Aromatase expression in the human fetal osteoblastic cell ... - CiteSeerX

533

Aromatase expression in the human fetal osteoblastic cell line SV-HFO M Watanabe1,2, E R Simpson1, N Pathirage1, S Nakajin2 and C D Clyne1 1

Prince Henry’s Institute of Medical Research, PO Box 5152,Clayton, VIC 3168, Australia

2

Department of Biochemistry, Hoshi University School of Pharmacy and Pharmaceutical Sciences, 2-4-42 Ebara, Shinagawa, Tokyo 142-8501, Japan

(Requests for offprints should be addressed to C D Clyne; E-mail [email protected])

Abstract A number of clinical studies have highlighted the importance of estrogen in bone growth and maintenance in men and postmenopausal women. In these instances, estrogen is synthesized locally within bone tissue by aromatase, encoded by the CYP19 gene. The mechanisms regulating aromatase expression in bone, however, are unclear. In this work we characterized the expression of aromatase activity and gene transcripts in the human fetal osteoblastic cell line, SV-HFO. Aromatase activity and gene transcript expression were stimulated by dexamethasone. Oncostatin M strongly stimulated aromatase expression in synergy with dexamethasone. These factors induced CYP19 transcripts that included the sequence of exon I.4 in their 5′UTR. Consistent with this, a reporter construct harboring the genomic sequence of the promoter region of exon I.4 (promoter I.4) was also activated by dexamethasone and oncostatin M. 5′ deletion and mutation analysis revealed important roles for a glucocorticoid response element, an interferon activating sequence and a putative binding site for Sp1. Transfection of exogenous glucocorticoid receptor, STAT3 or Sp1 increased promoter activity, indicating a potential role for these transcription factors in regulating aromatase expression in SV-HFO cells. These data suggest that the SV-HFO cell line is a valuable model with which to elucidate the mechanisms regulating local estrogen synthesis in osteoblasts. Journal of Molecular Endocrinology (2004) 32, 533–545

Introduction Studies of human patients with estrogen deficiency have highlighted the importance of estrogen for pubertal bone growth and establishment of peak bone mineral density (BMD) (Grumbach 2000). Both male and female patients with defects in the estrogen receptor (ER) gene or the CYP19 gene that encodes aromatase show very low BMD, tall stature, delayed bone age, absence of epiphyseal closure and absence of the pubertal growth spurt (Smith et al. 1994, Morishima et al. 1995, Carani et al. 1997, Mullis et al. 1997, Bilezikian et al. 1998). Administration of estrogen to aromatase-deficient men resulted in increased BMD and epiphyseal closure (Carani et al. 1997, Mullis et al. 1997, Bilezikian et al. 1998). Knockout mouse models have supported the importance of estrogen in bone metabolism: male and female ER (ERKO) and aromatase (ArKO) null mice exhibit low BMD (Oz Journal of Molecular Endocrinology (2004) 32, 533–545 0952–5041/04/032–533 © 2004 Society for Endocrinology

et al. 2000, Vidal et al. 2000, Miyaura et al. 2001), and treatment of the ArKO mice with estrogen improved BMD (Miyaura et al. 2001). Recently, a significant relationship between a common CYP19 polymorphism (tetranucleotide tandem repeat in intron 4) and osteoporosis and fracture risk of postmenopausal women has been demonstrated (Masi et al. 2001). Aromatase activity and gene transcripts are present in human bone tissue, primary osteoblastlike cells and osteoblastic osteosarcoma cell lines (Blanchard et al. 1991, Purohit et al. 1992, Nawata et al. 1995, Schweikert et al. 1995, Shimodaira et al. 1996, Shozu & Simpson 1998, Oz et al. 2001). In situ hybridization and immunohistochemistry have shown that aromatase immunoreactivity and gene transcripts are localized in lining cells, osteoblasts and chondrocytes of human bone tissue (Sasano et al. 1997, Oz et al. 2001). Some types of osteoclasts also express aromatase Online version via http://www.endocrinology.org

Printed in Great Britain

534

M WATANABE

and others

· Aromatase in osteoblasts

Figure 1 Partial structure of CYP19 gene. Nomenclature of multiple exons I by Simpson and Harada (upper) (Sasano & Harada 1998, Simpson et al. 2002) and main sites of expression (lower) are indicated. Location of first exons is based on the analysis of human genome sequence by Sebastian and Bulun (2001).

immunoreactivity (Oz et al. 2001). The human CYP19 gene has multiple first exons, which are selected in a tissue-specific manner (Simpson et al. 1997, Sasano & Harada 1998, Sebastian & Bulun 2001) (Fig. 1). RT-PCR analysis indicates that the most dominant 5 UTR in aromatase gene transcript in human bone tissue is derived from upstream exon I.4 (Sasano et al. 1997, Shozu & Simpson 1998). This suggests that the 5 region of exon I.4 (promoter I.4) is predominantly used in bone tissue to drive aromatase expression. The high levels of aromatase expression in these bone cells, together with the low circulating levels of estrogen in postmenopausal women and in men, lead to the conclusion that local aromatase expression in bone is the major source of estrogen acting on this tissue in these instances (Simpson & Davis 2001). The relationship between local aromatase expression and bone metabolism, however, is still not clear. An in vitro osteoblastic cell line would clearly be useful to define better the molecular mechanisms regulating aromatase expression and estrogen production in bone. Recently, an SV40-immortalized human fetal osteoblastic cell line (SV-HFO) was established from human fetal calvaria (Chiba et al. 1993). SV-HFO cells exhibit alkaline phosphatase activity, osteocalcin gene expression and calcium deposition into extracellular matrix in response to dexamethasone (Dex) and -glycerophosphate (-GP) (Iba et al. 1995). Expression of aromatase, 17hydroxysteroid dehydrogenase and steroid sulfatase have also been reported in SV-HFO (Janssen et al. 1999). In addition, SV-HFO cells express both ER and ER (Arts et al. 1997). These properties suggest that the SV-HFO cell line might be an ideal model with which to identify mechanisms of Journal of Molecular Endocrinology (2004) 32, 533–545

