when treatment commenced Journal of Molecular Endocrinology (2004) 33, 175–193

M34141 M24377 AW122933 U16959 Z29532 X59846 M96827 X66473 X66976 AI843580 U12403 AF064088 M62470 L07803

Cox-1

Krox20 (early growth response 2)

Enpp2

FK506 binding protein 5 (51 kDa)

Follistatin

Growth arrest specific 6

Haptoglobin

Matrix metalloproteinase 13

Procollagen, type VIII, alpha 1

RhoGAP8 (expressed sequence C78947)

Ribosomal protein L10A

TGF
inducible early growth response

Thrombospondin 1

Thrombospondin 2

58CAGTACAGCACACAGCATCC 58CTTTCCCACACGTGAGATATC 58CTCGACAACTACCAGTGTG 58CAGAAGCTGAACATCTGG 58ATCACAGGCAGGAGAGAGTC 58AGGCTGTGGTTGAAGCTG 58AGCTATTCGGAAATTCTGACCCTCA 58GGAGAAAAACAGCACAGTCATTCAG 58ACATTTGATTGCCGAGATG 58ACTCTCGGAGCTTCAGGTAG 58TCCACTTGTGTGGTGGATCAGACCA 58CAAAGGCTATGTCAACACTGAACAT 58AGAGGTGTTCGAGAACGAC 58GCAGAGGCAAGAGTAGGAG 58GAGTCATGCCTATCTGCCTGCCTTC 58ATCCCAGCTGCGTACCAGGTGTCCT 58GATGACCTGTCTGAGGAAG 58ATCAGACCAGACCTTGAAG 58CTCTACAGCTGCTGGGAATAC 58TTCCAGGCATTCCATGAC 58AGGTCTGACTGTGGCTTG 58GTCACTTTATACTCCCAGTCTG 58CGCCGCAAGTTTCTGGAGAC 58CTTGCCAGCCTTGTTTAGGC 58CGTCTAGTGTCTCAGTGCTC 58CTCTTTGAAAGGAGTGGC 58CCTCATTTGTTGTGTGACTGAGTAA 58TTGTTGTTCCTTGTACATAAGAAAC 58GAAGAATATTTTGTCCCTCTGGTGG 58TCCACAAAAATCTAAAAAGTGCAGG

Journal of Molecular Endocrinology (2004) 33, 175–193

58

58

55

61

60

55

54

58

55

58

55

58

65

54

55

Annealing temp (°C)

30

30

29

29

40

29

29

33

29

30

29

30

28

30

29

No. of cycles

201

566

490

327

254

561

357

340

442

543

460

245

554

378

402

Product size (base pairs)

and others ·

Primers usually anneal with two adjacent exons, to confirm lack of genomic DNA contamination. However, the primer pairs for Enpp2, Follistatin, Haptoglobin, and Thrombospondins 1 and 2 targeted the 38UTR sequences present on the microarray chip.

U21301

Gene C-mer proto-oncogene

Primers (Top-forward; bottom: reverse)

N LECLERC

Accession number

Table 1 Primers and conditions used for RT-PCR

178 Microarray analysis of DEX-treated osteoblasts

www.endocrinology.org

Microarray analysis of DEX-treated osteoblasts ·

Figure 1 DEX inhibition of calcium deposition is dependent on the dose and the time of treatment commencement. MC3T3-E1 cells were plated (day 0) in 12-well plates and treated with 10, 100, or 1000 nM DEX commencing on the indicated days. (A) Calcium deposition was biochemically quantitated on day 32. (B) Calcium deposition is demonstrated in day-13 cultures by alizarin red staining. In panel B, [DEX]=1000 nM.

on day 0 or day 1 (Fig. 1A). Cultures treated with 1000 nM DEX commencing on day 1 were less inhibited than those treated from day 2 or day 3. Figure 1B presents an independent experiment with 1000 nM DEX, demonstrating similar results of calcium deposition by alizarin red staining. In order to identify genes that link the GC inhibitory activity at the commitment stage to the abrogation of the osteoblast phenotype, we treated MC3T3-E1 cultures with 1000 nM DEX commencing on day 3, and performed a comprehensive gene expression analysis after 2·5 days. This experimental design provides a comparison between steady-state mRNA levels in osteoblasts undergoing normal, uninterrupted differentiation and those not allowed to undergo commitment. The morphological difference between differentiating and arrested cultures is demonstrated in Fig. 2A–B. Alizarin red staining of day-14 cultures, www.endocrinology.org

N LECLERC

and others 179

plated and treated in parallel to those used for the microarray analysis, is shown in Fig. 2C-F. Interestingly, the DEX-induced arrest was not associated with increased, but rather with decreased, apoptosis (Fig. 2G-J). In vitro inhibition of osteoblast apoptosis has been demonstrated by several investigators, although promotion of apoptosis in vivo and in vitro has been reported by others (reviewed by Zalavras et al. 2003). The total RNA collected from the day-6 cultures was interrogated by MG-U74Av2 gene chips (Affymetrix Inc.). Of the over 12 000 transcripts represented on the microarray, expression of the vast majority (99%) did not change by more than 2·5-fold (P^0·05) in response to DEX. In contrast, 74 genes were up-regulated and 17 were down-regulated by d2·5-fold (P^0·05). The expression levels of the 91 DEX-responsive genes, including 82 with known functions and 9 expressed sequence tags (ESTs), as well as the unedited list of raw data, are posted on our web site (http:// hacialab.usc.edu/supplement/frenkel, Tables S1, S2). The potential significance of some of these findings is discussed below. Dexamethasone-responsive genes encoding transcription factors

DEX treatment altered by d2·5-fold the expression of zinc finger transcription factors (Krox20, Kruppel-like factor 13, transforming growth factor (TGF)-
inducible early growth response), basic leucine-zipper DNA-binding proteins (CAAT/ enhancer binding protein  and
, nuclear factor I/B), and a basic helix-loop-helix transcription factor (Drosophila period homolog) (Table 2). Krox20 (Early growth response 2, Egr2)

The Egr genes (Krox for the murine homologs) are mitogen-inducible, immediate early genes encoding transcription factors with three zinc fingers. Egr proteins control cell proliferation and differentiation in a variety of tissues including the brain and peripheral nerves, the vasculature, the kidney and the skeleton (McMahon et al. 1990, Sukhatme 1992, Beckmann & Wilce 1997, Khachigian & Collins 1998). Although Krox20/Egr2 is expressed mainly in the adult thymus and in the embryonic central and peripheral nervous systems (Wilkinson et al. 1989, Topilko et al. 1994), the gene is also Journal of Molecular Endocrinology (2004) 33, 175–193

180

N LECLERC

and others ·

Microarray analysis of DEX-treated osteoblasts

expressed in differentiating osteoblasts (Levi et al. 1996). Krox20/Egr2 null mice have a severe osteoporotic phenotype, attributable to reduced bone formation (Levi et al. 1996). In our microarray analysis, Krox20/Egr2 mRNA levels were repressed

5·8-fold by DEX (Table 2) and this result was confirmed by RT-PCR (Fig. 3A). Furthermore, as demonstrated by Western analysis, DEX strongly reduced Krox20 protein levels in MC3T3-E1 cultures (Fig. 3B). Thus, decreased Krox20/Egr2 in response to GCs may contribute to bone loss by abrogating the transcriptional control of Krox20/ Egr2 target genes. Such targets may include the Biglycan,
1(I) and
2(I) Collagen genes, encoding components of the bone extracellular matrix, which have been shown to respond to another Krox family member, c-Krox (Galera et al. 1996, Heegaard et al. 1997). In addition, the decrease in Krox20/Egr2 may be involved in the DEXinduced up-regulation of Follistatin (Table 3, Fig. 4 and see below), which is negatively regulated by Krox20/Egr2 (Seitanidou et al. 1997). TGF- inducible early growth response (Tieg, Kruppel-like factor 10) and Kruppel-like factor 13 (Klf13)

DEX-treated MC3T3-E1 cells had 5·8-fold less Tieg mRNA compared with control cultures (Table 2) and the GC-mediated repression was confirmed by RT-PCR (Fig. 4). Tieg encodes a member of the Sp1/Kruppel-like transcription factor family. Tieg was originally identified in human osteoblasts as a transcript that was rapidly induced by TGF-
(Subramaniam et al. 1995) and was later shown to be stimulated, albeit to a lesser extent, by other TGF-
family members including activin and BMPs (Hefferan et al. 2000). Thus, the decreased Tieg mRNA levels may result from abrogated BMP signaling in the DEX-treated cultures as suggested by our previous (Luppen et al. 2003b) and current studies (see below). In contrast to Tieg, another Sp1/Klf mRNA encoding Klf13

Figure 2 The effects of DEX on cell morphology and apoptosis. MC3T3-E1 cells were plated (day 0) in 100-mm plates and treated with 1000 nM DEX commencing on day 3. (A, B) Brightfield micrographs of day-5 cultures (×200). (C, D) Micrographs of alizarin red-stained cultures on day 14 (×200). (E, F) Scans of the same stained plates. (G, H) Acridine orange staining of day-14 cultures. Bright nuclei denote cells undergoing apoptosis. (I, J) Apopercentage staining of cells undergoing apoptosis on day 7; reproduced with permission from CRC Press (Zalavras et al. 2003). Several stained cells are indicated by arrows. This method is based on the flip-flopping of plasma membrane phosphatidylserine (Biocolor). Journal of Molecular Endocrinology (2004) 33, 175–193

www.endocrinology.org

Microarray analysis of DEX-treated osteoblasts ·

N LECLERC

and others 181

Table 2 Transcription factor genes altered by DEX. Data are means± S.E.M. (n=3). Only genes altered by ≥2·5-fold are listed

Gene Period homolog (Drosophila) CCAAT/enhancer binding protein (C/EBP),  Nuclear factor I/B Zinc finger protein 99 Kruppel-like factor 13 (Klf13) CCAAT/enhancer binding protein (C/EBP),
TGF-
inducible early growth response (Tieg) Early growth response 2 (Egr2/Krox20)