aromatase expression in osteoblasts. In the present study, we have defined the regulation of aromatase activity, expression of gene transcripts and promoter usage in SV-HFO cells by a variety of hormonal factors, including glucocorticoids and cytokines.

Materials and methods Materials

Dexamethasone, forskolin, phorbol-12-myristate 13-acetate (PMA), cholecalciferol, recombinant human oncostatin M (OSM) and tumor necrosis factor (TNF) were from Sigma. Recombinant human bone morphogenetic protein-2, interleukin 1, interleukin 6, interleukin 6 soluble receptor, and interleukin 11 were from R&D Systems (Minneapolis, MN, USA). [1-3H] androstenedione was from Amersham Biosciences. Plasmids

A luciferase reporter construct containing 1018 nucleotides of CYP19 promoter I.4 was prepared by PCR amplification of the appropriate region of genomic DNA and subcloning into the vector pGL3 (Promega). A series of 5 -deleted constructs were then prepared by PCR. Mutations of the GRE, GAS and Sp1 binding sites within the 1018 nt promoter were introduced using QuikChange (Stratagene) according to the manufacturer’s instructions. The specific mutations introduced were (mutated nucleotides in lower case) as follows: GRE, AGAAGATTCTGTTCT < ctgcagTTC ctgcag; GAS, TTCCTGTGAAA < TTCCTcTG cAg; Sp1, GGGCGGGGC < ctGCaGGGC. www.endocrinology.org

Aromatase in osteoblasts ·

Expression vectors encoding human STAT1, STAT3 and GR were prepared by PCR amplification of the appropriate cDNA sequences from human adipose stromal cell cDNA and subcloning into the vector pCDNA3·1+ (Invitrogen). An expression plasmid encoding Sp1 (pPacSp1) was kindly provided by Dr R. Tjian. Cell culture, transient transfection and reporter gene assay

The SV40-immortalized human fetal osteoblast cell line (SV-HFO) was a generous gift from Dr Hideki Chiba (Sapporo Medical University, Sapporo, Japan). Cells were cultured in DMEM supplemented with 10% fetal bovine serum (FBS), penicillin G sodium (100 units/ml) and streptomycin sulfate (100 µg/ml) (Invitrogen). Cells were maintained in 5% CO2 at 37 C. Cells of passages 10–18 were used in the experiments described. SV-HFO cells were transfected with various promoter I.4 luciferase reporter constructs or the promoter-less luciferase vector (pGL3-Basic, Promega), along with a -galactosidase or seapansy luciferase (phRL-TK) internal control vector (Promega), using Fugene6 transfection reagent (Roche) according to the manufacturer’s instructions. After overnight transfection, cells were treated with experimental agents for 8 h at the concentrations indicated. Firefly luciferase activity and -galactosidase or seapansy luciferase activity were measured using the Luciferase Assay System or the Dual-Luciferase Reporter Assay System (Promega).

M WATANABE

transcriptase (Roche) and random primers (Invitrogen). PCR primer sequences for the human CYP19 and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) genes are described by Agarwal et al. (1995). PCR conditions were as follows: 94 C for 5 min; 1 cycle at 94 C for 30 s, T1 C for 30 s, 72 C for 30 s; 10 cycles at 94 C for 30 s, T2 C for 30 s, 72 C for 30 s; 10 cycles at 94 C for 30 s, T3 C for 30 s, 72 C for 30 s; 10 cycles at 72 C for 15 min; 1 cycle. Primers and temperatures were as follows: for coding region of aromatase gene transcript, RT7 and RT8 primers, T1 was 59, T2 was 57 and T3 was 55. For exon I.4-containing gene transcript, RT1 and RT8 primers, T1 was 65, T2 was 60 and T3 was 55. For exon II-containing gene transcript, RT3 and RT8, T1 was 60, T2 was 55 and T3 was 50. For exon I.3-containing gene transcript, RT2 and RT8, T1 was 65, T2 was 60 and T3 was 55. For human GAPDH, GAPDH1 and GAPDH2, primers were used with the following cycle conditions: 94 C for 5 min; 1 cycle at 94 C for 30 s, 63 C for 30 s, 72 C for 30 s; 25 cycles at 72 C for 15 min; 1 cycle. Amplification of expected gene transcripts was confirmed by direct sequencing of the PCR products. RT-PCR for human GR, STAT1 and STAT3 was performed according to published data (Sun et al. 1998, Fisher et al. 1999, Huang et al. 2001). RT-PCR for human Sp1 was performed using hSp1-S (5 -CCCTCC AGGCCCTCCAAGCAGC-3 and hSp1-AS (5 GCGCAAGTGTGCCCGCAGGTGAGAGG-3 ) primers.

Aromatase assay

Results

SV-HFO cells were grown to 80% confluence and then serum starved overnight. After serum starvation, cells were treated with experimental agents at the concentrations indicated. Endogenous aromatase activity in SV-HFO cells was measured by the detection of 3HOH released during aromatization of [1-3H] androstenedione, as previously described (Ackerman et al. 1981).

Aromatase activity in SV-HFO cells

Reverse transcriptase polymerase chain reaction (RT-PCR)

Total RNA was prepared from SV-HFO cells using the QIAamp system (Qiagen), according to the manufacturer’s instructions. First-strand cDNA was prepared from total RNA using AMV-reverse www.endocrinology.org

and others 535

Figure 2 shows the effect of various growth factors and hormones on aromatase activity in SV-HFO cells. Basal aromatase activity was low (0·3 pmol/mg protein per 6 h). Treatment with Dex for 24 h increased aromatase activity approximately seven-fold. Neither -GP nor FBS had any effect on aromatase activity on its own. However, in the presence of Dex, FBS caused a further two-fold increase in aromatase activity. This suggests that a factor or factors present in serum synergized with Dex in stimulating aromatase activity. We examined a variety of hormones and cytokines for their ability to induce aromatase activity in SV-HFO (Fig. 2). Of these cytokines, only OSM was Journal of Molecular Endocrinology (2004) 32, 533–545

536

M WATANABE

and others


Figure 2 Aromatase activity is stimulated in a manner dependent on dexamethasone in SV-HFO cells. Cells were treated with dexamethasone (Dex, 100 nM), -glycerophosphate (-GP, 10 mM), fetal bovine serum (FBS, 10%), oncostatin M (OSM, 5 ng/ml), interleukin-1 (IL-1, 5 ng/ml), interleukin-6 (IL-6, 5 ng/ml), interleukin-6 soluble receptor (IL-6sR, 20 ng/ml), interleukin-11 (IL-11, 5 ng/ml), tumor necrosis factor (TNF, 5 ng/ml), PTHRP, bone morphogenetic protein 2 (BMP2, 200 ng/ml), forskolin (FSK, 25 µM), phorbol 12-myristate 13-acetate (PMA, 4 nM) or cholecalciferol (VD3, 100 nM). After treatment for 24 h, aromatase activities were measured. Experiments were performed in triplicate, and similar results were obtained in at least two further experiments. Error bars represent standard error.

effective. OSM had a slight stimulatory effect on aromatase activity on its own, but markedly potentiated the effect of Dex. The combination of Dex and OSM induced aromatase activity approximately 50-fold. Since aromatase activity in human adipose stromal cells can be stimulated by activators of protein kinase A (PKA) and protein kinase C (PKC) (Zhao et al. 1996), we activated these pathways by treatment with forskolin (FSK) and phorbol ester (PMA) respectively. However, these agents did not alter the aromatase activity of SV-HFO cells either alone or in combination. Aromatase gene expression in SV-HFO cells

We next examined expression of the CYP19 gene in response to various hormones by RT-PCR, using primers designed to amplify the CYP19 coding region (Fig. 3). As a positive control, RNA from human adipose stromal cells treated with Dex and OSM, or FSK and PMA, was used. These agents induce CYP19 gene expression in adipose stromal cells from promoter I.4 (Dex + OSM) and promoter II (FSK + PMA) respectively. RT-PCR from these samples produced a band of the Journal of Molecular Endocrinology (2004) 32, 533–545

expected size (coding region, lanes 1 and 2). CYP19 was undetectable in SV-HFO under basal conditions (lane 3). In the presence of Dex, the CYP19 coding region was readily detected, and the level was increased upon addition of OSM, consistent with the effect of these agents on aromatase activity. This suggests that Dex and OSM act to induce aromatase activity by increasing CYP19 gene expression in SV-HFO. The human CYP19 gene has multiple first exons derived from alternative splicing through the use of tissue-specific promoters (Fig. 1). To examine the nature of the 5 ends of the CYP19 transcripts in SV-HFO, we performed RT-PCR using exon I-specific primers. As shown in Fig. 3, CYP19 transcripts containing untranslated exon I.4 were induced in the same profile as the coding region in response to experimental treatments. Neither exon II nor the two alternative forms of exon I.3 were detected in SV-HFO under any conditions, although these were readily detectable in adipose stromal cells. Using 5 -RACE for CYP19 transcripts in SV-HFO, no sequences other than exon I.4 were detected (data not shown). Therefore, exon I.4 is predominantly selected in SV-HFO, and the www.endocrinology.org


M WATANABE

and others 537

Figure 3 CYP19 mRNA expression in SV-HFO cells. RT-PCR of total RNA from SV-HFO or human adipose stromal cells was performed using the primers derived from the sequence of coding region, exon I.4, exon II and exon I.3 of human CYP19. RT-PCR using human GAPDH primers was also performed. RT (+) indicates first-strand cDNA synthesis with AMV reverse transcriptase. RT (–) indicates incubation without reverse transcriptase. Total RNAs from human adipose stromal cells were used as positive control reactions.