Affymetrix identification

Control

DEX

AF022992 X61800

93619_at 160894_at

236±17 1101±57

1116±50 4881±239

4·7 4·4

Y07685 U62906 AW125783 M61007

160859_s_at 104010_at 160617_at 92925_at

619±29 116±16 912±137 2136±203

2199±83 389±45 2758±187 6083±237

3·6 3·3 3·0 2·8

AF064088

99603_g_at

703±144

120±21

−5·8

M24377

102661_at

1028±189

177±38

−5·8

was up-regulated 3·0-fold by DEX (Table 2). Klf13 is a positive regulator of erythropoiesis (Asano et al. 2000), and is also expressed in the skeleton during mouse development (Martin et al. 2001). The role of Klf13 in osteoblasts, and its possible involvement in the effects of GCs in these cells, remains to be explored. Dexamethasone-responsive genes encoding cell surface and secreted proteins Follistatin and Dan (Differential screening-selected gene aberrant in neuroblastoma)

DEX stimulated the expression of two genes encoding secreted BMP antagonists. Follistatin

Figure 3 DEX represses the transcription factor Krox20/Egr2 at the mRNA and protein levels. MC3T3-E1 cells were plated (day 0) in 100-mm plates and treated with 1000 nM DEX commencing on day 3. (A) RT-PCR analysis of Krox20/Egr2 mRNA in representative cultures treated for 60 h with either DEX or vehicle. Ribosomal protein L10A (rpL10A) serves as control. (B) Western blot analysis of Krox20 in representative cultures treated for 72 h. The top panel shows the 56 kDa band representing Krox20 and the bottom panel shows two non-specific bands of 63 and 68 kDa demonstrating equal loading. www.endocrinology.org

Fold induction/ repression by DEX

Accession no.

mRNA level was increased by 3·1-fold (Table 3) and this increase, possibly related to the suppression of Krox20/Egr2 (see above), was confirmed by RT-PCR (Fig. 4). The mRNA for Dan (aka neuroblastoma suppressor of tumorigenicity 1) was increased by 3·3-fold (Table 3). Since both follistatin and Dan bind to and inactivate BMPs (reviewed by Canalis et al. 2003), and since BMPs are required for osteoblast differentiation in vitro and control bone mass in vivo (Yamaguchi et al. 1991, Sampath et al. 1992, Yamaguchi et al. 2000, Xiao et al. 2002, Styrkarsdottir et al. 2003, Yang et al. 2003), the stimulation of Follistatin and Dan mRNAs may contribute to GC-induced osteoporosis. We have recently shown that GCinhibition of MC3T3-E1 osteoblast differentiation was also associated with a 100-fold decrease in Bmp-2 gene expression, and that addition of recombinant BMP-2 rescued the osteoblast phenotype (Luppen et al. 2003a,b). The GC-mediated repression of Bmp-2 was not disclosed by the microarray analysis because the overall level of its mRNA was close to the background noise level of the microarray assay. The relationship between the down-regulation of Bmp-2 and the up-regulation of Follistatin and Dan remains to be explored. Thrombospondins

DEX increased the mRNA levels for Thrombospondin 1 and 2 by 5·3-fold and 2·9-fold respectively (Table 3), and these results were confirmed by RT-PCR (Fig. 4). Thrombospondins are non-collagenous Journal of Molecular Endocrinology (2004) 33, 175–193

182

N LECLERC

and others ·

Microarray analysis of DEX-treated osteoblasts

Table 3 Genes for cell surface and secreted proteins altered by DEX. Data are means± S.E.M. (n=3). Only genes altered by ≥2·5-fold are listed Accession Affymetrix no. identification Control Gene Haptoglobin M96827 Integrin binding sialoprotein (Bsp) L20232 Ectonucleotide pyrophosphatase/ AW122933 phosphodiesterase 2 (Enpp2) Collagen, type VIII, alpha 1 X66976 Thrombospondin 1 M62470 Matrix metalloproteinase 13 (collagenase 3) X66473 Cysteine-rich protein 61 (Cyr61) M32490 Neuroblastoma, suppression of D50263 tumorigenicity 1 (Dan) A disintegrin and metalloproteinase D50411 domain 12 (Adam12) Follistatin Z29532 Thrombospondin 2 L07803 Collagen, type XI, alpha 1 D38162 Chondroitin sulfate proteoglycan 2 D45889

secreted glycoproteins that associate with the extracellular matrix, but do not appear to contribute directly to tissue integrity. Rather, the proteins influence cell functions such as adhesion,

Figure 4 RT-PCR analysis. Expression level of the indicated genes was quantitated in day-6 MC3T3-E1 cultures by RT-PCR using primers listed in Table 1. Shown are results from a representative DEX-treated and a representative control culture. The arrow indicates the Tieg cDNA amplicon. Ribosomal protein L10A (rpL10A) amplification serves as internal control. Journal of Molecular Endocrinology (2004) 33, 175–193

Fold induction/ repression by DEX

DEX

96092_at 100476_at 97317_at

272±81 1143±838 205±69

2180±327 8410±809 1408±90

8·0 7·4 6·9

100308_at 160469_at 100484_at 92777_at 101969_at

1161±256 6909±471 915±121 4881±185 2241±1152 11601±558 942±102 3325±582 486±61 1584±14

6·0 5·3 5·2 3·5 3·3

92414_at

140±31

445±48

3·2

98817_at 94930_at 100481_at 100019_at

938±428 1203±364 645±106 708±106

2929±258 3500±85 1853±331 1758±165

3·1 2·9 2·9 2·5

motility, proliferation, apoptosis, differentiation, cytoskeletal organization and transcriptional regulation by modulating the assembly of cell surface multi-protein complexes. The two Thrombospondin genes are expressed with different developmental and spatial patterns (Iruela-Arispe et al. 1993, Tooney et al. 1998). Thrombospondin 1 is widely expressed and is involved in cytoskeletal organization during tissue repair, possibly mediating some of the effects of TGF-
1 (reviewed by Chen et al. 2000). Thrombospondin 2 has a more restricted expression pattern and is synthesized primarily in connective tissues (reviewed by Bornstein et al. 2000). Both thrombospondin 1 and 2 are found in sites of new bone formation and both are produced by cultured osteoblasts (Robey et al. 1989, Carron et al. 1999). Although Thrombospondin 1 mRNA level was shown to increase during MC3T3-E1 culture progression (Sherbina & Bornstein 1992), thrombospondins seem to negatively regulate osteogenesis, based on the increase in osteoblast number, endosteal bone formation and cortical bone thickness observed in Thrombospondin 2 knock-out mice (Hankenson & Bornstein 2002). Therefore, the GC-induced Thrombospondin mRNA levels may contribute to bone loss in vivo. Furthermore, because both thrombospondin 1 and 2 inhibit angiogenesis (reviewed by Bornstein 2000 www.endocrinology.org

Microarray analysis of DEX-treated osteoblasts ·

and by Chen et al. 2000), their stimulation by GCs may compromise the perfusion to bone and other organs, and contribute to GC-induced apoptosis in vivo. Collagens

Ninety percent of the organic bone matrix is collagen, mostly type I. GCs are known to inhibit the expression of type I collagen genes in several cell types (reviewed by Kucharz 1988), including osteoblasts (Choe et al. 1978, Delany et al. 1995a, Mahonen et al. 1998, Luppen et al. 2003b). In the present study, the genes encoding the
1(I) and the
2(I) Collagen chains were inhibited by 1·9fold and 1·7-fold respectively (Table S2; http:// hacialab.usc.edu/supplement/frenkel), missing the 2·5-fold cutoff set for the initial analysis of the microarray data. It is noteworthy that GCs further inhibit osteoblast collagen synthesis at the protein level (Luppen et al. 2003b). In contrast to type I collagen chains, DEX increased the mRNA levels for
1(VIII) and
1(XI) Collagen chains by 5·9-fold and 2·9-fold respectively (Table 3).
1(VIII) Collagen was further tested by RT-PCR analysis, which confirmed the microarray data (Fig. 4). Type VIII collagen, a short-chain network-forming collagen prominently expressed by endothelial cells of the vasculature and the Descemet’s membrane of the cornea, is also found in newborn mouse calvaria and skin (Muragaki et al. 1992). Type XI collagen, a fibril-forming collagen, is also expressed in the cornea but is mainly found in cartilage, where it copolymerizes with type II collagen and regulates chondrocyte differentiation and spatial organization of the growth plate (Li et al. 1995). The potential impact of increased collagens type VIII and XI on the physiological assembly of the bone extracellular matrix remains to be explored. It would also be interesting to test whether GCs induce collagens type VIII and XI in the cornea, because this could provide an insight into GC-induced glaucoma (Schacke et al. 2002). Matrix metalloproteinase 13 (Mmp13, aka collagenase 3)

Collagens are primarily degraded by collagenases, specific proteinases of the matrix metalloproteinases (MMP) family. Among all the MMPs www.endocrinology.org

N LECLERC

and others 183

represented on the microarray (Table S2; http:// hacialab.usc.edu/supplement/frenkel), DEX altered the expression of only one gene, Mmp13, which was increased by 5·2-fold (Table 3). This increase was confirmed by RT-PCR (Fig. 4) and is consistent with previous reports (Shalhoub et al. 1992, Delany et al. 1995b). Induction of Mmp13 mRNA may contribute to GC-induced osteoporosis by interfering with collagen type I homeostasis and should also be considered in the context of GC treatment of arthritis patients, given the high affinity of Mmp13 for type II collagen (Knauper et al. 1996). A disintegrin and metalloprotease domain 12 (Adam 12)

Adam12 gene expression was up-regulated 3·2-fold by DEX (Table 3). Adams are membraneanchored cell surface glycoproteins involved in cell fusion, cell adhesion, proteolysis and subsequent signaling. In MC3T3-E1 cultures, Adam12 is highly expressed prior to confluency and is downregulated immediately thereafter and throughout differentiation (Inoue et al. 1998). It remains to be seen whether persistence of Adam12 mRNA contributes to GC-inhibition of osteoblast differentiation. Ectonucleotide pyrophosphatase/phosphodiesterase 2 (Enpp2)

Enpps exist as both membrane and soluble proteins, and are expressed in cells from most tissues, including osteoblasts (Goding et al. 1998). They hydrolyze phosphodiester bonds in nucleoside triphosphates, leading to the generation of inorganic pyrophosphate, a physiologic calcification inhibitor (Bollen et al. 2000, Terkeltaub 2001, Johnson et al. 2003). Enpp2 mRNA level was up-regulated several-fold in response to DEX (Table 3, Fig. 4), potentially contributing to the decreased accumulation of calcium (Figs 1 and 2). Besides its role in regulating pyrophosphate levels, Enpp2 is also involved in the synthesis of lysophosphatidic acid (Tokumura et al. 2002), which can promote osteoblast proliferation and survival (Grey et al. 2001, 2002). While opposite actions of DEX apparently override the mitogenic potential of Enpp2, the putative increase in lysophosphatidic acid may help explain the inhibition of apoptosis observed in DEX-treated Journal of Molecular Endocrinology (2004) 33, 175–193

184

N LECLERC

and others ·

Microarray analysis of DEX-treated osteoblasts

MC3T3-E1 (Fig. 2) and other osteoblastic cell cultures (Zalavras et al. 2003). Cysteine-rich protein 61 (Cyr61)

The gene encoding Cyr61 is a growth factorinducible immediate early gene. Its product is an extracellular matrix-associated signaling protein that promotes cell proliferation, migration and adhesion (Kireeva et al. 1996, Yang & Lau 1991). Unexpectedly, our microarray analysis indicated a 3·5-fold DEX-mediated increase in Cyr61 mRNA level (Table 3). Cyr61 is expressed during embryogenesis of the circulatory system and the cartilaginous skeleton (O’Brien & Lau 1992). Cyr61 mRNA has been detected in human osteoblasts (Schutze et al. 1998, Lechner et al. 2000), and is up-regulated by cytokines such as tumor necrosis factor (TNF)- and interleukin (IL)-1
, which are important stimulators of bone resorption (Schutze et al. 1998). Interestingly, Cyr61 was the most up-regulated gene in a microarray analysis of mice fed with cadmium, an in vivo model for cadmiuminduced osteoporosis (Regunathan et al. 2003). It is possible that osteoblast-derived Cyr61 contributes to the increased bone resorption observed in the initial phase of GC-induced osteoporosis. Acute phase proteins

The microarray analysis revealed the stimulation by DEX of five genes encoding proteins with known functions in the immune system: haptoglobin, orosomucoid 1 and 2, the Ia-associated invariant chain of the major histocompatibility complex class II, and sterile alpha motif domain and hydrolase domain-containing protein 1 (Table S2; http://hacialab.usc.edu/supplement/frenkel). Of these, the first three are known as acute phase proteins, and they were induced by 8·0-, 17·8-, and 5·3-fold respectively (Table 3 and Table S2; http://hacialab.usc.edu/supplement/frenkel). Notably, a structural role for orosomucoid 1 in bone was suggested based on its presence in bone extracts (Mbuyi et al. 1982). Haptoglobin was shown to affect bone cells in an organ culture system, probably through induction of prostanoids (Lerner & Frohlander 1992). Specifically, treatment with haptoglobin resulted in higher levels of prostaglandin (PG) E2 and increased bone resorption. The GC-stimulation of osteoblastic Haptoglobin mRNA level was confirmed by RT-PCR (Fig. 4). Journal of Molecular Endocrinology (2004) 33, 175–193

Integrin binding sialoprotein (aka bone sialoprotein, Bsp)

Consistent with previous reports (Ogata et al. 1995, Chen et al. 1996), DEX stimulated Bsp gene expression by 7·4-fold (Table 3). Bsp is a late marker for osteoblast differentiation, and the gene product constitutes approximately 12% of the non-collagenous bone extracellular matrix proteins (reviewed by Ganss et al. 1999). It stimulates osteoblastic cell adhesion and proliferation (Mintz et al. 1993, Zhou et al. 1995) and has been proposed to function in the initiation of hydroxyapatite crystal formation (Hunter & Goldberg 1993). Given the general inhibitory effect of DEX on the osteoblast phenotype, including the downregulation of other late differentiation markers such as osteocalcin (Lian et al. 1997, Luppen et al. 2003b), the stimulation of Bsp mRNA level is somewhat surprising. While it cannot explain the inhibition of the osteoblast phenotype in this study, it may explain the positive effect that GCs have on cultured osteoblasts under different conditions (Tenenbaum & Heersche 1985, Bellows et al. 1987, Leboy et al. 1991, Shalhoub et al. 1992). Dexamethasone-responsive genes involved in signal transduction Protein kinases and phosphatases

Our microarray analysis revealed d2·5-fold alteration in the mRNA levels for nine protein kinases and phosphatases in response to DEX. The data provide evidence for increases in both protein tyrosine kinases and phosphatases, and for a decrease in MAP kinase (MAPK) activity. Protein tyrosine phosphatase (PTP) non-receptor type 16 (Ptpn16)

PTPs are either membrane receptors or nonreceptor proteins. DEX increased the expression of Ptpn16 mRNA level by 4·4-fold (Table 4). Ptpn16 is an immediate early gene that encodes the PTP non-receptor MAPK phosphatase-1 (MKP-1), also called dual specificity phosphatase 1 for its ability to dephosphorylate both phospho-tyrosines and phospho-threonines (Keyse & Emslie 1992). MKP-1 inactivates MAPKs (Alessi et al. 1993), and GCs have previously been shown to inhibit MAPKs by elevating Mkp-1 mRNA level www.endocrinology.org

Microarray analysis of DEX-treated osteoblasts ·

N LECLERC

and others 185

Table 4 Genes for signaling molecules altered by DEX. Data are means± S.E.M. (n=3). Only genes altered by ≥2·5-fold are listed

Gene C-mer proto-oncogene Growth arrest specific 6 (Gas6) Cdc42 guanine nucleotide exchange factor 9-like (EST C79050) Protein tyrosine phosphatase, non-receptor type 16 (Ptpn16) Leukemia inhibitory factor receptor Rho GTPase-activating protein 8 (expressed sequence C78947) Fibroblast growth factor receptor 2 p21 Ras GTPase-activating protein 3 (p21 Rasgap3) Aplysia ras-related homolog B (RhoB) Portein tyrosine phosphatase, receptor type, D (PtpD) Rho guanine nucleotide exhange factor 5 (Rhogef5)

Affymetrix identification

Control

DEX

U21301 X59846 AI647003

92698_at 99067_at 100958_at

208±55 1639±173 342±102

1051±83 7978±598 1584±151

5·0 4·9 4·6

X61940

1004598_at

192±16

850±126

4·4

D17444 AI843580

104658_at 103511_at

166±16 100±0

727±42 321±45

4·4 3·2

M63503 U20238

93091_s_at 93319_at

1271±194 714±94

3789±393 1947±184

3·0 2·7

X99963 D13903

101030_at 93896_at

1127±126 151±50

3002±142 387±54

2·7 2·6

AA726063

160977_at

274±40

100±0

in fibroblasts and mast cells (Kassel et al. 2001). Moreover, a recent study linked MKP-1 up-regulation to the GC-inhibition of osteoblast proliferation (Engelbrecht et al. 2003), and administration of the tyrosine phosphatase inhibitor sodium orthovanadate prevented bone loss in GC-treated rats (Hulley et al. 2002). The MAPK pathway plays a role in osteoblast differentiation by potentiating BMP signaling (Suzawa et al. 2002), parathyroid hormone/parathyroid hormonerelated peptide (PTH/PTHrP) signaling (Carpio et al. 2001, Miao et al. 2001), and by phosphorylating and activating the osteoblast master regulator Runx2 (Xiao et al. 2000). Thus, DEX stimulation of Ptpn16 mRNA probably plays an important role in GC-induced osteoporosis. Receptor protein tyrosine phosphatase type D (RptpD)

DEX also increased RptpD mRNA level by 2·6-fold (Table 4). RptpD belongs to the leukocyte antigen related (LAR) family of protein tyrosine phosphatases (Mizuno et al. 1993). Although RptpD expression is largely restricted to the brain, our data demonstrate that it is also expressed in osteoblasts. Known substrates for LAR phosphatases include the insulin receptor and the epidermal growth factor receptor (Hashimoto et al. 1992, Zhang et al. www.endocrinology.org

Fold induction/ repression by DEX

Accession no.

−2·7

1996). Because signals mediated by these receptors contribute to osteoblast proliferation and differentiation (Canalis 1985), increased RptpD expression may adversely affect osteoblast function. C-mer proto-oncogene and Growth arrest specific 6 (Gas6)

Along with the increase in PTP mRNA levels, DEX also stimulated the genes encoding C-mer, a receptor tyrosine kinase, and its ligand, Gas6 (Nagata et al. 1996, Chen et al. 1997). The 5·0-fold stimulation of both genes (Table 4) was supported by RT-PCR (Fig. 4). Gas6 is a member of the vitamin K-dependent family of growth factors, and has mitogenic, survival and adhesive properties in a variety of normal and cancer cell types (Goruppi et al. 1996, Li et al. 1996, Nakano et al. 1996, Yanagita et al. 2001, Funakoshi et al. 2002, Shankar et al. 2003). These effects are probably mediated through interactions with members of the Axl family of kinases, including C-mer (Varnum et al. 1995, McCloskey et al. 1997). Because Gas6/Axl signaling has been shown to inhibit osteogenic differentiation of vascular pericytes (Collett et al. 2003), it is possible that the DEX-induction of Gas6 and C-mer mRNA levels plays a role in GC-induced osteoporosis. Journal of Molecular Endocrinology (2004) 33, 175–193

186

N LECLERC

and others ·

Microarray analysis of DEX-treated osteoblasts

Fibroblast growth factor tyrosine kinase receptor 2 (Fgfr2)