5 -promoter region of exon I.4 (promoter I.4) appears to be the only promoter used to drive CYP19 expression in these cells. Activation of promoter I.4 in SV-HFO

To confirm direct activation of promoter I.4, a luciferase reporter construct containing 1018 nucleotides of this genomic sequence was transfected into SV-HFO (Fig. 4). Activity of this reporter gene was stimulated five- to six-fold in the presence of Dex. OSM further increased luciferase activity induced by Dex. However, no significant upregulation by -GP, FBS or IL-1 was observed. Treatment with FSK and PMA did not affect luciferase activity, consistent with their lack of effect on aromatase activity (Fig. 2) or expression (Fig. 3). Promoter I.4 contains a number of putative transcription factor binding sites, including sites for www.endocrinology.org

AP1, STAT3 (interferon activating sequence (GAS)), GR and Sp1. To identify the sequences that mediate induction of promoter I.4 by Dex and OSM, a series of 5 -deleted promoter I.4 constructs were transfected into SV-HFO (Fig. 5). Reporter activity in the presence of Dex and OSM decreased slightly as sequences between –1004 and –518 were deleted. Deletion of sequences between –518 and –408, which harbor the GAS element, resulted in a significant loss of reporter activity. A further significant reduction in reporter activity was observed on deletion of a 30-nucleotide sequence containing the putative GRE. Activity of the shortest construct (–244 /+14), containing only the Sp1-like element, was still significantly increased in the presence of Dex and OSM compared to the empty luciferase vector (pGL3 basic), which showed no hormonal induction. These data suggest important roles for sequences surrounding the Journal of Molecular Endocrinology (2004) 32, 533–545

538

M WATANABE

and others


Figure 4 Promoter I.4 is activated by oncostatin M and dexamethasone in SV-HFO cells. SV-HFO cells were transfected with –1004/+14 Luc vector (containing nucleotide –1004 to +14 of promoter I.4) and -galactosidase internal control vector, and treated with Dex (100 nM), -GP (10 mM), FBS (10%), OSM or IL-1 (both 5 ng/ml) for 8 h. Luciferase and -galactosidase activities were measured. Experiments were performed in triplicate, and similar results were obtained in at least two further experiments. Error bars indicate standard error.

GAS, GRE and Sp1 sites in regulating promoter I.4 in SV-HFO. We therefore mutated these putative response elements individually within the context of the full-length promoter to assess the contribution of each motif to the regulation of promoter I.4 activity. Role of GAS, GRE and Sp1 sites

To assess directly the role of each element in the regulation of promoter I.4, we mutated each site individually in the context of the –1004/+14 luciferase construct (Fig. 6). As before, activity of the wild-type construct (–1004/+14) was increased three-fold and 0·5-fold in the presence of Dex and OSM respectively, and approximately five-fold by the combination of Dex and OSM. Mutation of the GAS element (GAS mut) reduced both basal and hormone-induced reporter activity by approximately 60%, although the fold activation in the presence of stimulation was similar to that of the Journal of Molecular Endocrinology (2004) 32, 533–545

wild-type promoter. In contrast, mutation of the GRE reduced basal promoter activity by 70%, and abolished the response to Dex or OSM. In combination, however, Dex and OSM increased activity of the GRE mut construct four-fold. Similarly, mutation of the putative Sp1 site decreased basal promoter activity by approximately 50%, although this construct was still activated by Dex (two-fold), but not by OSM. These data confirm the importance of the GAS, GRE and Sp1 sites in regulation of CYP19 by promoter I.4 in SV-HFO, but further suggest that no one site by itself is obligatory for transcription, since each of the mutated constructs retained some residual inducibility in response to Dex and OSM. We next asked whether SV-HFO cells express some of the transcription factors that would be predicted to bind to these sites. Semiquantitative RT-PCR was performed using primers specific for human GR, STAT1, STAT3 and Sp1 (Fig. 7). www.endocrinology.org


M WATANABE

and others 539

Figure 5 Effect of dexamethasone and oncostatin M on 5′-deleted CYP19 promoter I.4 reporter vectors. SV-HFO cells were transfected with the 5’-deleted CYP19 promoter I.4 reporter constructs indicated along with a -galactosidase internal control vector. Cells were treated for 8 h with Dex (100 nM) and OSM (5 ng/ml). Experiments were performed in triplicate, and similar results were obtained in at least two further experiments. Error bars indicate standard error.