DEX increased the level of Fgfr2 mRNA by 3·0-fold (Table 4). Fgfr1–4 binds fibroblast growth factor (FGF), leading to cell growth, differentiation, migration and chemotaxis (reviewed by Galzie et al. 1997) and the Fgfr2 gene is expressed during osteogenesis in vivo (Eswarakumar et al. 2002). However, increased Fgfr2-mediated signaling could be detrimental to bone through mechanisms similar to those operative in patients with Crouzon, Apert and Pfeiffer syndromes, who have constitutively active mutant forms of Fgfr2 (Mansukhani et al. 2000). Small G proteins and their regulators

Of the 91 genes that responded to DEX by d2·5-fold, 22 are involved in intracellular signaling, and of these, five are small G proteins or their regulators. Small G proteins (also called small GTP binding proteins or small GTPases) are divided into five families: Ras, Rho/Rac/Cdc42, Rab, Sar1/ Arf, and Ran (reviewed by Takai et al. 2001). By transducing upstream signals to downstream effectors, small G proteins serve as molecular switches that regulate processes as diverse as cytoskeletal reorganization, intracellular vesicle trafficking, nucleocytoplasmic transport and gene expression. In the absence of activating signals, small G proteins are found in their inactive, GDP-bound form. Upon stimulation, GDP is exchanged for GTP, a process controlled by specific guanine exchange factors (GEFs), leading to activation of the downstream effector(s). The signal transduction is terminated by the intrinsic GTPase activity of the small G proteins, which is under the control of specific GTPase activating proteins (GAPs). p21 Rasgap3

DEX increased p21 Rasgap3 mRNA level by 2·7-fold (Table 4), which can be expected to inhibit Ras signaling and the downstream MAPK and phosphatidylinositol 3-kinase pathways (reviewed by Takai et al. 2001). As mentioned above, MAPKs contribute to osteoblast differentiation by supporting the actions of BMPs, PTH, PTHrP and Runx2 (Xiao et al. 2000, Carpio et al. 2001, Miao et al. 2001, Suzawa et al. 2002). The DEX-mediated inhibition of the MAPK pathway, both via MKP-1 Journal of Molecular Endocrinology (2004) 33, 175–193

(see above) and the up-regulation of p21 Rasgap3, could severely compromise osteoblast function. RhoB and Rho regulators

DEX increased RhoB mRNA level by 2·7-fold (Table 4). Despite this increase, our results suggest that overall Rho signaling is probably decreased in DEX-treated cells, because the Rhogef5 mRNA level was decreased by 2·7-fold and the Rhogap8 mRNA level was increased by 3·2-fold. The increase in Rhogap8 expression was confirmed by RT-PCR (Fig. 4). Rho belongs to the Rho/Rac/Cdc42 family, which regulates the organization of the actin cytoskeleton at focal adhesions, and in filipodia and lamellipodia. Therefore, the decrease in Rho signaling may be related to the morphological arrest of DEX-treated MC3T3-E1 cultures in the so-called cobblestone stage (Fig. 2B). EST C79050 (Cdc42 gef9-like gene)

Unlike the inhibition of Rhogef5, DEX stimulated the expression of a Cdc42 gef9-like gene by 4·6-fold (Table 4). Thus, whereas Rho and Ras signaling are probably attenuated in DEX-treated cells, Cdc42-dependent pathways might in fact be stimulated. This result is somewhat surprising because another Cdc42 gef gene, Fgd1, is specifically expressed in calcified tissues (Gorski et al. 2000), where it might contribute to bone cell differentiation through activation of c-jun N terminal kinase (Zheng et al. 1996). Dexamethasone-responsive genes encoding metabolic enzymes Prostaglandin-endoperoxide synthase 1 (aka cyclooxygenase 1, Cox1)

The two Cox enzymes, Cox1 and Cox2, are integral membrane proteins which localize primarily to the endoplasmic reticulum, and function in PG biosynthesis. Cox1 is constitutively expressed and is thought to be involved in cell signaling and tissue homeostasis. Cox2 is expressed in a limited number of cell types and is probably responsible for prostanoid biosynthesis in response to mitogens and pro-inflammatory factors (reviewed by Smith et al. 1996). The microarray analysis indicated a 2·8-fold DEX-mediated decrease in Cox1 mRNA level (Table 5), and this result was confirmed by www.endocrinology.org

Microarray analysis of DEX-treated osteoblasts ·

N LECLERC

and others 187

Table 5 Metabolic enzyme genes altered by DEX. Data are means± S.E.M. (n=3). Only genes altered by ≥2·5-fold are listed

Gene Aldehyde dehydrogenase family 1, subfamily A1 Glutamine synthetase Glutamate-cysteine ligase, catalytic subunit Pyruvate dehydrogenase kinase 4 Hydroxysteroid 17-
dehydrogenase 11 Xanthine dehydrogenase Galactokinase Ferredoxin reductase Cytochorome P450, 1b1, benz[a]anthracene inducible Carnitine palmitoyltransferase 1, liver Prostaglandin-endoperoxide synthase 1 (Cox-1) RIKEN cDNA 2610002H19 gene (cytochrome c-1) Platelet-activating factor acethylhydrolase, isobform 1b, 1

Accession no.

Affymetrix identification

M74570

100068_at

U09114 U85414 AJ001418 AA822174 X75129 AV217354 D49920 X78445

94852_at 99649_at 102049_at 102370_at 97950_at 161363_r_at 92754_at 99979_at

AF017175 M34141

5·8

158±29 269±28 718±66 930±40 558±95 210±55 376±50 1128±346

753±164 959±62 2541±132 3181±88 1898±80 595±42 1061±76 2871±315

4·8 3·6 3·5 3·4 3·4 2·8 2·8 2·5

93320_at 95597_at

1018±116 338±31

395±66 123±19

−2·6 −2·8

AV069997

162469_at

4556±310

1508±436

−3·0

U57746

100576_at

828±130

250±36

−3·3

Steroidogenic enzymes

DEX induced the expression of three genes involved in the synthesis of steroid hormones: cytochrome P450 1b1 (2·6-fold), Ferredoxin reductase (2·8-fold), and Hydroxysteroid 17- dehydrogenase 11 (3·4-fold), (Table 5). Cytochrome P450 1b1 (Cyp1B1) is involved in the metabolism of diverse substrates including steroids, fatty acids and xenobiotics. The gene is broadly expressed in bone (Sasano et al. 1997, Eyre et al. 1998), where it is responsible for the local production of estrogens (Nawata et al. 1995). The up-regulation of Cyp1B1 is consistent with the reported increase in its mRNA

100±0

DEX 578±97

RT-PCR (Fig. 4). Cox2 mRNA level was decreased by 2·1-fold, less than our 2·5-fold cutoff (Table S2; http://hacialab.usc.edu/supplement/frenkel). Both Cox1 and Cox2 genes are known to be expressed in osteoblastic cells, in association with osteogenesis (Sato et al. 1997), and PGs are known to contribute to BMP signaling (Chikazu et al. 2002) and bone formation (as well as resorption) (reviewed by Raisz 1999 and by Pilbeam et al. 2002). Thus, down-regulation of Cox genes in osteoblasts could contribute to the GC-mediated inhibition of osteoblast function and bone formation.

www.endocrinology.org

Control

Fold induction/ repression by DEX

and activity in GC-treated cultures of normal human osteoblast-like and osteosarcoma cells (Nawata et al. 1995). Ferredoxin reductase, a component of the mitochondrial electron transport chain, is involved in the first step of steroid synthesis. Hydroxysteroid 17-
dehydrogenase is also expressed in bone cells (Sasano et al. 1997, Eyre et al. 1998), where it converts estrone produced by Cyp1B1 to estradiol. Given the well-established anti-apoptotic properties of estrogens in osteoblasts (Gohel et al. 1999, Manolagas et al. 2002), the putative stimulation of estradiol synthesis may help explain the unexpected inhibition of apoptosis in GC-treated osteoblast cultures (Fig. 2G–J and Zalavras et al. 2003). In addition, GC-induced estradiol synthesis may contribute to the attenuation of bone resorption and the low bone turnover seen after long-term GC treatment. Additional genes altered by DEX FK506 binding protein 5 (aka Fkbp51)

The transcript encoding Fkbp51 was induced 7·0-fold after DEX treatment, and this result was confirmed by RT-PCR (Fig. 4). Fkbp51 is a Journal of Molecular Endocrinology (2004) 33, 175–193

188

N LECLERC

and others ·

Microarray analysis of DEX-treated osteoblasts

member of the immunophilin protein family and is involved in protein folding and trafficking, including the nuclear transport of the GC receptor following ligand binding (Davies et al. 2002). Fkbp51 mRNA is abundant in T cells and thymus, and increases during GC-induced apoptosis of murine thymoma cells (Baughman et al. 1995). The gene product also selectively accumulates during adipogenesis (Yeh et al. 1995). Noteworthy, the decrease in bone volume in GC-induced osteoporosis is accompanied by increased marrow fat (Meunier et al. 1971, Burkhardt et al. 1987, Nuttall & Gimble 2000). Thus, the increase in Fkbp51 mRNA level may reflect a GC-induced switch from an osteoblastic to an adipocytic phenotype (Wang et al. 1977, Nuttall et al. 1998).