Transcripts specific for each factor were detected in unstimulated cells, and there was no significant change in transcript levels of GR, STAT1 or STAT3 on stimulation of SV-HFO cells with Dex, -GP, OSM, FBS or FSK plus PMA. Sp1-specific transcripts, however, were slightly increased in the presence of FBS+Dex, or FSK+PMA. Role of GR, STAT and Sp1 proteins

To determine whether exogenous GR, STAT or Sp1 proteins are capable of modulating transcription from promoter I.4 through these consensus sequences, SV-HFO cells were transiently transfected with the –1004/+14 construct and expression vectors encoding these transcription factors www.endocrinology.org

(Fig. 8). In the presence of Dex, transfection of human GR increased promoter I.4 activity by five-fold (Fig. 8A). This effect of GR was completely abolished when the putative GRE was mutated. Co-transfection of STAT1 had no effect on promoter I.4 activity (Fig. 8B); however, STAT3 co-transfection increased promoter activity by two-fold (Fig. 8C). STAT3 had no effect on activity of a promoter I.4 construct containing a mutation in the GAS element. Finally, transfection of Sp1 increased wild-type promoter activity four-fold, but had no effect on activity of a promoter construct containing a mutated Sp1 site (Fig. 8D). Thus, exogenous GR, STAT3 and Sp1 can stimulate promoter I.4 activity in SV-HFO cells, in a manner dependent upon their respective cis-elements. Journal of Molecular Endocrinology (2004) 32, 533–545

540

M WATANABE

and others


Figure 6 Effect of mutation of putative CYP19 promoter I.4 cis-elements. –1004/+14 Luc and the indicated mutants were transfected into SV-HFO and stimulated with Dex (100 nM) and OSM (5 ng/ml), either alone or in combination. Gray symbols represent the mutated elements. Experiments were performed in triplicate, and similar results were obtained in at least two further experiments. Error bars indicate standard error.

Discussion Aromatase is expressed in a number of different tissues through the use of multiple alternative promoters. Several lines of evidence indicate that bone tissue, as well as osteoblast-like cells in primary culture, expresses aromatase and that this expression is modified in pathologic situations such as bone fracture or osteoporosis (Nawata et al. 1995, Lea et al. 1997). However, the factors that regulate aromatase expression in bone and the promoter(s) through which they act are poorly understood. Several potential regulators have been proposed, including TGF-, l,25-dihydroxycholecalciferol, IL-11, TNF, vitamin D and 17-estradiol (Tanaka et al. 1996, Shozu & Simpson 1998, Shozu et al. 2000, Enjuanes et al. 2003). Although several different promoters have been implicated in the action of these hormones, the primary promoter Journal of Molecular Endocrinology (2004) 32, 533–545

utilized appears to be promoter I.4, with relatively minor contributions from promoters II, I.3, I.2 and I.6 (Shozu & Simpson 1998, Enjuanes et al. 2003). The recent development of the human osteoblastderived SV-HFO cell line (Chiba et al. 1993) provides a valuable model in which to dissect the mechanisms regulating aromatase expression in osteoblasts. SV-HFO cells exhibit osteoblastic markers such as alkaline phosphatase and calcitonin, respond to the osteotropic factors vitamin D3 and retinoic acid, and undergo mineralization in response to glucocorticoids. In the present study we found that aromatase activity and gene transcript level in SV-HFO cells were stimulated by Dex in SV-HFO cells. This is consistent with previous studies in primary human osteoblast-like cells and THP-1 monocytic cells (Tanaka et al. 1996, Shozu et al. 2000, Takayanagi et al. 2002). Although glucocorticoid excess is well www.endocrinology.org


M WATANABE

and others 541

Figure 7 mRNAs encoding GR, STAT1, STAT3 and Sp1 are expressed in SV-HFO cells. Total RNA was extracted from SV-HFO treated with Dex (100 nM), -GP (10 mM), FBS (10%), OSM (5 ng/ml) and FSK (25 µM) plus PMA (4 nM) or vehicles. RT-PCR for the transcription factors indicated was performed as described in Materials and methods.

known to induce osteoporosis – mainly through inhibition of intestinal calcium absorption – physiological concentrations of glucocorticoids can produce osteoplastic effects (Canalis 1983). Thus, the induction of aromatase expression in osteoblasts by low concentrations of glucocorticoids, as shown here, is not inconsistent with their osteoporotic effects at high concentrations. Of the other agents tested in this study, OSM was the only factor found to induce aromatase activity on its own, and this effect was markedly enhanced by Dex. In this regard, mice with targeted gene disruption and transgenic mice have revealed an important relationship between gp130 cytokines and bone formation (Heymann & Rousselle 2000). Metallothionein promoter-driven bovine oncostatin M gene expression caused excessive growth of bone in the femur and phalanges (Malik et al. 1995). This suggests a potent positive effect of OSM on bone formation and probably on maturation of osteoblasts. From the results obtained here, these actions of OSM could be related, in part, to a stimulation of aromatase expression. Further analysis of the effects of these www.endocrinology.org

agents on alkaline phosphatase activity, osteocalcin expression or mineral deposition should provide insight into this relationship. The dominant promoter used to express aromatase in SV-HFO was found to be promoter I.4 and, consistent with the lack of effect of the PKA pathway or PKC pathway upon aromatase activity, promoter II was not utilized in SV-HFO (Fig. 3). These data suggest that the profile of aromatase regulation in SV-HFO is similar to that of primary human osteoblasts (Shozu & Simpson 1998). SV-HFO is therefore a relevant and useful model with which to investigate aromatase regulation in osteoblasts. In human adipose stromal cells, CYP19 gene transcripts are increased in response to either Dex + OSM or FSK + PMA; DEX + OSM induces CYP19 expression via promoter I.4, whereas FSK + PMA induces expression via promoter II (Zhao et al. 1995a,b). Promoter II was not activated in SV-HFO cells under any of the experimental conditions, and this is one important difference between aromatase expression in adipose stromal cells and osteoblasts. Journal of Molecular Endocrinology (2004) 32, 533–545

542

M WATANABE

and others


Figure 8 GR, STAT3 and Sp1 induce promoter I.4 reporter activity in SV-HFO. SV-HFO cells were co-transfected with –1004/+14 Luc, or derivatives thereof containing mutations in the indicated cis-elements (GRE, GAS or Sp1), and expression vectors encoding (panel A) GR; (panel B) STAT1; (panel C) STAT3 or (panel D) Sp1. The total amount of DNA was normalized to 1·5 µg with the parent pCNDNA3·1+ vector. Cells were then stimulated with Dex (100 nM) and OSM (5 ng/ml) for 8 h before being lysed and assayed for luciferase activity. Data represent the mean± S.E.M. of three independent experiments, and are expressed relative to control activity (100%).

Journal of Molecular Endocrinology (2004) 32, 533–545

www.endocrinology.org


The results of promoter deletion/mutation experiments suggest important roles for the GAS, GRE and Sp1 sites in promoter I.4 regulation in SV-HFO. In adipose stromal cells, activators of gp130-coupled receptors, such as OSM, and glucocorticoids induce promoter I.4 through these same elements. The action of OSM is associated with phosphorylation and activation of homo- or heterodimers of STAT1 and STAT3, which bind to the GAS element to induce specific genes (Gomez-Lechon 1999). Induction of promoter I.4 activity via binding of STAT3 to the GAS element in response to IL-11 in adipose stromal cells has been reported (Zhao et al. 1995b). Glucocorticoids, on the other hand, stimulate CYP19 transcription through a classical GR-GRE-mediated pathway in adipose stromal cells (Zhao et al. 1995a). Although we have not yet identified the transcription factors that bind the GAS element and GRE in SV-HFO, transfection of GR, STAT3 or Sp1 increased promoter I.4 activity, and RT-PCR analysis indicated that these transcription factors are expressed in SV-HFO cells. While it is therefore likely that the transcriptional mechanisms used to drive promoter I.4 in SV-HFO and adipose stromal cells are similar, the results of the present study have revealed subtle differences between the two cell types. In the first place, as mentioned above, promoter II is not activated in SV-HFO cells, in contrast to adipose stromal cells. Secondly, in adipose stromal cells, promoter I.4 has an obligate requirement for glucocorticoids, although glucocorticoids by themselves have no effect on aromatase activity. In contrast, glucocorticoids alone increase aromatase activity and gene expression in SV-HFO cells. Lastly, a number of class I cytokines activate promoter I.4 in adipose stromal cells, whereas in SV-HFO cells, only OSM is effective. These differences presumably reflect divergent mechanisms that dictate cell fate from mesenchymal stem cells. From a clinical standpoint, recognition that aromatase activity within bone cells is the major source of estrogen influencing this tissue leads to several considerations. First, it suggests that circulating androgens rather than estrogen are important precursors of estrogen within bone, since circulating testosterone levels in blood of postmenopausal women are an order of magnitude greater than circulating estrogen levels. Secondly, circulating levels of testosterone in men are one order of www.endocrinology.org

M WATANABE

and others 543

magnitude greater than these in postmenopausal women, and this high level of estrogen precursor could be one reason why men are less at risk of bone loss than are women. Finally, aromatase inhibitors are widely used for the treatment of breast cancer patients. However, these inhibitors inhibit aromatase in a global fashion, including bone, and this could lead to loss of bone mineralization. Although this might be of less consequence for elderly women with advanced postmenopausal breast cancer, the effects of estrogen deprivation in bone of young postmenopausal women could be significant. This is particularly relevant given the current interest in the use of aromatase inhibitors as preventative therapies. To avoid this kind of risk, the development of selective aromatase modulators (SAMs) that inhibit aromatase activity only in breast tissue, but not in bone tissue, would be useful (Simpson et al. 2000, Simpson & Davis 2001). The concept of SAMs is based on the fact that (a) local aromatase activity is the main source of estrogen in breast and bone, and (b) the mechanism of aromatase expression is different between breast adipose stromal cells and osteoblasts (for example, promoter II is activated in adipose stromal tissues of breast cancer patients, but not in osteoblasts). Thus, inhibition of promoter II-mediated transcription as breast cancer therapy (Clyne et al. 2002) would have little effect on local estrogen production in bone.

Acknowledgements We thank Dr Hideki Chiba, Sapporo Medical University (Sapporo, Japan), for his kind gift of the SV-HFO cell line. We thank Mr Fumihiko Yukinawa for his technical assistance. This work was supported by a grant from the Victorian Breast Cancer Research Consortium and by USPHS Grant no. R-37 AG-08174.