Discussion This study provides an extensive profile of the alterations in gene expression that occur when GCs block the progression of MC3T3-E1 osteoblast cultures through a commitment stage around the time of confluency. While some of the disclosed genes have previously been described as GCresponsive (e.g. Mmp13, Serum/GC-regulated kinase, Tieg, Fkbp51), the response of many others, such as Krox20/Egr2, Thrombospondin 1 and 2, and Follistatin, is described here for the first time. Based on these data, several potential mechanisms by which GCs impair osteoblast function and induce osteoporosis can be proposed. The strong repression of Krox20/Egr2 in GCtreated cells, shown here at both the mRNA and protein levels, most likely contributes to GCinduced osteoporosis, because Krox20 gene ablation in the mouse leads to osteoblast insufficiency and bone loss (Levi et al. 1996). Decreased Krox20 mRNA levels could result in decreased collagen synthesis (Galera et al. 1996) and may also contribute to the observed increase in the mRNA for Follistatin, a BMP antagonist negatively regulated by Krox20 (Seitanidou et al. 1997). BMP signaling plays a key role in regulating bone mass and skeletal integrity (Styrkarsdottir et al. 2003). Taken together, a number of changes observed following DEX treatment can be expected to ultimately inhibit BMP signaling. Most directly, DEX increased the expression of genes encoding the secreted BMP antagonists, Follistatin Journal of Molecular Endocrinology (2004) 33, 175–193

and Dan. DEX could also have inhibited BMP action through parallel pathways known to interact with BMP signaling. For example, DEX increased the expression of the genes encoding p21 Rasgap3 and the PTPs Ptpn16 and RptpD. These effects probably result in the inhibition of the MAPK pathway, which is known to synergize with Smad-mediated BMP signaling and to transmit Smad-independent signals (Lou et al. 2000, Gallea et al. 2001, Lai & Cheng 2002, Suzawa et al. 2002, Hassel et al. 2003). In addition, the DEX-induced decrease in Cox mRNA levels can be expected to inhibit BMP signaling because PGs, the products of Cox activity, support BMP action (Chikazu et al. 2002). Evidence for abrogated BMP signaling in the GC-treated cultures is provided by the 5·8-fold decreased expression of Tieg, a BMP-responsive gene (Hefferan et al. 2000). Along with the alterations that may contribute to osteoblast dysfunction, the expression of several genes was inconsistent with the overall adverse effect of GCs. For example, DEX stimulated the expression of genes involved in the synthesis of estrogens, known to promote osteoblast differentiation and preserve bone mass (Qu et al. 1998, Gohel et al. 1999, Manolagas et al. 2002). Cyr61, an extracellular matrix-associated protein that promotes cell proliferation (Kireeva et al. 1996), was also induced by DEX, as was Bsp, a late marker for osteoblast differentiation (Mintz et al. 1993, Zhou et al. 1995). Indeed, GCs can either stimulate or suppress osteoblast function in vitro, depending on their concentration, the specific culture system and the administration protocol. We performed our microarray analysis under conditions where the adverse effects apparently override the stimulatory ones. This study employed a cell line, MC3T3-E1, a novel culture system in which GCs inhibit commitment to the ultimate osteoblast phenotype, i.e. elaboration of mineralized extracellular matrix. This inhibitory effect is associated with alterations in pathways, which critically contribute to bone health in vivo, including the Wnt (Smith et al. 2002) and the BMP signaling pathways (Luppen et al. 2003b). Furthermore, some GC-responsive genes identified in the present study (e.g. Krox 20/Egr2 and thrombospondin 2) have been directly linked to bone health in vivo. However, extrapolation of observations made in the MC3T3-E1 cell line to GC-induced osteoporosis will require careful studies with other osteoblast culture systems, www.endocrinology.org

Microarray analysis of DEX-treated osteoblasts ·

animal models and/or human bone biopsies, which will have to demonstrate, for example, similar alterations in gene expression in vivo and a role for these alterations in mediating the adverse effects of GCs on bone mass and quality. Analysis of gene expression using microarrays suffers a number of inherent limitations. First, by setting our cutoff to a 2·5-fold change, we filtered out some smaller, yet possibly important effects. Such appears to be the case for the mRNAs encoding the type I collagen chains, which were down-regulated by only 1·7- to 1·9-fold in the DEX-treated cultures. A second obvious limitation is that the microarray analysis is confined to the RNA level. This limitation too is well demonstrated by the inhibition of type I collagen gene expression, which reaches a factor of 4- to 10-fold when measured in the same cultures at the protein level (Luppen et al. 2003b). Thirdly, we only reported robust effects (d2·5-fold, P^0·05) for genes whose expression levels were above a predetermined threshold (see Materials and methods). Along with many potentially misleading differences within the noise range, this approach could have filtered out important effects of DEX on rare transcripts. Our RT-PCR analyses generally confirmed the microarray results. Most of the primers used for RT-PCR amplified protein-coding sequences spanning adjacent exons, in order to disclose possible contamination of the cDNA-derived signal by amplification of genomic DNA. In contrast, the sequences represented on the microarrays usually interrogate the 3 -UTRs of mRNAs. The general agreement between the microarray and the RTPCR results, despite the difference in sequences targeted by the two methods, increases our confidence in the disclosed effects of DEX. However, the difference between the sequences targeted by the microarray versus the RT-PCR analysis can lead to disagreement between the two methods in cases where DEX affects, for example, alternative polyadenylation or alternative splicing. This could have been the case for four of the twenty genes tested, where the RT-PCR results were different from those of the microarray analysis. These four genes are denoted by asterisks in Table S1 (http:// hacialab.usc.edu/supplement/frenkel). In summary, we used oligonucleotide microarray analysis to identify DEX-responsive genes in MC3T3-E1 osteoblast cultures and highlight potential mechanisms of GC-induced osteoporosis. www.endocrinology.org

N LECLERC

and others 189

Two of the most attractive potential mechanisms emerging from this study are the suppression of Krox20/Egr2 and the abrogation of BMP signaling.

Supplemental material Supplemental material, including the complete set of raw microarray data and the list of the 91 genes most responsive to DEX, is available online at http://hacialab.usc.edu/supplement/frenkel.

Acknowledgements The authors thank Dr Jack Turman (University of Southern California) for generously providing the Krox20 antibodies. The authors are also grateful to Vijaya Rao, Sara Martinez and Jack Chen for their contribution to the calcium assays and the RT-PCR analyses. This study was supported by grants from the National Institutes of Health (AR47052), the Arthritis Foundation (Atlanta, GA, USA) and the Robert E and May R Wright Foundation (Los Angeles, CA, USA). N L is the recipient of an Arthritis Foundation postdoctoral training grant.

References Alessi DR, Smythe C & Keyse SM 1993 The human CL100 gene encodes a Tyr/Thr-protein phosphatase which potently and specifically inactivates MAP kinase and suppresses its activation by oncogenic ras in Xenopus oocyte extracts. Oncogene 8 2015–2020. Asano H, Li XS & Stamatoyannopoulos G 2000 FKLF-2: a novel Kruppel-like transcriptional factor that activates globin and other erythroid lineage genes. Blood 95 3578–3584. Baughman G, Wiederrecht GJ, Campbell NF, Martin MM & Bourgeois S 1995 FKBP51, a novel T-cell-specific immunophilin capable of calcineurin inhibition. Molecular Cell Biology 15 4395–4402. Beckmann AM & Wilce PA 1997 Egr transcription factors in the nervous system. Neurochemistry International 31 477–510. Bellows CG, Aubin JE & Heersche JN 1987 Physiological concentrations of glucocorticoids stimulate formation of bone nodules from isolated rat calvaria cells in vitro. Endocrinology 121 1985–1992. Bollen M, Gijsbers R, Ceulemans H, Stalmans W & Stefan C 2000 Nucleotide pyrophosphatases/phosphodiesterases on the move. Critical Reviews in Biochemistry and Molecular Biology 35 393–432. Bornstein P 2000 Matricellular proteins: an overview. Matrix Biology 19 555–556. Bornstein P, Armstrong LC, Hankenson KD, Kyriakides TR & Yang Z 2000 Thrombospondin 2, a matricellular protein with diverse functions. Matrix Biology 19 557–568. Journal of Molecular Endocrinology (2004) 33, 175–193

190

N LECLERC

and others ·

Microarray analysis of DEX-treated osteoblasts

Burkhardt R, Kettner G, Bohm W, Schmidmeier M, Schlag R, Frisch B, Mallmann B, Eisenmenger W & Gilg T 1987 Changes in trabecular bone, hematopoiesis and bone marrow vessels in aplastic anemia, primary osteoporosis, and old age: a comparative histomorphometric study. Bone 8 157–164. Canalis E 1985 Effect of growth factors on bone cell replication and differentiation. Clinical Orthopaedics 193 246–263. Canalis E 1996 Clinical review 83: Mechanisms of glucocorticoid action in bone: implications to glucocorticoid-induced osteoporosis. Journal of Clinical Endocrinology and Metabolism 81 3441–3447. Canalis E, Economides AN & Gazzerro E 2003 Bone morphogenetic proteins, their antagonists, and the skeleton. Endocrinology Reviews 24 218–235. Carpio L, Gladu J, Goltzman D & Rabbani SA 2001 Induction of osteoblast differentiation indexes by PTHrP in MG-63 cells involves multiple signaling pathways. American Journal of Physiology, Endocrinology and Metabolism 281 E489–E499. Carron JA, Bowler WB, Wagstaff SC & Gallagher JA 1999 Expression of members of the thrombospondin family by human skeletal tissues and cultured cells. Biochemical and Biophysical Research Communications 263 389–391. Chang DJ, Ji C, Kim KK, Casinghino S, McCarthy TL & Centrella M 1998 Reduction in transforming growth factor beta receptor I expression and transcription factor CBFa1 on bone cells by glucocorticoid. Journal of Biological Chemistry 273 4892–4896. Chen H, Herndon ME & Lawler J 2000 The cell biology of thrombospondin-1. Matrix Biology 19 597–614. Chen J, Sodek J, Thomas HF & Ranly DM 1996 Dexamethasone stimulates luciferase gene expression through the rat bone sialoprotein gene promoter in transgenic mice. Connective Tissue Research 35 33–39. Chen J, Carey K & Godowski PJ 1997 Identification of Gas6 as a ligand for Mer, a neural cell adhesion molecule related receptor tyrosine kinase implicated in cellular transformation. Oncogene 14 2033–2039. Chen TL, Cone CM & Feldman D 1983 Glucocorticoid modulation of cell proliferation in cultured osteoblast-like bone cells: differences between rat and mouse. Endocrinology 112 1739–1745. Chikazu D, Li X, Kawaguchi H, Sakuma Y, Voznesensky OS, Adams DJ, Xu M, Hoshio K, Katavic V, Herschman HR, Raisz LG & Pilbeam CC 2002 Bone morphogenetic protein 2 induces cyclo-oxygenase 2 in osteoblasts via a Cbfal binding site: role in effects of bone morphogenetic protein 2 in vitro and in vivo. Journal of Bone and Mineral Research 17 1430–1440. Choe J, Stern P & Feldman D 1978 Receptor mediated glucocorticoid inhibition of protein synthesis in isolated bone cells. Journal of Steroid Biochemistry 9 265–271. Collett G, Wood A, Alexander MY, Varnum BC, Boot-Handford RP, Ohanian V, Ohanian J, Fridell YW & Canfield AE 2003 Receptor tyrosine kinase Axl modulates the osteogenic differentiation of pericytes. Circulation Research 92 1123–1129. Davies TH, Ning YM & Sanchez ER 2002 A new first step in activation of steroid receptors: hormone-induced switching of FKBP51 and FKBP52 immunophilins. Journal of Biological Chemistry 277 4597–4600. Delany AM, Gabbitas BY & Canalis E 1995a Cortisol downregulates osteoblast alpha 1 (I) procollagen mRNA by transcriptional and posttranscriptional mechanisms. Journal of Cellular Biochemistry 57 488–494. Delany AM, Jeffrey JJ, Rydziel S & Canalis E 1995b Cortisol increases interstitial collagenase expression in osteoblasts by post-transcriptional mechanisms. Journal of Biological Chemistry 270 26607–26612. Engelbrecht Y, de Wet H, Horsch K, Langeveldt CR, Hough FS & Hulley PA 2003 Glucocorticoids induce rapid up-regulation of mitogen-activated protein kinase phosphatase-1 and Journal of Molecular Endocrinology (2004) 33, 175–193