References Ackerman GE, Smith ME, Mendelson CR, MacDonald PC & Simpson ER 1981 Aromatization of androstenedione by human adipose tissue stromal cells in monolayer culture. Journal of Clinical Endocrinology and Metabolism 53 412–417. Agarwal VR, Bulun SE & Simpson ER 1995 Quantitative detection of alternatively spliced transcripts of the aromatase cytochrome P450 (CYP19) gene in aromatase-expressing human cells by competitive RT-PCR. Molecular and Cellular Probes 9 453–464. Journal of Molecular Endocrinology (2004) 32, 533–545

544

M WATANABE

and others


Arts J, Kuiper GG, Janssen JM, Gustafsson JA, Lowik CW, Pols HA & van Leeuwen JP 1997 Differential expression of estrogen receptors alpha and beta mRNA during differentiation of human osteoblast SV-HFO cells. Endocrinology 138 5067–5070. Bilezikian JP, Morishima A, Bell J & Grumbach MM 1998 Increased bone mass as a result of estrogen therapy in a man with aromatase deficiency. New England Journal of Medicine 339 599–603. Blanchard O, Tsagris L, Rappaport R, Duval-Beaupere G & Corvol M 1991 Age-dependent responsiveness of rabbit and human cartilage cells to sex steroids in vitro. Journal of Steroid Biochemistry and Molecular Biology 40 711–716. Canalis E 1983 Effect of glucocorticoids on type I collagen synthesis, alkaline phosphatase activity, and deoxyribonucleic acid content in cultured rat calvariae. Endocrinology 112 931–939. Carani C, Qin K, Simoni M, Faustini-Fustini M, Serpente S, Boyd J, Korach KS & Simpson ER 1997 Effect of testosterone and estradiol in a man with aromatase deficiency. New England Journal of Medicine 337 91–95. Chiba H, Sawada N, Ono T, Ishii S & Mori M 1993 Establishment and characterization of a simian virus 40-immortalized osteoblastic cell line from normal human bone. Japanese Journal of Cancer Research 84 290–297. Clyne CD, Speed CJ, Zhou J & Simpson ER 2002 Liver receptor homologue-1 (LRH-1) regulates expression of aromatase in preadipocytes. Journal of Biological Chemistry 277 20591–20597. Enjuanes A, Garcia-Giralt N, Supervia A, Nogues X, Mellibovsky L, Carbonell J, Grinberg D, Balcells S & Diez-Perez A 2003 Regulation of CYP19 gene expression in primary human osteoblasts: effects of vitamin D and other treatments. European Journal of Endocrinology 148 519–526. Fisher GM, Iqball S & Knight SC 1999 Gene expression during differentiation of human dendritic cells from cord blood cd34 stem cells. Cytokine 11 111–117. Gomez-Lechon MJ 1999 Oncostatin M: signal transduction and biological activity. Life Sciences 65 2019–2030. Grumbach MM 2000 Estrogen, bone, growth and sex: a sea change in conventional wisdom. Journal of Pediatric Endocrinology and Metabolism 13 Suppl 6 1439–1455. Heymann D & Rousselle AV 2000 gp130 Cytokine family and bone cells. Cytokine 12 1455–1468. Huang L, Xu J, Kumta SM & Zheng MH 2001 Gene expression of glucocorticoid receptor alpha and beta in giant cell tumour of bone: evidence of glucocorticoid-stimulated osteoclastogenesis by stromal-like tumour cells. Molecular and Cellular Endocrinology 181 199–206. Iba K, Chiba H, Sawada N, Hirota S, Ishii S & Mori M 1995 Glucocorticoids induce mineralization coupled with bone protein expression without influence on growth of a human osteoblastic cell line. Cell Structure and Function 20 319–330. Janssen JM, Bland R, Hewison M, Coughtrie MW, Sharp S, Arts J, Pols HA & van Leeuwen JP 1999 Estradiol formation by human osteoblasts via multiple pathways: relation with osteoblast function. Journal of Cellular Biochemistry 75 528–537. Lea CK, Ebrahim H, Tennant S & Flanagan AM 1997 Aromatase cytochrome P450 transcripts are detected in fractured human bone but not in normal skeletal tissue. Bone 21 433–440. Malik N, Haugen HS, Modrell B, Shoyab M & Clegg CH 1995 Developmental abnormalities in mice transgenic for bovine oncostatin M. Molecular and Cellular Biology 15 2349–2358. Masi L, Becherini L, Gennari L, Amedei A, Colli E, Falchetti A, Farci M, Silvestri S, Gonnelli S & Brandi ML 2001 Polymorphism of the aromatase gene in postmenopausal Italian women: distribution and correlation with bone mass and fracture risk. Journal of Clinical Endocrinology and Metabolism 86 2263–2269. Journal of Molecular Endocrinology (2004) 32, 533–545