dephosphorylation of extracellular signal-regulated kinase and impair proliferation in human and mouse osteoblast cell lines. Endocrinology 144 412–422. Eswarakumar VP, Monsonego-Ornan E, Pines M, Antonopoulou I, Morriss-Kay GM & Lonai P 2002 The IIIc alternative of Fgfr2 is a positive regulator of bone formation. Development 129 3783–3793. Eyre LJ, Bland R, Bujalska IJ, Sheppard MC, Stewart PM & Hewison M 1998 Characterization of aromatase and 17 beta-hydroxysteroid dehydrogenase expression in rat osteoblastic cells. Journal of Bone and Mineral Research 13 996–1004. Funakoshi H, Yonemasu T, Nakano T, Matumoto K & Nakamura T 2002 Identification of Gas6, a putative ligand for Sky and Axl receptor tyrosine kinases, as a novel neurotrophic factor for hippocampal neurons. Journal of Neuroscience Research 68 150–160. Galera P, Park RW, Ducy P, Mattei MG & Karsenty G 1996 c-Krox binds to several sites in the promoter of both mouse type I collagen genes. Structure/function study and developmental expression analysis. Journal of Biological Chemistry 271 21331–21339. Gallea S, Lallemand F, Atfi A, Rawadi G, Ramez V, Spinella-Jaegle S, Kawai S, Faucheu C, Huet L, Baron R & Roman-Roman S 2001 Activation of mitogen-activated protein kinase cascades is involved in regulation of bone morphogenetic protein-2-induced osteoblast differentiation in pluripotent C2C12 cells. Bone 28 491–498. Galzie Z, Kinsella AR & Smith JA 1997 Fibroblast growth factors and their receptors. Biochemistry and Cell Biology 75 669–685. Ganss B, Kim RH & Sodek J 1999 Bone sialoprotein. Critical Reviews in Oral Biology and Medicine 10 79–98. Goding JW, Terkeltaub R, Maurice M, Deterre P, Sali A & Belli SI 1998 Ecto-phosphodiesterase/pyrophosphatase of lymphocytes and non-lymphoid cells: structure and function of the PC-1 family. Immunological Reviews 161 11–26. Gohel A, McCarthy MB & Gronowicz G 1999 Estrogen prevents glucocorticoid-induced apoptosis in osteoblasts in vivo and in vitro. Endocrinology 140 5339–5347. Gorski JL, Estrada L, Hu C & Liu Z 2000 Skeletal-specific expression of Fgd1 during bone formation and skeletal defects in faciogenital dysplasia (FGDY; Aarskog syndrome). Developmental Dynamics 218 573–586. Goruppi S, Ruaro E & Schneider C 1996 Gas6, the ligand of Axl tyrosine kinase receptor, has mitogenic and survival activities for serum starved NIH3T3 fibroblasts. Oncogene 12 471–480. Grey A, Banovic T, Naot D, Hill B, Callon K, Reid I & Cornish J 2001 Lysophosphatidic acid is an osteoblast mitogen whose proliferative actions involve G(i) proteins and protein kinase C, but not P42/44 mitogen-activated protein kinases. Endocrinology 142 1098–1106. Grey A, Chen Q, Callon K, Xu X, Reid IR & Cornish J 2002 The phospholipids sphingosine-1-phosphate and lysophosphatidic acid prevent apoptosis in osteoblastic cells via a signaling pathway involving G(i) proteins and phosphatidylinositol-3 kinase. Endocrinology 143 4755–4763. Hankenson KD & Bornstein P 2002 The secreted protein thrombospondin 2 is an autocrine inhibitor of marrow stromal cell proliferation. Journal of Bone and Mineral Research 17 415–425. Hashimoto N, Zhang WR & Goldstein BJ 1992 Insulin receptor and epidermal growth factor receptor dephosphorylation by three major rat liver protein-tyrosine phosphatases expressed in a recombinant bacterial system. Biochemical Journal 284 569–576. Hassel S, Schmitt S, Hartung A, Roth M, Nohe A, Petersen N, Ehrlich M, Henis YI, Sebald W & Knaus P 2003 Initiation of Smad-dependent and Smad-independent signaling via distinct BMP-receptor complexes. Journal of Bone and Joint Surgery of America 85 44–51. Heegaard AM, Gehron Robey P, Vogel W, Just W, Widom RL, Scholler J, Fisher LW & Young MF 1997 Functional characterization of the human biglycan 5 -flanking DNA and www.endocrinology.org

Microarray analysis of DEX-treated osteoblasts · binding of the transcription factor c-Krox. Journal of Bone and Mineral Research 12 2050–2060. Hefferan TE, Subramaniam M, Khosla S, Riggs BL & Spelsberg TC 2000 Cytokine-specific induction of the TGF-beta inducible early gene (TIEG): regulation by specific members of the TGF-beta family. Journal of Cellular Biochemistry 78 380–390. Hulley PA, Conradie MM, Langeveldt CR & Hough FS 2002 Glucocorticoid-induced osteoporosis in the rat is prevented by the tyrosine phosphatase inhibitor, sodium orthovanadate. Bone 31 220–229. Hunter GK & Goldberg HA 1993 Nucleation of hydroxyapatite by bone sialoprotein. PNAS 90 8562–8565. Inoue D, Reid M, Lum L, Kratzschmar J, Weskamp G, Myung YM, Baron R & Blobel CP 1998 Cloning and initial characterization of mouse meltrin beta and analysis of the expression of four metalloprotease-disintegrins in bone cells. Journal of Biological Chemistry 273 4180–4187. Iruela-Arispe ML, Liska DJ, Sage EH & Bornstein P 1993 Differential expression of thrombospondin 1, 2, and 3 during murine development. Developmental Dynamics 197 40–56. Ishida Y & Heersche JN 1998 Glucocorticoid-induced osteoporosis: both in vivo and in vitro concentrations of glucocorticoids higher than physiological levels attenuate osteoblast differentiation. Journal of Bone and Mineral Research 13 1822–1828. Johnson K, Goding J, Van Etten D, Sali A, Hu SI, Farley D, Krug H, Hessle L, Millan JL & Terkeltaub R 2003 Linked deficiencies in extracellular PP(i) and osteopontin mediate pathologic calcification associated with defective PC-1 and ANK expression. Journal of Bone and Mineral Research 18 994–1004. Karaman MW, Houck ML, Chemnick LG, Nagpal S, Chawannakul D, Sudano D, Pike BL, Ho VV, Ryder OA & Hacia JG 2003 Comparative analysis of gene-expression patterns in human and African great ape cultured fibroblasts. Genome Research 13 1619–1630. Kassel O, Sancono A, Kratzschmar J, Kreft B, Stassen M & Cato AC 2001 Glucocorticoids inhibit MAP kinase via increased expression and decreased degradation of MKP-1. EMBO Journal 20 7108–7116. Keyse SM & Emslie EA 1992 Oxidative stress and heat shock induce a human gene encoding a protein-tyrosine phosphatase. Nature 359 644–647. Khachigian LM & Collins T 1998 Early growth response factor 1: a pleiotropic mediator of inducible gene expression. Journal of Molecular Medicine 76 613–616. Kireeva ML, Mo FE, Yang GP & Lau LF 1996 Cyr61, a product of a growth factor-inducible immediate-early gene, promotes cell proliferation, migration, and adhesion. Molecular and Cellular Biology 16 1326–1334. Knauper V, Lopez-Otin C, Smith B, Knight G & Murphy G 1996 Biochemical characterization of human collagenase-3. Journal of Biological Chemistry 271 1544–1550. Kucharz EJ 1988 Hormonal control of collagen metabolism. Part II. Endocrinologie 26 229–237. Lai CF & Cheng SL 2002 Signal transductions induced by bone morphogenetic protein-2 and transforming growth factor-beta in normal human osteoblastic cells. Journal of Biological Chemistry 277 15514–15522. Leboy PS, Beresford JN, Devlin C & Owen ME 1991 Dexamethasone induction of osteoblast mRNAs in rat marrow stromal cell cultures. Journal of Cellular Physiology 146 370–378. Lechner A, Schutze N, Siggelkow H, Seufert J & Jakob F 2000 The immediate early gene product hCYR61 localizes to the secretory pathway in human osteoblasts. Bone 27 53–60. Lerner UH & Frohlander N 1992 Haptoglobin-stimulated bone resorption in neonatal mouse calvarial bones in vitro. Arthritis and Rheumatism 35 587–591. www.endocrinology.org

N LECLERC

and others 191

Levi G, Topilko P, Schneider-Maunoury S, Lasagna M, Mantero S, Cancedda R & Charnay P 1996 Defective bone formation in Krox-20 mutant mice. Development 122 113–120. Li R, Chen J, Hammonds G, Phillips H, Armanini M, Wood P, Bunge R, Godowski PJ, Sliwkowski MX & Mather JP 1996 Identification of Gas6 as a growth factor for human Schwann cells. Journal of Neuroscience 16 2012–2019. Li Y, Lacerda DA, Warman ML, Beier DR, Yoshioka H, Ninomiya Y, Oxford JT, Morris NP, Andrikopoulos K, Ramirez F, Wardell BB, Lifferth GD, Teuscher C, Woodward SR, Taylor BA, Seegmiller RE & Olsen BR 1995 A fibrillar collagen gene, Col11a1, is essential for skeletal morphogenesis. Cell 80 423–430. Lian JB, Shalhoub V, Aslam F, Frenkel B, Green J, Hamrah M, Stein GS & Stein JL 1997 Species-specific glucocorticoid and 1,25-dihydroxyvitamin D responsiveness in mouse MC3T3-E1 osteoblasts: dexamethasone inhibits osteoblast differentiation and vitamin D down-regulates osteocalcin gene expression. Endocrinology 138 2117–2127. Linkhart TA, Mohan S & Baylink DJ 1996 Growth factors for bone growth and repair: IGF, TGF beta and BMP. Bone 19 1S–12S. Lou J, Tu Y, Li S & Manske PR 2000 Involvement of ERK in BMP-2 induced osteoblastic differentiation of mesenchymal progenitor cell line C3H10T1/2. Biochemical and Biophysical Research Communications 268 757–762. Luckert B 1997 Glucocorticoid-induced osteoporosis. In Osteoporosis, pp 801–820. Eds R Markus, D Feldman & J Kelsey. San Diego: Academic Press. Luckert B Kream B 1996 Clinical and basic aspects of glucocorticoid action in bones. In Principles of Bone Biology, pp 533–548. Eds JP Bilezikian, LG Raisz & GA Rodan. San Diego: Academic Press. Luppen CA, Leclerc N, Noh T, Barski A, Khokhar A, Boskey AL, Smith E & Frenkel B 2003a Brief bone morphogenetic protein 2 treatment of glucocorticoid-inhibited MC3T3-E1 osteoblasts rescues commitment-associated cell cycle and mineralization without alteration of Runx2. Journal of Biological Chemistry 278 44995–45003. Luppen CA, Smith E, Spevak L, Boskey AL & Frenkel B 2003b Bone morphogenetic protein-2 restores mineralization in glucocorticoid-inhibited MC3T3-E1 osteoblast cultures. Journal of Bone and Mineral Research 18 1186–1197. McCloskey P, Fridell YW, Attar E, Villa J, Jin Y, Varnum B & Liu ET 1997 GAS6 mediates adhesion of cells expressing the receptor tyrosine kinase Axl. Journal of Biological Chemistry 272 23285–23291. McMahon AP, Champion JE, McMahon JA & Sukhatme VP 1990 Developmental expression of the putative transcription factor Egr-1 suggests that Egr-1 and c-fos are coregulated in some tissues. Development 108 281–287. Mahonen A, Jukkola A, Risteli L, Risteli J & Maenpaa PH 1998 Type I procollagen synthesis is regulated by steroids and related hormones in human osteosarcoma cells. Journal of Cellular Biochemistry 68 151–163. Mandavilli BS & Rao KS 1996 Accumulation of DNA damage in aging neurons occurs through a mechanism other than apoptosis. Journal of Neurochemistry 67 1559–1565. Manolagas SC & Weinstein RS 1999 New developments in the pathogenesis and treatment of steroid-induced osteoporosis. Journal of Bone and Mineral Research 14 1061–1066. Manolagas SC, Kousteni S & Jilka RL 2002 Sex steroids and bone. Recent Progress in Hormone Research 57 385–409. Mansukhani A, Bellosta P, Sahni M & Basilico C 2000 Signaling by fibroblast growth factors (FGF) and fibroblast growth factor receptor 2 (FGFR2)-activating mutations blocks mineralization and induces apoptosis in osteoblasts. Journal of Cell Biology 149 1297–1308. Martin KM, Metcalfe JC & Kemp PR 2001 Expression of Klf9 and Klf13 in mouse development. Mechanisms of Development 103 149–151. Journal of Molecular Endocrinology (2004) 33, 175–193

192

N LECLERC

and others ·

Microarray analysis of DEX-treated osteoblasts

Mbuyi JM, Dequeker J, Bloemmen F & Stevens E 1982 Plasma proteins in human cortical bone: enrichment of alpha 2 HS-glycoprotein, alpha 1 acid-glycoprotein, and IgE. Calcified Tissue International 34 229–231. Meunier P, Aaron J, Edouard C & Vignon G 1971 Osteoporosis and the replacement of cell populations of the marrow by adipose tissue. A quantitative study of 84 iliac bone biopsies. Clinical Orthopaedics 80 147–154. Miao D, Tong XK, Chan GK, Panda D, McPherson PS & Goltzman D 2001 Parathyroid hormone-related peptide stimulates osteogenic cell proliferation through protein kinase C activation of the Ras/mitogen-activated protein kinase signaling pathway. Journal of Biological Chemistry 276 32204–32213. Mintz KP, Grzesik MJ, Midura RJ, Robey PG, Termine JD & Fisher LW 1993 Purification and fragmentation of nondenatured bone sialoprotein: evidence for a cryptic, RGD-resistant cell attachment domain. Journal of Bone and Mineral Research 8 985–995. Mizuno K, Hasegawa K, Katagiri T, Ogimoto M, T & Yakura H 1993 MPTP delta, a putative murine homolog of HPTP delta, is expressed in specialized regions of the brain and in the B-cell lineage. Molecular and Cellular Biology 13 5513–5523. Muragaki Y, Shiota C, Inoue M, Ooshima A, Olsen BR & Ninomiya Y 1992 Alpha 1(VIII)-collagen gene transcripts encode a short-chain collagen polypeptide and are expressed by various epithelial, endothelial and mesenchymal cells in newborn mouse tissues. European Journal of Biochemistry 207 895–902. Nagata K, Ohashi K, Nakano T, Arita H, Zong C, Hanafusa H & Mizuno K 1996 Identification of the product of growth arrest-specific gene 6 as a common ligand for Axl, Sky, and Mer receptor tyrosine kinases. Journal of Biological Chemistry 271 30022–30027. Nakano T, Kawamoto K, Higashino K & Arita H 1996 Prevention of growth arrest-induced cell death of vascular smooth muscle cells by a product of growth arrest-specific gene, gas6. FEBS Letters 387 78–80. Nawata H, Tanaka S, Takayanagi R, Sakai Y, Yanase T, Ikuyama S & Haji M 1995 Aromatase in bone cell: association with osteoporosis in postmenopausal women. Journal of Steroid Biochemistry and Molecular Biology 53 165–174. Nuttall ME & Gimble JM 2000 Is there a therapeutic opportunity to either prevent or treat osteopenic disorders by inhibiting marrow adipogenesis? Bone 27 177–184. Nuttall ME, Patton AJ, Olivera DL, Nadeau DP & Gowen M 1998 Human trabecular bone cells are able to express both osteoblastic and adipocytic phenotype: implications for osteopenic disorders. Journal of Bone and Mineral Research 13 371–382. O’Brien TP & Lau LF 1992 Expression of the growth factor-inducible immediate early gene cyr61 correlates with chondrogenesis during mouse embryonic development. Cell Growth and Differentiation 3 645–654. Ogata Y, Yamauchi M, Kim RH, Li JJ, Freedman LP & Sodek J 1995 Glucocorticoid regulation of bone sialoprotein (BSP) gene expression. Identification of a glucocorticoid response element in the bone sialoprotein gene promoter. European Journal of Biochemistry 230 183–192. Pockwinse SM, Stein JL, Lian JB & Stein GS 1995 Developmental stage-specific cellular responses to vitamin D and glucocorticoids during differentiation of the osteoblast phenotype: interrelationship of morphology and gene expression by in situ hybridization. Experimental Cell Research 216 244–260. Pilbeam CC, Harrison JR & Raisz LG 2002 Prostaglandins and bone metabolism. In Principles of Bone Biology, pp 715–728. Eds JP Bilezikian, LG Raisz & GA Rodan. San Diego: Academic Press. Qu Q, Perala-Heape M, Kapanen A, Dahllund J, Salo J, Vaananen HK & Harkonen P 1998 Estrogen enhances differentiation of osteoblasts in mouse bone marrow culture. Bone 22 201–209. Journal of Molecular Endocrinology (2004) 33, 175–193