Miyaura C, Toda K, Inada M, Ohshiba T, Matsumoto C, Okada T, Ito M, Shizuta Y & Ito A 2001 Sex- and age-related response to aromatase deficiency in bone. Biochemical and Biophysical Research Communications 280 1062–1068. Morishima A, Grumbach MM, Simpson ER, Fisher C & Qin K 1995 Aromatase deficiency in male and female siblings caused by a novel mutation and the physiological role of estrogens. Journal of Clinical Endocrinology and Metabolism 80 3689–3698. Mullis PE, Yoshimura N, Kuhlmann B, Lippuner K, Jaeger P & Harada H 1997 Aromatase deficiency in a female who is compound heterozygote for two new point mutations in the P450 arom gene: impact of estrogens on hypergonadotropic hypogonadism, multicystic ovaries, and bone densitometry in childhood. Journal of Clinical Endocrinology and Metabolism 82 1739–1745. Nawata H, Tanaka S, 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. Oz OK, Zerwekh JE, Fisher C, Graves K, Nanu L, Millsaps R & Simpson ER 2000 Bone has a sexually dimorphic response to aromatase deficiency. Journal of Bone and Mineral Research 15 507–514. Oz OK, Millsaps R, Welch R, Birch J & Zerwekh JE 2001 Expression of aromatase in the human growth plate. Journal of Molecular Endocrinology 27 249–253. Purohit A, Flanagan AM & Reed MJ 1992 Estrogen synthesis by osteoblast cell lines. Endocrinology 131 2027–2029. Sasano H & Harada N 1998 Intratumoral aromatase in human breast, endometrial, and ovarian malignancies. Endocrine Reviews 19 593–607. 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. Schweikert HU, Wolf L & Romalo G 1995 Oestrogen formation from androstenedione in human bone. Clinical Endocrinology 43 37–42. Sebastian S & Bulun SE 2001 A highly complex organization of the regulatory region of the human CYP19 (aromatase) gene revealed by the Human Genome Project. Journal of Clinical Endocrinology and Metabolism 86 4600–4602. Shimodaira K, Fujikawa H, Okura F, Shimizu Y, Saito H & Yanaihara T 1996 Osteoblast cells (MG-63 and HOS) have aromatase and 5 alpha-reductase activities. Biochemistry and Molecular Biology International 39 109–116. Shozu M & Simpson ER 1998 Aromatase expression of human osteoblast-like cells. Molecular and Cellular Endocrinology 139 117–129. Shozu M, Zhao Y & Simpson ER 2000 TGF-beta1 stimulates expression of the aromatase (CYP19) gene in human osteoblast-like cells and THP-1 cells. Molecular and Cellular Endocrinology 160 123–133. Simpson E, Rubin G, Clyne C, Robertson K, O’Donnell L, Jones M & Davis S 2000 The role of local estrogen biosynthesis in males and females. Trends in Endocrinology and Metabolism 11 184–188. Simpson ER & Davis SR 2001 Minireview: aromatase and the regulation of estrogen biosynthesis – some new perspectives. Endocrinology 142 4589–4594. Simpson ER, Zhao Y, Agarwal VR, Michael MD, Bulun SE, Hinshelwood MM, Graham-Lorence S, Sun T, Fisher CR, Qin K & Mendelson CR 1997 Aromatase expression in health and disease. Recent Progress in Hormone Research 52 185–213. Simpson ER, Clyne C, Rubin G, Boon WC, Robertson K, Britt K, Speed C & Jones M 2002 Aromatase – a brief overview. Annual Review of Physiology 64 93–127. www.endocrinology.org

Aromatase in osteoblasts · Smith EP, Boyd J, Frank GR, Takahashi H, Cohen RM, Specker B, Williams TC, Lubahn DB & Korach KS 1994 Estrogen resistance caused by a mutation in the estrogen-receptor gene in a man. New England Journal of Medicine 331 1056–1061. Sun WH, Pabon C, Alsayed Y, Huang PP, Jandeska S, Uddin S, Platanias LC & Rosen ST 1998 Interferon-alpha resistance in a cutaneous T-cell lymphoma cell line is associated with lack of STAT1 expression. Blood 91 570–576. Takayanagi R, Goto K, Suzuki S, Tanaka S, Shimoda S & Nawata H 2002 Dehydroepiandrosterone (DHEA) as a possible source for estrogen formation in bone cells: correlation between bone mineral density and serum DHEA-sulfate concentration in postmenopausal women, and the presence of aromatase to be enhanced by 1,25-dihydroxyvitamin D3 in human osteoblasts. Mechanisms of Ageing and Development 123 1107–1114. Tanaka S, Haji M, Takayanagi R, Tanaka S, Sugioka Y & Nawata H 1996 1,25-Dihydroxyvitamin D3 enhances the enzymatic activity and expression of the messenger ribonucleic acid for aromatase cytochrome P450 synergistically with dexamethasone depending on the vitamin D receptor level in cultured human osteoblasts. Endocrinology 137 1860–1869.

www.endocrinology.org

M WATANABE

and others 545

Vidal O, Lindberg MK, Hollberg K, Baylink DJ, Andersson G, Lubahn DB, Mohan S, Gustafsson JA & Ohlsson C 2000 Estrogen receptor specificity in the regulation of skeletal growth and maturation in male mice. PNAS 97 5474–5479. Zhao Y, Mendelson CR & Simpson ER 1995a Characterization of the sequences of the human CYP19 (aromatase) gene that mediate regulation by glucocorticoids in adipose stromal cells and fetal hepatocytes. Molecular Endocrinology 9 340–349. Zhao Y, Nichols JE, Bulun SE, Mendelson CR & Simpson ER 1995b Aromatase P450 gene expression in human adipose tissue. Role of a Jak/STAT pathway in regulation of the adipose-specific promoter. Journal of Biological Chemistry 270 16449–16457. Zhao Y, Agarwal VR, Mendelson CR & Simpson ER 1996 Estrogen biosynthesis proximal to a breast tumor is stimulated by PGE2 via cyclic AMP, leading to activation of promoter II of the CYP19 (aromatase) gene. Endocrinology 137 5739–5742.

Received in final form 27 October 2003 Accepted 3 December 2003

Journal of Molecular Endocrinology (2004) 32, 533–545