Raisz LG 1999 Prostaglandins and bone: physiology and pathophysiology. Osteoarthritis Cartilage 7 419–421. Regunathan A, Glesne DA, Wilson AK, Song J, Nicolae D, Flores T & Bhattacharyya MH 2003 Microarray analysis of changes in bone cell gene expression early after cadmium gavage in mice. Toxicology and Applied Pharmacology 191 272–293. Robey PG, Young MF, Fisher LW & McClain TD 1989 Thrombospondin is an osteoblast-derived component of mineralized extracellular matrix. Journal of Cell Biology 108 719–727. Rogatsky I, Trowbridge JM & Garabedian MJ 1997 Glucocorticoid receptor-mediated cell cycle arrest is achieved through distinct cell-specific transcriptional regulatory mechanisms. Molecular and Cellular Biology 17 3181–3193. Rubin MR & Bilezikian JP 2002 Clinical review 151: The role of parathyroid hormone in the pathogenesis of glucocorticoid-induced osteoporosis: a re-examination of the evidence. Journal of Clinical Endocrinology and Metabolism 87 4033–4041. Sampath TK, Maliakal JC, Hauschka PV, Jones WK, Sasak H, Tucker RF, KH, Coughlin JE, Tucker MM, Pang RH, Corbett C, Özkaynak E, Oppermann H & Rueger DC 1992 Recombinant human osteogenic protein-1 (hOP-1) induces new bone formation in vivo with a specific activity comparable with natural bovine osteogenic protein and stimulates osteoblast proliferation and differentiation in vitro. Journal of Biological Chemistry 267 20352–20362. Sasano H, Uzuki M, Sawai T, Nagura H, Matsunaga G, Kashimoto O & Harada N 1997 Aromatase in human bone tissue. Journal of Bone and Mineral Research 12 1416–1423. Sato Y, Arai N, Negishi A & Ohya K 1997 Expression of cyclooxygenase genes and involvement of endogenous prostaglandin during osteogenesis in the rat tibial bone marrow cavity. Journal of Medical Dental Science 44 81–92. Schacke H, Docke WD & Asadullah K 2002 Mechanisms involved in the side effects of glucocorticoids. Pharmacology and Therapeutics 96 23–43. Schutze N, Lechner A, Groll C, Siggelkow H, Hufner M, Kohrle J & Jakob K 1998 The human analog of murine cystein rich protein 61 [correction of 16] is a 1 alpha,25-dihydroxyvitamin D3 responsive immediate early gene in human fetal osteoblasts: regulation by cytokines, growth factors, and serum. Endocrinology 139 1761–1770. Seitanidou T, Schneider-Maunoury S, Desmarquet C, Wilkinson DG & Charnay P 1997 Krox-20 is a key regulator of rhombomere-specific gene expression in the developing hindbrain. Mechanisms of Development 65 31–42. Shalhoub V, Conlon D, Tassinari M, Quinn C, Partridge N, Stein GS & Lian JB 1992 Glucocorticoids promote development of the osteoblast phenotype by selectively modulating expression of cell growth and differentiation associated genes. Journal of Cellular Biochemistry 50 425–440. Shankar SL, O’Guin K, Cammer M, McMorris FA, Stitt TN, Basch RS, Varnum B & Shafit-Zagardo B 2003 The growth arrest-specific gene product Gas6 promotes the survival of human oligodendrocytes via a phosphatidylinositol 3-kinase-dependent pathway. Journal of Neuroscience 23 4208–4218. Sherbina NV & Bornstein P 1992 Modulation of thrombospondin gene expression during osteoblast differentiation in MC3T3-E1 cells. Bone 13 197–201. Smith E, Redman RA, Logg CR, Coetzee GA, Kasahara N & Frenkel B 2000 Glucocorticoids inhibit developmental stage-specific osteoblast cell cycle. Dissociation of cyclin A-cyclin-dependent kinase 2 from E2F4-p130 complexes. Journal of Biological Chemistry 275 19992–20001. Smith E, Coetzee GA & Frenkel B 2002 Glucocorticoids inhibit cell cycle progression in differentiating osteoblasts via glycogen www.endocrinology.org

Microarray analysis of DEX-treated osteoblasts · synthase kinase-3 beta. Journal of Biological Chemistry 277 18191–18197. Smith WL, Garavito RM & DeWitt DL 1996 Prostaglandin endoperoxide H synthases (cyclooxygenases)-1 and -2. Journal of Biological Chemistry 271 33157–33160. Styrkarsdottir U, Cazier JB, Kong A, Rolfsson O, Larsen H, Bjarnadottir E, Johannsdottir VD, Sigurdardottir MS, Bagger Y, Christiansen C, Reynisdottir I, Grant SFA, Jonasson K, Frigge ML, Gulcher JR, Sigurdsson G & Stefansson K 2003 Linkage of osteoporosis to chromosome 20p12 and association to BMP2. PLoS Biology 1 351–360. Subramaniam M, Harris SA, Oursler MJ, Rasmussen K, Riggs BL & Spelsberg TC 1995 Identification of a novel TGF-beta-regulated gene encoding a putative zinc finger protein in human osteoblasts. Nucleic Acids Research 23 4907–4912. Sukhatme VP 1992 The Egr transcription factor family: from signal transduction to kidney differentiation. Kidney International 41 550–553. Suzawa M, Tamura Y, Fukumoto S, Miyazono K, Fujita T, Kato S & Takeuchi Y 2002 Stimulation of Smad1 transcriptional activity by Ras-extracellular signal-regulated kinase pathway: a possible mechanism for collagen-dependent osteoblastic differentiation. Journal of Bone and Mineral Research 17 240–248. Takai Y, Sasaki T & Matozaki T 2001 Small GTP-binding proteins. Physiological Reviews 81 153–208. Tenenbaum HC & Heersche JN 1985 Dexamethasone stimulates osteogenesis in chick periosteum in vitro. Endocrinology 117 2211–2217. Terkeltaub RA 2001 Inorganic pyrophosphate generation and disposition in pathophysiology. American Journal of Physiology - Cell Physiology 281 C1–C11. Tokumura A, Majima E, Kariya Y, Tominaga K, Kogure K, Yasuda K & Fukuzawa K 2002 Identification of human plasma lysophospholipase D, a lysophosphatidic acid-producing enzyme, as autotaxin, a multifunctional phosphodiesterase. Journal of Biological Chemistry 277 39436–39442. Tooney PA, Sakai T, Sakai K, Aeschlimann D & Mosher DF 1998 Restricted localization of thrombospondin-2 protein during mouse embryogenesis: a comparison to thrombospondin-1. Matrix Biology 17 131–143. Topilko P, Schneider-Maunoury S, Levi G, Baron-Van Evercooren A, Chennoufi AB, Seitanidou T, Babinet C & Charnay P 1994 Krox-20 controls myelination in the peripheral nervous system. Nature 371 796–799. Van Staa TP, Leufkens HG, Abenhaim L, Zhang B & Cooper C 2000 Use of oral corticosteroids and risk of fractures. Journal of Bone and Mineral Research 15 993–1000. Varnum BC, Young C, Elliott G, Garcia A, Bartley TD, Fridell YW, Hunt RW, Trail G, Clogston C, Toso RJ, Yanagihara D, Bennett L, Sylber M, Merewether LA, Tseng A, Escobar E, Liu ET & Yamane HK 1995 Axl receptor tyrosine kinase stimulated by the vitamin K-dependent protein encoded by growth-arrest-specific gene 6. Nature 373 623–626. Wang GJ, Sweet DE, Reger SI & Thompson RC 1977 Fat-cell changes as a mechanism of avascular necrosis of the femoral head in cortisone-treated rabbits. Journal of Bone and Joint Surgery of America 59 729–735. Weinstein RS, Jilka RL, Parfitt AM & Manolagas SC 1998 Inhibition of osteoblastogenesis and promotion of apoptosis of

www.endocrinology.org

N LECLERC

and others 193

osteoblasts and osteocytes by glucocorticoids. Potential mechanisms of their deleterious effects on bone. Journal of Clinical Investigation 102 274–282. Wilkinson DG, Bhatt S, Chavrier P, Bravo R & Charnay P 1989 Segment-specific expression of a zinc-finger gene in the developing nervous system of the mouse. Nature 337 461–464. Xiao G, Jiang D, Thomas P, Benson MD, Guan K, Karsenty G & Franceschi RT 2000 MAPK pathways activate and phosphorylate the osteoblast-specific transcription factor, Cbfa1. Journal of Biological Chemistry 275 4453–4459. Xiao G, Gopalakrishnan R, Jiang D, Reith E, Benson MD & Franceschi RT 2002 Bone morphogenetic proteins, extracellular matrix, and mitogen-activated protein kinase signaling pathways are required for osteoblast-specific gene expression and differentiation in MC3T3-E1 cells. Journal of Bone and Mineral Research 17 101–110. Yamaguchi A, Katagiri T, Ikeda T, Wozney JM, Rosen V, Wang EA, Kahn AJ, Suda T & Yoshiki S 1991 Recombinant human bone morphogenetic protein-2 stimulates osteoblastic maturation and inhibits myogenic differentiation in vitro. Journal of Cell Biology 113 681–687. Yamaguchi A, Komori T & Suda T 2000 Regulation of osteoblast differentiation mediated by bone morphogenetic proteins, hedgehogs, and Cbfa1. Endocrine Reviews 21 393–411. Yanagita M, Arai H, Nakano T, Ohashi K, Mizuno K, Fukatsu A, Doi T & Kita T 2001 Gas6 induces mesangial cell proliferation via latent transcription factor STAT3. Journal of Biological Chemistry 276 42364–42369. Yang GP & Lau LF 1991 Cyr61, product of a growth factor-inducible immediate early gene, is associated with the extracellular matrix and the cell surface. Cell Growth and Differentiation 2 351–357. Yang S, Wei D, Wang D, Phimphilai M, Krebsbach PH & Franceschi RT 2003 In vitro and in vivo synergistic interactions between the Runx2/Cbfa1 transcription factor and bone morphogenetic protein-2 in stimulating osteoblast differentiation. Journal of Bone and Mineral Research 18 705–715. Yeh WC, Li TK, Bierer BE & McKnight SL 1995 Identification and characterization of an immunophilin expressed during the clonal expansion phase of adipocyte differentiation. PNAS 92 11081–11085. Zalavras C, Shah S, Birnbaum MJ & Frenkel B 2003 Role of apoptosis in glucocorticoid-induced osteoporosis and osteonecrosis. Critical Reviews in Eukaryotic Gene Expression 13 221–235. Zhang WR, Li PM, Oswald MA & Goldstein BJ 1996 Modulation of insulin signal transduction by eutopic overexpression of the receptor-type protein-tyrosine phosphatase LAR. Molecular Endocrinology 10 575–584. Zheng Y, Fischer DJ, Santos MF, Tigyi G, Pasteris NG, Gorski JL & Xu Y 1996 The faciogenital dysplasia gene product FGD1 functions as a Cdc42Hs-specific guanine-nucleotide exchange factor. Journal of Biological Chemistry 271 33169–33172. Zhou HY, Takita H, Fujisawa R, Mizuno M & Kuboki Y 1995 Stimulation by bone sialoprotein of calcification in osteoblast-like MC3T3-E1 cells. Calcified Tissue International 56 403–407.

Received in final form 24 March 2004 Accepted 30 March 2004

Journal of Molecular Endocrinology (2004) 33, 175–193